catalysis - DigitalRefining

Transcription

catalysis
2012
ptq
cover and spine copy 6.indd 1
23/2/12 20:44:33
Stepping up performance
– next generation BRIM™ technology
WWW.TOPSOE.CO M
Are you looking to step up plant performance?
Topsøe’s next generation BRIM™ catalysts offer refiners the opportunity to increase
performance through an increase in catalyst activity.
Using the original BRIM™ technology Topsøe has developed several new catalysts, resulting
in higher activity at lower filling densities.
The next generation BRIM™ catalysts display
-
high dispersion
high porosity
high activity
We look forward to stepping up your performance!
haldor.indd 1
23/2/12 11:45:41
Security of
feedstock supply
catalysis
ptq
2012 Vol 13 No 2
www.eptq.com
2008
Editor
René G Gonzalez
editor@petroleumtechnology.com
Production Editor
Rachel Zamorski
production@petroleumtechnology.com
Graphics Editor
Mohammed Samiuddin
graphics@petroleumtechnology.com
Editorial
PO Box 11283
Spring TX 77391, USA
tel +1 281 374 8240
fax +1 281 257 0582
Advertising Sales Manager
Paul Mason
sales@petroleumtechnology.com
Advertising Sales
Bob Aldridge
Advertising Sales Office
tel +44 870 90 303 90
fax +44 870 90 246 90
Publisher
Nic Allen
publisher@petroleumtechnology.com
Circulation
Jacki Watts
circulation@petroleumtechnology.com
Crambeth Allen Publishing Ltd
Hopesay, Craven Arms SY7 8HD, UK
tel +44 870 90 600 20
fax +44 870 90 600 40
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
Box 437, Emigsville, PA 17318-0437
Back numbers available from the Publisher
at $30 per copy inc postage.
contents/ed com copy 8.indt 1
3 Onwards and upwards
Chris Cunningham
D
5 ptq&a
espite signs in 2007 of a slowdown in various sectors of the economy,
refiners remain a big play for prospective investors. It used to be
17 conventional
Evaluation of
a low that
rare higher
earth resid
FCC catalyst
wisdom
fuel prices
and a slowing economy
would
curb
demand
and
increase
supply,
but
for Catalysts
the past seven
years
Sabeeth Srikantharajah and Colin Baillie Grace
Technologies
that has not proved to be the case. While the rate of increase in world oil demand
Bernhard
Wieland
Wache
Bayernoilappears that
has declined
since theZahnbrecher
surprising 4%and
surge
in 2004,
it nevertheless
demand beyond 2008 will grow, along with prices. It is a safe bet that rapidly
increasing oil consumption by China, India and even the Middle East producers
23 Refinery fuel gas in steam reforming hydrogen plants
themselves will continue. It is also safe to assume that refinery and petrochemical
Peter
Broadhurst
and Graham
Hinton Johnson Matthey Catalysts
conversion
unit capacity
will need
to expand.
No massive new sources of energy are expected to come on stream for the
foreseeable future. The world will remain dependent on oil and gas for decades to
accumulation
in coker
naphtha
hydrotreaters
come 31
evenEstimating
though thesilicon
upstream
industry faces
increasing
challenges
in the
discovery
and
production
of new
sources.
fact,Kraus
some well-placed industry
Thienan
Tran, Patrick
Gripka
andInLarry
analysts think 2008 may be the year where there is no increase in crude supply at
Criterion Catalysts & Technologies
all from regions outside of OPEC. For this reason, we will continue to see significant
investment in refinery upgrades despite surging costs — security of feedstock
supply,
albeit
low-quality
feedstock,
takes precedence
over the
35
FCCunconventional
catalyst coolers
in maximum
propylene
mode
quality of feedstock supply.
Rahul Pillai and Phillip Niccum KBR
Feedstock options such as biomass (for biofuels production), Canadian tar sands
(for distillate production) and other types of unconventional crude sources require
reactor45
technology
thatcatalyst
allows for
the integration
of these analysis
operations
intosorting
existing
Decrease
costs
by regeneration,
and
process configurations. The quality of these types of feedstock are one important
whyPierre
Dufresne
reason
a wider
array of Eurecat
catalysts has been introduced into the market. For
example,
asFrancois
refiners Locatelli
cut deeper
into the
vacuum tower, the concentration of
Eurecat
France
metals in the VGO requires a properly designed guard bed system to protect active
catalysts in the hydrocracker. The characteristics of feedstock with low API gravity
53 high
Optimisation
of integrated
complexes
(eg, <10),
metals, nitrogen
and otheraromatic
undesirable
components is one of the
main reasons
hydrotreaters
andAG
hydrocrackers are becoming larger — to
Axelwhy
Düker
Süd-Chemie
accommodate not only higher volumes of catalyst, but also a wider variety of
catalyst with specific formulations.
Non-catalytic
processes are also
playing
59 Troubleshooting
a FCC
unit a significant role in the refiner’s ability
to process whatever unconventional crude sources become available. For example,
Chiranjeevi Thota, Shalini Gupta, Dattatraya Tammanna Gokak,
some refiners processing higher volumes of resid and atmospheric tower bottoms
Ravi kumar
P VofC solvent-extraction
Rao and Viswanathan
Poyyamani
have considered
addingVoolapalli,
certain types
processes
in addition
to overall
improvements
to crude unit
(eg, vacuum
tower revamps) and delayed
Swaminathan Bharat
Petroleum
Corporation
coker operations. Improvements in furnace technology, such as with olefin steam
cracker operations, have resulted in significant increases in worldwide ethylene
capacity.
However, any expansion of the value chain (eg, ethylene-to-propylene via
dehydrogenation) requires investment in catalytic-based processes, as discussed in
the following articles authored by experts in the field of downstream process
technology. PTQ wishes to extend its gratitude to the authors who provided
Marathon
Oil’sresponded
Catlettsburg refinery,
Kentucky,
USA Marathon Oil
editorial
and
to the
Q&A
published
in this issue of PTQPhoto:
Catalysis,
as
well as to those respondents who addressed the online questions (www.eptq.com)
that addressed the specifics of certain reactor and catalytic issues of importance to
the industry.
©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.
René G Gonzalez
24/2/12 09:54:43
(9(5:21'(5:+$70$.(6
285&$7$/<67662$'9$1&('"
,1'8675</($',1*0,1'62)&2856(
(YHQ ZLWK D ZLGH UDQJH RI SURYHQ FDWDO\VWV OLNH &(17(5$ LQ RXU SRUWIROLR DQG QHDUO\ F\FOHV
RI FRPPHUFLDO 8/6' RSHUDWLRQV DURXQG WKH ZRUOG DW &5,7(5,21 ZH VWLOO WKLQN WKH XOWLPDWH NH\ WR
SHUIRUPDQFHLVRXUSHRSOH2XUUHVHDUFKDQGGHYHORSPHQWWHDPUHSUHVHQWVDVHOHFWIRUFHRIREVHVVLYHO\
GHGLFDWHGWKLQNHUV¤LQGXVWU\OHDGLQJVFLHQWLVWVZLWKWKHDELOLW\WRWUDQVIRUPDQLGHDLQWRDEUHDNWKURXJK
VROXWLRQ 5HVW DVVXUHG WKH QH[W JHQHUDWLRQ RI FDWDO\VW WHFKQRORJ\ LV LQ JRRG KDQGV DQG KHDGV
CRITERION: Leading minds. Advanced technologies.
www.CRITERIONCatalysts.com
criterion.indd 1
23/2/12 11:52:11
Security
of
Onwards and
feedstock
upwards supply
catalysis
ptq
Vol 17 No
Vol213 No 2
2012
2008
Editor
Editor
René G Gonzalez
Chris Cunningham
Production Editor
Production Editor
Rachel Storry
Rachel Zamorski
Graphics Editor
Rob Fris
Graphics Editor
Mohammed Samiuddin
Editorial
tel +44 844 5888 773
fax +44 844 5888 667
Editorial
PO Box 11283
Business Development Director
Spring TX 77391, USA
Paul Mason
tel +1 281 374 8240
fax +1 281 257 0582
Advertising Sales
Bob Aldridge
Advertising Sales Manager
Paul Mason
tel +44 844 5888 771
Advertising Sales
fax +44 844 5888 662
Bob Aldridge
Publisher
Nic Allen
tel +44 870 90 303 90
fax +44 870 90 246 90
Circulation
Jacki Watts
Publisher
Nic Allen
Crambeth Allen Publishing Ltd
Circulation
tel +44 844 5888 776
Jacki Watts
fax +44 844 5888 667
ISSN 1362-363X
Crambeth
Allen Publishing Ltd
tel +44 870 90 600 20
Petroleum Technology Quarterly (USPS 0014-781)
faxquarterly
+44 870
90 600
40 edition
is published
plus annual
Catalysis
by Crambeth Allen Publishing Ltd and is distributed
in the USA by SPP, 75 Aberdeen Rd, Emigsville, PA
ISSNpostage
1362-363X
17318. Periodicals
paid at Emigsville PA.
Postmaster: send address changes to Petroleum
Technology Quarterly c/o PO
Back numbers available
from theQuarterly
Publisher
Petroleum
Technology
(USPS
at $30 per copy
postage.
0014-781)
is inc 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.
DT
signs
in 2007business
of a slowdown
in various
of the
economy,
heespite
refining
catalysts
is nothing
if not sectors
responsive.
Sometimes
refiresponse
ners remain
big
for prospective
investors.
It used
to be
that
has toa be
in play
the immediate
term, reference
some
significant
conventional
wisdom
that
higher
fuel
prices
and
a
slowing
economy
price hikes for FCC catalyst in particular during 2011, although there
would curb
and increase
supply,
for the
past seven
is a continuing
anddemand
strong technical
riposte
frombut
catalyst
developers
to years
the
thatunlegislated
has not proved
to
be
the
case.
While
the
rate
of
increase
in
world
oil
demand
rise in rare earth metals prices that chiefly caused the hikes. For the
hasmost
declined
since the
4% surge
in 2004,
it nevertheless
appears
that
part, though,
thesurprising
catalyst firms’
technical
development
and business
focus
demand
beyond
2008
will
grow,
along
with
prices.
It
is
a
safe
bet
that
rapidly
is determined in the longer term by the twin drivers of economic development
increasing
oil consumption
by China, India and even the Middle East producers
and environmental
regulation.
themselves
will
continue.
It
is
also safe
to assumestandards
that refinery
and petrochemical
In the US and Western Europe,
engineering
are delivering
increasconversion
unit
capacity
will
need
to
expand.
ingly efficient road vehicles, while the post-recession market is applying a
No
massive
sources
of energy
are expected
to come on
stream
for the
more
generalnew
brake
to growth
in demand
for transportation
fuels.
As a result,
foreseeable future. The world will remain dependent on oil and gas for decades to
the world’s developing economies are determining the future shape of growth
come even though the upstream industry faces increasing challenges in the
in demand for petroleum products.
discovery and production of new sources. In fact, some well-placed industry
To illustrate, demand for transport fuels in developing economies may rise
analysts think 2008 may be the year where there is no increase in crude supply at
by as much as 300% by 2050, according to the World Energy Council. Half a
all from regions outside of OPEC. For this reason, we will continue to see significant
decade ago,
vehicle
was at 11
cars per
1000
China
and
investment
in refi
nery ownership
upgrades despite
surging
costs
— people
securityin of
feedstock
about
20%
higher
in
India.
The
world
average
was
around
110
cars
per
1000
supply, albeit unconventional low-quality feedstock, takes precedence over the
capita.
But China’s
car ownership has been growing by 12% per annum in
quality
of feedstock
supply.
recent years,
whilesuch
the equivalent
India is
9%. More immediately,
Feedstock
options
as biomassrate
(for in
biofuels
production),
Canadian tarChina
sands
expectedproduction)
to cut its and
sulphur
fuels from
a street-chocking
(forisdistillate
otherlimit
typesfor
of vehicle
unconventional
crude
sources require
350 ppm
to 50 ppm,
and theofArabian
Gulf area into
are moving
reactor
technology
that while
allowsBrazil,
for theIndia
integration
these operations
existing
towards
dieselofregulation,
which are
should
a
process
confiultra-low
gurations.sulphur
The quality
these typesall
of of
feedstock
one deliver
important
strong
upsurge
in array
demand
for hydrotreating
in the
near
Taken
reason
why
a wider
of catalysts
has beencatalysts
introduced
into
theterm.
market.
For
together,
trends
strongly
bothtower,
catalystthe
sales
effort and theof
example,
asthese
refiners
cutare
deeper
intoinfluencing
the vacuum
concentration
siting
new
catalyst
production
centres.
metals
inof
the
VGO
requires
a properly
designed guard bed system to protect active
If not
on the scaleThe
of demand
for caroffuels
in developing
countries,
catalysts
in quite
the hydrocracker.
characteristics
feedstock
with low API
gravity
(eg,growth
<10), high
metals,for
nitrogen
and other
undesirable
components
is one more
of the
in demand
ships’ bunker
fuels
is strong and,
geographically,
main
reasons
why
hydrocrackers
— to
even.
Most of
thehydrotreaters
world’s trade and
is done
by ship andare
thebecoming
global fleetlarger
continues
accommodate
not
only
higher
volumes
of
catalyst,
but
also
a
wider
variety
to grow in line with populations and their trade. Although bunkers’ share ofof
catalyst
withmarket
specificfor
formulations.
the total
fuel oils continues to increase, there is uncertainty about
Non-catalytic
processes
also playing
a signifi
cant role
in the suppliers.
refiner’s ability
the rate of increase. The are
reasons
are of special
interest
to catalyst
The
to process
whatever
unconventional
crude
sources
become
available.
Forare
example,
International Maritime Organisation’s Marpol Annex VI regulations
delivsome
refianers
processing
higher
resid and
atmospheric
towerships.
bottoms
ering
steady
reduction
in thevolumes
level of of
sulphur
oxide
emissions from
In
have
considered
adding
certain
types
of
solvent-extraction
processes
in
addition
its initial stages, the legislation chiefly affected coastal and semi-enclosed
to overall
improvements
crudeeffect
unit in
(eg,
tower revamps) and delayed
seawaters,
but now it istotaking
thevacuum
open seas.
cokerFrom
operations.
Improvements
in
furnace
technology,
suchcap
as with
olefisulphur
n steam
the start of this year, a reduction in the global
on the
cracker operations, have resulted in significant increases in worldwide ethylene
content of bunker fuels, from 4.50% to 3.50%, came into force. A progressive
capacity.
reduction in the allowable level of sulphur in ships’ fuel will see the cap fall to
However, any expansion of the value chain (eg, ethylene-to-propylene via
0.5% in 2020, subject to a review in 2018. For coastal waters and sulphur emisdehydrogenation) requires investment in catalytic-based processes, as discussed in
sion control areas, the allowable sulphur level is set to fall to 0.10%, from a
the following articles authored by experts in the field of downstream process
current 1%, in 2015. Uncertainty arises in how the IMO expects to apply the
technology. PTQ wishes to extend its gratitude to the authors who provided
more stringent
levels of
VI. published
The IMO does
notissue
favour
stackCatalysis,
emissionsas
editorial
and responded
to Annex
the Q&A
in this
of PTQ
cleaning
on
board
ships,
but
ship
owners
are
not
especially
in
favour
of the
well as to those respondents who addressed the online questions (www.eptq.com)
premium
radicalreactor
drop inand
fuelcatalytic
sulphurissues
levels.ofDepending
thatprice
addressed
the implied
specificsby
of acertain
importanceonto
price balance, operators of new-build ships may opt for distillate as their
thethe
industry.
fuel of choice. In any event, a whole lot more hydrotreating to meet maritime
demand is implied.
chris cunningham
René G Gonzalez
CATALYSIS 2012
contents/ed com copy 8.indt 2
3 23/2/12 12:35:32
Worried about the cost of rare earth?
Grace has the solution:
REp
R
TM
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.
We’re also investing in our plants to bring these products to the refining industry
quickly and globally.
So if you’re concerned about rare earth pricing and availability, but need optimal
FCC performance, call the technical experts at Grace. We’ll customize a solution
using one of our new zero/low rare earth catalysts that delivers the yields you expect.
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
ptq&a
