Anda di halaman 1dari 60

catalysis

2010

ptq

cover and spine copy 3.indd 1 2/2/10 13:43:15


Looking for
Clean Fuel Solutions?
W W W. T O P S O E . C O M

Check out Topsøe Catalysts and Technologies


Hydroprocessing Hydrogen
Major supplier of clean fuels catalyst and licensed tech- Leading supplier of hydrogen plant catalysts and the
nology. Low pressure, low cost revamps for ULSD. most efficient and feed flexible technology.

WSA Selective Catalytic Reduction (SCR)


Cost effective and environmentally friendly method for For FCC and all other flue gas streams. Superior
turning sulphur containing waste streams and spent NOx removal with low SO2 oxidation. High poison
alkylation acid into clean concentrated sulphuric acid. tolerance and low pressure drop.

www.ptqenquiry.com
for further information

haldor.indd 1 29/1/10 12:08:55


catalysis Security of
ptq
feedstock supply
2010 Vol 13 No 2 3 Metals supply may impede catalysis growth
www.eptq.com
2008 Chris Cunningham

D
5 ptq Catalysis Roundtable
Editor espite signs in 2007 of a slowdown in various sectors of the economy,
René G Gonzalez refiners remain a big play for prospective investors. It used to be
editor@petroleumtechnology.com
11 conventional
Maximising wisdom fuels from thatthe higherFCCfuel unitprices and a slowing economy
would
Bruce Adkins and Ken Bruno Albemarlebut for the past seven years
curb demand and increase supply,
Production Editor that has not proved to be the case. While the rate of increase in world oil demand
Rachel Zamorski

has declined Yen Yung
since theAlbemarle
surprising Catalysts
4% surge in 2004, it nevertheless appears that
production@petroleumtechnology.com 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
21 Active site developments for improved productivity
themselves will continue. It is also safe to assume that refinery and petrochemical
Graphics Editor
conversion Kevin Carlsonwill
unit capacity Criterion
need toCatalysts
expand. & Technologies
Mohammed Samiuddin No massive new sources of energy are expected to come on stream for the
graphics@petroleumtechnology.com
foreseeable future. The world will remain dependent on oil and gas for decades to
come 25 evenRing-shaped
though the upstream catalystsindustrymake the faces gradeincreasing challenges in the
Editorial
discovery Brianproduction
and Moyse Haldor of new Topsøe
sources. In fact, some well-placed industry
PO Box 11283 analysts think 2008 may be the year where there is no increase in crude supply at
Spring TX 77391, USA all from regions outside of OPEC. For this reason, we will continue to see significant
tel +1 281 374 8240 31 Reducing
investment in refinery sulphur
upgrades oxide emissions
despite surgingfrom costs the FCC unitof feedstock
— security
fax +1 281 257 0582 supply,
albeit Colin unconventional
Baillie, Renaud low-quality
Kieffer and feedstock,
Maria Luisa takesSargenti
precedence over the
quality of feedstock supply.
Grace Davison Refining Technologies Europe
Feedstock options such as biomass (for biofuels production), Canadian tar sands
Advertising Sales Manager (for distillate production) and other types of unconventional crude sources require
Paul Mason
reactor37 technology
Maximising that allows for the integration
FCC distillate production of these operations into existing
sales@petroleumtechnology.com
process configurations. The quality of these types of feedstock are one important
Advertising Sales reason whyJoe McLean
a wider array BASF Catalysts
of catalysts has been introduced into the market. For
Bob Aldridge example, as refiners cut deeper into the vacuum tower, the concentration of
sales@petroleumtechnology.com metals in the VGO requires a properly designed guard bed system to protect active
47 Advances in methanol synthesis
catalysts in the hydrocracker. The characteristics of feedstock with low API gravity
Advertising Sales Office
(eg, <10), Terry
high Fitzpatrick
metals, nitrogen andand Tom otherHicks Johnson components
undesirable Matthey Catalysts is one of the
tel +44 870 90 303 90 main reasons why hydrotreaters and hydrocrackers are becoming larger — to
fax +44 870 90 246 90 accommodate not only higher volumes of catalyst, but also a wider variety of
catalyst54 withCatalytic steam generation
specific formulations.
Publisher
Nic Allen
Non-catalyticFranck Letellier
processes areand alsoDave
playing Wardlea signifi Oxford
cant roleCatalyst
in theGroup refiner’s ability
publisher@petroleumtechnology.com to process whatever unconventional crude sources become available. For example,
some refiners processing higher volumes of resid and atmospheric tower bottoms
Circulation have considered adding certain types of solvent-extraction processes in addition
Jacki Watts to overall improvements to crude unit (eg, vacuum tower revamps) and delayed
circulation@petroleumtechnology.com coker operations. Improvements in furnace technology, such as with olefin steam
cracker operations, have resulted in significant increases in worldwide ethylene
Crambeth Allen Publishing Ltd capacity.
Hopesay, Craven Arms SY7 8HD, UK However, any expansion of the value chain (eg, ethylene-to-propylene via
tel +44 870 90 600 20
dehydrogenation) requires investment in catalytic-based processes, as discussed in
fax +44 870 90 600 40
the following articles authored by experts in the field of downstream process
ISSN 1362-363X technology. PTQ wishes to extend its gratitude to the authors who provided
editorial and responded
Cover caption: to Western
BP Kwinana refinery, the Q&A published
Australia: a filter suppliedin this
by Pall issue of
Corporation PTQ Catalysis,
cut emissions of catalyst dust as
from the FCC unit.
well as to those respondents who addressed the online questions Photo: BP
(www.eptq.com)
Petroleum Technology Quarterly (USPS
that addressed the specifics of certain reactor and catalytic issues of importance to
0014-781) is published quarterly plus the industry.
annual Catalysis edition by Crambeth Allen
Publishing Ltd and is distributed in the USA ©2010. The entire content of this publication is protected by copyright full details of which are available from the publishers. All rights
by SPP, 75 Aberdeen Rd, Emigsville, PA 17318. reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means – electronic,
Periodicals postage paid at Emigsville PA. mechanical, photocopying, recording or otherwise – without the prior permission of the copyright owner.
Postmaster: send address changes to The opinions and views expressed by the authors in this publication are not necessarily those of the editor or publisher and while every care
Petroleum Technology Quarterly c/o PO has been taken in the preparation of all material included in Petroleum Technology Quarterly the publisher cannot be held responsible for any
Box 437, Emigsville, PA 17318-0437 statements, opinions or views or for any inaccuracies.
Back numbers available from the Publisher
at $30 per copy inc postage. René G Gonzalez

contents/ed com copy 6.indt 1 2/2/10 11:27:15


)&&
&DWDO\VWV

‡ 3URS\OHQH0D[LPL]DWLRQ
‡ 5HVLG&RQYHUVLRQ
‡ *DVROLQH0D[LPL]DWLRQ
)&&
‡ /&20D[LPL]DWLRQ $GGLWLYHV
‡ &RNH6HOHFWLYLW\

‡ &RQWUROOLQJ62[12[ &2
‡ 5HGXFLQJ*DVROLQH6XOIXU
‡ %RRVWLQJ2FWDQH
‡ ,PSURYLQJ)XHO4XDOLW\
‡ 3URGXFLQJ3HWURFKHPLFDO
 )HHGVWRFNV
‡ 5HGXFLQJ(QYLURQPHQWDO
 (PLVVLRQV

:HGRQ¶WMXVWPDNH)&&FDWDO\VWV
ZHPDNH)&&FDWDO\VWVIRU\RX
2QO\*UDFH'DYLVRQRIIHUVUH¿QHUVDIXOOSRUWIROLRRIKLJKWHFKQRORJ\VWDWHRIWKHDUW)&&FDWDO\VWVDQGDGGLWLYHV
'ULYHQE\RXUZRUOGFODVVUHVHDUFKDQGGHYHORSPHQWLQWKH86DQG(XURSHRQO\*UDFH'DYLVRQFDQFXVWRPL]HSURGXFWV
GHVLJQHGVSHFL¿FDOO\WRRSWLPL]H\RXURSHUDWLRQ

2XULQGXVWU\OHDGLQJWHFKQLFDOVHUYLFHSURYLGHVVXSSRUWLQFOXGLQJXQLWPRQLWRULQJDQGWURXEOHVKRRWLQJEDFNHGE\RXUKLJKOHYHO
RIDQDO\WLFDOVHUYLFHVLQFOXGLQJ(FDWVSHQWFDWDO\VWDQG¿QHVDQDO\VLV

:HKDYHEHHQWKHOHDGHULQ)&&FDWDO\VLVVLQFHLQYHQWLQJWKH¿UVW)&&FDWDO\VWRYHU\HDUVDJRDQGZHDUHFRPPLWWHGWR
WKHIXWXUHRIWKHUH¿QLQJLQGXVWU\
www.ptqenquiry.com
+DYH*UDFH'DYLVRQGHVLJQ\RXURSWLPDOFDWDO\VW&RQWDFWXV for further information

*UDFH'DYLVRQ5H¿QLQJ7HFKQRORJLHV *UDFH'DYLVRQ :5*UDFH6LQJDSRUH37(/WG


:5*UDFH &R&RQQ *UDFH*PE+ &R.* 2UFKDUG5RDG
*UDFH'ULYH ,QGHU+ROOHUKHFNH   :KHHORFN3ODFH
&ROXPELD0'86$ :RUPV*HUPDQ\ 6LQJDSRUH
3KRQH    3KRQH   3KRQH   

ZZZJUDFHFRPāZZZHFDWDO\VWVFRP

57B*OREDO$Q]$LQGG 
grace.indd 1 29/1/10 12:11:04
catalysis Security of may
Metals supply
T
he European Union has arguably
been the global leader in biodiesel
production ptq and use, with overall
biodiesel production increasing from 1.9
million tonnes in 2004 to nearly 10.3 million
feedstock supply
impede catalysis
tonnes in 2007. Biodiesel production in the
Vol 15 No
2010
past
Vol213 No 2
US has also increased dramatically in the
2008
few years from 2 million gallons in 2000
growth
to approximately 450 million gallons in 2007.

DR
According to the National Biodiesel Board,
171 companies own
Editor Editor
biodiesel manufacturing ecently
espite published
signs in 2007 estimates for the future
of a slowdown of catalyst
in various markets
sectors of thesuggest that
economy,
René G Gonzalez
plants and are actively marketing biodiesel.1.
Chris Cunningham therefivalue
ners of industrial
remain a big chemical
play for catalysis
prospectiveby the mid-point
investors. of It this
useddecade
to be
editor@petroleumtechnology.com
editor@petroleumtechnology.com
The global biodiesel market is estimated conventional
will be $7.5 billion,wisdom
of which thatthe higher
market fuelforprices
refinery and a slowing
catalysts will economy
account
to reach 37 billion gallons by 2016, with an would curb demand and increase supply,
for some $4.3 billion. For the refinery sector alone that represents annual but for the past seven years
growth
Production Editor thatupwards
has notof proved to be the case. While the rate of increase in world oil demand
average Production
annual growthEditor
rate of 42%. Europe 5% — a more cheerful view of the future than most businesses can at
Rachel Zamorski
will Rachel
continue to Zamorski
be the major biodiesel hasleast
declined
currentlysince thefor.
hope surprising 4% surge in 2004, it nevertheless appears that
production@petroleumtechnology.com
production@petroleumtechnology.com
market for the next decade, followed closely demand beyond 2008
This level of rising market will grow, along
value with
is no prices.Catalyst
surprise. It is a safe bet that rapidly
developments have
by the USEditor
Graphics market. increasing oil consumption by China, India and
enabled a fast and effective response, time and again, to whatever even the Middle East producers
form of
AlthoughSamiuddin
Mohammed high energy prices, themselves will continue. legislation
environment-mending It is also safe to assume
is thrown in the that refineryofand
direction petrochemical
fuels producers.
Graphics Editor
graphics@petroleumtechnology.com conversion unit capacity will need to expand.
increasing global demand, drought
Mohammed Samiuddin Catalyst developments are key to the preservation of margins when the crude
and other factors are the primary driv- No massive
supply newtakes
situation sources
one of of energy
its moreare expected to
disconcerting come
turns. And onthey
streamhelpfor the
refin-
graphics@petroleumtechnology.com
Editorial foreseeable future. The world will remain dependent on oil and gas for decades to
ers for higher food prices, food ers to derive greater flexibility from their line-ups of processing units when
tel +44 844 5888 773 come even though the upstream industry faces increasing challenges in the
competitive feedstocks
fax +44 844 5888 667 have long been variation in demand for products and quality of raw material inputs present
Editorial to be a major discovery and production of new sources. In fact, some well-placed industry
and will continue different types of challenges. The catalyst producers’ continuing high spend on
POtheBox 11283 analysts think 2008 may be the year where there is no increase in crude supply at
concern for
Advertising development
Sales Manager of biofu- research and development indicates more of the same in the future, by way of
Spring TX 77391, USA all from regions outside of OPEC. For this reason, we will continue to see significant
els. Totel compete,
Paul Mason the
+1 281 374 8240 industry has response to
investment in challenges faced by the
refinery upgrades refining
despite industry.
surging costs — security of feedstock
sales@petroleumtechnology.com
responded by281
developing
fax +1 257 0582methods to However, the technological consensus
supply, albeit unconventional low-quality feedstock, between producers and consumers
takes precedence of
over the
increase process efficiency, utilise or catalysts is obviously
quality of feedstock supply. not the only significant factor in the marketplace. Variables
Advertising Sales
upgrade by-products and operate with as unpredictable
Feedstock optionsassuch priceasvariations
biomass (for in the crudeproduction),
biofuels slate are lurking on thetar
Canadian supply
sands
Bob Aldridge
Advertising
lower SalesasManager
quality lipids feedstocks. (forside, and production)
distillate current thinking is that
and other theofnext
types five years may
unconventional crude put unwelcome
sources require
sales@petroleumtechnology.com
Paul Mason pressure on the thatpriceallows
ratherfor than
reactor technology the the size of the
integration market
of these for catalysts,
operations as the
into existing
sales@petroleumtechnology.com
Feedstocks
Advertising Sales Office cost of
process commodity
confi gurations. raw
The materials
quality inflates
of theseand typessources become tighter.
of feedstock are one important
Biodiesel
tel +44 844refers to aSales
5888 771
Advertising diesel-equivalent reason Fewwhy producing
a wider industries can relyhas
array of catalysts more been than petroleuminto
introduced refining on suchFor
the market. a
fax +44 844 5888 662
fuel consisting of short-chain alkyl broad range
example, as refinersof commodity
cut deepermetals into theto catalyse
vacuumitstower, processes, from relativelyof
the concentration
Bob Aldridge
(methyl or ethyl) esters, made by the
sales@petroleumtechnology.com minor
metals transition
in the elements
VGO requires such as designed
a properly cobalt and guardmanganese
bed system to rare earths active
to protect and
Publisher
transesterification of triglycerides, precious
catalysts metals.
in the During theThe
hydrocracker. pastcharacteristics
decade, the price of such with
of feedstock metals low hasAPIrisen by
gravity
Nic Allen
commonlyAdvertising knownSales Office oils or
as vegetable (eg,around
<10), high 300%.metals,
The chief reasonand
nitrogen for other
this fluctuation
undesirable is components
much the same as itofwas
is one the
publisher@petroleumtechnology.com
tel +44 870
animal fats. The most common 90 303 90 form mainfor reasons
shifts in why the prices of a wider
hydrotreaters and base of commodity
hydrocrackers arematerials:
becoming demand
larger from
— to
fax +44 870 90 246 90 accommodate not only of higher
uses methanol, the cheapest alcohol
Circulation the “hot” economies Asia, volumes
China in of catalyst,The
particular. but vagaries
also a wider variety of
of commodity
available,
Jacki Watts to produce methyl esters. catalyst with specifi c formulations.
pricing aside, there are also growing fears about the future availability of the
Publisher
circulation@petroleumtechnology.com
The molecules NicinAllenbiodiesel are primar- Non-catalytic
types of metals processes
found inare also playing
refinery a signifi
catalysts. The cant role in the
US National refiner’s
Research ability
Council
ily fatty acid methyl esters (FAME),
publisher@petroleumtechnology.com to process whatever unconventional crude sources become
has compiled a list that includes a number of these metals, whose supply, it available. For example,
Crambeth Allen Publishing Ltd some refiisners
usually created by transesterification says, at aprocessing higher
highly critical volumes
level. of resid
Difficulties in and
sourcingatmospheric towersaid
metals could, bottoms
the
Hopesay, Craven Arms SY7 8HD, UK have considered adding certain types of solvent-extraction processes in addition
between Circulation
fats and
tel +44 844 5888 776methanol. Currently, council, expose a range of US industries to political, economic and other risk.
biodiesel Jacki Watts from various to overall improvements
in Europetofollows crude unit (eg, vacuum tower revamps)
2010, theand delayed
fax +44 844is5888 produced
667 Thinking similar lines. During European
circulation@petroleumtechnology.com
vegetable and plant oils. First-genera- coker operations. Improvements in furnace technology,
Commission is thought likely to produce its own list of at-risk, economically such as with olefi n steam
cracker operations, have resulted in significant increases in worldwide ethylene
tion
ISSN food-based
Crambeth 1362-363X feedstocks are straight
Allen Publishing Ltd important metals and to propose measures such as stockpiling. Meanwhile,
capacity.
vegetable
Hopesay, oils such
Craven Arms as SY7
soybean8HD,oilUK and Japan is known to be actively stockpiling uncommon metals as a safeguard for
However, any expansion of the value chain (eg, ethylene-to-propylene via
animal fats
tel +44 such
870 as90 tallow,
600 20lard, yellow its technologically intensive industries.
Petroleum Technology Quarterly (USPS 0014-781) dehydrogenation) requires investment in catalytic-based processes, as discussed in
grease, fax
is published +44 870
chicken
quarterly 90
and600
fatannual
plus the40 by-products
Catalysis edition There have been alarms over the supply of so-called strategic metals in the
by Crambeth Allen Publishing Ltd and is distributed the following articles authored by experts in the field of downstream process
of the production
SPP, 75 Aberdeenof Omega-3 fatty past and instances of stockpiling that have backfired on eventually over-
in the USA by
ISSNfish 1362-363X
Rd, Emigsville, PA technology. PTQ wishes to extend its gratitude to the authors who provided
acids from
17318. Periodicals postageoil. Soybean
paid at Emigsville PA.oil and supplied
editorial andhoarders.
responded However,
to the Q&A there published
are fairly tight in thisgeographical
issue of PTQ andCatalysis,
geologicalas
Postmaster: send address changes to Petroleum
rapeseeds oil are
Technology Quarterly c/o POthe common source limits to the supply side of many of the metals in
well as to those respondents who addressed the online questions (www.eptq.com) question. It would be as well
for biodiesel
Box 437, Emigsville,production
PA 17318-0437 in the US and
thatfor some reassurance
addressed the specifiabout their future
cs of certain availability.
reactor and catalytic issues of importance to
Back numbers available
Petroleum from theQuarterly
Publisher
Europe inisTechnology
at $25 per copy
0014-781)
quantities
inc postage.
published
that can (USPS
quarterly plus
produce Among the advances in catalyst technology that refiners have to look
the industry.
enough
annual biodiesel
Catalysis edition byto Crambeth
be used Allenin a forward to in the next five years and beyond, it may be that none will draw as
Publishing Ltd and is distributed in the USA
commercial market with currently
by SPP, 75 Aberdeen Rd, Emigsville, PA 17318.
much favourable attention as advances in recycling metals from spent catalyst.
applicable
Periodicals technologies.
postage paid at Emigsville PA.
Postmaster: send address changes to
First-generation
Petroleum Technology Quarterly c/o PO
feedstocks for
Box 437, Emigsville, PA 17318-0437
chris cunningham
Back numbers available from the Publisher
at $30 per copy inc postage. René G Gonzalez
CATALYSIS 2010 3

contents/ed com copy 6.indt 2 1/2/10 12:14:59


Travel with us,
and you’re always
on the road
to innovation.

THE CLEAN FUELS TEAM

www.criterioncatalysts.com
© 2008 Criterion Catalysts & Technologies L.P. (cri815_0508)
www.ptqenquiry.com
for further information

cri815_A_roadInnovation_ptq.indd 1 5/29/08 12:07:17 PM


criterion 1 9/3/09 10:13:45
PTQ Catalysis Roundtable

Q I understand that high levels of sodium or potassium in Q We are at the initial stages of planning a revamp of our
the feed to the FCCU can have a seriously deleterious effect HF-based alkylation unit. What’s the current federal view of
on the acid sites of the catalyst. Are there any pre-treatment the technology and should we be looking for alternatives?
steps I should apply?
A Edwin van Rooijen, Business Manager, Albemarle
A Yen Yung, Global Technical Specialist, Fluid Cracking Catalysts
Catalysts, Albemarle Catalysts Solid acid alkylation by AlkyClean is one alternative.
FCC catalyst contamination by potassium is rare. The The technology and proprietary solid acid catalyst
measures to avoid the deleterious effect of potassium AlkyStar are developed and demonstrated by Lummus
contamination are similar to the more commonly found Technology, Albemarle Catalysts Corporation and
phenomenon of contamination by sodium. Neste Oil. The process significantly improves the safety
Sodium salts enter the refinery with the crude oil. of refinery alkylation over conventional liquid acid-
Desalting is generally applied to remove the sodium based processes, reducing the potential hazards
from the crude. Any remaining sodium ends up in the associated with the transportation and handling of
heaviest distillates and the residue, and can find its liquid acids. The process gives refiners a competitive,
way into the FCC or thermal conversion units. The cleaner and inherently safer alkylation technology. No AlkyC
operation of the crude desalter must therefore be moni- acid-soluble oils or spent acids are produced, and there Award
tored and optimised cautiously. In several refining is no need for product post-treatment to remove traces from
steps of refineries processing sour crudes, caustic is of acid. The process has also proven to be economic which
applied for corrosion control by acids. Proper monitor- and robust, requiring minimal maintenance. as the
ing of the separation of sodium from the oil stream is nolog
required too.
Some FCCUs processing vast amounts of long resi- Q What sort of run length should I reasonably expect from
dues are equipped with an FCC feed desalter to limit the guard bed in my hydrotreater?
not only sodium contamination of the FCC catalyst,
but also to prevent the formation of salts in the frac- A Steve Mayo, Global HPC Applications Technology
tionator overhead vessel. We recommend routine Manager, Albemarle Catalysts
measurement of the sodium content of FCC feed, espe- There are three uses for guard beds in hydrotreaters:
cially import feed. If large quantities of sodium are collection of solids to prevent pressure drop build-up,
found in the import feed, leave it to settle in a tank to controlled removal of reactive species, and adsorption
remove the sodium with the water, if possible, or adapt of inorganic poisons to protect downstream catalysts.
the catalyst addition policy to avoid excessive catalyst Failure of any of these functionalities can result in
contamination. premature shutdown. In general, guard beds should be
Modern resid FCC catalysts such as Upgrader and sized so that the capacity of the bed for solids collec-
Coral SMR have excellent resistance to the deleterious tion and/or poison adsorption is similar in cycle length
effect of metals poisons such as sodium, nickel, vana- to that of the main bed catalyst. Data from spent cata-
dium, calcium and, when present, potassium. lyst analysis and canisters can be used to optimise
subsequent guard bed configurations for maximum
A Kevin Carlson, Global Business Manager, FCC Pretreat cycle life.
& Sentry Products, Criterion Catalysts & Technologies
Feed contaminants such as sodium or potassium can,
indeed, adversely affect FCC operations. Pre- Q Is Fischer-Tropsch becoming a significant and efficient
treatment of the feed in a hydrotreater results in option for the economic conversion of flare gas? Are there any
improvements in FCC performance, but can also result catalyst developments I should be keeping an eye on?
in catalyst deactivation and pressure drop build-up in
the pre-treated unit. Proper application of top-bed A Frans Plantenga, Technical Director, Albemarle AFT
grading and demetallisation products, such as those The key challenge when converting flare gas to liquid
found in Criterion’s Sentry portfolio, can mitigate these is the scale of the operation. So far, to be economical,
issues, but often improved crude handling and installations have been built on a large scale. So, the
desalting can provide an effective long-term solution. solution to this problem lies in the successful scaling

Additional Q&A can be found at www.eptq.com/QandA

www.eptq.com CATALYSIS 2010 5

Q&A copy 3.indd 1 1/2/10 10:03:08


down of the integral process. Several companies are represents the operating window of the catalyst stand-
working on this concept, including Velocys and pipe. Operating the FCCU at a high Umb/Umf is
Compact GTL. Currently, such processes are operated important to ensure stable and high catalyst circulation
at pilot plant scale and plans are in progress to rates. This ratio is a function of the properties of the
construct a large-scale demo facility. fluidisation medium and the catalyst. A high e-cat
0–40 µm particulate content and a low average particle
A Alan Del Paggio, VP Renewables, CRI/Criterion size helps to increase the Umb/Umf.
In the current pricing environment, the economics for a Controlling and monitoring the e-cat catalyst particle
Fischer-Topsch (FT)-based conversion scheme for flare size distribution is critical to ensure smooth and high
gas would be dominated by capital considerations. catalyst circulation rates. A high equilibrium apparent
Normally, natural gas-based FT processes are large bulk density can also aggravate standpipe circulation
(tens of thousands of barrels per day) to maximise problems. R Mott1 described how equilibrium catalyst
economic return. Advances in smaller-scale gasification with a high ABD and low 0–40 µm particle content can
and FT technology driven by renewable energy defluidise in a relatively short vertical distance in the
concerns may alter this landscape slightly, but not with catalyst standpipe.
the same economies of scale offered by the larger, Monitoring cyclone performance is important to
world-class operations. ensure low catalyst losses, but is also critical for main-
taining smooth catalyst circulation. Catalyst fines tend
to be lost from the unit before larger catalyst particles
Q We are having reliability issues with our catalyst circul- when cyclone performance degrades. As a result, the
ation. What key process or other variables should we be e-cat 0–40 µm content drops and the average particle
looking at to better understand these issues? size increases. The Umb/Umf then falls, and the unit
loses catalyst and stops circulating smoothly. Standpipe
A Yen Yung, Albemarle Catalysts flux rate should be monitored to ensure smooth cata-
Problems with catalyst circulation might arise if the lyst circulation. Flux rates of ~125–300 lb/s/ft2 can be
FCCU is hydraulically limited, has a non-proper achieved in many standpipes. Catalyst circulation is at
reactor/regenerator pressure balance, non-optimal aera- risk at flux rates much above and below these rates.
tion of standpipe, too dense catalyst or too little catalyst Pressure can also impact catalyst circulation. At lower
fines (due to malfunctioning cyclones) in the circulating reactor or regenerator pressures, the relative volume
equilibrium catalyst inventory. Albemarle’s technical change of the catalyst emulsion along the standpipe is
specialists can help customers to greater than at higher pressures. A
restore smooth catalyst circulation. high-volume change can result in
If the problem is caused by a lack of If combustion air in the catalyst defluidising between
fines, Smoothflow can be used to aeration taps. This phenomenon is
quickly restore flawless operation. the regenerator or described in detail by Mott.1
Air and steam distributor Dp is
A M Zehender, R Schiller and D steam in the stripper another critical variable to monitor.
Hunt, W R Grace
Standpipe aeration rate and design
is not distributed If combustion air in the regenerator
or steam in the stripper is not
are critical to reliable circulation in properly, catalyst distributed properly, catalyst circu-
many FCCUs. R Mott described lation can be at risk. A damaged
proper aeration techniques that are circulation can be (low Dp) air grid can result in zones
largely accepted by the industry.1 It of defluidised or partially deflu-
is important to remember that aera- at risk idised catalyst. If the regenerator
tion rates must be adjusted in catalyst withdrawal well is located
response to operational changes. A common mistake in this zone, defluidised catalyst will enter the regener-
made by refiners is to not their adjust aeration rates ated catalyst standpipe. Once defluidised catalyst
when feed rates and catalyst circulation rates change. enters a standpipe, it is almost impossible to refluidise
The Umb/Umf ratio is the ratio of the velocity at which it. The minimum dp of an air or steam distributor is
bubbles begin to form and the minimum fluidisation dependent on design.
velocity: High e-cat iron also can result in poor catalyst circu-
Umb = Minimum bubbling velocity, m/s lation. Iron tends to form nodules on the surface of the
Umf = Minimum fluidisation velocity, m/s catalyst particles, and poor catalyst circulation has been
F = 0–45 µm fraction in catalyst observed in some units when e-cat iron increases. In
µg = Gas viscosity, kg/ms this case, it is necessary to flush the iron out of the
Dp = Mean particle diameter = m inventory with increased catalyst additions.
tp,g = Particle and gas properties, kg/m3 Routine monitoring of these process variables will
g = Gravitational constant = 9.8 m/s2 help maintain reliable catalyst circulation.
This value is reported in Grace’s FCC equilibrium
catalyst (e-cat) reports. In a standpipe, the catalyst must 1 Mott R W, Troubleshooting FCC standpipe flow problems, Catalagram,
be fluidised, but not bubbling. As a result, this ratio 83, 1993.

6 CATALYSIS 2010 www.eptq.com

Q&A copy 3.indd 2 1/2/10 10:03:49


When the right
reaction matters ...
Trust BASF Process Catalysts
and Technologies

At crucial moments, the right reaction matters. When you are looking
for the right reaction from your process catalyst, turn to BASF. Our
technical experts will recommend the right catalyst from our innovative
product line that will achieve the desired reaction. The end results
may also include greater yield and better end product properties.
When the catalyst is right, the reaction will be right. Trust BASF.
� Adsorbents � Chemical catalysts � Polyolefin catalysts
� Custom catalysts � Refining catalysts
For more information, please visit www.catalysts.basf.com/process

www.ptqenquiry.com
for further information

basf.indd 1 29/1/10 12:14:49


A Ye Mon Chen, FCC Licensing Americas, Shell Global A David Hunt and Kevin Sutovich, W R Grace
Solutions US With the correct equipment, optimal operating
Several key variables can affect catalyst circulation in conditions and proper catalyst formulation, the FCCU
FCCUs. The first variable to look into is the fines can successfully process a wide range of feedstocks,
(0–40 µm) content in the equilibrium catalyst. Some including vegetable oils and other renewable
FCCUs can tolerate a lower fines content than others. If feedstocks.
the fines content falls below 5%, A full evaluation is required
some units could suffer from catalyst before processing used vegetable
circulation problems. If it is below Coker gas oil is oils or other renewable feedstocks.
3%, most units will have problems. While most vegetable oils contain
If the unit cannot retain catalyst fines best processed almost no sulphur or nitrogen,
in the system, catalyst circulation is many contain significant amounts
only the symptom; the real problem in a medium- to of sodium, calcium, phosphorus
is the cyclone system. and particulates.1 Another concern
Another variable to look into is high-pressure when cracking vegetable oils is the
aeration along the standpipe. Too release of water due to the cracking
much or too little aeration, or aera-
hydrotreater with of the oxygen-rich triglyceride
tion in the wrong places, can cause molecules.
catalyst circulation problems.
a properly designed We recommend you provide
Some FCCUs have an inlet hopper catalyst system your catalyst supplier with a
at the entrance to the standpipe. The sample of the vegetable oil together
design of the hopper and aeration with the primary feedstocks.
around it could also be a key contributor to catalyst Physical properties of the vegetable oil and the pres-
circulation problems. ence of any harmful contaminations relative to your
primary feedstock are important when designing the
proper catalyst. Cracking the vegetable oil blend in a
Q Our catalyst losses in hydrotreating coker gas oil are pilot plant with multiple catalyst candidates will
high. What should we be looking for in catalyst type and provide the yield data necessary to identify the best
performance to mitigate this? catalyst system with minimal compromises.
B Watkins1 describes the FCC yield effects of co-
A Kevin Carlson, Criterion Catalysts & Technologies processing rapeseed, soybean and palm oil with a
Coker gas oil is a highly aromatic feedstock that often composite resid feedstock. Pilot plant yield data
contains high levels of feed contaminants such as showed that co-processing vegetable oil improves
sulphur and nitrogen, as well as silicon, nickel, vana- coke selectivity and reduces hydrogen yield. Improved
dium and concarbon. This more difficult feed is best coke selectivity will allow increased conversion,
processed in a medium- to high-pressure hydrotreater which will increase gasoline production. Data reported
with a properly designed catalyst system. by Watkins suggests that additional catalyst
Key components to such a catalyst system would be activity may be required to successfully process vegeta-
a graded top-bed component, to mitigate pressure drop bles oils.
and reactor bed fouling due to possible coke fines, a Every situation is unique in terms of the feedstock,
demetallisation section to remove the feed metal unit constraints and objectives. Sharing information
contaminants, such as silicon, nickel and vanadium, and samples with an engaged supplier of industry-
followed by the more active section of the system to leading FCC catalyst technology will insure a successful
remove nitrogen and sulphur, and saturate feed application.
aromatics. Criterion has had great success with
such feeds utilising products from both the Sentry top- 1 Watkins B, et al, New opportunities for co-processing renewable feeds
bed and demetallisation portfolio with Ascent in refinery process, Catalagram,103, 2008.
technology.
A Easwar Ranganathan, FCC R&D Program Leader, Shell
Global Solutions US
Q Our studies indicate that it would be feasible to co-process Our studies indicate the feasibility of co-processing
waste vegetable oil in the FCCU for gasoline production, but waste vegetable oils in the FCCU. We do not believe
we have yet to identify a suitable compromise catalyst. there is any need to tweak the catalyst or use a compro-
mise catalyst for this purpose. Acid sites in a typical
A Frans Plantenga, Albemarle AFT FCC catalyst do the job, and the metals that come
Several studies have been done on the co-processing of with the waste vegetable oils (in ppm levels) do not
VO-VGO mixtures in the FCCU, which indicate that appear to have a significant negative impact on catalyst
this is possible with normal FCC catalysts and addi- activity. What we need worry about is the distribution
tives if the intake of vegetable oil is not too high. of the metals in the FCC products (notably gasoline
Limitations are set by the by-products formed and the and LCO streams), which may make the product
level of oxygenates in the products. off-spec.

8 CATALYSIS 2010 www.eptq.com

Q&A copy 3.indd 3 1/2/10 10:03:35


www.ptqenquiry.com
for further information

albemarle 3.indd 1 16/9/09 10:51:40


Prepared for Winter

Süd-Chemie – Contributing to a Cleaner Environment


Improve cold flow properties of your ultra low sulfur diesel (ULSD)
with Süd-Chemie's Hydex-G!
Major benefits of catalytic dewaxing with Hydex-G:
• Outstanding cold flow properties
• Flexible process / reactor configuration
• High resistance against nitrogen compounds
• Reduced capital investment and maximized profitability
Please contact us for further information!
Süd-Chemie – your partner for a sustainable future!