Q Is there significant commercial experience with solid acid
alkylation catalysts? What sorts of advantages are experienced
or expected over liquid acid catalysts?
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 significantly improves the
safety of refinery alkylation over conventional liquid
acid-based processes. It reduces potential hazards associated with the transportation and handling of liquid
acids. Relying on patented technology, combined with
Albemarle’s durable AlkyStar catalyst, the AlkyClean
process gives refiners a competitive, cleaner and inherently 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.
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 selectively maximise sulphur removal while minimising
olefin saturation and mercaptan recombination reactions. The best catalysts for the application are
formulated to maximise direct-route desulphurisation
with minimum hydrogenation activity. Cobaltmolybdenum catalysts are the preferred catalyst type.
Olefins are readily saturated under typical naphtha
hydrotreating conditions, so catalysts alone are usually
insufficient to prevent significant loss of octane in these
units. Licensed FCC gasoline post-treatment process
technology combined with proprietary catalyst technology 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 SCANfining,
selective FCC gasoline desulphurisation, process. This
catalyst offers exceptionally high HDS activity with
even better octane retention than the first-generation
SCANfining catalyst, RT-225. RT-235 can be used in
any SCANfiner and is also available for use in selective
FCC naphtha desulphurisation units licensed by
others.
Q How
in FCC gasoline hydrotreaters?
effective are NOx-reducing additives at cutting
regenerator stack emissions, and is using them a cost-effective
option?
A Brian Watkins, Manager of Technical Service and
A Alan Kramer, Global FCC Additives Specialist,
Q What catalyst types are best for minimising octane losses
Laboratory Evaluations, Advanced Refining Technologies,
brian.watkins@grace.com
Octane loss in FCC gasoline hydrotreaters occurs with
the saturation of the olefins 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/Al2O3) catalyst selective for hydrodesulphurisation is recommended. The goal is to be able to provide
the required sulphur removal with limited olefin and
aromatic saturation. Nickel molybdenum (NiMo/
Al2O3) catalyst, although having high hydrodesulphurisation activity, also has a much higher olefin and
aromatics conversion activity, making it unsuitable for
this application. Generally, to minimise olefin saturation, lower pressure and high liquid space velocity are
recommended in order to limit octane loss.
Albemarle, alan.kramer@albemarle.com
There are two additive types that can lower FCC regenerator stack NOx emissions. The first type is low-NOx
combustion promoters such as Albemarle’s ElimiNOx.
These replace conventional platinum-based promoters
used in full-combustion FCC units. ElimiNOx has been
shown to be very effective over the past 15 years in
lowering NOx emissions while maintaining CO and
afterburn control in the regenerator. In the 2007 NPRA
annual meeting,1 it was reported that US refineries
using low-NOx promoters, in accordance with EPA
consent decrees, usually saw between 20% and 80%
reductions in NOx after switching promoter type.
The second type of additive is a non-promoting NOx
reduction additive, such as Albemarle’s DuraNOx.
These additives are also only used in full-combustion
FCC units. The performance of these additives varies
Additional Q&A can be found at www.eptq.com/QandA
www.eptq.com
Q&A copy 10.indd 1
Catalysis 2012 5
23/2/12 12:51:18
greatly from unit to unit and is often difficult to
predict. During the 2007 NPRA annual meeting, it was
also reported that NOx reduction additives used by US
refiners engaged in consent decree trials averaged 26%
reduction in NOx. Of the full-burn FCC units reporting
results, five saw no effect, 10 observed reductions up
to 30%, and eight saw reductions between 50% and
80% when NOx additives were used.
When combined with tighter controls on regenerator
excess oxygen levels, Albemarle’s NOx reduction additives allow refiners to completely avoid the capital
expenditure of installing hardware to reduce NOx emissions, 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, Refining Additives Manager, BASF,
Jason.k.smith@basf.com
NOx-reducing additive performance and cost effectiveness is highly dependent on the unit, as both
equipment characteristics and operational variables
play a role in NOx formation and reduction. Controlling
NOx emissions is probably one of the most difficult
applications in the FCC unit.
There are several interacting factors that influence
NOx emissions. These include: type and level of CO
promotion, air flow distribution, excess oxygen, regenerator temperature, regenerator pressure, regen bed
level, stripping steam rate, catalyst circulation, and
type and quantity of NOx reduction additive.
There are two options to reduce NOx emissions. The
first approach is to reduce the level of NOx generated.
This can be achieved by replacing the platinum CO
promoter with one that generates lower levels of NOx
while maintaining the capability of oxidising CO to
CO2. Low NOx Promoter (LNP) is used by many refineries in controlling afterburn and CO emissions, with a
limited amount of NOx produced. Typically, switching
from platinum to LNP will result in a reduction of NOx
of approximately 30%.
A second approach is to use a NOx additive specifically designed to chemically reduce NOx to inert
nitrogen. Several NOx-reducing products are being
offered in the marketplace, including CleaNOx, BASF’s
NOx reduction additive. CleaNOx has been most
successful in the US, where it has been used to address
EPA consent decrees. In one example, a US refinery in
the midwest was able to reduce NOx by more than 70%
from an average of 200 ppm to 60 ppm using 1.4 wt%
of CleaNOx in its inventory.
CleaNOx has also been used and proven in applications where a refinery wanted to reduce NOx from an
already low average base of 27 ppm on the East Coast
of the US. Even in such a demanding application,
CleaNOx demonstrated a 33% reduction in NOx.
A Eric Griesinger, Marketing Manager, Environmental
NO, ppm
Additives, Grace Catalysts Technologies, eric.griesinger@
grace.com
NOx reduction additives generally fall under two categories: standalone NOx reduction additives and low
NOx combustion promoters.
Standalone NOx reduction additives are catalyticbased NOx control technologies that provide NOx
reduction without any combustion promotional activity. Generally, this NOx control technology has
provided a slow response to mitigating elevated NOx
concentrations. Grace Davison has developed a catalytic NOx reduction additive, GDNOx 1, which shows
prospect of providing a quicker ability to curb NOx
emissions. Further, GDNOx 1 technology, which has
been patented, provides greater NOx
reduction with a correspondingly
greater dosing rate (see Figure 1),
350
yet with diminished FCC unit yield
300
penalties often encountered when
utilising previous-generation NOx
250
reduction additives. Additionally,
2.5% GDNOX 1
200
GDNOx 1 has not been vulnerable to
5.0% GDNOX 1
material surcharges, thus making the
10.0% GDNOX 1
150
product a cost-effective option.
Current generation of low NOx
100
combustion promoters are typically
50
formulated with a noble metal other
than platinum. Historically, the use
0
of platinum has been demonstrated
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
to exhibit a correlation with elevated
Time, hrs
and prolonged NOx concentrations
in regenerator flue stack gases.
Applications of Grace Davison’s
GDNOX1
NOx after GDNOX 1 Percentage NOx
current-generation low NOx combusAddition rate % of inventory Base line NOx, ppm
addition, ppm
reduction, %
tion promoter, CP P, when dosed in
2.5% GDNOX 1
292
139
50
5.0% GDNOX 1
287
110
60
higher than normal rates, whether
10.0% GDNOX 1
287
62
80
intentionally to correct other FCC
unit conditions or unintentionally,
Figure 1 Pilot plant testing: NOx reduction with multiple GDNOx 1 additions
has shown a shortened duration of
6 Catalysis 2012 Q&A copy 10.indd 2
www.eptq.com
23/2/12 12:51:35
de use our hydropro
i
w
d
l
r
o
w
s
r
cessin
e
n
f
g
Re
a
c
t
a
d
l
y
n
s
a
t
s
s
e
t
i
o
g
deliver
lo
o
n
h
c
e
t
ts from low-quality f
c
u
d
o
r
p
r
e
eeds.
clean
you how. (001) 510.242.3
w
o
h
s
s
u
177
Let
www.clg-clean
.com
ISOCRACKING
VGO
ICR 512
ICR 180
ICR 185
ICR 250
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
Learn how CLG’s tailored hydroprocessing catalyst systems
maximize coversion to clean products — visit clg-catalysts.com
clg.indd 1
23/2/12 11:55:22
elevated NOx emissions is likely. This observation of a
shortened NOx emission excursion interval can provide
benefit to refiners when striving to satisfy a rolling day
average or other time-based NOx emission limit
constraints, while still providing CO promotion
performance similar to prior medium-activity platinum-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 final amendments to its New Source
Performance Standards for Petroleum Refineries
(NSPS)1 on 24 June 2008. Within this amendment, the
EPA states that the currently Best Demonstrated
Technology (BDT) to NOx emission control now
includes the use of additives in conjunction with an
upwardly revised NOx 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 NOx 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 environmental impact that many of the hardware solutions
inflict upon the environment, inherent in their operation to achieve a 20 ppmv maximum NOx emission
limit. These secondary impacts include PM (Particulate
Matter) as well as additional SO2 and NOx 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 formulated 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 NOx
control technologies to the refining industry in agreement with the EPA’s NSPS1 conclusions, whereby the
combination of non-platinum-formulated oxidation
promoters and advanced oxidation controls typically
are anticipated to provide the least overall environmental impact, and do so at a reasonable cost in many
applications.
1 New Source Performance Standards (NSPS) for Petroleum Refineries,
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 fluid catalytic cracking units (FCCU), fluid coking units
(FCU), sulphur recovery plants (SRP), and fuel gas combustion devices
for facilities that were newly constructed, modified 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 NOx 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 catalyst’s base price plus rare earth surcharges. The
adjustment formula to calculate the rare earth
surcharge is the result of mutual agreement between
suppliers and refiners 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 developed 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 example 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 conversion, 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 catalyst’s
rare earth content without compensating via advanced
low rare earth technology modifications. 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 stabilisation technology, improved porosity, reduced mass
transfer limitations (higher accessibility), advanced
active matrices and zeolites grown to a high silica-toalumina 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 profitability for the
FCC unit.
www.eptq.com
23/2/12 16:21:58
PTQ_Ad_CBI- jan-press.pdf 1 1/20/2012 11:18:33 AM
CB&I’s performance catalysts …
they cause quite the reaction.
C
M
Y
CM
MY
CY
CMY
K
www.CBI.com
Lummus Technology offers a multitude of performance
catalysts for refining and petrochemical processes, including:
t
t
t
t
t
t
Hydroprocessing
Alkylation
Hydrogenation
Hydrocracking
Olefins Metathesis
FCC
Visit our website for a complete list of catalysts and services.
Engineering Solutions . . . Delivering Results
cbi.indd 1
24/2/12 12:56:10
A Solly Ismail, Modeling Specialist, BASF, solly.ismail@ The biggest challenge in fluid catalytic cracking is
basf.com
While the reduction of rare earth in catalyst formulations 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 customers with evaluating the impact of lower rare earth
catalytic options and to limit its downside, BASF works
closely with each refiner to understand specific unit
parameters and objectives. A customised strategy can
include a combination of levers such as increasing catalyst addition rates, increasing total surface area or a
combination of both in order to restore activity and
achieve desired product specifications 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 fits the
needs of the specific user.
As of the end of Q4 2011, 40% of our customers had
made the switch to a lower rare earth formulation. Of
these, five went through multiple reductions. All
customers were happy with BASF’s approach of tailoring new solutions based on either increased surface
area with a minimum of additional catalyst usage. The
company also worked closely with refiners in monitoring 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 catalysts 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 significant catalyst
cost savings associated with high rare earth prices.
Refineries moving to zero rare earth REpLaCeR catalysts 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
10 Catalysis 2012
Q&A copy 10.indd 4
converting as much material in the feed with an atmospheric boiling point above 370°C bottoms to more
valuable LPG, gasoline (hydrocarbon molecules boiling
between about 40°C and 221°C) and light cycle oil
(LCO — hydrocarbon molecules boiling between 221°C
and 370°C). To maximise the yield of LCO, it is imperative to maximise the conversion of bottoms to LCO
while minimising the conversion of LCO to lighter
products and coke. It is generally accepted that mesopore 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 catalyst. Accessibility is the property that allows primary
products to escape promptly from the reaction sites.
Outstanding performance of highly accessible catalysts, as measured by our internally developed
Albemarle Accessibility Index (AAI) method, has been
confirmed 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 recommended for cracking residual feedstocks. For
applications requiring flexibility, the company’s
bottoms conversion additive, BCMT-500, is recommended for all types of feedstocks. In addition,
Albemarle’s 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, Albemarle’s technical specialists will assist their customer in optimising
their operating strategy for maximum LCO production,
as discussed in the following two examples.
The first example includes an FCC unit processing
vacuum gas oil with a typical API of 23°C and a
sulphur content of about 1 wt%. The catalyst used in
this example is Amber MD. The FCC unit was operating at a unit riser outlet temperature of 518°C, a
combined feed temperature of 226°C and a catalyst-tooil ratio of 7.0 kg/kg. In this example, the gasoline end
point is minimised to an ASTM D-86 end point of
149°C, while the LCO end point is very high at an
ASTM D-86 end point of 379°C. 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 first example,
www.eptq.com
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.catalysts.basf.com/process
Adsorbents
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 Polyolefin Catalysts
n Refining Catalysts
n Syngas Catalysts
n Custom Catalysts
n
n
basf.indd 1
23/2/12 11:58:16
extremely low severity. Despite the high bottoms yield