SÜD-CHEMIE www.ptqenquiry.com
www.sud-chemie.com for further information

SC Winterdiesel Anz_A4_F05.indd 1 01.09.2008 15:40:24 Uhr


sud chemie.indd
Prozessfarbe 1
CyanProzessfarbe MagentaProzessfarbe GelbProzessfarbe Schwarz 29/1/10 13:44:01
Maximising fuels from the FCC unit

Developments in catalyst pore structure engineering aim for higher yields of


transportation fuels from FCC operation

Bruce Adkins and Ken Bruno Albemarle


Yen Yung Albemarle Catalysts

T
o grasp the true complexity
of the FCC process, it is
important to recall the inter- $RYGAS
( (3 #(
play between chemical reactions, #( #(
thermodynamics and fluid mechan-
ics occurring simultaneously in an
FCCU, and the roles they play in #OKE
influencing yields.
The role of catalyst can best be &EED ,0'
visualised with the aid of a lumped ##
kinetic model. A great many
lumped kinetic models for oil refin-
ing processes can be found in the
literature. The most elaborate are
structure oriented lumping (SOL) "OTTOMS
,#/
models, wherein the lumps are SLURRY
²# ²n²# 'ASOLINE
based more on chemical similarity #n²#
than boiling point and can number
in the hundreds or even thousands.1
However, for our purpose, a much ²#
CONVERSION
simpler network, such as the one
shown in Figure 1, will suffice. The
lumps shown in Figure 1 are the Figure 1 FCC reaction network
traditional ones used in petroleum
refining, defined primarily on the pathway, or even group of selectivi- Interfacial transport phenomena
basis of boiling point: coke, bottoms, ties, without considering the entire introduce additional transfer
LCO, gasoline, LPG and dry gas. reaction network. Contributing resistances
While cracking is the main reac- additional complexity to the overall • Isothermal vs adiabatic operation
tion implied by the arrows in Figure reaction network is the interplay of: Most laboratory tests are operated
1, many other solid-catalysed reac- • Diffusion Reaction rates in FCC isothermally or, at best, semi-
tions take place in the FCC process: catalyst particles are strongly influ- adiabatically. Commercial units
isomerisation, dehydrogenation, enced by diffusion. Classical operate adiabatically. Translation of
condensation, transalkylation and diffusion-reaction theory2 maintains results from laboratory to commer-
hydrogen transfer. that local reaction rates are depend- cial scale can be challenging.
Consider gasoline selectivity. ent on the flux of reactants as well
According to Figure 1, increasing as diffusivity, which can differ Producing middle distillates in
the rates of the dot-dash arrow greatly for various compounds the FCCU
reactions relative to the rates of the • Mechanistic similarity For reac- Based on the reaction network
dash arrow reactions will cause the tions with mechanistic similarities, postulated in Figure 1, Figure 2
gasoline yield to increase. However, it can be extremely difficult to influ- shows the yields of bottoms
higher gasoline selectivity (gasoline ence one without influencing the (slurry), LCO and gasoline as a
yield per unit conversion) also other. This is a prime reason for the function of 221°C+ conversion. The
requires that the dotted arrow reac- prevalence of auto-correlations in corresponding TBP (true boiling
tion rates be reduced. FCC test data point) cutpoints are: gasoline,
In practice, it is difficult to alter • Multiphase phenomena FCC is a C5-221°C; LCO, 221–343°C; and
the selectivity of any single reaction complex multiphase process. bottoms, >343°C. These cutpoints

www.eptq.com CATALYSIS 2010 11

albermarle.indd 1 1/2/10 10:17:55


consuming reactions (dot-dash
arrow to gasoline, dotted arrow to

coke).
LCO

It is generally accepted that
mesopore and macropore activity,
 the so-called (alumina) matrix activ-
ity, favours bottoms cracking, while
9IELD WT

Bottoms
 zeolite provides for higher LPG and
gasoline selectivity.3,4 Therefore,
  
 middle distillate production is
generally favoured by higher matrix
Gasoline
 cracking (as evidenced by higher
meso surface area) and reduced

   
zeolite cracking. In other words,
#ONVERSION WT middle distillate production
increases as the zeolite-to-matrix
ratio (Z/M) decreases. The benefits
Figure 2 Overcracking of LCO of this approach with respect to
increasing LCO at a constant
are typical for a gasoline mode of to overcracking. In all three regions, conversion of 70 wt% are shown in
operation, which is characterised by the gasoline yield increases with Table 1.5
a gasoline endpoint as high as conversion. Not shown in the graph For greatest bottoms conversion,
possible, at the expense of the LCO is zone 4, the gasoline overcracking the feed molecules need to reach
initial boiling point. zone, where the net production of the active sites quickly. Conversion
Most FCCUs have been designed gasoline drops. Clearly, when the of the desirable products in the
for operation in the LCO overcrack- LCO yield is the main objective, the diesel boiling range and other
ing mode for the sake of high gasoline yield suffers. The greatest secondary reactions, such as
gasoline yields. Nonetheless, a benefits are obtained with a catalyst hydrogen transfer, aromatisation
substantial increase of LCO in the having a high bottoms cracking and condensation, have to be
30–40 wt% range can be obtained. selectivity and a somewhat lower avoided. This is achieved by
A challenge to overcome is to limit gasoline selectivity. increasing the accessibility of the
the potential loss in bottoms conver- catalyst. Accessibility is the prop-
sion. The actual yield of gasoline Maximising catalytic conversion of erty that allows primary products
and LCO will depend on operating bottoms to middle distillates to escape promptly from the reac-
severity, feed type, and quality and The catalytic approach for produc- tion sites.
catalyst. Figure 2 illustrates what is ing the greatest yield of middle Outstanding performance of
achievable with a conventional distillates in FCC is easy to visual- highly accessible catalysts, as meas-
catalyst at low 221°C+ conversion ise using Figure 1. The challenge is ured by our internally developed
(= 100 - LCO - bottoms) levels. to selectively crack the large slurry Albemarle Accessibility Index (AAI)
In zone 1, the LCO yield is low molecules to LCO, while keeping method, has been confirmed in
due to insufficient bottoms conver- down the coke yield and avoiding several applications.6-8 A typical
sion. In zone 2, the LCO yield overcracking of LCO molecules into example of the benefits of high
increases with conversion, as the lighter products. In terms of Figure accessibility is shown in Figure 3.
cracking of bottoms to LCO exceeds 1, this means increasing the reac- This is data from an FCCU process-
the cracking of LCO to lighter prod- tion rates of the LCO-producing ing resid and operating in
ucts. In zone 3, the net production reactions (solid arrows) while low-severity LCO mode.
of LCO drops with conversion due decreasing the rates of the LCO-
Hydrotreated VGO operation
Reduced Z/M ratio and higher meso surface area results in a greater
with high middle distillate and
LCO yield and better bottoms conversion propylene yields
The second maximum LCO exam-
ple is an FCCU processing large
Yields at 70 wt% conversion, wt% High Z/M Low Z/M
Dry gas 2.7 3.1
amounts of hydrotreated vacuum
C3s 5.4 5.3 gas oil in a market favouring high
C4s 9.3 9.1 propylene yields. The refiner has
Gasoline 45.4 45.3 switched from a competitive cata-
LCO 16.9 18.7
Bottoms 13.2 11.3
lyst to an Amber MD catalyst
Coke 7.1 7.2 featuring higher mesoporosity,
(Lab SCT-RT data; resid feed) lower Z/M ratio and higher AAI
than the previous catalyst. At
Table 1 constant operating conditions in

12 CATALYSIS 2010 www.eptq.com

albermarle.indd 2 1/2/10 10:18:03


o r l dw id e use ou r hyd ro
er sw pro c
es s
f i n g i es and Cataly st s to
Re ech nol oducts fro m low - d eliv er ing
o
t e r pr qu al it
an us sh ow y ou h o w. 0
yf
ee d
le
c Let
( 0 1) 5 1 0
www. cl g . 24 2. 3 s.
- cl ea 1 77
n. com

clean transportation fuels


ultra-low sulfur diesel (<5 ppm)
ISOCRACKING
ICR 179
high smoke point jet (25-30 mm)
VGO ICR 180 Ultra-low sulfur naphtha (<0.5 ppm)
ICR 183 FCC feed
ICR 245

Learn how CLG’s tailored hydroprocessing


catalyst systems maximize coversion to
clean products — visit clg-catalyst.com

www.ptqenquiry.com
for further information

clg.indd 1 29/1/10 12:26:25


Technology and EPC:
Who says you can’t have it all?

CB&I offers proven process technologies and catalysts from Lummus Technology,
CDTECH and Chevron Lummus Global (CLG). Combined with a full range of EPC
services, you can have it all from CB&I.

» CB&I Lummus - Full-scope EPC for upstream and downstream energy process facilities
» CB&I Steel Plate Structures - The world’s most experienced tank builder
» Lummus Technology - Licensed process technologies and catalysts for refineries, petrochemical facilities
and gas processing plants www.ptqenquiry.com
for further information

www.CBI.com Engineering Solutions . . . Delivering Results


cbi.indd 1 29/1/10 12:28:45
maximum conversion mode, the
Amber MD catalyst made 1.1 wt%
more propylene, 0.5 wt% more LCO 

and 0.6 wt% less bottoms than the Highly accessible Albemarle
competitive catalyst. Despite a  catalyst with high matrix activity
2.5 wt% drop in conversion, the
bottoms yield stayed virtually 

constant and the LCO yield

,#/ ,6
increased by 2.4 wt% (see Table 2). 

Propylene yield stayed constant,



despite the lower LPG yield, thanks Competitor
to the much higher LPG olefinicity, catalyst

a characteristic of high accessibility
catalysts.

With economics turning ever         
more favourably towards distillate #ONVERSION ,6
products, the refiner took the
opportunity to make more LCO by
dropping the riser outlet tempera- Figure 3 Higher accessibility increases LCO yield in low severity operation
ture by 18ºC. Also, the feed
throughput was increased slightly. Increased LCO yield in a hydrotreated vacuum gas oil operation by switching
to an Amber MD grade designed to maximise propylene and LCO
FCC operation for ultra-high
LCO yield Competitor catalyst Amber MD
The important features of an “MD” Feed specific gravity 0.895 0.896
catalyst for maximum LCO opera- Feed sulphur content, wt% 0.05 0.08
tions have been described in the Feed UOP-K 12.15 12.11
Feed rate, bbl/d 30 950 32 000
preceding sections. In addition to Combined feed temperature, ºC 256 252
using such a catalyst, process condi- Riser outlet temperature, ºC 539 521
tions can be further adjusted to Regenerator temperature, ºC 708 705
dramatically enhance the yield of Dry gas, wt% 2.5 2.1
LPG, wt% 25.6 23.9
LCO. Two examples illustrate that Gasoline (C5-221ºC), wt% 45.8 45.7
very high yields of LCO are LCO (221–366ºC), wt% 15.0 17.4
possible. Slurry (370+ºC), wt% 5.8 5.9
The first example concerns an Coke, wt% 5.3 5.0
Conversion, wt% 79.2 76.7
FCCU processing vacuum gas oil Propylene, wt% 9.4 10.5
with a typical API of 23° and a
sulphur content of about 1 wt%.
The catalyst used is Amber MD. Table 2
The special unit feature is that the
gasoline endpoint is minimised to
an ASTM D-86 endpoint of 149ºC, Typical main operating conditions Typical main operating conditions and
while the LCO endpoint is very and yields from an FCCU maximising yields from a unit maximising LCO
high at an ASTM D-86 endpoint of LCO yield with Amber MD yield with Amber MD at low severity
379ºC. Thanks to these cutpoints
and the use of Amber MD, a yield Amber MD Amber MD
of 44 wt% LCO is obtained, having Feed API 23.1 Feed API 25.5
Feed sulphur content, wt% 1.0 Feed sulphur content, wt% 0.5
a typical cetane index of 34, as
Feed Concarbon, wt% 0.57 Feed Concarbon, wt% 0.35
shown in Table 3. This unit applies Combined feed temperature, ºC 226 (Recycle rate)/(fresh feed rate), vol/vol 0.5–1
no bottoms recycle. Riser outlet temperature, ºC 518 Combined feed temperature, °C 368
The second example is an Amber Regenerator temperature, ºC 710 Riser outlet temperature, °C 499
Cat-to-oil ratio, lb/lb 7.0 Catalyst-to-oil ratio, lb/lb 4.0
MD application processing vacuum
Catalyst addition rate, lb/bbl 0.35 Dry gas, wt% 2.9
gas oil. The severity is very low: Dry gas, wt% 3.4 LPG, wt% 4.9
low reaction temperature, high LPG, wt% 11.8 Gasoline, wt% 25.7
combined feed temperature and Gasoline (ASTM D-86 FBP=149°C), wt% 24.9 LCO, wt% 42.4
LCO, wt% 44.2 Bottoms, wt% 21.2
low catalyst-to-oil ratio (see Table
Bottoms, wt% 10.5 LCO cetane index 34
4). Bottoms recycle is applied to LCO cetane index 34 Ni, ppmwt 1600
enhance the production of LCO. Ni, ppmwt 1600 V, ppmwt 3500
Note that the volume of recycle can V, ppmwt 3500
be as high as the fresh feed intake.
In this example, gasoline endpoint Table 3 Table 4

www.eptq.com CATALYSIS 2010 15

albermarle.indd 3 1/2/10 10:18:15


is also minimised. LCO yield and
cetane index are very high: 42.4
wt% and 34, respectively. Bottoms
yield is higher due to extremely
low severity.

Maximising gasoline and


Go-Ultra
minimising coke
Diffusion is important in determin-
D6DLOG$

ing the middle distillate yield of an


FCC catalyst. Not surprisingly,
diffusion is equally important in
the production of gasoline from gas
oil. Gas oil molecules must diffuse
Ruby into the interior of a catalyst parti-
cle through the pore structure in
    order to access the active sites (both
0OREDIAMETER NM
matrix and zeolite). At the same
time, products must counter-diffuse
out of the particle.
Figure 4 Improved PoSD of Go-Ultra compared to Ruby Rapid product diffusion is an
absolute necessity in minimising
secondary reactions. For gasoline
and coke selectivity, three classes of
secondary reactions are particularly
important: coke formation, hydro-
gen transfer and secondary
cracking. Coke formation, in addi-
tion to changing the FCC heat
balance, alters the active sites and
their corresponding activities and
selectivities. Secondary hydrogen
Go-Ultra
transfer can lead to saturation of
olefins and reduced octane values,
reduced crackability to LPG olefins
(important when using ZSM-5) and
other yield debits. Overcracking,
Ruby another form of secondary cracking,
is also important. Gasoline can be
cracked to LPG when it stays in
contact with catalyst too long.
Major difference: Accessibility, as measured by
“AlO4/5” species
AAI, is an indicator of an MD cata-
lyst’s most important diffusional
characteristics. However, a catalyst
optimised for gasoline will usually
Figure 5 27Al MAS NMR spectra depend heavily on the naphtha
selectivity of zeolite and much less
so on the cracking selectivities of
Feed properties, deactivation and testing conditions matrix. Thus, a gas oil gasoline
catalyst will typically have higher
Z/M ratio and a lower absolute
Feed properties Deactivation amount of matrix than a gas oil MD
API 22.2 Method CD
S, wt% 0.81 Ni, ppm 1000 catalyst. For these kinds of formula-
Total/basic N, wt% 1508/488 V, ppm 1000 tions, we have found that AAI is
D1160 T5, °C 284 Testing not the best measure of accessibil-
T50, °C 428 Unit ACE ity, for several reasons:
T95, °C 558 RxT, °C 540
CCR 0.43 CTO 3, 4, 5, 6 • First, the levels of AAI required
tend to be lower for gas oil gasoline
catalysts than for resid and gas oil
Table 5 MD catalysts. This is because gas

16 CATALYSIS 2010 www.eptq.com

albermarle.indd 4 1/2/10 10:18:48


oil feed molecules tend to be
smaller, and the mild process condi-
tions used to maximise LCO require  'ASOLINE #OKE ,0' $RYGAS $RYGAS
the highest AAI possible to crack ( (3
the large bottoms molecules #( #( #(

• Second, AAI measurements

9IELDIMPROVEMENT WT
depend on the adsorption of dark-
,0'
coloured bodies, which are usually #OKE
##
asphaltene molecules. In the case of 

gas oil gasoline catalysts, smaller


molecules can be as important to
the gasoline and LPG yields as
²# 'ASOLINE
asphaltenes, or even more CONVERSION #n²#
important n
• Finally, and most importantly,     
3AMPLENUMBER
for gasoline catalysts, accessibility
to the zeolite crystals may be more
critical than bulk accessibility. AAI Figure 6 Go-Ultra yield improvements at constant conversion
is not the most discerning test for
examining differences in accessibil- achieve the advances demanded to The 27Al MAS NMR spectra in
ity to the zeolite particles embedded maximise gasoline. To reach that Figure 5 show that ADM-25 induces
in the catalyst particle. goal, Albemarle also developed a the formation of a unique
For this reason, we also use pore new catalyst component, called aluminium species, referred to here
size distributions (PoSDs) as ADM-25, which fine-tunes the cata- as AlO4/5 in conjunction with the
indicators of internal accessibility: lyst active sites by creating a new more commonplace tetrahedral
either as a complimentary technique type of silico-aluminate interaction. (AlO4) and octahedral (AlO6)
to AAI or as a standalone tech- To demonstrate the existence of species. The significance of AlO4/5
nique. It is true that FCC catalyst these sites, a brief foray into the is that it gives rise to Lewis and
pore structures are notoriously world of magic angle scattering Brønsted centres that modulate the
difficult to measure; suffice it to nuclear magnetic resonance (MAS- strength and distribution of overall
say, there is certainly not enough NMR) is required. MAS-NMR is a catalyst acidity. These new active
room in this article to discuss all complex but proven analytical tech- sites, coupled with the aforemen-
the problems. nique that provides insights into tioned improvements in PoSD, are
the coordination state and composi- responsible for enhanced coke-
Design features of Go-Ultra tional structure of solid-state selective bottoms cracking and
With the recent introduction of Go- materials.10,11 In particular, 27Al improved gasoline yield.
Ultra,9 Albemarle has taken the next MAS NMR has been used to char-
step in catalyst pore structure engi- acterise aluminium species in
Selectivity improvements
neering. Figure 4 compares the framework and extra-framework
Test comparisons have consistently
PoSDs of the company’s FCC gas zeolitic aluminium and matrix
verified the performance advan-
oil catalyst (Ruby) and the same alumina. tages of Go-Ultra. The examples
nominal formulation
in a Go-Ultra version.
The Go-Ultra sample
has enhanced meso- 'ASOLINE "OTTOMS ,0' ,#/
and macro-porosity 
compared with Ruby.
These PoSD improve-
ments result from a
9IELDIMPROVEMENT WT

new manufacturing &EED ,0'


##
technology, called

Onyx, used to manu-
²#
facture Go-Ultra. The CONVERSION
Onyx-derived PoSD
improvements are 3LURRY
,#/
²n²# 'ASOLINE
essential to the #n²#
improved coke and n
    
gasoline selectivity of
3AMPLENUMBER
Go-Ultra.
This improved PoSD
is not sufficient to Figure 7 Yield improvements at constant coke for Go-Ultra

www.eptq.com CATALYSIS 2010 17

albermarle.indd 5 1/2/10 10:18:58


shown in Figure 1
reduces to the arrange-
 
ment shown at the
right-hand side of
 Figure 6. From this,
 we arrive at a simple,
 intuitive explanation
for the yield pattern:
9IELD WT

the extent of second-

9IELD WT
  ary reactions (long
dash arrows) is
 reduced, preventing

gasoline from convert-
ing into LPG and coke.

This is consistent with
the reduced diffusion

#OKE CONV "OTTOMS COKE

'ASOLINE CONV 'ASOLINE COKE
resistance associated
with the improved
2UBY 'O 5LTRA #OMPETITOR #OMPETITOR #OMPETITOR PoSD. A more robust
interpretation is that
the combination of
Figure 8 Comparison of commercially available “maximum gasoline” catalysts PoSD and more selec-
tive active sites is
used in this section are based on At constant conversion (Figure 6), responsible for these yields.
the deactivation and feed condi- Go-Ultra increases the gasoline Figure 7 shows yield improve-
tions shown in Table 5. yield by as much as 3%, at the ments at constant coke. Here, the
Figures 6 and 7 are based on expense of coke and LPG. The trend is also clear: 4% more gaso-
four testing comparisons between magnitude varies between the four line and 2% more LPG, at the
four Ruby formulations and the comparisons, but the pattern is the expense of bottoms (slurry).
same formulations in Go-Ultra same: reduced coke and an equal- (Differences in dry gas are very
form. Results are plotted as yield to-smaller reduction in LPG. At small and are neglected here.) Also
differences between the catalysts. constant conversion, the scheme shown in Figure 6 is the simplified
reaction network for the case of
constant coke. In this case, the
Feed and e-cat properties and base case FCCU conditions improved yields must come from
the dot-dash reaction arrows. (Note:
Feed properties E-cat properties FCCU base conditions bottoms, that is material boiling
API 25.0 MAT activity 70.0 Feed rate, bpd 70 000 above 343°C, exist in both feed and
S, wt% 0.5 Ni, ppm 500 CTO 5.2 slurry.) This demonstrates the
Total N, wt% 1000 V, ppm 1000 CAR, tpd 7
coke-selective bottoms cracking
D1160 T5, °C 303 Cat inv, ton 400
T50, °C 398 RxT, °C 540 advantage of ADM-25 as well as
T95, °C 527 Max RgT, °C 735 the improved PoSD.
CCR 0.7 221°C + conv, wt% 72.3
Economic case study
Table 6 An economic case study was
performed for a 70 000 bpd gas oil
Economic case study
FCCU running against a maximum
regenerator temperature constraint
of 735°C. Table 6 gives the feed and
Base Case 1 Case 2 Case 3 e-cat properties and base case
Go-Ultra Go-Ultra Go-Ultra
conditions for this FCCU.
base cond. higher RxT higher feed
Feed rate, bpd 70 000 70 000 70 000 82 000 Table 7 shows the results of the
CTO 5.2 5.5 5.8 5.1 case study. In Case 1, the liquid
Coke, wt% 5.58 5.54 5.87 5.52 yield improvements alone of Go-
Delta coke, wt% 1.08 1.01 1.01 1.07
Ultra translate to a $28 000/day
RxT, °C 521 521 535 521
RgT, °C 735 724 735 735 improved margin. Note that this
221°C + conv, wt% 72.3 72.7 75.6 71.0 case does not take full advantage of
Margin, $/bbl Base +$0.40 +$0.68 +$0.10 the catalyst’s coke benefits; the
Gross margin, $/day Base +$28M/day +$48M/day +$68M/day
regenerator is loafing at 724°C, fully
11°C below maximum.
Table 7 In Case 2, the riser temperature is

18 CATALYSIS 2010 www.eptq.com

albermarle.indd 6 1/2/10 10:19:09


increased to 535°C until the regen-
erator temperature limit of 735°C is
At constant 5 Peterman R W, Distillate yield from the
FCCU: process and catalysts changes for maxi-
again reached. This improves
conversion to 75.6% and consequent
conversion, mization of LCO, Catalyst Courier, 32, 1998.
6 Yung K Y, Jonker R J, Meijerink B, A Novel
liquid yields result in a margin Go-Ultra increases and Fast Method to Quantify FCC Catalysts
Accessibility, ACS Preprints, Boston, MA, Aug
improvement of $48 000/day. In
Case 3, the feed rate is increased to the gasoline yield by 2002, 47, 3, 270–280.
7 Bruno K, Hakuli A, Fletcher R, Imhof P, FCC
82 000 bpd, again up to the maxi-
mum regenerator temperature of as much as 3%, at catalyst selection in diffusion limited operating
regimes, AM-03-58, NPRA Annual Meeting,
735°C. This scenario nets a total
margin improvement of $68 000/ the expense of coke 23–25 Mar 2003, San Antonio.
8 Yung K Y, Pouwels A C, Fluid Catalytic
day. Of course, Cases 2 and 3
assume the other equipment in the
and LPG cracking — a diesel producing machine,
Hydrocarbon Processing, Feb 2008, 79–83.
FCCU complex is capable of 9 Adkins B D, Knowles W, Bruno K, Francis J,
handling higher liquid yields, containing aluminium atoms in a Go-Ultra: a new line of catalysts, Hydrocarbon
Engineering, Sept 2009, 33–40.
higher air flows and so on. special coordination state. This
10 Duer M J, Introduction to Solid-State NMR
delivers as much as a 3% higher
Spectroscopy, Wiley-Blackwell, Oxford, 2004.
Conclusions gasoline yield, mainly from coke 11 Fukushima E, Roeder S B W, Experimental
For diesel and jet fuel production, and LPG (at iso-conversion). The Pulse NMR: A Nuts and Bolts Approach, Perseus,
Albemarle has developed the MD key to this is a careful balance of Reading, MA, 1981.
catalyst family. High accessibility, diffusional characteristics and the
as measured by AAI, is a key selectivity of the new active sites. Bruce Adkins is Global FCC VGO Specialist
feature of MD catalysts to maximise with Albemarle, Houston, Texas. After working
in government-funded coal and energy
bottoms conversion and LCO yield.
research, he became a researcher and research
Another key feature is a Z/M ratio AMBER MD, GO-ULTRA and RUBY are marks of
manager with Albemarle for 20 years before
specifically matched to accessibility Albemarle.
becoming Hydroprocessing Technical Service
and unit constraints. The use of MD Manager for the Americas. He has a PhD in
catalysts and appropriate process References materials engineering from the University of
conditions has been shown to 1 Jaffe S B, Extension of structure-oriented Kentucky. Email: bruce.adkins@albemarle.com
produce FCCU LCO yields as high lumping to vacuum residua, 2005, Ind. Eng. Ken Bruno is Global Applications Technology
as 42 wt%. Chem. Res., 44, 26, 9840–9852. Manager, FCC, for Albemarle, Houston, Texas.
On the gasoline side, Albemarle 2 Aris R, The Mathematical Theory of Diffusion He was a specialist in FCC catalyst technology
has introduced Go-Ultra, a catalyst and Reaction in Permeable Catalysts, Clarendon at the Amoco Research Center, then at Amoco’s
for maximum gasoline and superior Press, 1975. (BP) Whiting refinery before joining Albemarle.
3 Imhof P, Baas M, Gonzalez J, Fluid catalytic He has a PhD in chemical engineering from the
coke selectivity, while still preserv-
cracking catalyst evaluation: the short contact University of Notre Dame, Indiana.
ing high bottoms conversion. Its
time resid test, Catal. Rev., 2004, 46, 2, 151– Email: Ken.Bruno@albemarle.com
key elements are superior pore size 161. Yen Yung is Global Technical Specialist for
distribution tailored specifically to 4 Humphries A,Wilcox J R, Zeolite components Albemarle, Amsterdam, The Netherlands. He
gas oils and a matrix component, and matrix composition determine the FCC has over 20 years’ experience in FCC catalysts,
ADM-25, that leads to the forma- catalyst performance, Oil & Gas Journal, Feb technical service and troubleshooting.
tion of a new type of active site 1989. Email: yen.yung@albemarle.com

˜`Ê
܅>Ì 1
 ‡
-Êq

V>˜ , ̅iÊÃiÀۈViÊëiVˆ>ˆÃÌ
vœÀÊV>Ì>ÞÃÌʅ>˜`ˆ˜}
>˜`ʓiV…>˜ˆV>ÊܜÀŽ\

`œÊvœÀÊޜն
,   ]
ÊÊÊÊÊÊÊ 
 /]
ÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊ
ÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊ  , /°
1
 ‡
-ʘ`ÕÃÌÀˆ>‡
>Ì>ÞÃ̇-iÀۈViÊ“LÊ±Ê “`i˜iÀÊ-ÌÀ°ÊÓÇnÊ±Ê ‡xäÇÎxÊœi˜Ê±Ê*…œ˜i\ʳ{™ÊÓÓ£ÊÇ£ÇLJÓäÓÊ±Ê “>ˆ\Ê, JLÕV…i˜‡ˆVð`i

PTQ1-4-4c_X3.pdf 1 23.02.2009 10:42:13 Uhr


www.ptqenquiry.com
for further information
www.eptq.com CATALYSIS 2010 19

albermarle.indd 7 1/2/10 10:19:24


JM_725 Ad Concepts 2009 9/9/09 16:20 Page 1

Johnson Matthey Catalysts offer a range of


specialist, high performance products and
services for refineries.

Activate your potential

Catalysts for hydrogen Services to improve plant Hydroprocessing catalysts in Specialist measurement and
production, the market leaders. performance and profitability. conjuction with SINOPEC. diagnostic services; from
column scans to FCC studies.

Speciality purification solutions, A complete service package


absorbents and process for PURASPECJM™ catalysts.
technology for sulphur, chloride,
and mercury removal.

www.ptqenquiry.com
www.jmcatalysts.com/refineries for further information

UK USA
Tel +44 (0)1642 553601 Tel +1 630 268 6300
Fax +44 (0)1642 522542 Fax +1 630 268 979

j matthey.indd 1 10/9/09 09:21:48


Active site developments for improved
productivity
Developments including improved active site dispersion and assembly are designed
for higher-valued operation in a range of processing units

Kevin Carlson
Criterion Catalysts & Technologies

W
hile global economies
begin to show signs of Centera
recovery, refinery margins 
2ELATIVE($3VOLUMETRICACTIVITY 

have continued to challenge even


the most prescient planning and Centinel

Gold
scheduling groups in our industry.
As global/regional margins and Centinel Ascent

Plus
demands fluctuate, and capital and Ascent
operating budgets decline, the 
global drive towards clean fuels
continues resolute, requiring refin-

ers to squeeze the most from their       
existing units. In this environment,
flexibility becomes one of the
desired process capabilities to Figure 1 Relative performance comparisons
maximise refinery economics as
margins continue to shift with feed produced economic value through Gold developments and further
and product slates. Given that over- improved run lengths and reduced increases performance through
all unit performance is determined generation of low-value products. innovations in active site design
by capability, reliability and dura- Criterion Catalysts & Technologies and manufacturing technology.
bility, technology providers are has developed Centera technology This combines improvements in
challenged to assist refiners to to meet the current challenges. In catalyst support with advances in
increase performance and flexibility refining applications, including Type II active phases. Further
without sacrificing run length. ULSD production, hydrocracker advancements in the manufacture
In recent years, the ultra-low pre-treat (HCPT) and fluid catalytic of Type II technology catalysts
sulphur diesel (ULSD) unit has cracker pre-treat (FCCPT), the prod- generate gains in both initial kinetic
proven to be a versatile refining uct range has significant increases activity and enhanced retention of
asset, demonstrating the ability to in activity and stability, and increas- activity, resulting in a product range
maximise diesel volume when ing unit capabilities to provide with the following characteristics:
margins are favourable and the additional flexibility without sacri- • Increased active site dispersion
flexibility to alter operations when ficing reliability. Improved support technology lead-
the opportunities shift. However, Combined with customised cata- ing to better dispersion of MoS2
the activity of many ULSD units is lyst system design, this technology crystallites in the active catalyst
constrained, so higher-performance can generate higher performance. • Maximum sulphidation Manufact-
catalysts are required to unlock the Relative catalyst performance uring technology leading to the
full potential of the unit. With the comparisons are shown in Figure 1. highest degree of sulphidation,
reduction of cracking margins, FCC These solutions can improve overall forming ideal Type II active sites
pre-treat operations have required refinery economics by providing • Optimal assembly of active sites
the flexibility to shift from a role in the flexibility to process heavier Optimised decoration of MoS2
maximum hydrogenation to a feed feeds, maximise LCO blending into edges with promoter metals leading
uplift service, enabling less valuable the diesel pool and/or extend unit to increased catalytic activity.
products to be shifted to clean fuels cycle length. These conclusions are based on
production. Increased catalyst activ- detailed characterisation of the
ity combined with the ability to Centera catalyst technology active catalysts using various
process difficult feeds with high The technology builds on elements analytical techniques. The ability to
levels of contaminants have of Criterion’s Ascent and Centinel perform molecular active site

www.eptq.com CATALYSIS 2010 21

criterion.indd 1 1/2/10 10:28:12


MoS 2 Dispersion of Criterion Catalysts
Centera exceeds Centinel Gold MoS 2 dispersion
the degree of stacking was deter-
mined by counting the number of
layers in each individual supported
particle of MoS2.
A typical TEM image of the NiMo
DN-3630 is shown in Figure 2.
Evaluation of the image shows that
the sample consists of supported
(75%) and unsupported (25%)
domains, each having particle sizes
of around 35 Å. The average degree
of stacking is 1.5 for the supported
Particle Diameter
and 2.1 for the unsupported parti-
Centinel: 40 – 45 Å
cles. An estimation of the particle
Centinel Gold: 30 – 35 Å
size based on additional extended
Centera: 25 – 35 Å
X-ray absorption fine structure
(EXAFS) analysis leads to diameters
10 nm
of around 25 Å, indicating that the
range given in the figure (25–35 Å)
Figure 2 Molybdenum sulphide dispersion in Centera and Centinel catalysts is a representative characterisation
of the improvement seen with the
analysis and characterisation gives turing protocols, both of which are technology when compared with
the framework for continued key to the technology’s performance. earlier catalyst generation. The
advances. During the development pro- double-layered particle displayed in
gramme, a TEM protocol, Figure 3 is thus a realistic descrip-
Increases in MoS2 particle dispersion specifically developed for character- tion of the typical structure of
Catalyst performance is a function ising the catalyst, provided many sulphided DN-3630.
of many parameters, one of which digital images and measurements
is metal particle interactions with for each preparation, giving accu- Maximised sulphidation with
the alumina support. Transmission rate analysis of metal sites. increased activity
electron microscopy (TEM) meas- During development, the quanti- The technology applies innovative
urements and imagery can allow fication of dispersion of active manufacturing techniques to
for detailed observation of particle metals sites was detailed with an provide for complete sulphidation
formations of active metals and observation of MoS2 crystal slab of the active metals. A critical
their dispersion on the alumina lengths. Measurements of the slab element in achieving full sulphida-
support. This observation and lengths and the degree of stacking tion is to convert the Mo-O in the
measurement provides the ability (layers) of the slabs were performed interior of the particle, thus mini-
to further develop and optimise by examining hundreds of individ- mising its interaction with the
EXAFS: Full Sulfidation of Molybdenum
catalyst formulations and manufac- ual slabs per sample. Furthermore, alumina support. Observations and
measurements of the Mo-S struc-
tures within the active particles
First shell Mo-S have been detailed utilising EXAFS
coordination number = 6 spectroscopy. During the catalyst
technology’s development, EXAFS
spectra of metals particles were
obtained, to provide the Mo-S
geometry and the extent of interac-
tions with the support material. The
DN-3630
Bulk MoS2 inset in Figure 3 shows the Fourier
transform of the respective V func-
tion from the sample. The first shell,
at about 2 Å, is attributable to Mo-S
contributions. It is identical to that
of bulk MoS2 and corresponds to
the ideal coordination number of
six, indicating that every Mo atom
is surrounded by six sulphur atoms
in trigonal prismatic fashion. Thus,
Degree of sulfidation as in bulk MoS 2 the catalysts are fully sulphided
with only weak van der Waals
Figure 3 Full sulphidation of molybdenum interactions with the support,

22 CATALYSIS 2010 www.eptq.com

criterion.indd 2 1/2/10 10:28:20


NOChemisorption
: Promoter Edge
Decoration
allowing maximum performance
and complete utilisation of active NO adsorbs to
Ni and Mo Ni only
metals.
Full sulfidation
Optimal use of promoter metals Incomplete sulfidation Optimal edge decoration
An additional feature of the tech-
nology is the ability to “design” the
active metals sites by influencing
the placement of the promoter
metals into the resultant Mo-
Promoter Metal-S structure,
allowing for enhanced performance
of the active sites.
NO (nitric oxide) adsorption is a
precise method used to characterise
accessibility to the catalyst active
sites and to determine the state of
the sulphided catalyst Mo-S sites.
CENTINEL CENTERA
The method is based on the obser-
vation that the N-O stretching Figure 4 NO chemisorption: promoter edge decoration
vibration, as measured with Fourier
transform infrared spectroscopy for HDS targeted units, providing developed to address the capabili-
(FTIR), is structurally sensitive and either extended cycle lives or ties and the operational constraints
enables discrimination between NO reduced product sulphur for heavy of processing units.
molecules adsorbed to Ni, Co or VGO processing. When comparing The technology and manufactur-
Mo. With Centera DN-3630, we can previous generations of CoMo ing innovations applied in the
observe that the edge surface of a FCCPT products, the move from initial Centera product releases
sulphided metals particle exposes DC-200 to Centinel DC-2118 have already shown capability for
only coordinatively unsaturated Ni provided improved FCC feed qual- further gains in refinery perform-
centres. In this regard, DN-3630 ity via increased HDN, while the ance. As development continues,
displays unique surface structural move to Ascent DC-2551 provided additional product applications will
properties. Molecular modelling increased performance through service hydroprocessing needs in
calculations of different surface increased contaminant uptake and the refinery.
ensembles have shown that the reduced deactivation rates. Centera
Centinel, Ascent and Centera are marks of
active site edges have at least two, DC-2650 enables continued unit
Criterion Catalysts & Technologies.
but possibly three, unsaturated stability with gains in HDS capabil-
promoter Ni centres in a row. It is ity. Figure 5 shows the relative Kevin D Carlson is Global Business Manager,
the full presence of these specific performances of these catalysts. FCC Pretreat & Sentry Products, with Criterion
active structures and the absence of Catalysts & Technologies in Houston, Texas. He
less active, unsaturated Mo-based Conclusion has worked in the industry for over 18 years and
edge structures present in all other With changing objectives and eco- has been an author of a number of technical
catalysts that is responsible for the nomics, catalyst systems have been papers. Email: Kevin.Carlson@cri-criterion.com
increased activity observed with the
catalysts. Figure 4 shows the differ-
ences in the edge structures of 
Ni-promoted Centera and Centinel  ($3PERFORMANCE
catalysts. ($.PERFORMANCE

2ELATIVEVOLUMETRICACTIVITY

FCC pre-treat gains 


Centera was first introduced in 
ULSD service with DC-2618 and 
DN-3630, and its applications have

been expanded into heavier feed
applications, such as FCC pre-treat, 

with the initial release of DC-2650. 