unit, the economics were much improved as the market
favoured a high LCO yield.
0.7
Pore volume, cm3/g
0.6
A Rosann Schiller, Senior Marketing Manager, rosann.
0.5
0.4
0.3
0.2
0.1
0
10
100
1000
Pore diameter, Å
Figure 1 Porosity of LCO maximisation catalysts
European maximum distillate trial of BASF Stamina catalyst
Unit composition
Competitor, % Stamina, % Early trial
85
15
Late trial
31
69
Unit operating conditions
Total feed rate, ton/d Feed preheat temp, C ROT, C Regenerator bed temp, C C/O, wt/wt 1723 178 520 721 7.0 1672
232
515
720
5.8
Feed quality
Specific gravity Conradson carbon, wt% Basic nitrogen, ppm TBP90, C
0.93 0.71 450 547 0.92
1.14
410
571
Equilibrium catalyst properties
V + Ni, ppm TSA, m2/g MSA, m2/g Z/M FACT activity, wt%
4100 120 53 1.3 73
4240
127
64
1
70
FCC unit yields (with cutpoint adjustments)
Gasoline (C5-160C), wt% 37.19 LCO (160-340C), wt% 29.03 Slurry (340C+), wt%
13.57
Coke, wt% 5.79 LCO/slurry, % 68.1
34.06
33.87
13.14
5.09
72.1
Summary:
•Achieved >4.8 wt% greater distillate yields at partial turnover
•Maintained lower bottoms at reduced reactor severity and dirtier
feed conditions.
Table 1
Amber MD is used. In this example, there is a low
reaction temperature (499°C), high combined feed
temperature (368°C) 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, respectively. Bottoms yield (21.2 wt %) is higher due to
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 catalysts from Grace incorporate proprietary pore
restructuring functionality, which results in more pores
with the diameter range of 100-600 Å (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. Grace’s 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 cokeselective 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 flexibility to swing between
maximum conversion and maximum distillate operations with the same catalyst should the market change
rapidly. BASF’s Prox-SMZ (Proximal Stable Matrix and
Zeolite) technology is an example that addresses all of
www.eptq.com
23/2/12 12:52:08
Crude unit
Gas / liquid
purification
S, Hg, Cl removal
Hydrotreating catalysts
Hydrogen catalysts
Naphtha HDS
Hydrogen plant
Process
diagnostics
Diesel HDS
Vacuum tower
VGO HDS
FCC unit
FCC additives
SOx NOx removal
Light olefin production
Bottoms conversion
Metals traps
Activity boosters
CO oxidation
Value adding catalysts, absorbents, additives and
process technology for oil refining processes.
www.jmcatalysts.com/refineries
UK
Tel
+44 (0)1642 553601
1
j JM_2485_RefineriesAd_ART_210x297.indd
matthey.indd 1
USA
Tel +1 630 268 6300
Tel +1 732 223 4644
09/02/2012
15:59
23/2/12
11:59:38
the
above
(see
NPRA-AM-09-34
and
NPRA-AM-10-17).
With BASF’s unique manufacturing process,
40
26
the zeolite and the matrix are not physically
35
22
blended, as in a traditional low Z/M catalyst,
but are created in-situ during a single manu30
18
facturing
step.
This
process
creates
25
14
unprecedented proximity of matrix and zeolite,
enabling reduced diffusion path length and
20
10
best-in-class coke selectivity among the low
15
6
Z/M family of catalysts. Further, the extremely
low sodium content (below 0.1 wt%) achiev10
2
able in this manufacturing process not only
Time
enables high stability of both zeolite and
Figure 1 LCO and slurry yields with BASF Stamina. LCO yields were increased matrix to reduce the opex, but also helps to
with reduced reactor severity without impacting the slurry
reduce the hydrogen transfer reactions,
improving the LCO cetane. Maximum LCO
catalysts
from the Prox-SMZ technology have
520
been commercially established and are avail518
able for both VGO (under the name of HDXtra)
and resid feeds (under the name of Stamina).
516
In several classic distillate maximisation
514
trials with BASF’s Stamina catalyst, LCO yields
512
have increased while either holding steady or
dropping slurry yields. In one case, a European
510
FCC unit was processing resid feed and
508
wanted to move to LCO maximum mode while
506
dropping slurry yields and improving coke
504
selectivity. The unit severity was dropped by
Time
lowering the reactor outlet temperature by 5°C
and increasing the feed preheat. The reduced
20
Competitor
severity conditions together with the highly
Stamina
18
stable Stamina MSA generated a 4.84 wt%
improvement in LCO yields while maintaining
16
bottoms conversion. All this was achieved in
spite of dirtier feed conditions (ie, higher
14
Conradson carbon and feed metals).
The second case was an Asian trial that also
12
underwent reduced reactor severity to maximise LCO yields. While using Stamina catalyst,
10
the lowest bottom yields on record were
achieved together with record throughputs due
8
52
54
56
58
60
62
64
66
to the coke-selective bottoms upgrading.
30
Slurry (340ºC+), wt%
LCO
HCO + slurry
Slurry, wt%
Temperature, ºC
LCO (160-340ºC), wt%
45
Total feed rate, tons/day
Conversion, wt%
10500
Q Are there any dedicated catalyst developments
10000
geared towards favouring FCC propylene yield?
A Stuart Foskett, Regional Technology Manager,
9500
9000
8500
8000
7500
Time
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
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 define how much propylene can ultimately be produced. Propylene yields
eventually reach a plateau as ZSM-5 content is
increased to high levels; therefore, a holistic
www.eptq.com
23/2/12 12:52:21
approach to catalyst design requires
attention to additional factors
beyond the ZSM-5. It is the generation and preservation of gasoline
olefins, as precursors to ZSM-5
cracking, that is the defining 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
olefins for cracking to propylene by
ZSM-5. We have also invested heavily 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). BASF’s MPS
(maximum propylene solution) technology was first introduced in 2005
(see NPRA-AM-05-61) and has
undergone continuous improvement
since then. MPS has been operating
continuously in the world’s largest
propylene-focused FCC unit since
2006. Based upon extensive circulating pilot plant evaluations, we are
anticipating the latest MPS developments 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 progressing. 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
unit’s objectives. Crucial to the
success of reaching record propylene yields is the ability to minimise
www.eptq.com
Q&A copy 10.indd 7
hydrogen transfer while having
sufficient cracking activity.
Albemarle’s AFX catalyst has
been developed to meet the desired
objectives, through its unique
features of high catalyst accessibility and strong matrix activity.
Hereby, maximum slurry conversion is achieved while generating a
maximum of gasoline precursors,
which are converted in record
propylene yields. The high accessibility of AFX enables fast diffusion
of primary cracking products away
from the acid sites, thus minimising
unwanted hydrogen transfer.
Tower packings,
catalyst support
material and
column equipment.
Q
What is the impact of vanadium level
on E-cat affecting FCC gasoline sulphur?
3
4
A Alan
Kramer, Global
FCC
Additives Specialist, Albemarle, alan.
kramer@albemarle.com
Generally, refiners 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 indicated the vanadium mechanism
primarily reduces the saturated
sulphur content of the gasoline and
has little to no effect on benzothiophene, 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
justifiable. Depending on local regulations, equilibrium catalyst with
high levels of vanadium may be
classified as hazardous or toxic
waste and can be very difficult and
expensive to dispose of properly.
Refiners do have other options
besides increasing E-cat vanadium
or using vanadium-based products
for reducing gasoline sulphur. For
example, Albemarle’s R-975 and
Scavenger catalyst additives are
designed to remove gasoline sulphur
and do not contain any vanadium.
5
6
Please visit us
Hall: 4.0
Stand: D66
10
7
8
9
4
6
1
4
2
10
7
8
9
1
4
2
1
DURANIT® inert ceramic balls
2
special reformed packings
3
droplet separators / demisters
4
support plates / grids
5
feed devices: gas / liquids
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
For further information
please visit:
www.vff.com
P. O. Box 552, D - 56225 Ransbach-Baumbach
Phone +49 26 23 / 895 - 0, info@vff.com
Catalysis 2012 15
23/2/12 12:52:32
dupont.indd 1
23/2/12 15:04:26
Evaluation of a low rare earth resid
FCC catalyst
A zero rare earth catalyst blended with a rare earth-based resid catalyst enabled a
refinery to reduce its FCC catalyst rare earth requirement by 80%
SABEETH SRIKANTHARAJAH and COLIN BAILLIE Grace Catalysts Technologies
BERNHARD ZAHNBRECHER and WIELAND WACHE Bayernoil
R
are earth metals have played
an important role in the refining industry since the 1970s,
when it was discovered that they
could be used to stabilise the
zeolite-Y component of FCC catalysts to provide higher activity, as
well as being used to influence
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 reduction 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 somewhat, but remain significantly
higher than historical levels.
The rare earth market is incredibly 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 first commercially 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
www.eptq.com
grace.indd 1
most FCC operations due to the
lower activity and worsening product 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 catalysts. 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 lowmetal 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-ofthe-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 catalysts incorporating these new
zeolites provide similar and even
improved performance compared
to conventional rare earth-containing 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 development of rare earth-free catalysts
is much more challenging due to
the additional demands placed on
zeolite stability. However, significant advances have been made by
applying processing technology
involving metals resistance functionality
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 associated. There are currently 22
refineries using the REduceR catalyst, and typically they are applying
a stepwise approach to implement
the rare earth-free catalyst. Refiners
are starting with a blending level of
30% REduceR catalyst and, upon
confirming 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 refinery is
located in the Bavarian region of
southern Germany and, along with
the nearby Bayernoil Neustadt
refinery, contributes to a total refining 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
Catalysis 2012 17
23/2/12 12:56:58
E-Cat MAT
CAR
6
REduceR
78
76
74
CAR, t/d
72
5
70
68
4
66
64
3
E-Cat MAT, wt%
7
62
11
20
11
08
/0
9/
20
11
14
/0
7/
20
11
19
/0
5/
20
11
24
/0
3/
20
10
27
/0
1/
20
10
02
/1
2/
20
10
0/
/1
07
12
/0
8/
20
10
10
17
/0
6/
20
10
20
4/
20
2/
/0
22
/0
25
31
/1
2/
20
09
2
Figure 1 Activity retention of the REduceR catalyst blend at 30%
G a s f a c t or
/FLUPS
3&EVDF3
Ni equivalents, mg/kg
/FLUPS
3&EVDF3
Coke f act or
Ni equivalents, mg/kg
Figure 2 Coke and gas factors of the REduceR catalyst blend at 30%
has a Conradson carbon content of
3 wt%, and the e-cat metals levels
are approximately 4500 ppm vanadium, 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
18 Catalysis 2012
grace.indd 2
earth, which performed well. In
April 2011, the refinery began to
blend 30% of REduceR catalyst with
the Nektor catalyst, with the simple
objective of reducing rare earth
while maintaining high performance. A certain misconception about
rare earth-free catalysts is that they
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 similar 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 catalyst shows lower gas factors and
similar coke factors, further demonstrating 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 regenerator bed temperature by about
10°C. This allowed the refinery to
achieve higher conversion at
constant feed atmospheric residue
content, or to process an increased
amount of atmospheric residue at
constant conversion.
The refinery considered the
performance of the REduceR catalyst to be such a success that they
increased the blending ratio from
30% to 50%, thus reducing the overall 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 conditions. The key objective of the
refinery 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
www.eptq.com
23/2/12 12:57:07
pcs 1.indd 1
23/2/12 20:37:57
3&EVDF3
/FLUPS
3&EVDF3
/FLUPS
De l t a coke , X t %
Sl u r r y yi e l d wU %
3&EVDF3
/FLUPS
Appare n t con ve r si on , wt %FF
Re ge n e r at or be d t e mpe r at u re , ºC
Apparent conversion, wt%FF
Conversion, wt%FF
3&EVDF3
/FLUPS
AtRes in feed, wt%
ROT, ºC
Figure 3 FCC unit data of the REduceR catalyst blend at 30%
Calculated FCC unit data of the REduceR catalyst blend at 50%
Cat-to-oil, g/g
Conversion, wt%
Hydrogen, wt%
C1+C2s, wt%
Propylene, wt%
C4 olefins, wt%
LPG, wt%
Gasoline, wt%
LCO, wt%
Slurry, wt%
Coke, wt%
Delta coke, wt%
CAR, MT/D
Feed Ni, ppm
Feed V, ppm
Regen bed temp, °C
Nektor
50% REduceR
Base
Base
Base
Base
Base
Base
Base
Base
Base
Base
Base
Base
Base
Base
Base
Base
50% Nektor
Base + 0.4
Base + 0.5
Base + 0.02
Base + 0.2
Base + 0.4
Base + 0.6
Base + 2.0
Base - 1.6
Base - 0.2
Base - 0.2
Base - 0.1
Base - 0.09
Base
Base
Base
Base - 15°C
Table 1
provided a similar coke yield but
an improved delta coke, and
allowed the regen bed temperature
to be decreased by 15°C. The higher
LPG yield at the expense of gasoline is a consequence of the lower
hydrogen transfer from REduceR.
This is a positive yield shift for the
refinery and was anticipated.
Bayernoil Vohburg was highly
20 Catalysis 2012
grace.indd 3
satisfied with the REduceR catalyst
trial, and subsequently became the
first refinery to move to a 70%
blending level, further reducing the
rare earth content to 1.0 wt%. In
December
2011,
the
refinery
increased the blending ratio of the
REduceR catalyst up to 80%, thus
reducing rare earth to 0.6 wt%.
Despite high nickel and vanadium
levels, the refinery continues to see
excellent performance. It is estimated that the switch to the
REduceR catalyst has provided the
refinery 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
master’s 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
Refinery in Vohburg. He holds a diploma in
chemistry from the Technical University
Aachen and PhD in chemical engineering from
University Bayreuth.
www.eptq.com
23/2/12 12:57:19
pcs2.indd 1
23/2/12 20:39:00
eurecat.indd 1
23/2/12 12:02:36
Refinery fuel gas in steam reforming
hydrogen plants
Fuel gas is an attractive feedstock for hydrogen production, but appropriate
catalysts and temperature control are needed to address high olefin levels
Peter Broadhurst and Graham Hinton
Johnson Matthey Catalysts
O
perators of steam reforming-based hydrogen plants
want feedstock options to
minimise operating costs and
maximise operational flexibility.
Consequently, new-build hydrogen
plants are often designed for a
number of hydrocarbon feeds. It is
common to have three or four feedstocks ranging from offgases
through to naphtha requiring full
operational flexibility 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 modification of
the plant’s operating conditions,
hardware,
equipment,
catalyst
selection and so forth.3-6 The quest
for cheaper feedstock options is
undoubtedly heightened by the
significant 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 feedstock option being considered
increasingly by hydrogen plant
operators associated with oil refineries is the refinery fuel gas (RFG)
pool. Relative to imported natural
gas and many other hydrocarbon
streams and offgases in the refinery,
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 difficulties in processing in the feedstock purification
and steam reforming sections of
the hydrogen plant flowsheet. In
this article, we will explore these
www.eptq.com
j matthey.indd 1
difficulties and the strategies for
hydrogen plant design and operation, 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
difficulties
RFG is a combination of refinery
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 processed elsewhere in the refinery
either because of the composition
or because there is an excess available. The offgases in the RFG come
from various refinery unit operations (catalytic reforming, FCC,
hydrotreating, coking), the availability and amount of which will
depend on the refinery operation.
Hydrogen-containing offgases may
be partly or fully used in hydrogenconsuming 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 olefin levels
may be treated to recover olefins,
with the olefin lean gas going to the
RFG pool. Thus, RFG can differ
significantly between refineries.
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%; methane, from 26–65 mol%; and olefins,
from 2.6–15.9 mol%.
RFG feeds often contain quite
high levels of sulphur compounds.
Up to 100 ppmv can be found and
this can contain quite a significant
proportion of mixed organic
sulphur compounds. The level
and speciation are necessarily
RFG compositions considered for hydrogen plant feed
Mol%Example 1Example 2Example 3Example 4Example 5Example 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
~
~
~
~
~
Carbon monoxide ~
0.3
0.3
0.3
1.1
Trace
Carbon dioxide
0.4
0.8
0.2
0.4
~
0.5
Total
100.0
100.0
100.0
100.0
100.0
100.0
Table 1
Catalysis 2012 23
23/2/12 13:02:52
dependent on the blend of gases
going to the RFG pool.
Additionally, the RFG composition can fluctuate significantly in a
given refinery 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 difficulties can be summarised:
• High olefin levels
• Variability in the RFG composition as the blend of offgases
changes
• High hydrogen levels
• Significant 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 difficulties 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 existing plant, the extent to which there
is a problem will be influenced by
the original design basis.
Cooler
Recirculator
Figure 1 HDS converter with recirculation
hydrogenation reaction, other hydrogen-consuming reactions and to
provide the target hydrogen level
specified at the HDS converter exit.
Johnson Matthey recommends a
different level of hydrogen be
present in the HDS effluent, depending on the feed composition and
how heavy it is. For the gases given
as examples 5 and 6 in Table 1,
insufficient hydrogen is present in
the RFG on its own to hydrogenate
the significant olefin content. This
means that additional hydrogen
must to be added, usually recycled
to the purification 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.
Olefins hydrogenate exothermically over the HDS catalyst and the
Feedstock purification section
temperature rise can be 20+°C
RFG feeds can cause various issues (36°F) per mol% olefin, depending
in the feedstock purification on the heat capacity of the feed gas.
section.
The inlet temperature must be
adjusted to ensure that the maxiHigh olefin levels
mum HDS bed temperature remains
Olefins need to be removed from the below 400°C (752°F). However,
hydrocarbon feed in a hydrogen using standard HDS catalysts, such
plant to a level below 1 mol% to as Katalco 41-6T or Katalco 61-1T,
minimise possible olefin-derived the inlet temperature needs to be
carbon formation in the steam above 300°C (572°F) to provide
reformer, although higher levels sufficient activity for reactions to
may be acceptable where there is a initiate. Given the need for some
pre-reformer in the flowsheet. The operating margin inside these
hydrodesulphurisation (HDS) cata- restrictions, this limits the olefin
lyst is also an effective olefin that can be processed to a few
hydrogenation catalyst and removes mol% in a once-through reactor
olefins almost completely as long as system.
there is sufficient hydrogen present.
To process higher olefin levels, a
Thus, for RFGs with significant recirculation system is usually
olefin levels, the hydrogen available employed around the HDS reactor
in the feed and as recycle must (see Figure 1). The reactor effluent
be
sufficient
for
the
olefin is olefin free so the recycle dilutes
24 Catalysis 2012
j matthey.indd 2
the inlet olefin level to an acceptable level. This approach has some
disadvantages. Depending on the
inlet olefin level and therefore the
extent of dilution required, the
recycle can be substantial. This
increases capital expenditure in a
number of ways: the recycle flow
usually increases the vessel size to
stay within design space velocity
design limits and additional equipment is needed such as a recycle
cooler and circulator.
Without a recycle system, there is
limited flexibility in terms of the
olefin level using standard HDS
catalyst types, so approaches that
widen the operating envelope without installation of a recycle system
will be beneficial. 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 temperatures down to ~250°C (462°F) with
the same exit temperature limit of
400°C (752°F). Thus, the amount of
olefins that can be processed without installation of a recycle system
is increased to 7–8 mol%, depending on the precise composition of
the feed in terms of heat release
and specific heat.
This concept has been extended
to form part of the technology
claimed in recent patent applications filed jointly by Johnson
Matthey and Air Products &
Chemicals7 and which is available
for licence. Detailed evaluation of a
typical range of RFG feed compositions established that the extremely
active pre-sulphided Katalco product allows the olefin hydrogenation
reaction to strike below 150°C
(302°F). This widens the operating
temperature envelope to over 250°C
(450°F) and allows the processing
of feeds containing well in excess of
10 mol% olefins.
It is imperative, however, that the
consequences of the changes in
operating temperatures are fully
considered if this is to be retrofitted
into an existing reactor to accommodate a higher olefin 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,
www.eptq.com
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 sufficiently fast for the temperature to rise in the reaction zone
within the tubes before the coolant
brings the temperature back down.
This approach has been commercialised in at least one plant.
Variability in the RFG composition
As refinery unit operations change
rate, come off line or back on line,
the RFG composition can fluctuate.
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 olefin content in the
feed. If the olefin level increases
sharply, it is possible that the
400°C (752°F) upper limit for HDS
operating temperature might be
exceeded. Also, if the inlet temperature has been lowered to well
below 300°C (572°F) to allow for
the hydrogenation exotherm, and
the olefin 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 temperature 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
insufficient to hydrogenate olefins
(and any traces of oxygen and the
expected organo-S compounds) in
the feed. Additionally, the recommended HDS exit hydrogen level
should ideally be maintained. If a
rapid decrease in feed hydrogen
level occurs, another source of
hydrogen, probably recycle hydrogen, needs to be rapidly established.
If not, the consequences could be
slippage of olefins and/or sulphur
compounds to the steam reforming
section, with resulting carbon
formation and/or poisoning.
26 Catalysis 2012
j matthey.indd 3
High hydrogen levels
Some RFG blends may contain a
significant 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 reactions are methanation, which would
also need a substantial amount of
carbon oxides present, and hydrocracking, 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 flow, very low
sulphur and excess temperature).
Methanation and hydrocracking
activity
can
be
significantly
suppressed by pre-sulphiding the
HDS, an effect that is also achieved
if there is a reasonable level of
sulphur in the feed.
Significant sulphur levels
RFG blends can contain significant
sulphur levels, of which much may
be present as organo-sulphur
compounds. As long as the maximum 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 difficult organic sulphur
compounds so that conversion in
the existing HDS converter may not
be possible. There are three retrofit
options depending on the additional 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 H2S
absorption (such as Katalco 33-1) in
the downstream H2S 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 H2S removal beds will be shortened. If the design features lead-lag
H2S removal beds, allowing changeout on-line, the consequence is
more frequent change-out. Selection
of the highest capacity H2S removal
absorbents (for instance, Katalco 325) can lessen the change-out
frequency. The problem can be
greater if the design has only a
single H2 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 retrofit
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 catalyst with a high-voidage shape,
such as Dypor 604, which has more
capacity for the carbon foulant,
before the pressure drop becomes
critical.
Less usual trace and minor
components
There are a number of other species
that can be present in a refinery
offgas stream and which, therefore,
could find their way into the RFG.
These include chlorides, arsine,
oxygen, acetylenes and dienes.
Chloride can be dealt with by
hydrogenation of any organochlorides 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
H2S removal absorbent. In existing
plants, it may be necessary to retrofit this into the purification section,
usually by putting it on top of the
H2S 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
www.eptq.com
23/2/12 13:03:22
cat tech2.indd 1
23/2/12 12:06:09
Impact of varying feed composition on feed flow and reformer heat load
Natural Example Example Example Example Example
gas
1
2
3
4
5
Reformer heat load, % 100
93.4
97.8
94.2
89.6
99.9
Feed flow, %
100
76.2
88.4
81.6
109.7
69.6
Average carbon no.
1.06
2.09
1.34
1.76
1.42
1.71
Example
6
99.5
64.5
1.85
Table 2
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 olefin 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 deposition. 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 associated equipment need to be
considered.
High olefin levels
With proper design and operation,
olefins should be converted in the
feed purification section. However,
levels of up to 1 vol% olefins can be
accommodated in the feed to a
steam reformer.
High hydrogen levels
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 flow 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
28 Catalysis 2012
j matthey.indd 4
that the tendency for carbon formation 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-tocarbon ratio and reformer outlet
pressure). The second row of the
table shows how the molar flow of
the RFG in question will compare
to the molar flow of a typical natural 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 existing 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 first, and most
obvious, is that the firing on the
reformer box must be reduced. This
means that the bridge wall temperature will reduce, the flow rate of
combustion gases will reduce and
the amount of heat available to be
recovered in the flue gas duct will
be less. Hence, there is the possibility that some of the coils may
become pinched, leading to a reduction in steam production or in
combustion air preheat temperature. 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
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 fire this
new fuel type with good control
and flame stability.
Significant sulphur levels
sulphur should be removed in the
feed purification 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 flowsheets,
the impact of a heavier hydrocarbon 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.
Less usual trace and minor
components
these should be removed in the
feed purification section.
Variability in the RFG composition
This presents possible problems for
the steam reforming section if there
is not sufficient control. If there is a
sudden decrease in olefins causing
the HDS exit temperature to
decrease, slip of organo-sulphur is
possible, which would poison the
pre-reformer or steam reforming,
whichever is first in the flowsheet.
If there is a sudden decrease in
hydrogen, there may be insufficient
to hydrogenate olefins and/or
organo-sulphur. The resulting slip
could lead to sulphur poisoning
and carbon formation in the prereformer or steam reforming,
whichever is first in the flowsheet.
Conclusions
While RFG is an attractive economic
www.eptq.com
23/2/12 13:03:34
option
for
hydrogen
plants
associated with refinery operations,
its processing presents challenges
in the feedstock purification and
steam reforming sections of the
flowsheet. These challenges result
from the composition of the RFG
and also from its variability.
In the feedstock purification
section, the range of options for
handling the high olefin levels often
found in the RFG pool has been
expanded in recent years by considering more active catalysts and heat
exchange reactors. In conjunction,
control systems have been developed to allow the HDS converter
operating temperature to be
adjusted as the feed composition
changes to maintain the temperature 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 necessary 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 Refinery 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.
Elemental Analysis
of Fuels
Determination of Sulfur and other elements
at-line and in the laboratory
www.eptq.com
j matthey.indd 5
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 Olefins, World Patent WO
2009/123909 A2, 8 Oct 2009.
8 Musich N, Natarajan R S, Klein H, Process
and System for Reducing the Olefin 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 purification product team.
He was previously Technical Sales Manager
for hydrogenation catalysts, hydrogen plant
catalysts and refinery absorbent products, and
Technical Manager for oil refinery 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 master’s
degree in engineering science, specialising in
chemical engineering, from the University of
Oxford, UK.
To keep pace with the demanding quality requirements of
modern fuels, advanced, precise and easy to use analytical
technology is required. With a complete range of XRF and
ICP spectrometers, SPECTRO’s unique solutions for at-line
and laboratory elemental analysis are capable of meeting
the most demanding product specification testing
requirements.
Measure your fuels
- at sub 10-15 ppm levels of sulfur to ensure federal
government agency compliance
- at trace ppm levels for metal elements such as Cu, Ca, Mg,
Na, K, and P, to ensure low engine emissions as well as
trouble-free motoring
- with reliability, speed and accuracy for any particular application
- manually or fully automatically
Discover more exciting details,
visit SPECTRO’s
e-Learning center
or contact us for
additional information
about the SPECTRO
solutions for fuels
analysis at www.spectro.com/fuels
spectro.info@ametek.com and
Tel +49.2821.892-2102.
Catalysis 2012 29
23/2/12 16:23:05
0HVKQXPHURXVDOOR\SODVWLFV
,QVHUWLRQ0LVWIL[0LVW(OLPLQDWRU