DC-2650 was designed for enhanced 
HDS activity in low- to moderate- 
pressure VGO units, and is also $#  $#  $#  $# 
combined with Ascent catalysts in a #ENTINEL !SCENT #ENTERA

custom-designed catalyst system.


DC-2650 delivers significant gains Figure 5 Relative performances of Criterion catalysts

www.eptq.com CATALYSIS 2010 23

criterion.indd 3 1/2/10 10:28:31


- 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. www.ptqenquiry.com


for further information
Tricat GmbH Catalyst Service Bitterfeld
260 Schilling Circle OT Greppin
Hunt Valley, MD 21031 Tricat-Str. (ChemiePark Areal B-Ost)
USA 06803 Bitterfeld-Wolfen
Phone: (410) 785 7900 Germany
Fax: (410) 785 7901 Phone: +49 3493 75910
Internet: www.tricatgroup.com Fax: +49 3493 75999
E-mail: sales@tricatgroup.com Internet: www.tricatgroup.de
E-mail: info@tricatgroup.de

tricat.indd 1 29/1/10 12:46:05


Ring-shaped catalysts make the grade

Shape-optimised topping materials, together with ring-shaped hydroprocessing


catalyst, provide a solution for units that are pressure drop limited

Brian Moyse
Haldor Topsøe

O
il refineries in the US,
Canada and other regions
of the world are showing
their age. Within the worldwide
oil refining system, there are refin-
eries in operation that are more
than 100 years old. These are,
indeed, the senior citizens of our
industry. So, it comes as no
surprise, then, that corrosion is a
large and expensive issue for refin-
ers to address on a daily basis. A
refinery can end up spending
millions of dollars every year deal-
ing with corrosion issues.
Obviously, fouling affects many
different aspects of refinery opera-
tions.1 However, this article will
deal with only one aspect of foul- Figure 1 Debris from distributor tray
ing, which inevitably results from
corrosion. We will deal exclusively fouling sources besides iron corro- fairly large-sized flakes and deposit
with the subject of fouling of fixed- sion products: on top of the catalyst bed.
bed catalytic reactors used for • Catalyst fines and dust (FCC- Even if feeds containing cracked
hydroprocessing of the whole range derived streams) stocks are fed directly to the unit
of refinery streams. • Coke fines (coker-derived and not via a tank, there are still
streams) issues to be faced. In these cases,
The problem • Sediment (tar sands, inefficient the olefins and diolefins react read-
As regards catalytic hydroprocess­ desalter operation) ily in the reactor inlet to form
ing units, it is desirable to maximise • Salts (Na, K, P, etc) polymeric materials. These gummy
on-stream time and achieve the • Carbonaceous material spalled reaction products provide the
desired cycle length without from heater (polymeric) “glue” to cement together any
out­ages or interruption. It is not • Metal sulphides (Ni, V, etc). foreign matter entrained in the feed
unusual to see some operations It should also be pointed out that and can ultimately form a crust on
halted due to unfavourable pres- treating cracked feedstocks or top of the bed. Of course, this
sure drop across all or part of the blends containing cracked material results in high delta P, flow maldis-
catalytic reactor. Most often, this is presents an additional complication. tribution and possibly hot spots in
caused by feedstock contaminants, This is because cracked stocks the catalyst bed.
which usually include a very well- contain olefins and, in particular, As everyone is no doubt aware, a
known corrosion product — iron. diolefins. If such feeds are exposed catalyst bed makes for a great filter.
This is shown in Figure 1, a photo- to oxygen, as for example when Unfortunately, this undesirable
graph of debris recovered from the using non-blanketed feed tanks, facet brings with it an increase in
top of a gas/liquid distributor tray. gums will form. The gums can pressure drop as the catalyst bed
Note the variation in the size cause fouling of the heater itself fouls. Feed filtration is a good idea
of particles. and the heat exchange train. in such cases, because whatever
It should also be mentioned that, Deposits can build up in the heater material can be removed in this
in hydroprocessing, there are other and will eventually slough off as step means that much less is passed

www.eptq.com CATALYSIS 2010 25

haldor.indd 1 1/2/10 10:34:25


column shows the useful void or
space available for deposition of
 contaminants before pressure drop
increases exponentially. This
 number is derived by subtracting

22%, the point at which delta P
rises significantly, from the SOR
clean bed value. This is the actual
2EACTORD0 PSI


maximum void space available to

deposit debris in the top catalyst

layers of the bed.
Spheres are the worst possible
 choice in cases where pressure drop
is an issue. It is permissible to use

spherical support media at the
 bottom of the bed or reactor, but
              they should definitely not be used
-ONTHSON LINE
as bed topping in any units experi-
encing pressure drop issues.
Figure 2 Unit pressure drop in a naphtha hydrotreating unit We might also mention that even
units that do not have pressure
to the reactor. The average size of hydroprocessing units worldwide. drop issues could actually benefit
the filter element employed is 25 The only difference is the actual from replacing spherical topping
micron. This means that even units values of psi and on-stream time. with a high-void type product. This
currently employing feed filtration The shape of these curves is exactly would protect the unit from unfa-
are not necessarily “out of the the same. vourable pressure drop excursions
woods”, since particles smaller than Figure 2 also shows that, after a in cases of plant trips, refinery
the nominal filter element size are little more than a year on-line, the upsets and power outages.
going to pass through to the reac- reactor pressure drop rises expo- Normally, if a refinery experiences
tor. Also, it appears that for any nentially and requires that the cycle an emergency trip due to a power
given industrial application, the be terminated and the unit shut outage, for instance, the pressure
micron size of the filter is selected down for corrective action. This drop across the hydrotreater reactor
not for its efficacy but rather on situation can and will occur when will be higher on restart than it was
what the refiner considers to be an the bed void fraction, either overall prior to the event.
acceptable replace­ment or back- or in a particular layer, is reduced
wash frequency. from its start-of-run (clean bed) Early solutions
Figure 2 shows an example from value down to 22%. The range is One of the early approaches to miti-
an actual naphtha hydrotreating actually between 20% and 25%, so gate this kind of problem was the
unit, depicting the development of we selected 22% as representative. use of scale traps or “trash baskets”
the reactor pressure drop as a func- When this occurs, pressure drop in the top of the reactor. It is obvi-
tion of months on stream. It can be takes off exponentially, resulting in ous that these baskets could only
seen that normally the reactor pres- channelling, catalyst bypassing, hot hold back particles larger than the
sure drop is predictable and follows spots and so on. mesh size used in their construc-
the variations typically encountered Table 1 shows start-of-run, clean tion.2 Anything of a smaller size
as a result of the usual fluctuations bed void fraction of materials that would naturally pass through into
in feed rate. This is normal and to are typically loaded into a reactor, the catalyst bed.
be expected. This plot is also typical be they catalytic or inert topping/ Also, some primitive grading
of hundreds that we have seen from support media. The right-hand systems were initially used on the
top of the lead catalyst bed. These
Clean bed void fraction and pressure drop all involved the use of layers of
inert spheres on top of the catalyst
bed, larger on top and smaller
Catalyst shape SOR void, % SOR - 22 = useful void, %
Spheres 33 11
below, actually the opposite of
Cylinders 40 18 what was used on the bottom of
Threelobes 42 20 the bed.
Quadralobes 45 23 In those cases where feed filtra-
Ring catalysts 53 31
High-void inerts 55–65 33–43
tion was employed, the size of the
filter openings chosen would affect
Spheres provide lowest void fraction its efficacy, on-stream time and how
often it had to be replaced. The
Table 1 average size of feed filters used in

26 CATALYSIS 2010 www.eptq.com

haldor.indd 2 1/2/10 10:34:42


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 -PXQSFTTVSF ýYFECFEQSPDFTT—JNQSPWFTQSPEVDU
  ZJFMEBOERVBMJUZ

 t *ODSFBTFETVMGVSSFNPWBMBOESFEVDFEOBQIUIBZJFMET

 t 1SPQSJFUBSZDBUBMZTU JTPNFSJ[FTXBYZOQBSBGýOTJO
  HBTPJM UPMFSBOUUP)24 BNNPOJB BOEPSHBOJD4BOE/

 t 4UBOEBMPOFPSJOUFHSBUFXJUIPUIFST

&YYPO.PCJM3FTFBSDIBOE&OHJOFFSJOH
$PNQBOZT.*%8 5FDIOPMPHZ
*NQSPWFT1SPEVDU:JFMEBOE2VBMJUZ

* MIDW is a trademark of ExxonMobil Research and Engineering Company


** Available from ExxonMobil Catalyst Technologies LLC

www.ptqenquiry.com
for further information
3225 Gallows Road, Fairfax, Virginia 22037-001, USA
www.exxonmobil.com/refiningtechnologies • +1-703-846-2568 • fax +1-703-846-3872 • technology.licensing@exxonmobil.com

exxon.indd 1 29/1/10 12:47:28


Table 1). For this reason, the
company pioneered the use of high-
void inert topping materials for the
top of the bed, for a total void
advantage from the reactor inlet
down.
Haldor Topsøe’s standard topping
material is called TK-10. It consists
of a 5/8 inch domed tablet with
seven axial holes. This material has
55% void content, and is inert and
robust. TK-15 is intended for units
with high gas velocities and/or no
gas/liquid distributor tray such as
naphtha hydrotreaters. The 1 1/2
Figure 3 Topsøe high-void products inch size TK-15 is also inert, with
65% void, 10% higher than TK-10,
the industry is around 25 microns. shape of the catalyst particle. and a loading density that is almost
Since iron rust particles are around Initially, the rings were made in 50% greater than TK-10.
5 microns or less in size, a lot of 1/4 and 3/16 inch OD size, but, as As Figure 1 shows, corrosion
material is going to pass on through experience was gained and the products, in particular iron, range
to the process. technology improved, 3/16 and 1/8 in size from sub-micron to rather
These solutions worked to a inch became the standard sizes. large-sized pieces. So even units
degree in some cases, and in others, The selection of ring catalysts employing a feed filter will have iron
while helping to prolong the on- from Haldor Topsøe includes nickel scale entering the unit as particulate
stream factor, really did not solve only, molybdenum only, nickel matter entrained in the liquid.
the problem or provide the desired molybdenum and cobalt molybde- To address this type of situation,
long-term effect. num types. They are also available TK-25 TopTrap was developed.
In our experience, almost all units with different activity levels, pore This material has the form of 1/2
with pressure drop issues can bene- size, surface area, porosity and pore inch Quadralobe with an axial hole.
fit from the use of a graded catalyst size distribution, as well as func- It has a 61% particle void fraction
bed without resorting to the use of and loads in the reactor at 30 lb/ft3.
trash baskets. This is actually good This product is intended to pick up
news, since installation and retrieval Almost all units with particulates in the feed. Larger-
of trash baskets in and from the sized material deposits in the void
reactor is a maintenance nightmare. pressure drop issues in the interstices between the indi-
In addition, trash baskets contribute vidual quadralobes.
to maldistribution of reactants can benefit from Fines or smaller-sized materials
in the catalyst bed, which is enter the TK-25 TopTrap pore
undesirable. the use of a graded system and are trapped within the
There are certain situations where structure of the particle itself (see
conditions dictate that trash baskets
catalyst bed without Figure 4). The internal particle void
are required as part of the overall resorting to the use of TK-25 TopTrap is 25%, so that
solution to the delta P problem. the total void in this product is
However, these cases are rare. of trash baskets greater than 86%.
This was the state of the technol-
ogy until the late 1970s, when How they work
Haldor Topsøe invented ring- tionality for picking up poisons. The problem we are dealing with
shaped hydroprocessing catalyst, In addition, there are shape-opti- here is, basically, the deposition of
which, in turn, revolutionised bed mised topping materials for the particulates on and between the
grading systems (see Figure 3). upper layers of the first bed. This is catalyst particles in the bed. When
Researchers were able to take the placed on top of the graded bed, so the void fraction is reduced to 22%,
conventional high surface area these kinds of materials should also or thereabouts, saturation level is
alumina catalyst carrier, promoted be carefully selected so that they achieved and delta P rises
with either nickel molybdenum or are compatible with, and part of, exponentially.
cobalt molybdenum, and produce a the overall grading strategy. Any solution to this problem
catalyst in the shape of a small This is a very important and inte- must therefore be based upon
Raschig ring. These catalysts were, gral part of a graded catalyst bed “opening up” the void volume in
and are, identical in properties to system, since it makes no sense to the uppermost layers of the catalyst
their cylindrical or lobed counter- use ceramic spherical hold-down on bed. This includes both the catalyst
parts, the only difference being the top of high-void ring catalyst (see and hold-down or topping material.

28 CATALYSIS 2010 www.eptq.com

haldor.indd 3 1/2/10 10:34:59


Such an approach will provide an
increase in the inventory between
the particles as well and in the pore
volume of the material to store
contaminants entering the reactor
with the feedstock (see Table 1).
Since the catalyst bed also functions
as a deep bed filter, more void
volume equals more on-stream time
before the same reactor pressure
drop limit is reached.

Conclusion
Graded catalyst beds have come of
age. They have clearly demon-
strated that the combined use of
shape-optimised topping materials Figure 4 TK-25 TopTrap topping material
and ring-shaped hydroprocessing
catalyst provides the optimal solu- ing unit monitoring and analyses of hydrotreating unit, AM-00-33, NPRA Annual
tion for units that are pressure drop samples taken from the reactor at Meeting 26–28 Mar 2000, San Antonio, Texas.
limited. end-of-run. This offers the refiner
Ideally, the graded bed should be the most viable approach to under- Brian Moyse is Technical Manager for
Hydroprocessing Catalysts and Technology in
composed jointly between the standing and thus the opportunity
the Refinery Sales Group at Haldor Topsøe. He
refiner and catalyst supplier, and to solve any delta P issues.
is an international expert in the application of
grading should be selected based grading technology and hydrotreating. He has
References
upon feedstock analyses, actual 1 Wright B, No room for foul play, published and presented numerous papers in
contaminants and poisons encoun- Hydrocarbon Engineering, Sept 2009. the field of hydroprocessing. He was awarded
tered and optimised over time. This 2 Johnson B, Basket case: graded catalyst beds the NPRA Lifetime Service Award in 2006.
requires a planned approach includ- solve pressure drop problems in a naphtha Email: bmm@topsoe.com

Catalyst handling
and mechanical work
With more than 25 years
of experience
Mourik International
is a world leading supplier
for catalyst handling,
vessel and reactor services.
We think globally, but act Mourik International B.V.
locally, as we operate Nieuwesluisweg 110
through a world-wide 3197 KV Botlek-Rotterdam
network of subsidiary The Netherlands
organizations. Tel. +31-10-296 56 00
Fax +31-10 296 56 09
Offices: Europe, Middle East, Asia, E-mail: mint@mourik.com
North America, South America

mourik.com
www.ptqenquiry.com
for further information
www.eptq.com CATALYSIS 2010 29

haldor.indd 4 1/2/10 10:35:11


• Mesh numerous alloys & plastics Wide variety of random packing types, • Valve (numerous options) • Baffle
• Vane (Chevron) • Double Pocket Vane sizes and materials in stock! Ask us how • Sieve or perforated • Dual flow
our patented SuperBlend™ 2-Pac can
• Candle or Fiberbed • Insertion MistFix ®
increase your capacity and efficiency. • Bubble cap trays • Cartridge trays

MIST ELIMINATORS RANDOM PACKING TRAYS

Need A Catalyst Bed Support....NOW!


Catalyst Bed Supports Why wait 10 ,15, 20 weeks for delivery. Ask AMISTCO Separation
Basket Strainers
Products about our Fast Track delivery for our full line of profile
Nozzles
Outlet/Inlet Baskets wire screen products. With in-house engineering support and
Distributors fabrication, we can use your existing drawings or modify them to
Hub and Header Laterals
improve your process.

www.ptqenquiry.com
for further information

www.amistco.com • 24hr EMERGENCY SERVICE • 1-800-839-6374

DISTRIBUTORS & SUPPORTS COALESCERS STRUCTURED PACKING


Manufactured to customer • Oil water separations • Haze removal Woven, sheet metal, and knitted
specifications or engineered to from fuels • Removal of tower wet reflux structured packing. Built to spec
meet performance requirements • Caustic treater applications or performance requirement.