3ODWH3DFN0XOWL3RFNHW9DQHV
9DQHV&KHYURQ6LQJOH'RXEOH3RFNHW
WLGHYDULHW\RIUDQGRPSDFNLQJW\SHV
VL]HVDQGPDWHULDOVLQVWRFN $VNXVKRZ
RXUSDWHQWHG6XSHU%OHQG3DFFDQ
LQFUHDVH\RXUFDSDFLW\DQGHI¿FLHQF\.
)L[HG)ORDWLQJ9DOYH7\SH
6LHYHRU3HUIRUDWHG
%XEEOHFDS
'XDOIORZ%DIIOH
3"/%0.1"$,*/(
53":4
.*45&-*.*/"5034
ACS-AMISTCO Catalyst Bed Supports
Catalyst Bed Supports
Basket Strainers
Nozzles
Outlet/Inlet Baskets
Distributors
Hub and Header Laterals
For separation or mass transfer applications you will not find a
manufacturer with a wider range of internals and factory
capability that can deliver like ACS-AMISTCO, Inc.
With in-house engineering support and fabrication, we can use
your existing drawings, or modify them to improve your process.
wwwBDTBNJTUDPDPNtIS&.&3(&/$:4&37*$&t
%*453*#6503446110354
-*26*%$0"-&4$&34
4536$563&%1"$,*/(
Manufactured to customer
VSHFL¿FDWLRQVRUHQJLQHHUHGWR
meet performance requirements
2LOZDWHUVHSDUDWLRQV+D]HUHPRYDO
IURPIXHOV5HPRYDORIWRZHUZHWUHÀX[
Caustic treater applications
WRYHQZLUHVKHHWPHWDODQGNQLWWHG
VWUXFWXUHGSDFNLQJ%XLOWWRVSHF
or performance requirement.
amistco.indd 1
23/2/12 12:08:58
Estimating silicon accumulation in coker
naphtha hydrotreaters
Improved sampling and analysis of silicon in the feed enable a significant gain in
the cycle life of coker naphtha hydrotreater catalysts
Thienan Tran, Patrick Gripka and Larry Kraus
Criterion Catalysts & Technologies
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 siliconcontaining oils, polydimethylsiloxane
(PDMS), to suppress foaming in the
coker drums. At the elevated
temperature inside the coker drums,
these high molecular weight, siliconcontaining oils crack to form lighter
silicon oil fragments, such as dimers
and trimers of the dimethylsiloxane.
The majority of these silicon oil fragments boil in naphtha range and
therefore are routed to the downstream
naphtha
hydrotreaters
together with the coker naphtha.
Under the operating conditions of
the naphtha hydrotreaters, the silicon oil fragments present in the
feed transform to modified 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 silicon is bound to the catalyst surface
it cannot be removed and results in
an irreversible loss of catalyst activity. 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 silicon capacity of the selected catalyst
system and the silicon accumulating rate. The silicon capacity of a
selected catalyst system is known.
However, the silicon accumulating
www.eptq.com
criterion copy.indd 1
rate often cannot be determined
due to a lack of accurate feed characterisation data.
Refiners 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 sufficient
to determine the actual amount of
silicon being fed to the unit. In addition 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 refiners
changing out coker naphtha hydrotreater catalyst based on fixed 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 hydrotreater to obtain composite samples.
Unit description
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
Table 1
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 catalysts from the unit were analysed to
determine the amount of silicon
accumulated on the catalyst. Results
indicated the silicon deposition estimated using hot loop sampling and
ICPDIN was within 10% of the Si
deposition determined from spent
catalyst analysis. Due to the accuracy of the estimate, the cycle life of
the unit could have been extended
up to 4.5 months beyond the scheduled 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 reactors, which are in parallel. The guard
reactor contains OptiTrap grading
materials and DN-200. The primary
function of the guard reactor is to
saturate diolefins. The main reactors
contain 45% MaxTrap[Si], 20% DN140 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 significant silicon
uptake capacity and hydrotreating
activity for the load. DN-3531 is a
high-activity NiMo hydrotreating
catalyst, which provides the majority of the HDS and HDN activity
requirements for the unit. The catalyst system had a silicon capacity of
Catalysis 2012 31
23/2/12 13:08:39
7
Si content, wt ppm
6
5
4
3
2
1
0
0
5
10
15
20
25
30
35
40
Week on stream
Figure 1 Weekly average Si content in feed
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
&TUJNBUFEBDDVNVMBUFE4J
-JOFBSFTUJNBUFEBDDVNVMBUFE4J
#VEHFUFE4JBDDVNVMBUJPO
-JOFBSBDUVBM4JBDDVNVMBUJPO
"DUVBM4JBDDVNVMBUJPO
*GUIFVOJUIBEOPUCFFOTIVUEPXOEVFUP
GVSOBDFGPVMJOHUIFDZDMFDPVMEIBWFCFFO
FYUFOEFEVQUPNPOUITCFZPOEUIF
TDIFEVMFENPOUIDZDMFMFOHUI
Si l i con , l bs
4JCSFBLUISPVHI
DBQBDJUZMCT
5IFFTUJNBUFE4JBDDVNVMBUJPO
JTMPXFSUIBOUIFBDUVBM
ýHVSF5IJTEJGGFSFODFJTXJUIJO
UIFFSSPSNBSHJOPGUIFBOBMZUJDBM
UFTUVTFEJOUIJTTUVEZDBTF
Days on stream
Figure 2 Silicon accumulation
8750 lb and was scheduled for a 12month 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 silicon 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 defined
32 Catalysis 2012
number of feed samples throughout
the week to obtain weekly composite 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 refiners, cannot
accurately measure the silicon
content in the coker naphtha.
Depending on the form of the silicon compound and the dynamics of
the sample introduction system of
the ICP method being used, the
silicon concentration determined
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 deposition profile to be constructed.
Samples of the spent catalyst were
collected to determine the silicon
uptake capacity of each type of catalyst and the total amount of silicon
accumulated on the catalyst system.
Results and discussion
Results of the weekly silicon analysis indicated the silicon content in
the feed varied from 0.26 to 6.5
wppm, with 62% of the data population 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 significantly affected the estimated 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 analyses indicated the actual amount of
silicon accumulated on the catalyst
www.eptq.com
23/2/12 13:08:46
OptiTrap
products
Silicon uptake, wt%
25
DN-200
MaxTrap (Si)
DN-140 DN-3531
Guard reactor − silicon
Main reactors − silicon
20
Our new clean-fuels
plant is straining the
auxiliary units!
15
10
5
0
0
10
20
30
40
50
60
Feet down Rx system — Guard inlet = 0
Reactor inlet = 30
Figure 3 Spent catalyst silicon profile
was 5220 lb, which equated to 9%
higher compared to the estimated
number. Results of the spent catalysts also confirmed 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 (<450°F) 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
www.eptq.com
ACS-AMISTCO
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 deposition and could have extended the
cycle length up to 4.5 months
beyond the scheduled 12-month
cycle length.
How can we boost their capacity
without major construction?
TO MEET CLEAN FUELS requirements for gasoline, a medium-sized
Midwestern re¿nery added a lowsulfur 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.
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
On studying the hydrogen, amine, and
sulfur units, ACS-Amistco found many
opportunities for improving separation
ef¿ciency 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.
Read more on this topic at
www.amistco.com
3KRQH)D[
amistco@amistco.com
KU(PHUJHQF\6HUYLFH
Catalysis 2012 33
24/2/12 10:03:13
Reaching our customers globally
through innovation
At Süd-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 finite raw materials and energy are used efficiently.
Discover what the future holds for your business at www.sud-chemie.com
SC Syngas Anz_A4_2012-01-18 f01.indd 1
sud chemie.indd 1
18.01.12 14:39
23/2/12 12:11:02
FCC catalyst coolers in maximum
propylene mode
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
M
arket demand for propylene
has placed a strong emphasis 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 unacceptable
levels,
resulting
in
accelerated catalyst deactivation,
degraded cracking selectivity and a
need for exotic mechanical design to
avoid equipment damage.
This article presents FCC modelling that demonstrates the utility of
continuous heat removal from the
regenerator for maximum propylene
operations. Developments in FCC
catalyst cooling technology have
given refiners a flexible and reliable
option to confront the heat balance
challenges of maximum propylene
FCC operations.
In an unconstrained environment,
increasing FCC propylene production 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, regenerator coke burning and the vapour
recovery unit limit capacity, increasing the reactor temperature without
first 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
www.eptq.com
kbr.indd 1
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
balance1
• FCC operators can effect a reduction 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 propylene production.
History of FCC propylene production
The first commercial FCC unit was
built by The M W Kellogg Company
in Standard Oil of New Jersey’s
Baton Rouge, Louisiana, refinery
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 first 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
Disengager with internal
reactor cyclone system
Primary feed riser
2nd riser for
naphtha recycle
Regenerator
Atomax-2
fresh feed injection
Recycle naphtha injection
Figure 1 Maxofin unit
Catalysis 2012 35
23/2/12 13:15:37
Water in
Tubesheet
Catalyst in
Inner tube
Water and
steam out
Tubesheet
Fluidisation
air
Catalyst
return
Scabbard −
outer tube
Slide
valve
Figure 2 KBR dense-phase catalyst cooler
Water in
Water and
steam out
High heat
transfer coefficient
Catalyst in
Flow-through design,
high mean temperature
differential
No fluidisation impingement
on tubes
KBR dense-phase catalyst cooler
Upflow boiling with natural
boiler feed water circulation
Tube bundle easily removed
Fluidisation
air
High turndown capability
Commercially proven design
Catalyst out
Figure 3 Key features of catalyst cooler
from the FCC units, but the catalyst’s high hydrogen transfer
characteristic greatly reduced the
light olefin 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
olefin yield, primarily with a higher
reactor operating temperature and
riser cracking.2 Increasing reactor
temperatures increased the light
olefin yield, but this came at the
expense of an increased yield of dry
gas — a lower-valued FCC product.
During the 1980s, Mobil introduced two new technologies with
application to increasing the produc-
36 Catalysis 2012
kbr.indd 2
it simultaneously relieved constraints
on both vapour recovery unit capacity and regenerator operating
temperature.
The Maxofin FCC Process introduced by M W Kellogg and Mobil
in 1985 (see Figure 1) is designed to
maximise the production of propylene, ethylene and aromatics from
traditional FCC feedstocks by
combining the effects of FCC
catalyst, ZSM-5 additive and a highseverity second riser designed to
crack surplus naphtha and C4s into
incremental light olefins and
aromatic naphtha.6 Like closed
cyclones, the Maxofin FCC Process
also provides some relief to the heat
balance while operating at high reactor temperatures because of the
limited delta coke from recracking
the recycled naphtha and C4 feedstocks. The recycled naphtha and
C4s essentially act as a regenerator
refrigeration system while simultaneously
serving
to
increase
propylene production in the highseverity second riser.
tion of light olefins and octane while
limiting incremental dry gas production: Mobil developed the ZSM-5
catalyst additive to crack low-octane
(linear) gasoline-boiling-range olefins
and paraffins into light olefins, 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 termination system reduced delta coke,
especially on units that previously
employed low catalyst separation
efficiency riser termination devices.
Therefore, the closed cyclone system
was especially adept at increasing
FCC propylene production because
For many decades and until recently,
FCC catalyst coolers have been
considered only as a means to effectively
process
high-carbonresidue FCC feedstocks, where the
impact of Conradson carbon residue
(CCR) on delta coke is a fundamental 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 lowering of the catalyst-to-oil ratio and
accelerated catalyst deactivation. To
mitigate the impact of increasing
feed CCR on regenerator temperature, 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 regenerator operation in partial CO
combustion mode.
www.eptq.com
23/2/12 13:15:45
- TRICAT
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, 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.indd 1
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
23/2/12 12:15:06
Regenerator temperature, ºF
The KBR dense-phase catalyst
cooler (see Figures 2 and 3) was
commercialised in 1991 based on
extensive KBR experience in hightemperature ammonia applications
and cold flow modelling of the catalyst side at KBR’s Houston
Technology Development Center.7
Two distinguishing features of the
KBR dense-phase catalyst coolers
impart flexibility in heat removal
duty and resistance to tube failure
from erosion by the catalyst. The
first feature is a gas vent line at the
top of the cooler fluid bed that
prevents
catalyst
backmixing
between cooler and regenerator
whenever the cooler catalyst circulation is stopped, thereby providing
complete heat removal turndown
capability. Without the vent, cold
flow modelling has demonstrated
that backmixing between the cooler
and regenerator (and therefore heat
transfer in a commercial unit) will
Feedstock properties
Feed rate, BPSD °API Molecular weight Sulphur, wt% Total nitrogen, ppmw Watson K Conradson Carbon, wt% Distillation type D2887, °F
10% 30% 50% 70% 90% 40 000
28.00
445
0.05
2
12.27
0.20
700
754
804
883
997
Table 1
occur due to fluidisation 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 defluidise whenever catalyst
1420
1400
1380
Use of catalyst cooler for propylene
production
1360
1340
1320
1300
980
990
1000
1010
1020
1030
1040
1050
Riser outlet temperature, ºF
Figure 4 Impact of ROT on regenerator temperature (w/o catalyst cooler)
Cooler duty, mmBTU/hr
90
80
70
60
50
40
30
20
10
0
980
990
1000
1010
1020
1030
1040
1050
Figure 5 Required catalyst cooler duty for 1350°F max regenerator temperature
38 Catalysis 2012
kbr.indd 3
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 important ramifications:
• Downward-hanging tubes allow
the cooler shell fluidisation air to be
introduced well below the tubes,
preventing any possibility of cooler
fluidisation air jet impingement on
the tubes, which could cause an
erosion-related tube failure
• Since steam is generated upflow
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 densephase catalyst coolers in operation,
and there have been no reports of
erosion-related tube failure in any of
these installations.
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 propylene production is demonstrated by
way of a hypothetical example using
KBR’s proprietary FCC yield modelling 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 650°F
(343°C).
Figure 4 shows that, while operating at the base case 980°F (527°C)
riser outlet temperature, the
www.eptq.com
23/2/12 16:24:15
Alternative regenerator temperature control strategies
Alternatives for Catalyst cooler Limited ROT Reduced activity
achieving a 1350ºF 1050°F ROT 1010°F ROT 1050°F ROT
regenerator
& &
&
temperature
72 MAT 72 MAT 58 MAT
FCC yields
Wt%
Vol%
Wt%
Vol%
Wt%
Vol%
C2 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 C4s
15.61
23.35
14.30
21.48
11.64
17.32
FCC gasoline (C5-204°C) 51.03
61.22
52.73
63.35
50.61
60.89
Light cycle oil (204-360°C) 10.48
10.01
13.52
13.01
16.50
15.99
Slurry oil (360°C+)
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 @ 204°C
84.8
86.3
81.1
82.5
76.8
78.1
Table 2
13.50
Propylene production, vol%
regenerator
temperature
heat
balances at about 1310°F (710°C), 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 temperature 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 1350°F (732°C). As
Figure 5 shows, the maximum allowable riser outlet temperature will be
approximately 1010°F (543°C) when
limiting the regenerator temperature
to 1350°F (732°C).
The increasing regenerator temperature is primarily driven by the
increasing temperature of the spent
catalyst being returned to the
regenerator.
Utilising a variable-duty densephase catalyst cooler, it is possible to
keep the regenerator temperature
from exceeding 1350°F (732°C) at
riser outlet temperatures exceeding
the 1010°F (543°C) value. For
purposes of this example, the riser
outlet temperature is increased to as
high as 1050°F (566°C). Figure 5
shows the catalyst cooler duty
required to maintain the regenerator
13.00
12.50
12.00
11.50
11.00
10.50
10.00
9.50
9.00
8.50
980
990
1000
1010
1020
1030
1040
1050
Figure 6 Impact of ROT on propylene
bed temperature of 1350°F (732°C)
at riser outlet temperatures greater
than 1010°F (543°C).
The point of this exercise is shown
in Figure 6, which demonstrates
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% without the use of regenerator heat
HANDLING THE WORLD’S MOST
IMPORTANT PRODUCTS EVERYDAY
CHEP Catalyst and Chemical Containers provides Catalyst-Bins to the petroleum refining,
gas processing and petrochemical manufacturing industries.
Our Catalyst-Bins offer you the following advantages:
- Efficient transportation
- Optimization of storage around reactor
- Efficient handling
- Optimized HSSE
- Reduced reactor down time
- Minimization of waste generation
For more information please visit our website www.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
www.eptq.com
kbr.indd 4
Catalysis 2012 39
23/2/12 16:25:06
TRANSFORMING TODAY’S PRODUCTS
FOR TOMORROW’S INNOVATIONS
Market Leader in Catalyst Services, Products and Tolling
Catalyst Services:
• Regeneration • Reactivation • Presulfiding • Density Grading • Length Grading
• Catalyst Acquisition • Metal Reclaim • Drying • Resale • Quanta
Products:
• Claus Catalysts • Adsorbents • Purification • Dehydration • FCC-additives
• Inert Support Media
Tolling:
• Extrusion • Calcination • Impregnation • Forming • Drying • Milling • Blending
• Classifying
GLOBAL PRESENCE, WORLDWIDE SATISFACTION
porocel.indd 1
Porocel_adv_FC_210x297mm.indd 1
23/2/12 12:16:42
20-02-12 09:26
removal while honouring the 1350°F
(732°C) 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 regenerator temperature.
The lessons learned from the
examples presented here can be
further demonstrated by comparing
some yield cases representing alternative
routes
for
controlling
regenerator temperature in search of
increasing the propylene yield. Table
2 shows two cases taken from the
examples presented above, contrasting
the
difference
between
controlling the regenerator temperature with a catalyst cooler versus
simply limiting the reactor temperature. 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
1050°F (566°C).
Comparing the three cases,
depending on a refiner’s objectives,
different cases may be the preferred
alternative:
• If the only objective is the maximisation of propylene, the first case
with the catalyst cooler is the clear
winner
• If the primary objective is the
maximisation of propylene and the
secondary objective is the maximisation 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 refiner 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.
www.eptq.com
kbr.indd 5
Fundamentals of fluid bed to tube heat transfer
The basic science of heat transfer between a fluidised bed and a confining wall was
described by Chin-Yung Wen and Max Leva in a 1956 paper:8
•For a fluidised bed confined in a pipe, there will always be a laminar fluid film in
immediate contact with the pipe. The heat transfer mechanism from the pipe wall
to the laminar flow is via conduction, and this constitutes the major resistance
between the fluid bed and the pipe wall
• The heat transfer from the laminar film to an adjacent buffer layer is through
turbulent mixing due to the eddy movement of the fluid
•From the boundary layer to the core of the fluid 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 flowing fluid. Therefore, up to a
point, an increase in velocity of the fluid in the bed that decreases the laminar layer
thickness will leads to an increase in heat transfer coefficient
•As the velocity of the fluid 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 flow and increases the solid particle
spacing, reducing the rate of heat transfer via turbulent mixing of the particles.
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 operation. 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 catalyst at 25 ft/s moving through the
tubes, transferring heat to water/
steam in the shell. A catalyst standpipe 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 standpipes and carrier lines
• Erosion of the carrier line at the
air/catalyst mixing point.
The run lengths obtained with
these dilute-phase coolers were typically three to four months, at the
end of which tube bundle repair or
replacement was required. As the
flowing catalyst was in dilute phase
(density ~3 lb/ft3), the heat transfer
coefficient was relatively low. An
analysis of commercial catalyst
cooler data showed the heat transfer
coefficient ranging from 9-35 Btu/hft2-°F, with most of the coefficients
falling between 18 and 34 Btu/h-ft2°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 regenerator bed coils. The first residue
cracker, built in 1960 for Phillips
Petroleum, had 15 layers of horizontal hairpin coils, which covered 320
degrees of the vessel circumference
near the wall of the regenerator. The
high-pressure steam for use in the
refinery was generated by passing
boiler feed water through the coils.
In the original design of these bed
Catalysis 2012 41
23/2/12 13:16:13
Mixtures in vertical transport
Flue gas
Regenerator
Steam
FR
PR
TR
T1
Catalyst
to reactor
LLR
Feed
water
Steam
drum
Recycle
catalyst
cooler
Catalyst
from reactor
T1
PR
TRC
PR
Standpipe
T1
T1
FRC
PR
Air
Figure 7 Arrangement and instrumentation of recycle catalyst cooler
Kellog 11/2 in. tubes, cooling
Kellog 17/8 in. tubes, cooling
Farbar and Morley heating
(averaged ordinates)
40
30
20
10
8
6
/
(hD/k) (ws/wg)0.45
100
80
60
4
3
2
1
2
3 4
×102
6 8 1
2
3 4
×103
6 8 1
2
3 4
×104
6 8
DG/µ
Figure 8 Correlation of data on heat transfer of air-cracking catalyst mixtures in
vertical upward transport
coils, a heat transfer coefficient of 60
Btu/h-ft2-°F was used.7 This proved
to be conservative and the steam
generated was greater than expected.
This high heat transfer coefficient
was the result of the coils immersed
in a relatively high-density (25 to 35
lb/ft3) fluidised bed in the regenerator. Unlike the dilute phase, the low
42 Catalysis 2012
kbr.indd 6
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 lowquality boiler feed water resulted in
a mineral deposition and eventual
failure due to pitting. Also, the low
In a 1963 paper,W J Danzinger presented
the development of a heat transfer
coefficient correlation for a fluidised
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
coefficients of air-fluidised 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 Morley10 for heating,
is:
• (h*d/k) = 0.0784 * ((D*G)/µ)0.68 *
(Wa/Wg)0.45
• The correlation covers Reynolds
numbers from 178 to 25 400, solids-togas 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, removablebundle, fire-tube boilers with the
air-catalyst mixture flowing upward
through the tubes. The steam drum
was elevated sufficiently to limit
vapourisation in the boiler shell to
about 10% of the thermosyphon water
flow. 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 flow 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 coefficient with Reynolds
number.
velocity in the coils resulted in a
slug or stratified flow, ultimately
leading to failure due to thermal
stresses. To avoid this problem, the
flow inside the tube was maintained
in the bubble flow regime, restricting the flexibility of these coolers to
cope with changes in feed rate and
feed quality.
www.eptq.com
23/2/12 13:16:24
The research continued for a catalyst cooler that was more flexible and
operated with a low catalyst velocity
to avoid the erosion, which compromised 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 erosionrelated tube failure.
Conclusions
A refiner should consider investing
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.
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 debottlenecking project.
Once a decision has been taken to
install an FCC catalyst cooler, careful consideration should be given to
aspects of the cooler technology that
impact the cooler’s operating flexibility 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 olefins 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 refinery — FCC’s role in
refinery/petrochemical integration, 2001 NPRA
Meeting, Mar 2001.
www.eptq.com
kbr.indd 7
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 fluidized
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
flowing gas-solid mixtures in a circular tube,
Ind. Eng. Chem., 1957, 49(7), 1143-1150.
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 KBR’s Fluid
Catalytic Cracking (FCC) team in 1989
following nine years of related work for a major
oil company.
Independent Catalyst Test
Reports 2010, 2011 & 2012
OUR REPORT WILL HELP YOU:
t4JNQMJGZDBUBMZTUTFMFDUJPO
t$IPPTFUIFSJHIU
DBUBMZTUXJUI
DPOĕEFODF
t4BWFIVOESFETPG
UIPVTBOETPO
ZPVSOFYUDBUBMZTU
QVSDIBTF
We have tested catalysts from all major suppliers. Our
reports provide full test data, analysis, and activity rankings. Our reports are available for immediate delivery.
Our team brings the industry’s 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
Catalysis 2012 43
23/2/12 13:16:34
Aim for More Diesel, Better Diesel
ExxonMobil’s 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:
t -PXQSFTTVSFýYFECFEQSPDFTT—JNQSPWFTQSPEVDU
ZJFMEBOERVBMJUZ
t *ODSFBTFETVMGVSSFNPWBMBOESFEVDFEOBQIUIBZJFMET
t 1SPQSJFUBSZDBUBMZTUJTPNFSJ[FTXBYZOQBSBGýOTJO
HBTPJMUPMFSBOUUP)24BNNPOJBBOEPSHBOJD4BOE/
t 4UBOEBMPOFPSJOUFHSBUFXJUIPUIFST
&YYPO.PCJM3FTFBSDIBOE&OHJOFFSJOH
$PNQBOZT.*%85FDIOPMPHZ
*NQSPWFT1SPEVDU:JFMEBOE2VBMJUZ
* MIDW is a trademark of ExxonMobil Research and Engineering Company
** Available from ExxonMobil Catalyst Technologies LLC
3225 Gallows Road, Fairfax, Virginia 22037-0001, USA
www.exxonmobil.com/refiningtechnologies • +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
Use of regenerated catalyst can be maximised by careful sampling and analysis of
spent catalyst to construct a reactor contamination profile
PIERRE DUFRESNE Eurecat
FRANCOIS LOCATELLI Eurecat France
O
il refining and petrochemical plants are currently
undergoing serious challenges 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 sulphurfree clean fuels and the need for
deeper heavy ends conversion. At
the same time, low refining margins
are driving industry to cost optimisation,
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 refineries 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 distillates 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 hydroprocessing 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 specifications. In
www.eptq.com
eurecat.indd 1
this last case, activity decrease is
mainly due to partial blockage of
the active sites by coke and sometimes 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 oxidising atmosphere at a temperature of
450-550°C, carbonaceous species
can be removed from the catalyst.
Regeneration has the benefit of
eliminating the first cause of deactivation: 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.
regeneration, the sulphides are
converted back to oxides, also
promoting some alumina sulphatation. 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).
How are hydroprocessing catalysts
contaminated?
Na
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
refining operations (silicon, lead
and phosphorous) or from corrosion (iron).
Ni + V
The nickel- and vanadium-containing 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
As
Organoarsenic compounds contained
in some crudes are extremely reactive
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.
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 introduced to the unit via heat exchanger
leakage. Sodium then adsorbs
on the catalyst surface and has a
significant
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%.
Catalysis 2012 45
23/2/12 13:18:42
4J
/B
"T
Case study 1 : SAS saved around 75%
of the catalyst batch (equivalent to
€1 million savings as fresh catalyst)
wt %
Bin number
Figure 1 SAS service. Contaminants analysis (sodium, arsenic, silicon) vs bin number on a
load of spent HDS catalyst
Si
Silicon originates essentially from
lighter fractions of coker or
visbreaker operations, where polydiméthylsiloxanes 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 tetraethyl 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 fine 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 interstices between catalyst granules.
How to maximise the quantity
of reusable regenerated catalyst.
Sorting is a solution.
Regeneration, however, will not
46 Catalysis 2012
eurecat.indd 2
used in another unit of similar service, cascaded down to a less severe
service, or sent for metals
reclamation.
cure anything as regards deactivation by contaminants, which remain
adsorbed after thermal treatment.
This does not mean that a contaminated batch cannot be regenerated.
What is necessary is to profit from
the concentration gradient that is
always observed in fixed-bed reactors 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 gradients throughout the reactor bed and
is thus not sufficient. 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 shutdown. 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 interpretation of the analysis of each
container by drawing the contaminant profile and then deciding on
the subsequent segregation of good
product from contaminated material. 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,
The following is an interesting
example of Eurecat’s SAS service,
which resulted in considerable
savings for the refiner. 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 catalyst. With these results in hand, the
customer’s 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 contamination profile. Figure 1 shows the
contamination profile, with a rather
clear cut between clean and contaminated fractions.
The high level of contaminants in
the first containers illustrates the
so-called “chimney effect”, where
top layers come first 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 contaminants 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 contamination analysis coupled with HDS
testing made the customer’s decision 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-
www.eptq.com
23/2/12 13:19:06
refining your profit
UOP helps you exceed your goals with innovative technology,
catalysts and optimization solutions specifically 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 specifications including
ultra-low sulfur diesel standards while improving your operational efficiency.