WedgeWire PTQ indd 1 1/28/08 6:57:07 PM


amitsco.indd 1 2/2/09 14:23:42
Reducing sulphur oxide emissions from
the FCC unit
Application of a SOx-reduction additive enabled a refiner to meet a pressing
requirement to reduce emissions from its FCCU

Colin Baillie, Renaud Kieffer and Maria Luisa Sargenti


Grace Davison Refining Technologies Europe

A
lthough the amount of
sulphur oxide (SOx) emitted
SOx removed Treated with H2S
from FCC units (FCCUs) is from flue gas in the sour gas
relatively small, it still accounts for
Products (with H2S)
most of the SOx released from oil
refineries. The major options for
reducing SOx emissions from the
Stripper:
FCCU include flue gas scrubbing Hydrolysis of Mg sulphide
Flue gas
(FGS), hydrodesulphurisation (with SOx) MgS (s) + H2O (g) = MgO (s) + H2S (g)
(HDS) and the use of a SOx-
reduction additive. FGS requires a Regenerator: Stripping steam sulphate
significant capital investment and Formation of SOx
the operating costs are high. In S (coke) + O2 (g) = SO2 (g)
addition, discarding the spent treat- SO2 (g) + 1/2 O2 (g) = SO3 (g)

ing compounds such as lime, Riser:


caustic soda or other compounds Reduction of Mg
Formation of
can be problematic. HDS requires Mg sulphate MgSO4 (s) + 4 H2 (g) = MgS (s) + 4 H2O (g)
MgSO4 (s) + 4 H2 (g) = MgO (s) + H2S (g) + 3 H2O (g)
the highest capital investment, but SO3 (g) + MgO (s) = MgSO4 (s)

also provides improved yields as


well as lower SOx emissions.
Therefore, refiners often choose Regenerator air
Feed (with sulphur)
SOx-reduction additives, as they
Oxidising Reducing
require very little capital invest- environment environment
ment and are extremely effective in
reducing SOx emissions.
Due to new, more stringent, local
environmental legislations, Refinery Figure 1 Catalytic SOx reduction using Super Desox
A, located in western Europe, was
confronted with the challenge of magnesium-alumina spinel struc- in the FCCU. A schematic diagram
reducing SOx emissions from its ture, which has been successfully of the oxidation and reduction reac-
FCC regenerator by 30%. This is used by more than 70 refiners tions of sulphur compounds in the
highlighted in Table 1, which shows worldwide. To understand why this FCCU is shown in Figure 1.
daily and yearly SOx bubble limits magnesium-alumina spinel struc- In the regenerator, sulphur
for the refinery before and after ture works so well, it is necessary contained in the coke is oxidised to
1 January 2009. to examine the mechanism for the SO2 and SO3. Both of these sulphur
Faced with the various options catalytic reduction of SOx emissions oxides are absorbed on the SOx
for tackling SOx reduction, the refin- additive, where SO2 is further
ery chose the additive route and SOx bubble limits for Refinery A oxidised to SO3 in the presence of
began using Super Desox in oxygen and an oxidation catalyst ­­in
December 2008. The results of Average SOx bubble limit
the additive. In the regenerator, the
this trial will be discussed in this Daily, Yearly, SO3 reacts with magnesium oxide
article. mg/Nm3 mg/Nm3 in the alumina spinel and is
Before 1 Jan 2009 1700 1400 converted to magnesium sulphate.
After 1 Jan 2009 1000 850
Super Desox Compared to free magnesium
Super Desox was commercialised in oxide, the spinel in Super Desox is
2003 and is based on a patented Table 1 more effective for this sulphate

www.eptq.com CATALYSIS 2010 31

grace.indd 1 1/2/10 11:09:49


Matrix plot of SOx.raw.3%O2; S.charge (%); RSV (t/h); O2 (%)

        




SOx.raw 
Our knockout drum needs 3% O2 
a mist eliminator! 
S.charge (%) 



RSV (t/h)


O2 (%)

Can we add one without Figure 2 Linear dependency of SOx emissions with feed sulphur content, vacuum residue
overhauling the vessel? throughput and flue gas oxygen content
IT’S A MORE COMMON problem than
you might think. A vertical knockout formation and a stable sulphate feed sulphur content and SOx
drum removes free liquid from a is formed under regenerator concentration.
certain gas stream. But at the time the conditions. Various scientific studies have
plant was built, a mist eliminator was In the reactor, the magnesium shown that the fraction of
not considered necessary. Now mist sulphate is reduced by hydrogen to thiophenic sulphur in the feed has
is carrying over and causing trouble
downstream. There is no manway, so
form magnesium sulphide. Super a direct impact on the coke sulphur
adding a conventional mist eliminator Desox plays an important role here content deposited on spent catalyst
would require cutting the vessel open. too because the magnesium in the and thus on SOx emissions.
alumina spinel is less basic than Therefore, changes in uncontrolled
free magnesium oxide, thereby SOx emissions can be accurately
making the sulphate easier to estimated by measur­ing thiophenes
reduce. In addition, vanadium on in the feed. However, it is uncom­
the spinel structure further mon for refiners to be able to
decreases the stability of the magne­ measure thiophenic feed content on
sium sulphate, especially under a regular basis. It is therefore
reducing conditions and in the recommended to estimate uncon­
presence of steam. The magnesium trolled SOx emissions by measuring
sulphide is then hydrolysed to a feed sulphur or even better with
magnesium oxide in the stripper, slurry sulphur (the slurry being the
which returns to the regenerator to most aromatic product of the
complete the catalytic cycle. FCCU).
0LVW)L[ŒLQVHUWLRQPLVWHOLPLQDWRU
Due to the operating variables
 ,QVWHDGLIWKHÀDQJHGH[LWQR]]OH Establishing a baseline for that can affect SOx additive
LVDWOHDVWVL[LQFKHVLQGLDPHWHU\RX uncontrolled SOx emissions performance, it is necessary to
FDQVLPSO\KDQJDSDWHQWHG0LVW)L[ΠSeveral operating variables have establish a baseline of uncontrolled
mist eliminator inside, securing it been identified as having significant SOx emissions to be able to calculate
EHWZHHQWKHÀDQJHV effects on the performance of SOx- the true SOx reduction caused by
Even when there is a manway,
reduction additives.1-3 Some of these the additive. Such a baseline is
0LVW)L[DYRLGVHQWU\ZHOGLQJDQG
UHFHUWL¿FDWLRQLQUHWUR¿WV)RUQHZ include the presence of combustion calculated by performing correl­
and old vessels, it drastically cuts promoters, the ratio of catalyst ations for every relevant parameter
installation cost and downtime. circulation rate to unit catalyst based on pre-trial unit data. To
inventory, unit temperatures, avail­ establish a baseline for Refinery A,
Read more on this topic at ability of oxygen in the regenerator, three months’ unit data were used.
www.amistco.com
SO2* = 291 + 1729 S.feed (%) + 5.04 RSV (t/h) - 64.5 O2 (%)

S = 34.1255 R-Sq = 84.2% R-Sq(adj) = 82.9%


3KRQH‡)D[ * Raw data obtained by Refinery A
amistco@amistco.com
KU(PHUJHQF\6HUYLFH
Figure 3 Correlation of SOx emissions

www.ptqenquiry.com
32 CATALYSIS 2010 for further information www.eptq.com

grace.indd 2 2/2/10 12:56:57


It was observed that, in general,
parameters such as slurry density
and slurry sulphur content corre- 1.8 wt% Super Desox
 0.5 wt% Super Desox
lated well with SOx emissions. 29% SOx reduction 66% SOx reduction
However, due to a lack of slurry 
measurements, it was decided to 

3/MG.M
use feed properties instead. In this
respect, the main parameters turned 
out to be feed sulphur content (S. 
charge), vacuum residue through- 0.8 wt% Super Desox

put (RSV) and flue gas oxygen 24% SOx reduction
content, as shown in Figure 2.  1.5 wt% Super Desox
Using the data from the refinery, 52% SOx reduction

the correlation shown in Figure 3

 









  

 






  

 

 

  

   

 
was established using the raw SO2

























values from the pre-trial period






























(from 21 October 2008 to 12 January
.ORMALISED3/ / #ALCULATED3/
2009). The high value of R2 (84.2)
signifies the good correlation
obtained.
As would be expected, it can be Figure 4 Flue gas SO2 emissions before and during Super Desox additions. Flue gas SOx
seen that increased feed sulphur was measured as SO2 by an online analyser, and the SO2 values are normalised to 3%
plays a major role in the above excess oxygen in flue gas
correlation; SOx emissions increase
with increasing feed sulphur. In around 19 January, indicated by the 0.5 wt% during 16 February to 29
addition, increased residue through­ subsequent and obvious drop in March. The average SOx-reduction
put can be expected to increase SOx flue gas SOx emissions illustrated in performance during this period
emissions, which is indeed observed Figure 4. remained high at 29%. In the subse-
in the correlation. The incorporation The effect was immediate, with quent period, between 30 March
of more residue results in increased average SOx reductions of 24% and 30 June, it was decided to
SOx emissions due to the more observed within the period 19 increase pre-blending levels to
refractive nature of its sulphur- January to 15 February, during 1.5 wt% to gain an idea of just how
containing compounds, which will which the additive was pre-blended much SOx reduction could be
thus end up more readily in the at 0.8 wt%. This can be effectively achieved, and during this period
coke on catalyst. The other impor- considered as a base-loading period, the average reduction rate increased
tant variable in the correlation is during which a stationary concen- to 52%. From 28 September,
flue gas oxygen content. As oxygen tration of additive is reached in the the refinery started pre-blending
increases, the coke yield decreases inventory. Super Desox at 1.8 wt% due to
and, therefore, less sulphur reaches It was subsequently decided to increased feed and slurry sulphur
the regenerator with coke. reduce pre-blending levels to levels, resulting in impressive

Trial of Super Desox at Refinery A


The performance of Super Desox at
Refinery A is estimated by compar- 

ing measured SOx emissions with 


predicted uncontrolled SOx emis-
3ULPHURCONTENT


sions; the latter are based on the
correlation in Figure 3. In order to 
reach the target of reducing SOx

emissions by 30%, Grace Davison
recommended using Super Desox 
at 0.8 wt%, with a prior period of 
base loading to accelerate the intro-
duction of the additive in the 

inventory. However, the refinery


 

 
 

 
 

 

 
 

 

 

 

 

   

 

 

 

 

 

  

 

   
 

 

 
   

 

  




decided not to have a base-loading














period and began using 0.8 wt% of


Super Desox directly. The additive &EEDSULPHUR 3LURRYSULPHUR

pre-blended with fresh catalyst was


delivered to Refinery A on 12
January 2009 and entered the unit Figure 5 Feed and slurry sulphur content

www.eptq.com CATALYSIS 2010 33

grace.indd 3 1/2/10 11:10:03


ExperienceDifference-CE.qxd 3/2/06 4:52 PM Page 1

Experience
the Sabin difference
for precious metal
catalyst recovery
and refining.
We turn science
into art for highest
possible returns
and added value.
The science of refining precious metal catalysts
is straightforward: we simply use the most
technically advanced analytical, assaying, and
processing equipment and procedures available.
The art of refining precious metal catalysts to
return maximum value is an entirely different
story. That involves knowledge, experience, and
expertise gained from nearly 60 years of
successfully serving thousands of organizations
around the world. We’d like to count you among
them. Please call to discuss your application,
arrange for a personal tour of our processing
facilities, or meet with us to learn how we might
fit into your precious metals asset recovery
program. Let us hear from you today.

Corporate Headquarters:
300 Pantigo Place, Suite 102
East Hampton, NY 11937
Telephone 631-329-1717 • Fax: 631-329-1985
Main Plant/Sales Office:
1647 Wheatland Center Road • Scottsville NY 14546
Telephone 585-538-2194 • Fax: 585-538-2593
Web: www.sabinmetal.com • Email: sales@sabinmetal.com
Additional Facilities: Williston, ND • Cobalt, Ontario, Canada
Latin America • Mexico • Europe

www.ptqenquiry.com
for further information

sabin 1 2/2/09 13:53:39


 



 

6ANADIUMPPM
-ma

-!4

 


 K^Ëgbg`<abgZ


 <hg_^k^g\^
 

+-&+/FZk\a+)*)
 

 
 

  
 

 

 
 

 

 

 

 

   

 

 

 

 

 

   

 

   
 

 

 
   

 

  












K^`^gmAhm^e%;^bcbg`


-!4ACTIVITYPPM 6ANADIUMONE CATPPM

Figure 6 Inventory deactivation by vanadium contamination

average SOx-reduction levels of catalyst could be evaluated. The


66%. trial was considered a success,
As can be seen in Figure 4, the achieving the refinery’s objectives
SOx values measured over time with even lower amounts of addi-
were somewhat volatile. This is tive than were first predicted.
attributable to variations in feed Refinery A is now using Super
and slurry sulphur content during Desox continuously as a solution to
+-&+.F:K<A&<hg_^k^g\^
the trial period, which is reflected reduce flue gas SOx emissions.
+.F:K<A&OblbmmhLbghi^\k^Ëg^kr
in the predicted uncontrolled SOx
+/F:K<A&;^bcbg`lb`aml^^bg`mkbi
levels. As shown in Figure 5, feed
sulphur varied between 0.31 and Super Desox (DESOX®) is a trademark of Grace
0.52 wt%, while slurry sulphur Davison Refining Technologies.
D>RGHM>LI>:D>KL
varied between 0.6 and 0.9 wt%. SA:HCNG@NBIZkmr<hffbmm^^h_
In terms of retention levels, the References <abgZI^mkhe^nfZg]<a^fb\ZeBg]nlmkr
1 Rheaume L, Ritter R E, ACS Symposium Series
refinery acknow­ledged excellent :llh\bZmbhg
375, Fluid Cracking Catalyst — Role in Modern P:G@CBFBG@<abg^l^:\Z]^frh_
perfor­mance. During the period
Refining, Ocelli M O, Ed, 1988, 9, 146–161. >g`bg^^kbg`Zg]Lrlm^fl>g`bg^^kbg`
when Super Desox was added at 2 Grace Davison Guide to Fluid Catalytic Lh\b^mrh_<abgZ
1.5 wt%, a serious inventory deacti- Cracking Part Two, Dougan T J, Ed, 1996, 163. CB:G@D>CNG>g^k`rLrlm^fl:gZerlbl
vation occurred due to increasing 3 DESOX Manual, Advanced FCC Flue Gas FZkd^m:gZerlbl<^gmk^%>g^k`rK^l^Zk\a
vanadium contamination of the Desulfurization Technology, Version 2.1, Grace Bglmbmnm^%GZmbhgZe=^o^ehif^gmZg]
e-cat, as shown in Figure 6. Even Davison Publication, 1994. K^_hkf<hffbllbhg
with dramatically increased e-cat SA:G@?NJBG<abgZI^mkhe^nfIeZggbg`
>g`bg^^kbg`Bglmbmnm^
vanadium levels, and the subse-
LNGEBEBLbghi^\>g`bg^^kbg`
quent drop in MAT activity, Bg\hkihkZmbhg
Colin Baillie is Marketing Manager, Grace
excellent levels of SOx reduction Davison Refining Technologies Europe, Worms, PNJBG@<GHH<ÃlAnbsahnK^Ëg^kr
were still achieved. Germany. He has a doctorate in chemistry AN:P>B;^bcbg`RZglaZgI^mkh\a^fb\Ze
from the University of Liverpool, UK. <hfiZgr
Conclusions Email: Colin.Baillie@Grace.com
As a consequence of increasingly Renaud Kieffer is Regional Technical Service
stringent local environmental regu- Manager, Grace Davison Refining Technologies ;hhdghpZg]]hpgehZ]
lations, Refinery A was forced to Europe, Worms, Germany. He has a degree in Z\hg_^k^g\^[kh\ank^Zm
reduce SOx emissions by 30%, and chemical engineering from Liège University,
Belgium.
ppp'k^Ëgbg`\abgZ'\hf
Grace Davison recommended the
Email: Renaud.Kieffer@Grace.com
use of Super Desox as the quickest
Maria Luisa Sargenti is Technology Coor-
and most cost-effective route to
dination Manager, Grace Davison Refining
achieve this. Using data from the Technologies Europe, Worms, Germany. She
refinery, a baseline for uncontrolled has a masters degree in chemical engineering
SOx emissions was established so from the Simon Bolivar University, Caracas,
that an accurate evaluation of the Venezuela.
SOx-reducing performance of the Email: Maria.Sargenti@Grace.com

www.eptq.com CATALYSIS 2010 35

china ad.indd 1

grace.indd 4 3/2/10 12:39:57


MacroTrap_Ad_rev_297mmx210mm 8/11/08 5:55 PM Page 1

MacroTrap ®

Guard Bed Media


Increases in pressure drop due to
fouling of the catalyst bed lead to
higher energy consumption and
compromise reactor performance.

Catalyst protection MacroTrap® guard bed media is


a performance-proven product that

that lasts.
protects your catalyst. Particulates
in the incoming feedstock and iron
sulfide gums are trapped in the
interconnecting pores of the ceramic
media, preventing them from
reaching the catalyst bed and causing
premature pressure drop rise.

MacroTrap® media delivers proven


benefits in several key areas:

• Premature shutdowns are avoided.

MACROTRAP® MEDIA MITIGATES • Life of catalyst bed is significantly


PRESSURE DROP INCREASE
100%
extended — in some cases to
as long as five-plus years.
Limit
• Reduced operating costs.

• Overall, more efficient


Pressure Drop

reactor operation.