Backed by over 50 years of hydroprocessing innovations, UOP offers the best and most advanced solutions
to keep your business one step ahead.
For more information about UOP, visit www.uop.com
©2010 UOP. All Rights Reserved.
UOP_Refining Your Profit_Ad_A4_PRINT.indd 1
uop.indd 1
5/18/10 2:10:44 AM
23/2/12 12:23:31
ad copy 2.indt 1
23/2/12 16:19:20
0.7
Some catalyst can be saved even in
the event of highly polluted feed
Another example illustrates why
the SAS procedure is always justified. 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 reactivity of arsenic species contained
in crude are so high that, even in
the event of highly polluted feed,
adsorption takes place in the first
www.eptq.com
eurecat.indd 3
Fe
As
Si
Na
Regeneration
0.5
0.4
0.3
0.2
0.1
0
0
10
20
30
40
50
60
Bin number
Figure 2 Contaminants analysis (Fe, As, Si, Na) vs bin number on a load of spent HDS
catalyst
Spent catalyst analysis can assess
average liquid feed quality over
the run
"T
"T
"S TF O J DX U The following example demonstrates that the SAS service is useful
for understanding the average feed
quality with regards to contaminants over the run. A simple way
of assessing average feed contaminant 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 velocity of one will result in 1 wt%
deposit of contaminant.
Figure 3 shows the arsenic
concentration profile 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.
Recycling
0.6
wt%
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 contamination profile.
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.
Figure 3 Arsenic profile of an HDS reactor for two consecutive cycles
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
profile within the reactor
Unit parameters, such as temperature, 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
profile of the reactor, which then
can be useful for helping to define
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 specific case.
In the same example, Figure 4
shows the coke profile: 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 catalyst 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
Catalysis 2012 49
23/2/12 13:19:20
12
11
10
9
8
7
6
5
4
3
2
1
0
Arsenic
3.5
Coke
Element, wt%
3.0
2.5
2.0
1.5
Iron
1.0
Silicon
0.5
80
10
0
12
0
14
0
16
0
18
0
20
0
22
0
24
0
26
0
28
0
30
0
60
0
20
0.0
40
Sodium
increase from top to bottom (6 to 14
wt%)
Carbon, wt%
4.0
Bin number
Figure 4 Analysis (As, Fe, Si, Na and C) vs bin number on a highly contaminated spent
catalyst: partial recovery is possible
E l e me n t , wt %
$PLF
7BOBEJVN
4PEJVN
4JMJDPO
"STFOJD
Bin number
Figure 5 Analysis (V, Na, As, Si and C) vs bin number on spent FCC pretreat catalyst:
partial recovery is not possible
105
RWA, %
100
95
90
85
80
130
140
150
160
170
180
190
200
210
SA, m2/g
Figure 6 Variation of Surface Area (SA) and HDS Activity (RWA) for various regenerated
batches of a new generation CoMo catalyst.
the bottom. Organic compounds
of those two elements are much
less
reactive
than
arsenic
compounds and diffuse more easily
50 Catalysis 2012
eurecat.indd 4
into the entire bed. In this particular case, the whole catalyst lot had
to be sent for metal reclaim. The
carbon profile showed a strong
Activity testing is necessary for
assessing catalyst reuse
Until a couple of years ago, contaminant and surface area analysis were
sufficient 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 recovery of 90%. This corresponds to a
delta start-of-run temperature of
around 3°C.
However, the situation has since
changed drastically with the
progressive replacement of conventional 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 various new-generation CoMo catalysts.
Some points are clearly outside
direct correlation. Using the criterion 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 testing is absolutely necessary for
assessing the safe reuse of regenerated 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 mechanical properties. The first of these is
assessed by contamination studies
followed by sorting, as previously
www.eptq.com
23/2/12 13:19:31
Gas phase pressure drop,
Pa/m
60000
www.eptq.com
eurecat.indd 5
40000
30000
20000
10000
0
1
2
3
4
5
6
7
Airflow rate, Nm3/hr
Figure 7a Effect of fines of less than 1 mm up to 2, 5 and 10 wt% on pressure drop
generation
10
9
8
7
6
5
4
3
2
1
Recycling spent non-reusable
catalyst
8
9.
2
9.
6-
6
9.
9.
0-
0
8.
8.
4-
4
8.
7.
7.
8-
8
7.
2-
2
6.
6.
6-
8
6.
6.
0-
0
5.
5.
5.
4-
4
4.
8-
8
4.
4.
2-
2
3.
3.
6-
8
3.
2.
3.
0-
0
42.
4
2.
1.
8-
2-
1.
8
0.
1.
6-
0-
0.
2
0
0.
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 fluctuate a lot and
are now high enough to give spent
catalysts a positive value. Thus, the
debate between landfilling 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: hydrometallurgy and pyrometallurgy. Eurecat
has developed a pyrometallurgical
route with its partner Valdi, now
part of Eramet. This method has
several advantages over competitive processes: the recycling is
performed not far from the regeneration plant in France, thus
limiting transport and guaranteeing
Base case
Base case + 2% fines
50000
0
0.
discussed, and activity testing is
performed on the segregated clean
catalyst component. With this in
place, catalyst activity is well
defined 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 particle size distribution, attrition level
and content of various sizes of
fines. 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 implemented such equipment in order to
better assess catalyst reusability.
Figure 7 shows how fines content
influences the pressure drop of a
catalyst batch. Catalyst batches
have been selected with fines (<1
mm) contents of 2, 5 and 10 wt%.
Pressure drop increases by about
60% and 130% respectively for a
fines contents of 5 and 10 wt%.
Length (mm)
Figure 7b Particle size distribution of a catalyst containing 10% fines below 1 mm
that the recycling operation meets
the highest environmental standards. In addition, this process
brings value to all of the catalyst
components: molybdenum, cobalt
and nickel are used to produce
speciality steels, while the recovered alumina is used in the
production of rock wool.
Conclusion
Sound regeneration practice can
greatly contribute to an overall
reduction in hydroprocessing catalyst costs. Careful analysis of spent
catalyst makes it possible to cherrypick 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 Français
du Pétrole dealing with hydroprocessing and
zeolites before joining Eurecat. He has coauthored around 70 patents.
Email: p.dufresne@eurecat.fr
François 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
Catalysis 2012 51
23/2/12 16:26:58
Experience
Experience
the
Sabin difference
thePGM
Sabin
difference
for
catalyst
recovery
for PGM
catalyst recovery
and
refining.
and refining.
We turn science into art for highest
possible
returns
and
value.
We turn science
into
artadded
for highest
Learn more atvalue.
sabinmetal.com
possible returns and added
The “science” of recovering and refining precious metal
catalysts is straightforward: state of the art technology. The
“art” of this process, however, is what makes Sabin different
The “science” of recovering and refining precious metal
from all others: that’s the knowledge, experience, and expertise
catalysts is straightforward: state of the art technology. The
gained from seven decades of successfully serving thousands of
“art” of this process, however, is what makes Sabin different
organizations around the world. We’d be pleased to count you
from all others: that’s the knowledge, experience, and expertise
among them.
gained from seven decades of successfully serving thousands of
organizations around the world. We’d be pleased to count you
among them.
Learn more at sabinmetal.com
Sabin ExpTheDif.indd 2
1/15/12 11:12 AM
Sabin ExpTheDif.indd 2
1/15/12 11:12 AM
sabin.indd 1
23/2/12 12:26:26
Optimisation of integrated aromatic
complexes
A novel class of purification catalysts provides improved selectivity in aromatic
complex service
Axel Düker
Süd-Chemie AG
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 olefins 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 confirms
superior performance compared to
standard, granular-activated clay
products.
The general flow scheme of an
aromatic complex is shown in
Figure 1. To simplify the overview,
only three major sections are
shown: the treatment of the reformate bottom; the purification of the
benzene/toluene stream; and the
fine purification of the p-xylene
rerun. All three services are different with regard to catalyst activity,
which is important for the reduction in the amount of olefins, and
catalyst selectivity, which is critical
for the purification of the p-xylene
rerun.
Before having a closer look at the
services, an understanding of the
mechanism and the specific differences of available catalytic systems
is needed. The conversion of naphtha into BTX streams in the reformer
is always accompanied by the
formation of non-aromatic olefins
(see Figure 2).
These olefins are either poisons to
downstream units, such as p-xylene
extraction units or benzene alkylation units, or they make the
produced BTX streams off spec and
lower their value. The removal
www.eptq.com
sud chemie.indd 1
Crude
reformate
Raffinate
BT
extraction
BT
treatment
BT
p-Xylene
Light
ends
p-Xylene
extraction
Xylene
isomerisation
Reformate
bottom
p-Xylene
rerun
C9+
Figure 1 Schematic overview of an aromatic complex
of olefins 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 downstream distillation unit. Granular
Benzene
Toluene
Mixed
xylenes
H2
Impurities:
Aliphatic olefins
Figure 2 Formation of olefins in the
catalytic reformer
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 olefins
and benzene or toluene needs
primarily Lewis acids and little
Brønsted acids. An excess of
Brønsted acids catalyses the polymerisation reaction of olefins into
gum, which blocks the catalyst
pores and reduces its effective operating lifetime (see Figure 3).
Depending on the operating
temperature, the catalysts can form
both species of acids. At temperatures below 140°C, mainly Brønsted
acidity prevails, whereas at temperatures above 165°C the desired
Catalysis 2012 53
23/2/12 13:51:07
Alkylation
Catalysed by
Lewis acids
+
Oligomerisation
Catalysed by
Brønstedt acids
2
Figure 3 Alkylation versus oligomerisation — equations
Alkylation
Polymerisation
40
60
80
100
120
140
160
180
200
220
Temperature, ºC
Figure 4 Alkylation versus oligomerisation as a function of temperature
Characteristics of services for Tonsil CO + APT products
Type of service
BI of feed
BI of product
Tonsil CO product
Tonsil APT product
Reformate bottom
High activity
500-1200
<20
Tonsil CO 616GS
Tonsil APT mX
BT unit
Medium activity
70-200
<10
Tonsil CO 616G
Tonsil APT BT
p-Xylene rerun
High selectivity
20-50
<5
Tonsil CO 610G
Tonsil APT pX
Table 1
Lewis acidity is active. During
normal operation at temperatures
well above 170°C, the reaction
mechanism favours alkylation catalysed by Lewis acids (see Figure 4).
During startup conditions, especially during the dry-out phase, the
operating temperature remains
below 140°C and the undesired
polymerisation reaction occurs. This
can only be avoided by increasing
the operating temperature during
the dry-out phase as fast as possible. 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
54 Catalysis 2012
sud chemie.indd 2
of non-aromatic olefins and the
necessity of providing high selectivity. The olefin concentration is
measured as bromine index (BI);
the selectivity is mainly determined
by the need to minimise the formation of benzene by the dealkylation
of toluene and xylene. This is
particularly crucial in the fine purification of p-xylene rerun. Table 1
summarises the different features
and categorises the three services
from high-activity service to highselectivity service.
Table 1 lists the Tonsil CO products that are the most suitable for
the preferred services, with Tonsil
CO 616GS the most active product
and Tonsil CO 610G the most selective one.
The cost blocks
may vary with plant
location, but the
message is clear: any
catalyst replacement
is accompanied by
high costs
Standard catalytic clay products
are granules of 0.5 mm on average,
as are almost all competitive products. 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 pressure drop over the reactors is not
negligible.
Tonsil APT are new types of catalysts 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 comparison to the standard products. For
easier reference, the designations of
the products reflect the major service they are designed for: BT for
benzene/toluene; mX for mixed
xylenes; and pX for para-xylene.
The use of Tonsil APT products is
beneficial with regard to loading
pattern, performance and cycle life.
A typical loading pattern of a reactor loaded with Tonsil CO
compared to one loaded with Tonsil
APT is shown in Figure 6. The most
striking difference is the replacement 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 products can be demonstrated by the
example of the treatment of reformate bottom cuts. This service is
the most demanding (see Table 1)
and the catalyst lifetime is normally
the shortest of all services.
www.eptq.com
23/2/12 13:51:19
Reactor Turnkey Services
HPA offers all reactor turnkey services from
blinds to blinds using state of the art equipment.
We specialize in inert-entry using the latest life
support units. All our catalyst technicians are
trained in vessel rescue, first aid, and CPR.
Hydropac Catalyst Dense Loading
The Hydropac allows the sprinkling of catalyst
in a continually uniform pattern at a rate slow
enough to let each particle settle, but fast
enough for acceptable loading time. The
Hydropac sits just six inches below the trays,
is able to rotate both directions, and can
load around transfer tubes and other internal
obstructions.
Catalyst Unloading Services
Our catalyst unloading services make use of
powerful vacuum units with 26 inches of vacuum.
Vessel Repairs and Retro-Fits
We have our U and R stamp allowing us to do
complete vessel welding, repairs, and inspection.
Hydroprocessing Associates is ISO 9001 and
OHSAS 18001 accredited, and ISNetworld and
PICS compliant.
Hydroprocessing Associates is located in the
U.S.A. and Singapore, and can mobilize to any
country.
Phone: +1 832-794-7942
E-mail: peter@hpa-usa.com
www.hpa.sg
BDGBK\GURFDUERQZRUOGBLQGG
hpa.indd 1
www.hpa-usa.com
23/2/12 12:28:06
Crude
reformate
Raffinate
BT
extraction
Tonsil APT
BT
BT
p-Xylene
Light
ends
p-Xylene
extraction
Xylene
isomerisation
Tonsil APT
mX
Tonsil APT
pX
C9+
Figure 5 Tonsil APT catalysts in aromatic plant service
Relative costs
Manpower
Loss of product
Figure 6 Advantageous reactor loading
with Tonsil APT
Disposal
Utilities
(nitrogen/steam
Figure 7 Typical cost breakdown of the
catalyst change
Comparison of Tonsil CO 616 GS and Tonsil APT mX
Tonsil CO 616 GS
Relative catalyst activity, ton feed/ton Tonsil
100
Bulk density, kg/m3
720
Relative catalyst activity, ton feed/m3 Tonsil
720
Loaded volume, m3
100
Catalyst lifetime, months
4
Catalyst consumption, batches per year
3
Catalyst consumption, ton per year
288
Tonsil APT mX Variance
115
+15%
750
+4%
863
+20%
130
+30%
6.3
+60%
2
195
-30%
Table 2
It is therefore of utmost importance 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
56 Catalysis 2012
sud chemie.indd 3
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 catalyst 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
significantly 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 catalyst. 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 approximately 15% more active acidic sites