The cost of shutting down a


0% reactor and adding replacement
Time Skim Skim Catalyst
Recharge
catalyst is expensive, so you
Without MacroTrap® Media With MacroTrap® Media can’t afford to rely on ineffective
filtering methods. MacroTrap®
guard bed media has been installed
Main Office: +1 330 673 5860 +971 50 6515205
in over 200 reactors worldwide.
Ohio, USA Dubai, UAE
Sales Offices:+86 10 6501 4172 +58 212 793 1911
MacroTrap® Media –
Beijing, China Caracas, Venezuela
Guarding Your Catalyst.
+49 6435 9657 0 norpro.saint-gobain.com
Steinefrenz, Germany oilandgas.saint-gobain.com Protecting Your Profits.
+81 3 3263 0334
www.ptqenquiry.com
Tokyo, Japan for further information
+7 812 740 16 80
Saint-Petersburg, Russia

www.ptqenquiry.com
for further information

norpro 1.indd 1 8/9/09 10:47:16


Maximising FCC distillate production

A novel catalyst manufacturing process forms matrix material and zeolite in a


single step. Trials indicate high matrix stability and low sodium content

Joe McLean
BASF Catalysts

I
ncreased demand on refiners to
process heavier crudes and
maximise diesel yields (see
Figure 1)1 means FCC units (FCCUs)
are continually pushed to their /THER ,0' 2ESIDUALFUEL $ISTILLATE *ETFUEL
.APHTHA 'ASOLINE KEROSENE
operating limits. Higher diesel
yields are often achieved through
the use of selective, bottoms       

upgrading FCC catalysts that      

0ER CENT

     
feature active matrices. However,
these conventional bottoms upgrad-       
ing catalysts also increase the yield
of undesirable products such as dry 



   

   
gas and coke. This often pushes the 
FCCU up against the limits of       
performance in terms of gas       )NCREMENTn
compression and metallurgy,
preventing the refiner from taking
full advantage of the potential for
increased diesel yields.
Figure 1 Global refined product demand focused on middle distillates, 2007–2030
Maximising distillate: fundamentals
The refining industry would like to
convert feed/bottoms in an FCCU
to light cycle oil (LCO) without
further cracking to gasoline.
Adjustments to the operating condi- 
tions that can be applied to an 
FCCU include reducing riser

temperatures and/or catalyst activ- LCO
ity by lowering the catalyst 
make-up rate (see Figures 2 and 3). 
9IELDS WT

However, both of these options lead 


to an increase in bottoms yield Bottom

greater than the increase in yield of
LCO. This clearly is not desirable. 
Another way to increase the 
production of LCO is to recycle 
slurry/bottoms. Second-pass yields

are typically less selective and lead           
to higher coke and gas production. 2EACTORTEMPERATURE ²#
Use of recycle in an FCCU
constrained by air blower capacity
can result in a reduction in feed
rate, which can reduce the total Figure 2 LCO and bottom yields’ response to decreasing reactor temperature shows that
production of LCO. No adjustments the bottom yield response is more pronounced than LCO

www.eptq.com CATALYSIS 2010 37

basf.indd 1 1/2/10 11:16:39


Brønsted and Lewis acidity of
different matrix materials

Brønsted acidity Lewis acidity
(μmol/m2) (μmol/m2)
 BASF novel matrix 0.2 2.51
Matrix A 0 2.59
LCO Matrix B 0 1.57
 Matrix C 0 1.81
9IELDS WT

 Table 1


Bottom to the FCCU, or combination of
adjustments, will alone result in
increased output of LCO without at

the same time increasing bottoms

production. Only an improvement
       in catalytic performance offers a
% CATACTIVITY WT pathway to increased LCO produc-
tion while maintaining bottoms
conversion.
Figure 3 LCO and bottom yields’ response to lowering the equilibrium catalytic (e-cat) With BASF’s Novel Matrix
activity shows that the bottom yield response is more pronounced than LCO Material, LCO yield can be
increased by controlling the depth
of cracking. The conventional cata-
lyst approach to increasing LCO


yield has been to lower the Z/M

ratio, typically by both increasing


the matrix surface area (MSA) and
Z/M = 0.4
 Z/M = 0.4 lowering the zeolite surface area
  (ZSA).
BASF has introduced a manufac-
#OKE WT
,#/ WT


 turing process for an advanced

Z/M = 4 matrix material. This is based upon
 
a novel technology platform that
 Z/M = 4 differs from more conventional

 matrix materials available to the
  industry. Figure 5 compares the
        hydrothermal stability of the new
#ONVERSION WT #ONVERSION WT
matrix material with a variety of
commercially available matrix
Figure 4a LCO yield vs conversion Figure 4b Coke vs conversion materials. It shows that the novel
manufacturing process has resulted
in a matrix material that demon-
strates improved hydrothermal
 stability.
"!3&nPROPRIETARYMATRIX
Most conventional matrix materi-
#OMPETITIVE&##MATRIX" als, such as alumina, predominantly
 #OMPETITIVE&##MATRIX! have Lewis acid sites, measured by
3UR FACEAREARETENTION 

#OMPETITIVE&##MATRIX# pyridine adsorption. The novel


manufacturing process generates
 not only Lewis acid sites, but also a
small but significant fraction of
Brønsted acid sites (see Table 1).
 The role of Brønsted acid sites in
zeolitic cracking is well known. The
Brønsted sites on the matrix mate-
 rial, although weaker than zeolitic
        sites, are believed to act synergisti-
3TEAMTIME²& STEAM HR
cally with the Lewis sites to enhance
catalytic properties.
However, a lower Z/M ratio
Figure 5 Stability of different matrix materials under hydrothermal conditions results in accentuated matrix

38 CATALYSIS 2010 www.eptq.com

basf.indd 2 1/2/10 11:16:52


cracking with poor coke and gas
selectivities. Figures 4a and 4b show
a comparison of LCO and coke
 "!3&nPROPRIETARYMATRIX
selectivities for two conventional #OMPETITIVE&##MATRIX"
"!3&nPROPRIETARYMATRIX 
catalysts at two different Z/M #OMPETITIVE&##MATRIX"
#OMPETITIVE&##MATRIX!
ratios. As preceding literature indi- #OMPETITIVE&##MATRIX!
#OMPETITIVE&##MATRIX#

cates, lowering Z/M does increase #OMPETITIVE&##MATRIX# 

LCO yield, but at the expense of

#OKE WT

,#/SELECTIVITY 
higher coke production. 

This demonstrates a need for a
fundamentally different catalytic
approach to increasing LCO yield. 

The need is to have a selective
matrix that would increase the LCO 
yield without the accompanying 
coke yield.      
#ONVERSION WT     
#OKE WT
New technology platform
Development of the novel matrix
material was extended to include Figure 6a Coke vs conversion Figure 6b LCO selectivity vs coke
both the crystallisation of Y zeolite
and the formation of the matrix in a
single step. The manufacturing
process forms both the matrix mate-  
rial and zeolite in a single step, and 30% low Z/M
it brings them into very intimate 
catalyst + baseline
30% low Z/M
contact, with an ultra-low sodium catalyst + baseline
content. The hexagonal crystallites  
of Y zeolite, smaller than 1 µm, are
,#/SELECTIVITY 
in the immediate vicinity of, if not
#OKE WT

 Baseline
in intimate contact with, the amor-
phous matrix materials. 

Catalytic properties of the differ- Baseline




ent matrix materials were compared 


30% non-zeolitic
by preparing physical blends of catalyst + baseline
30% non-zeolitic
matrix materials with separate 
catalyst + baseline
particle US-Y zeolite, which had no 

matrix. Both were separately deac-     
    
#OKE WT
tivated at 1500°F (815°C), four #ONVERSION WT

hours and 100% steam. US-Y zeolite


had 2.6 wt% rare earth oxides
(REO). The blend ratio was adjusted Figure 7a Coke vs conversion Figure 7b LCO selectivity vs coke
to maintain a constant Z/M ratio.
The blends were evaluated for
cracking performance in the



Development of the 

novel matrix material


:- RATIO



was extended to 

include both the 


crystallisation of 


R


Y





L








T











AR P A N U G P C OV E C N
! *U * U 3E / *A
Y zeolite and the 
-
 
-
  
!
  
.

$

3AMPLETAKENDATE
formation of the
matrix in a single step Figure 8 Z/M ratio for HDXtra from the e-cat data at Refinery 1

www.eptq.com CATALYSIS 2010 39

basf.indd 3 1/2/10 11:17:01


with lower coke yield. Figure 6b
Benefit of Prox-SMZ technology on coke yield and LCO selectivity
shows a plot of LCO selectivity
versus coke yield for the four
Coke at 71 wt% LCO selectivity at
conversion 4 wt% coke
blends. The other three matrix
Base 3.76 69.6 materials had similar LCO yields,
Base + 30% Prox-SMZ catalyst 3.82 71.5 whereas the proprietary matrix
Base + 30% competitive matrix only material 4.0 70.2 showed higher LCO selectivity.
While other catalyst technologies
Table 2 can incorporate zeolite and matrix
materials in the same catalyst parti-
cle, they do not have the capability
Produced yields from Refinery 1
to bring them together in such inti-
mate contact. It is this synergy
Produced yields, vol% Base HDXtra
Gasoline 56.9 51.5 created by the close proximity of
LCO 20.7 30.9 ⇐ + 10.2% zeolite and matrix that leads to
CSO 5.7 5.1 rapid transfer between reactant and
Corrected LCO (430–650°F) 16.8 20.8 ⇐ + 4% feed molecules from zeolitic acid
sites to matrix acid sites. The
Table 3 enhanced transfer results in the
coke precursors produced by matrix
Summary of results from Refinery 2 cracking being stabilised by zeolite,
leading to higher LCO production
with lower coke.
Before After
Feed concarbon, wt% 0.4 1.2
The benefit of zeolite and matrix
Feed rate, bpd Base Base synergy was evaluated in a circulat-
RXT temp, °F 970 970 ing riser unit (CRU) with a mildly
Catalyst addition Base >15% decrease hydrotreated gas oil (API gravity
Total C3s + C4s, vol% Base -1.2
Gasoline, vol% Base -3.3
23.2º and Conradson carbon residue
LCO, vol% Base +3.3 content of 0.4 wt%). The base cata-
Bottoms, vol% Base +0.4 lyst was a Distributed Matrix
LCO cetane index Base +3 Structures (DMS) based, high Z/M
E-cat metals, Ni + V, ppm 220 ppm & 840 ppm V 260 ppm & 1320 ppm V
catalyst targeted for making gaso-
line. The base catalyst was blended
Table 4 to a 30% level with either the new
BASF catalyst or a commercially
available matrix material designed
for making LCO. The results are
 shown in Table 2.
The catalytic performances of the

two blends are compared to the
base catalyst in Figures 7a and 7b.
:- RATIO


The blend containing the new tech-

nology platform had lower coke
 yield and higher LCO selectivity
than the blend containing commer-

          
cially available matrix material.
           This new technology platform is
Y  N L  G G T  V  V N  N B
A
*U *U U U /
C O O *A *A &E

-
  
!

!
 
.

.
   called Promixal Stable Matrix and
3AMPLETAKENDATE Zeolite (Prox-SMZ), and all four of
those features play an integral role
in its catalytic performance. Prox-
Figure 9a Z/M ratio for HDXtra from the e-cat data at Refinery 2 SMZ is a flexible platform that can
be tailored to individual applica-
advanced cracking evaluation other FCC catalysts were compared tions, including use with different
(ACE) unit at 970°F (520°C), with to the new matrix material. feeds.
hydrotreated feed containing 3.88 Figure 6a shows the plot of coke Using the Prox-SMZ platform, we
wt% Conradson carbon residue. versus conversion for the four have developed and commercial-
LCO selectivity is defined as the matrix materials tested. The ised HDXtra, a catalyst specifically
ratio of LCO to the sum of LCO competitive matrix materials had targeted for vacuum gas oil (VGO)
and heavy cycle oil (HCO). Three very similar coke yields. In contrast, operations. Other Prox-SMZ cata-
different matrix materials used in the proprietary matrix stood out lysts, including Stamina for resid

40 CATALYSIS 2010 www.eptq.com

basf.indd 4 1/2/10 11:17:11


We keep the world
a little bit cleaner.

Since 1818, it has always been the goal of Zeochem to develop and manufacture products
which do not additionally load our air and environment with impurities. ZEOcat® products are
the answer whenever it comes to remove volatile organic compounds (VOC) from waste air
and gas streams, the removal of odours and smell from thermoformed plastics in the car-
and food industry, as catalyst carrier in gasoline and diesel exhaust catalyst, as catalyst for
organic and process reactions in chemical, fine chemical- and petrochemical processes.
ZEOcat® is the answer and our commitment for a cleaner world, made in Switzerland by
experts and under very hard environment regulations. ZEOcat® products are available in forms,
such as Pentasil, ZSM-5, Mordenite, Beta and L-zeolite in powder form and a variety as
beads. For more information please go to www.zeochem.com
www.ptqenquiry.com
for further information

zeochem.indd 1 29/1/10 12:53:38


The HDXtra catalyst
 was specifically
 engineered

for distillate
.A WT


maximisation and is

targeted at VGO feed


AY



N


UL 


G 


G 


C T


OV 


OV 


N


N 


B


operations
*U * U U / * A * A &E
-  ! ! . .
         
3AMPLETAKENDATE
feed operations, are currently in the
process of being commercialised.

Figure 9b Na content for HDXtra from the e-cat data at Refinery 2 HDXtra: the catalyst for LCO
maximisation in VGO operations
HDXtra is the first commercialised
catalyst developed using the Prox-
SMZ technology platform.2 This
 FCC catalyst was specifically engi-


neered for distillate maximisation
 and is targeted at VGO feed opera-
tions. Its features include high
:- RATIO



matrix activity with good coke

 selectivity, moderate zeolite activity
 to maximise LCO, and low H trans-
 fer activity to preserve hydrogen in

           the LCO for minimal cetane penalty.
          
 N L  G G T V  V N N B HDXtra has been successfully
AY U *U U U C O O *A *A E
- *  ! ! / . . &
          demonstrated in three commercial
3AMPLETAKENDATE trials over the past year.

Trial at Refinery 1
Figure 10a MSA for HDXtra at Refinery 3 The first trial of HDXtra was at
Frontier in El Dorado, Kansas,
which operates as a hydrotreated
VGO refinery. HDXtra replaced a
NaphthaMax-II catalyst. This switch
 can be most effectively seen in
Figure 8, where the e-cat Z/M ratio
 drops from approximately 2.2 to
less than 1.
Another noticeable shift was in

Na, which on e-cat Na dropped
#OKE WT

from 0.22 wt% to below 0.1 wt%.


 Highlights of the trial are shown in
HDXtra trial Table 2. Operating conditions were
changed and conversion was

dropped from about 73 wt% to 64
wt%. At roughly 50% change-out,
 an increase in LCO yield of 10.2
  
    

P
 

V 

N

B
  wt% was measured. After correct-
/C
T C AR
3E .O $E * A &E - ing for conversion, the LCO yield
3AMPLETAKENDATE increased by about 4 wt%.

Trial at Refinery 2
Figure 10b ACE coke levels of HDXtra over time at Refinery 3 Refinery 2 produced similar results,

42 CATALYSIS 2010 www.eptq.com

basf.indd 5 1/2/10 11:17:28


shown in Table 4. The LCO volume
increased by 3.33 vol%, with an

increase in the LCO cetane index of
3. Notable reductions in e-cat Z/M
and Na are shown in Figures 9a 
BASF HDXtra
and 9b. Competitor


#OKE WT
Trial at Refinery 3
At Refinery 3, the objectives were 
to maximise feed rate and maintain
gasoline motor octane number 
(MON). The refinery was previ-
ously using a competitive catalyst 
designed for high LCO yields. The 





















base catalyst already had a rela- U L G  P  P
C T  V  V  C  N N B
* !U 3E 3E / .O .O $E *A *A &E
tively high MSA. Switching from
the competitive catalyst to HDXtra
increased the MSA from approxi-
mately 85 m2/g to 100 m2/g (see Figure 10c FCCU coke levels of HDXtra compared with the competitor at Refinery 3
Figure 10a). As in previous trials, e-
cat Na dropped dramatically during
the trial from about 0.25 wt% Na to
less than 0.1 wt%. This decrease in 
Na not only improves the stability
of zeolite and the matrix, but also 
decreases the hydride transfer,
which positively affects cetane.
 HDXtra trial
Although the MSA increased, it
did not detrimentally impact the
WT


coke yield (see Figure 10a). This
reinforces the experimental obser-

vation during product development
that not all matrices are equal in

their catalytic behaviour. In Refinery
3, the ACE and FCCU coke data

remained at the same levels (see
 




    

Figures 10b and 10c). Unlike tradi- P

CT
 V EC
 N

B

AR

3E / .O $ *A &E -
tional high-MSA catalysts, HDXtra
did not lead to coke penalties.
Further observations of e-cat
change-over at Refinery 3 show Figure 10d Na levels over time at Refinery 3
how HDXtra has the lowest
commercially available Na (see
Figure 10d). Refinery 3’s normalised



BASF’s e-cat
benchmarking 
WT

BASF HDXtra

database has shown 


Competitor

that HDXtra has some 

of the highest MSA 


         
         
L  P P  V V    N
and matrix stability *U !U
G
3 E 3 E /C
T
.O .O $E
C
*A
N
*A

available and the


lowest Na content Figure 10e Normalised LCO and HCN levels at Refinery 3

www.eptq.com CATALYSIS 2010 43

basf.indd 6 1/2/10 11:17:41


diesel production operating data
show that the FCCU reached

unprecedented maximum levels
 (see Figure 10e), while slurry yield
BASF HDXtra
remained constant (see Figure 10f).

Table 5 shows an almost 2 wt%
increase in LCO at Refinery 3 when
WT


Competitor
using HDXtra over the base

catalyst.

Refinery trials: conclusion

         
BASF’s e-cat benchmarking data-
         
UL G

P P T  V V C  N N base has shown through trials that
* !U 3 E 3 E /C .O .O $E *A *A
HDXtra has some of the highest
MSA and matrix stability available
and the lowest Na content (see
Figure 10f Normalised slurry levels at Refinery 3 Figure 11). Both factors result in
higher gasoline octane and better
LCO cetane in the unit. All three
Customer operating data at Refinery 3 trials show a significant increase in
normalised LCO yield, while the
Cases corrected for cut point, feed rate, feed quality and operating conditions coke and slurry levels remain
Catalyst supplier Competitor BASF ∆ constant. ACE testing of e-cats
Case title Base case HDXtra
confirmed the results.
Case date Oct 08 Jan 09
Conversion
Corrected fresh feed conversion, wt% 76.10 74.47 -1.63 Further developments
Product yields BASF is designing a set of solutions
Weight per cent basis
based on the Prox-SMZ technology
H2+C1+C2+C2=, wt% 2.12 2.05 -0.07
Total C3 + C4, wt% 22.38 19.69 -2.69 platform to address the needs of
Corrected C5+ gasoline, wt% 46.50 47.73 1.23 the refining industry. Stamina, an
Corrected light cycle oil, wt % 21.78 23.60 1.82 FCC catalyst for distillate maximi-
Corrected slurry, wt% 2.12 1.93 -0.19
sation targeted at resid feed
Coke, wt% 5.09 5.00 -0.09
Product properties operations, is currently undergoing
Gasoline RON, RON 92.5 93.2 0.7 extensive internal manufacturing
Gasoline MON, MON 80.4 80.6 0.1 trials. This catalyst was designed
with more macroporosity for heavy
molecule access, together with
Table 5 proven metals tolerance features.

Distributed Matrix Structures, Stamina,


  NaphthaMax and HDXtra are marks of BASF.

  References
Cumulative, %
  1 Hart’s World Refining and Fuels Service
(WRFS) 2009.
  2 McLean J B, The role of FCC catalyst

technology in maximizing diesel production —

two North American refinery success stories,
&REQUENCY

0ER CENT

Frequency
  2009 NPRA paper.

 

 The highest MSA 


are all BASF Joe McLean is Global Technology Manager for
 catalysts  Refining Catalysts for BASF Catalysts (formerly
Engelhard), based in Houston, Texas. He has
 
more than 30 years’ industrial experience,
  including positions with ARCO, Research-
        -ORE Cottrell and Hydrocarbon Research, before
-3! -G joining Engelhard. He holds a bachelor’s degree
from Princeton University and a master’s in
chemical engineering from the University of
Figure 11 Histogram showing MSA from all BASF benchmarking data California, Berkeley.