compared to Tonsil CO 616 GS,
which means it can process approximately 15% more feed on a weight
basis. This is a breakthrough in the
technology of acid-activated clay
products, as Süd-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 extrusions is a significant extension of
www.eptq.com
23/2/12 13:51:33
$IBOHFPVU
$IBOHFPVU
$IBOHFPVU
$BUBMZTU
Be n ze n e make , ppm wt
3F M B U J WF D PT U T
$BUBMZTU
$BUBMZTU
$IBOHFPVU
$IBOHFPVU
$BUBMZTU
$BUBMZTU
5POTJM$0(4
CBUDIFTQFSZFBS
1SPEVDU"
1SPEVDU"
5POTJM"15Q9
5POTJM"15N9
CBUDIFTQFSZFBS
1SPEVDU"
Time on-stream
Figure 8 Cost comparison of Tonsil CO 616GS and Tonsil APT mX
Figure 9 Benzene make over Tonsil APT pX
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 influence 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 defined specification. 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 170°C. 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 operator 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 catalyst, 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 fine purification of p-xylene
rerun streams is defined by the formation of benzene.
Benzene needs to be avoided, as it reduces the efficiency 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
specification within a very short time. Figure 9 shows
the evolution of the benzene concentrations downstream different catalytic clay products.
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.
www.eptq.com
sud chemie.indd 4
Tonsil is a registered trademark of Süd-Chemie.
Axel Düker is Director of Sales EMEA/Refinery, Süd-Chemie AG,
Munich. He has over 20 years’ experience in catalysis and holds a PhD
in chemistry from the University of Munich.
570002.indd 1
4.-0.0#1130#
.0-!#11,*78#0
-).)6!0/NLINE
¾5/NEF@RNKHMDBQTCDNHK@MC+/&
¾ 23,#
$-(/
¾'HFGDRS@BBTQ@BXENQADRS
ONRRHAKDAKDMCHMFSNNEEHBH@KKHLHSR
¾4OSNR@LOKDRSQD@LR
¾ TSNL@SHBB@KHAQ@SHNM
¾%@RS@MCD@RXL@HMSDM@MBD
¾ TSNL@SHBKTAQHB@SHNM -$6Ü
&Q@AMDQ(MRSQTLDMSR,DRRSDBGMHJ&LA' 5HDMM@ TRSQH@#Q.SSN-DTQ@SG&@RRD
/GNMD%@WHMENFQ@AMDQHMRSQTLDMSR@LDSDJ@S
12/7/2010 10:30:28 AM
Catalysis 2012 57
23/2/12 13:51:47
14 - 16 May 2012
RE
NO GIS
SA W TER
VE AN
£3 D
80
COLOGNE, GERMANY
www.wraconferences.com/globalpetrochemicals
GLOBAL
PETROCHEMICALS
CONFERENCE
9 Annual Meeting
th
STRATEGIES TO ACHIEVE PROFIT OPTIMISATION IN A VOLATILE MARKET
WORKSHOP
COCKTAIL
RECEPTION
14 May,
Radisson Blu Hotel
The 9th 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 Officer, CEPSA Quimica
Andreas Gupper, Strategic Planning Manager - Strategy & Group Development,
Borealis
Fred du Plessis, Executive Advisor, Corporate Strategy & Planning, Sabic
NETWORKING
LUNCH
15 May,
Radisson Blu Hotel
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
GPC GALA
DRINKS
15 May,
Radisson Blu Hotel
Sponsored by
wra.indd 1
To confirm 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.
Supported by
23/2/12 12:29:50
Troubleshooting a FCC unit
The source of a fouling problem in a FCC CO boiler was identified 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
F
luid catalytic cracking (FCC) is
one of the major processes in
the refining 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 unit’s operation, 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 reactor 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 flue gas
(containing CO and CO2) holds an
appreciable amount of energy. In
most units, the flue gas is routed
through a steam generating boiler
(referred to as a CO boiler), where
the carbon monoxide in the flue gas
is burned as fuel to provide steam
for use in the refinery as well as to
comply with any applicable environmental regulatory limits on carbon
monoxide emissions.
FCC units generally experience
catalyst-related problems such as
circulation, catalyst loss and activity
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 fines
were found in the CO boiler stack
during cleaning. Subsequent to startup 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 fines deposition in
the CO boiler stack section, as well
as catalyst activity loss, was investigated systematically by analysing
particle size, attrition index, surface
area and chemical composition in
terms of metals, to arrive at a possible solution to enable smooth
operation of the plant.
Samples of e-cat, fresh catalysts
and CO boiler samples from FCC
units were obtained from refineries,
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 concentration 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 D451397 method using a sieving
procedure. Typically, about 50g of
FCC catalyst sample was sieved
through a test sieve column comprising 180, 150, 106, 90, 75, 63, 53, 45,
32, and <20µm sieves for 30 minutes
using an auto-sieve shaker. After
sieving, fractions collected over each
sieve were weighed and PSD calculated as per the ASTM method.
Average particle size decreased
from 72 microns to 67 microns. This
drop corresponds to an increased
amount of fines in the catalyst. In
order to substantiate this further,
e-cat samples collected at different
time intervals were also characterised for particle size (see Figures 1
and 2).
Particle size distribution of fresh catalysts and additives (ASTM D4513)
Sample code, wt%
Fresh Cat-1
Fresh Cat-2
Fresh Cat-3
Fresh Cat-4
Fresh Cat-5
Fresh Add-1
-180
99
100
99
100
99
98
-150
98
98
98
99
99
97
-106
83
87
88
90
90
81
-90
69
75
77
75
77
53
-75
34
61
61
58
62
33
-63
37
44
41
44
44
19
-53
25
31
24
25
28
12
-45
12
20
15
16
19
6
-32
5
10
3
3.5
5.7
1.2
-20
1
3
0.1
0.27
0.6
0.1
(APS) µ
72
69
69
71
67
88
Table 1
www.eptq.com
bpcl.indd 1
Catalysis 2012 59
23/2/12 16:28:11
The Global Refining Summit 2012 the must attend conference for anyone involved in
the refining and processing of hydrocarbons. The event may also be of interest to those
involved in the health, safety and environment aspects of the refining sector.
In Association With:
21st - 23rd May 2012 - Hesperia Tower, Barcelona, Spain
www.refiningsummit.com
“Great opportunity to meet and learn from other refining
people who share their views and approaches to current
and future challenges”
Head of Technical Services, Petrobrazi Refinery, OMV Petrom
“Fruitful opportunity to exchange views and learn new
ideas for optimising refinery operations.”
Senior Expert, Refining, Organisation of Arab Petroleum
Exporting Countries (OAPEC)
Leading experts include:
B.R. Mehta, Senior Vice President,3FMJBODF*OEVTUSJFT-JNJUFE
Paulo Roberto Costa,.FNCFSPGUIF&YFDVUJWF#PBSEBOE
%PXOTUSFBN0GàDFS1FUSPCSBT#SB[JM
The 2012 Summit will look at:
r )PXUPBDIJFWFPQFSBUJPOBMFYDFMMFODF
Philippe Marchand,%JSFDUPSPG#JPGVFMTBOE3FHVMBUJPOTGPSUIF
3FàOJOHBOE.BSLFUJOH5PUBM
r #FTUQSBDUJDFTGPSCPUUPNPGUIFCBSSFM
Frigyes Lestak,(FOFSBM.BOBHFS$0&GàDJFODZ4IFMM
r &YDFMMJOHDMFBOGVFMQSPEVDUJPO
r *NQSPWJOHFOFSHZFGàDJFODZJOSFàOFSZ
r .BYJNJTJOHEJFTFMQSPEVDUJPO
Pierre Dechamps,"EWJTFSGPS&OFSHZ$MJNBUF$IBOHFBOEUIF
&OWJSPONFOU&VSPQFBO$PNNJTTJPO
His Excellency Abbas Ali Al-Naqi,4FDSFUBSZ(FOFSBM0"1&$
,VXBJU
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
global ref.indd 1
Researched and produced by:
23/2/12 12:32:32
Attrition index data (ASTM D5757)
74
Attrition loss, wt%
5.0
6.9
9.3
10.0
4.7
5.7
7.1
70
APS
05
05
20
b
27
8
Fe
ov
Ja
n
20
20
04
04
N
19
11
30
Se
Au
g
p
20
20
04
04
20
Ju
n
22
3
14
M
M
ar
ay
20
20
04
04
50
Figure 1 APS variation with respect to time
Surface area data for fresh catalyst, additive, e-cat and CO boiler samples
S.No
1
3
5
7
8
9
Catalyst
Fresh Cat-1
Fresh Add-1
E-cat-4
CO boiler fines surface area
Additives fresh fines, heated to 1000°C
Fresh Cat-3, attrited fines, heated to 1000°C
S.A, m2/g
349
100
204
9
30
40
Table 3
fines, the surface areas of fresh FCC
catalyst fines (<40 micron), additive,
attrited fresh catalyst and CO boiler
fines were analysed. The surface
areas of e-cat, fresh catalyst and fresh
additive fines were comparable with
those of the original sample. Fresh
additive fines and attrited fresh catalyst were heated to 1000°C, and the
measured surface area was found to
be in the range 30-40 m2/g. A sample
from the CO boiler was found to
have a surface area of 9 m2/g. The
very low surface area of CO boiler
fines was due to their exposure to
very high temperatures (>950°C),
which could have resulted in pore
sintering and formation of nonporous material; hence, it could not
be attributed to either additive or
28
−45 fraction
24
20
16
12
8
4
20
05
27
Fe
b
20
n
Ja
8
19
N
ov
20
05
04
04
p
30
Se
g
11
Au
n
Ju
22
20
04
20
20
04
20
04
M
ay
ar
20
04
0
14
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-cat2 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.
54
M
Attrition index
62
58
Table 2
The average particle size distribution 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 addition
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 fines in the unit.
The ‘-45’ fraction also fluctuated in
average particle size after the introduction
of
additive,
before
stabilising. During this period, a lot
of fluctuation 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.
66
3
Catalyst
Fresh Cat-1
Fresh Cat-3
Fresh Cat-6
Fresh Add-1
E-cat-1
E-cat-2
E-cat-3
78
Surface area measurements
To further understand the source of
www.eptq.com
bpcl.indd 2
Figure 2 Variation in ‘-45’ fraction with respect to time
Catalysis 2012 61
23/2/12 14:10:09
0.8
Mass balance
Actual
0.7
Ma, %
0.6
0.5
0.4
0.3
0.2
0
30
60
90
120
Days
Figure 3 Comparison of actual and calculated Mg content in e-cat
Microwave system parameters for digestion
Stage
Power
Max
%
1200 W
100
I
RAMP minuets
Control pressure, psi
Hold time, min
30.00
120
30.00
Table 4
FCC catalyst fines. Surface area measurements for catalyst samples were
carried out using nitrogen adsorption/desorption measurements in an
Autosorb-1MP unit. Nitrogen adsorption/desorption isotherms were
measured at -196°C after degassing
about 50 mg of sample below 10-3
torr at 300°C for three hours. BET
specific 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 fines by measuring 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 fines. A microwave digestion 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 first vigorous
reaction has taken place, the pres-
80
70
Activity
60
50
40
30
20
10
0
Dec
May
Aug
Figure 4 Activity variation with respect to time
62 Catalysis 2012
bpcl.indd 3
Dec
May
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 onestage 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 ecat. 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 fines with additive
may be contributing to this lower
retention. Hence, it may be necessary to improve the attrition
resistance of the additive.
Based on magnesium analysis of
CO boiler fines alone, the problem
cannot be attributed solely to either
catalyst or additive, as the fines
could have accumulated over a
period of time and the addition of
additive may have started meanwhile. Magnesium and rare earth
oxide distribution in different-sized
fractions of the e-cat was analysed.
Magnesium and RE2O3 distribution
with different-sized fractions of ecat 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 fines leads to a
decrease in activity. The effect is
shown in Figure 4.
Systematic analysis of fresh FCC
catalyst, equilibrium catalysts and
fines collected from the CO boiler
at different time intervals, and analysis of fresh additive samples, was
employed to identify problems and
enable corrective actions to be
www.eptq.com
23/2/12 14:11:06
Mg and RE2O3 analysis of different fractions of e-cat
E-cat -2
32 to 45 µ
45 to 63 µ
63 to 90 µ
>90 µ
Mg, wt%
0.5850
0.4855
0.4384
0.2780
RE2O3, wt%
1.62
1.55
1.45
1.02
Table 5
recommended for the smooth operation of the FCC
unit.
Conclusions
Systematic studies were conducted on different refinery
samples to troubleshoot operational problems in the
FCC unit. The studies led to the following conclusions:
• The CO boiler fines problem confirmed that in-use
fresh catalysts and additive are more prone to attrition,
so they contributed more to the fines loss from the unit
and ultimately to the fouling of the boiler. More fines
loss also affected the conversion loss, because fines have
more zeolite and additive content
• Even though the attrition index of additive is on the
high side, which may contribute to the total fines, the
amount of additive used is relatively small compared to
the catalyst; hence, the additive alone cannot contribute
to the total fines generated. This was established by
correlating the magnesium content of CO boiler fines
with additive
• Since both fresh catalyst and additive can contribute
www.eptq.com
bpcl.indd 4
to fines, 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 fines helped to diagnose
the CO boiler fines and fouling problem, and enabled
suitable remedial measures to be suggested to
operations.
Acknowledgement
The authors gratefully acknowledge BPCL’s management
encouragement and permission to publish the paper.
for
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 master’s 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 refining 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.
Catalysis 2012 63
23/2/12 14:10:44
Alphabetical list of advertisers
ACS-Amistco www.amistco.com
Axens
www.axens.net
11
Buchen-ICS
www.buchen-ics.com
63
CB&I
www.cbi.com
Chep Catalyst & Chemical Containers
www.chepccc.eu
Grabner Instruments
www.grabner-instruments.com
Grace Davison
www.grace.com
57
4
Haldor Topsøe
www.topsoe.com
IFC
Hoekstra Trading www.hoekstratrading.com
43
Hydroprocessing Associates
www.hpa.sg
55
Johnson Matthey Catalysts
www.jmcatalysts.com
13
Porocel
www.porocel.com
40
25 & 27
9
39
Chevron Lummus Global
www.clg-catalysts.com
7
Criterion Catalyst & Technologies
www.criterioncatalysts.com
2
Process Consulting Services
www.revamps.com
19 & 21
52
IBC
Sabin Metal Corporation
www.sabinmetal.com
29
DuPont Sustainable Solutions
www.cleantechnologies.dupont.com
16
Spectro Analytical Instruments
www.spectro.com
Eurecat France
www.eurecat.fr
22
Süd-Chemie, a Clariant Group Company
www.sud-chemie.com
34
Euro-Petroleum
www.europetro.com
48
Tricat
www.tricatgroup.de
37
ExxonMobil Research and Engineering Company 44
www.exxonmobil.com/refiningtechnologies
UOP
www.uop.com
47
Global Petrochemicals 2012
www.wraconferences.com
Vereinigte Füllkörper-Fabriken
www.vff.com
15
Digital Refeining
www.digitalrefining.com
64
OBC
BASF Corporation, Catalysts Division
www.catalysts.basf.com
Cat-Tech
www.cat-tech.com
om
30 & 33
Catalysis 2012
ad index copy 8.indd 1
58
www.eptq.com
24/2/12 13:44:48
◗ www.digitalrefining.com is a new,
easy-to-use website for the refining, gas
and petrochemical processing industries.
◗ Readers can download technical articles
and other information from over 40
categories.
◗ A simple search facility enables you to
find and display relevant information
across the whole spectrum of articles,
literature, news, events and companies.
◗ Find relevant industry events worldwide.
dig ref.indd 1
◗ Read news items focusing on contracts,
company news and new products.
24/2/12 13:30:07
The winning catalyst combination
for your hydrocracker
)3,)%,):,TFSJFTUIFFYQFSUUSJPUPNBYJNJ[F
DZDMFMFOHUIBDUJWJUZBOENJEEMFEJTUJMMBUFTTFMFDUJWJUZ
4JOHMFTPVSDF*40t*40t0)4"4
www.axens.net
Hydro_cats_ad_fin.indd 1
axens.indd 1
26/7/11 12:35:06
23/2/12 12:33:46

catalysis - DigitalRefining

Transcription

Similar documents

Case Study - BJ`s Wholesale Club Oaks, PA

Leaflet Catalyst Recycling

5 Reasons You Should Attend the Economic Forum

This is a - MacDonald

nissan nissan - Catalyst Outdoor

Maxcat

Dear sirs,

Volunteerism - Catalyst Life Services

Labor is not a dying cause Boat`s activism hits home tn Sarasota

radnor radnor