44 CATALYSIS 2010 www.eptq.com

basf.indd 7 1/2/10 11:18:49


E T H Y L E N E O X I D E C ATA LY S T S | H Y D R O G E N & S P E C I A L T I E S | S T Y R E N E C ATA LY S T S |
H Y D R O G E N AT I O N C ATA LY S T S | E N V I R O N M E N TA L C ATA LY S T S & S Y S T E M S |

With the men and women of CRI beside you, you’ll never take a wrong
turn in catalysts. We’ll help you innovate when you need to move in a
With our expert catalyst new direction, and we’ll be there with expert support down the road.

guidance, you’ll stay on Whether you need catalysts for ethylene oxide, distributed hydrogen
products, styrene, environmental applications, selective hydrogenation
the path to success. catalysts from KataLeuna™, or custom products, the people of CRI will
keep you on track.

www.ptqenquiry.com
for further information
+1 281 874-2600
catalystsales@cri-criterion.com
www.cricatalyst.com
© 2007 CRI Catalyst Company. KataLeuna is a trademark of CRI Catalyst Company. (CCC711)

ccc711_PTQ_sep07.indd 1 11/26/07 1:41:41 PM


cri 1 2/2/09 13:50:50
www.ptqenquiry.com
for further information

mbi.indd 1 11/6/09 20:35:37


Advances in methanol synthesis

Catalysts with higher and more stable activity enable cost savings and boost
output in methanol production

Terry Fitzpatrick and Tom Hicks


Johnson Matthey Catalysts

F
or many years, methanol has
been used primarily as a
Competitor
chemical intermediate in Com
pet A
itor B
manufacturing plastics and resins, JM
then more recently in the manufac-
Others
ture of methyl tertiary butyl ether
(MTBE) for use as a lead anti-knock
replacement and octane enhancer,
allowing a methanol derivative to
enter the transportation fuel chain
in a significant way for the first Figure 1 Methanol technology market shares
time. However, now methanol is
being seen as a product that can be methanol synthesis catalysts and series catalysts — an improvement
introduced directly into the gaso- discuss the various changes that that is a step change in methanol
line pool by blending, allowing have occurred in the Katalco range synthesis catalysis.
indigenous resources to be used of catalysts against the backdrop of
and providing a diversity of supply changing industry requirements. Methanol synthesis catalysts
that can help to reduce dependence Since the initial development of the
on crude oil and attempt to break Methanol production first copper-zinc low-pressure
the cycle of apparently ever- ICI initiated work on catalysts for methanol synthesis catalyst, Katalco
escalating oil prices. methanol synthesis in the 1920s, 51-1, continuing development
China has introduced a national when the only commercial process pro­gram­mes have improved
M85 standard that sees gasoline operated at high pressure. Follow­ perform­ance in terms of activity,
blended with 85% methanol, which ing early research on copper-zinc by-products production, strength,
has been manufactured using catalysts, ICI announced the Low shrinkage and overall life. The orig-
China’s cheap and abundant Pressure Methanol (LPM) process inal catalyst was designed for
supplies of coal, helping to reduce in 1963 and the first single-train application in the multi-bed ICI
its dependence on expensive production unit started operation in Quench lozenge converter, and an
imported oil. 1966. early variant, Katalco 51-2, quickly
In the US, there is considerable JM Catalysts has recently devel- became the industry standard. As
support for the Open Fuel Standard oped a new generation of copper additional technologies were devel-
Act, which, if passed, would call for zinc methanol synthesis catalysts oped, different types of converter
car manufacturers to introduce flex- called Katalco Apico. This extends were used, the most noteworthy
ible fuel vehicles that can the performance of the Katalco 51 being gas-cooled and steam-raising
run on methanol/ethanol/gasoline in both axial and radial flow config-
mixtures. Currently, there is little urations. These often impose
methanol production left in North Now methanol is different requirements on the cata-
America, but the development of lyst, so JM Catalysts has developed
shale gas is set to reduce natural being seen as a a range of synthesis catalysts.
gas prices significantly in North It is worth considering the vari-
America. And, like China, the US
product that can be ous changes that have occurred in
has abundant coal reserves, which, introduced directly Katalco catalysts against the back-
through methanol, could be drop of changing industry
used to displace oil imported from into the gasoline pool requirements. These changes do not
abroad. come from any one aspect of the
In this article, we will look at by blending catalyst. The enhancements have

www.eptq.com CATALYSIS 2010 47

jm.indd 1 1/2/10 11:24:35


such as sulphur and, in some cases,
 iron and nickel carbonyls brought
-ETNANOLSYNTHESISACTIVITY MOL3nGn8 into the loop with fresh syngas also
 contribute to deactivation or die off.
Thus, key formulation requirements
 are stabilisation of the copper
surface area and self-guarding

against poisons.
One of the major contributors to
a significantly increased in-service

#U/3I/ activity was the incorporation of
#U/-N/ magnesia (MgO) into the formula-
 #U/-G/
#U/:N/!L/
tion during the early 1990s. This
gave rise to Katalco 51-7 and has

been incorporated in subsequent
     
#UMETALAREA MG variants Katalco 51-8 and Katalco
51-9. The benefit from incorporating
MgO is evident from Figure 4, and
Figure 2 Variation in synthesis catalyst activity with copper surface area the significant improvement rela-
tive to Katalco 51-2 in terms of both
been generated by identifying and be achieved with the highest CuO initial and final activities is illus-
understanding the role of the key content in the fresh formulation, trated in Figure 5.
components in the formulation and but this ignored the impact of Activity testing is a specialised
the catalyst manufacturing process formulation. As Figure 3 shows, technique comparing aged activities
itself, as well as improvements in variations in the CuO:Al2O3 ratio to the catalyst Katalco 51-2. Ageing
manufacturing control. have a marked effect on the is reliably simulated by deactiva-
relative activity, as shown in accel- tion in a controlled and reproducible
Catalyst activity erated life tests. manner using elevated tempera-
The methanol synthesis reaction is High initial activity, while impor- tures and pressure plus a
an example of a structure insensi- tant, is not paramount, as the representative synthesis gas
tive catalytic reaction — one in effective useful life of the catalyst mixture, before measuring activity
which the activity is wholly will be governed by its stability under standard conditions. A typi-
dependent on the total exposed with time, so the formulation must cal test regime measures the activity
copper area and not affected by the also stabilise the copper surface after 144 hours on-line, represent-
structure of the crystallites. Figure 2 area under the process conditions ing approximately three months in
illustrates this direct relationship to which it is exposed. Thermal an operating methanol plant. The
between activity and copper surface sintering is a key mechanism for results have been validated over the
area for catalyst operating under synthesis catalyst deactivation with years using data from operating
industrial conditions. operation at temperatures as high charges in plants and side-stream
This relationship led to sugges- as 315°C, depending on reactor reactors on our own plants.
tions that maximum activity would type. Commonly found poisons Activities are regularly compared
with the leading competitive offer-
ings, and the most recent
comparison in Figure 6 clearly
 shows the relative beneficial
Optimum ratio
performance of Katalco 51-9S. The
 higher and, more critically, stable
activity allows operation at lower
 temperatures, favouring the reac-
2ELATIVEACTIVITY

Activity
tion thermodynamics and loop
!L/ 

 carbon efficiency, minimising ther-


mal sintering and giving benefits in
Al2O3 increased methanol output and

reduced by-product formation. The
reduced rate of activity loss trans-

lates into a longer period of

operation between catalyst changes.
#U/ 
Catalyst strength and shrinkage
Declining strength and activity
Figure 3 The impact of catalyst formulation on catalyst activity were originally the limiting factors

48 CATALYSIS 2010 www.eptq.com

jm.indd 2 1/2/10 11:24:47


Anzeigenaenderung_Jointheleader_A4:Layout 1 09.12.08 17:48 Seite 1

Come Join the Leader in Precious Metals:


Refining, Manufacturing,
Compounds and Trading.
Since 1851.

Refining: Catalysts:
Petroleum Production of chemicals
& petrochemical catalysts Gas purification
Carbon catalysts Petrochemistry
Homogeneous catalysts Special catalysts
Precious metal alloys
Chemical compounds Chemical Products –
E-scrap Inorganics and Organometallics:
Precious metal chlorides, nitrates,
Platinum Group Metal hydroxides, carbonates, acetates
Alloys Manufacturing: "Ammonium Perrhenate" (APR) and
Nitric gauzes "Perrhenic Acid" (PRA)
Thermocouple wires Homogeneous catalysts for petrochemistry,
Crucibles & labware e.g. CARAC, ROPAC, [RhH(CO)(PPh3)3],
Other custom alloys Rhodium Acetate
Comprehensive range of homogeneous catalysts
Trading – Buy/Sell/Lease: for other applications, e.g. Pd(dba)2, "Wilkinson's",
Platinum, Palladium, Rhodium, [Pd(PPh3)4], [{Ir(cod)}2(µ-Cl)2], "Karstedt's"
Ruthenium, Iridium, Silver, Gold Tailor-made compounds on an NDA basis

Heraeus Metal Processing, LLC. W. C. Heraeus GmbH


15524 Carmenita Road Chemicals Division, Business Unit Recycling
Santa Fe Springs, CA 90670, USA Heraeusstr. 12 – 14, 63450 Hanau, Germany
Phone +1 (562) 921.7464 Phone +49 (0) 6181.35-3939
Fax +1 (562) 926.5333 Fax +49 (0) 6181.35-164503
hmpi.customer.service@heraeus.com recycling@heraeus.com

www.ptqenquiry.com
for further information
of the catalysts’ operational life.
 Due to the high copper content,
initial catalyst reduction and devel-
 opment of the copper surface area
leads to major changes in the physi-

cal structure, which is manifested
#OPPERSUR FACEAREA MG

 in terms of shrinkage and reduced


strength. Low strength during oper-
 ation, especially during upset
conditions, can lead to physical

breakage of the pellets, giving
 increased pressure drop that
reduces efficiency as well as affect­
 ing gas distribution through the
 catalyst. High shrinkage also leads
       to distribution problems and a
-G/  reduced volume of active copper in
the reactor. These properties are
Figure 4 Effect of incorporating magnesia on copper surface area critical, for instance, in a steam-
raising reactor such as the axial
flow catalyst-in-tube design,
where the catalyst duty is quite
arduous both in terms of the
#ALCULATEDINITIALACTIVITY VV
volume of material charged and the
#ALCULATEDFINALACTIVITY VV
crushing forces to which the cata-
lyst is exposed during thermal
cycling.
2ELATIVEACTIVITY

Initially, in the oxidised state, the


catalyst must be strong enough to
withstand the rigours of charging
into the chosen reactor design with-
out breakage. Too high an initial
“as received” strength derived from
a high pellet density can be a disad-
vantage, leading to diffusional
limitations within the catalyst,
+ATALCO  +ATALCO 3 affecting overall activity. Through
an understanding of the formula-
Figure 5 Katalco 51 series activity enhancement tion and manufacturing parameters,
Katalco 51-9S has been designed
with a high pellet density, giving a
much enhanced strength both
initially and in operation without
#ALCULATEDINITIALACTIVITY VV adding any diffusion limitations, as
#ALCULATEDFINALACTIVITY VV shown in Berty reactor tests and
commercial experience.
The most readily obtainable
measure of strength is the mean
2ELATIVEACTIVITY

horizontal crush strength (MHCS),


appropriately measured across the
weakest pellet dimension. Occasion­
ally, the mean vertical crush
strength (MVCS) is reported, but
this can be misleading, being an
order of magnitude higher in the
“as received” state.
As a result of improvements to
+ATALCO 3 #OMPETITOR the catalyst formulation and manu-
facturing process, pressure drop
increase is no longer a limiting
Figure 6 Comparison of commercial catalyst activities feature of normal plant operation.

50 CATALYSIS 2010 www.eptq.com

jm.indd 3 1/2/10 11:24:59


In particular, the introduction of
post-pellet treatment to the manu-

facturing process in 2001 has
resulted in catalysts that have much +ATALCO 3ACTIVITY
greater reduced strength and as 
+ATALCO PREDICTEDACTIVITY

little as 5% shrinkage on reduction.


Catalyst lives of four to six years,

and occasionally as long as eight

!CTIVITY
years, are commonplace, even in
the arduous catalyst-in-tube steam- 
raising duty discussed in the case
study below.


Choice of catalysts
JM Catalysts currently offers four 
        
methanol synthesis catalysts;
4IMEON LINE MONTHS
namely, Katalco 51-8, Katalco 51-
8PPT and Katalco 51-9S and the
premium product Katalco Apico. Figure 7 Activity of Katalco 51-9S
Providing a range of catalysts, and
not relying on a single universal reaction heat removal systems. each charged with 60m3 of Katalco
product, caters for the different These include, but are not limited 51-9S. Experience has shown that
methanol synthesis technologies to, the direct quench-cooled adia- with this type of reactor careful
and enables a choice of product for batic converter design, the axial initial catalyst charging is essential
the specific duty. This is particu- steam-raising catalyst-in-tube to ensure minimum pressure drop
larly pertinent in a climate where design or the catalyst-in-shell-side, variation between tubes, to give an
feedstocks are changing. Natural gas-cooled design such as the tube- even flow distribution and maxi-
gas has been the principal feed for cooled converter. Furthermore, the mum use of the relatively small
synthesis gas generation (account- savings through economies of scale volume of catalyst. The catalyst has
ing for ~80% of world methanol that might be accrued from build- been producing record production
production); synthesis gas genera- ing world-scale methanol plants can levels of 2600 tpd since May 2005,
tion historically has largely been benefit from the use of radial flow giving significant financial benefit.
based on pure steam methane converters such as the Davy Process As a result of the enhanced
reforming, whereas more recently Technology radial steam-raising performance of the catalyst, an
there are circumstances when converter. extended run of four years is
combined reforming, which planned. JM Catalysts has provided
includes an oxygen-fired secondary Case study: Titan Methanol detailed technical support through­
or ATR, has its merits. This Lurgi-designed plant has a out, carrying out frequent
Coal gasification is another nameplate capacity of 2500 tpd and perform­ance optimisations. Figure
method of syngas generation seeing the synthesis loop comprises two 7 confirms the activity achieved,
rapid expansion in the number of parallel steam-raising converters, which is 20% greater than Katalco
medium- and large-scale methanol
projects in China based on a rela-
tively cheap and plentiful resource,

with many more plants likely to
appear in the future. The US,
Australia, India and Russia also
-EASUREDPRESSUREDROP BAR


have abundant supplies of coal. In
relation to the actual converter type,
and without going in to too much
detail, the different syngas- 
generating technologies yield
different syngas compositions (for
instance, CO content), each compo- 
sition yielding different reaction
rates over the synthesis catalyst.
Consideration of the exothermic 
synthesis reactions means that some         
4IMEON LINE MONTHS
converter designs are better suited
than others to the different
synthesis gases, by virtue of their Figure 8 Pressure drop stability of Katalco 51-9S with time

www.eptq.com CATALYSIS 2010 51

jm.indd 4 2/2/10 13:48:27


activity of Katalco Apico catalyst
(see Figure 10).
TMETHANOLPRODUCTIO As a direct result of the higher
-OS N
activity and stability with time on-
line, the catalyst can be operated
Least - for twice as long as any comparable

OR
by-products commercial catalyst, leading to

EM
fewer catalyst change-outs. On a

ETH
ANO
2500 tpd plant currently achieving
UP

L
ART

AND
four years between change-outs, the
T
&ASTESTS

IMPR
Pre-reduced Highest typical time saved by doubling the
catalyst activity catalyst life is about nine-and-a-half

OVEDEFFICIENCY
Apico days, equivalent to at least 23 750
tonnes of product methanol. This is
derived from savings on oxidation
of the previous catalyst charge prior
Strongest to discharge, time to discharge and
slowest

ES
product refill and then reduce the new

NG
deactivation
,OW

available catalyst charge.

HA
ES

TC
TR

LYS
AT

OF

TA
A
E

P TC By-product formation


RE
SSU WES
FE
RE
D ND The results of tests shown in Figure
ROP ESA
INCRE STLIV 11 confirm an even lower level of
ASE ,ONGE
higher alcohol and other oxygenate
by-products with Katalco Apico at
Figure 9 Sources of benefits to operators typical operating conditions. The
benefit is further enhanced by being
51-8 at end of life. Figure 8 shows • Fewer catalyst change-outs. able to operate the catalyst at the
the pressure drop has remained Simply stated, in existing plants, lowest temperature possible due to
stable with time, confirming the Katalco Apico will make the most stable activity.
high strength retention and resist- methanol, give the longest life On a 2500 tpd unit, using the
ance to breakage in this duty. and the fastest start-up possible. current generation of catalysts,
For a newly designed plant, it 98.5% of the crude methanol
Benefits offers the smallest reactor and the coming from the synthesis loop is
Since ICI first developed the LPM highest achievable efficiency. The converted to product methanol.
process, copper-based catalysts following sections show compara- With a 50% reduction in higher
have improved in small incremental tive data for Katalco Apico and the alcohol by-products, this figure is
steps at regular intervals until now. current industry standard Katalco increased to 99.15% — an increase
The immediate benefits to methanol 51-9S. of 0.65% in methanol produced.
plant operators are significant:
• Increased production Highest and most stable Greater strength
• Lower by-product formation catalyst activity Figure 12 shows the measured in-
• Faster start-ups using pre- Comparative activity tests and situ radial pellet strength compared
reduced catalyst projections clearly show the higher with the best currently available
product.
The 50% increase in operating
strength ensures the catalyst is
better able to withstand upset
conditions without physical break-
Katalco Apico age. This results in a more stable
pressure drop so that efficiency and
!CTIVITY

gas distribution are maintained.

Katalco 51-9S Pre-reduced catalyst


Katalco Apico catalysts are
supplied in the reduced form,
 which ensures the catalyst has been
    
%QUIVALENTYEARSON LINE
activated to achieve maximum unit
activity. Since there is no shrinkage
associated with normal catalyst
Figure 10 Enhanced activity of Katalco Apico reduction, this also maximises the

52 CATALYSIS 2010 www.eptq.com

jm.indd 5 2/2/10 13:49:09


amount of active copper charged.
By using a pre-reduced catalyst,
there is a typical saving of around

30 hours for a new charge, repre-
senting at least 3000 tonnes of %THANOL
product methanol on a 2500 tpd  0ROPANOL

plant.  0ROPANOL
-%+
 "UTANOL

#ONCENTRATIONINCRUDEMETHANOL PPM
Overall benefit  -ETHYLPROPAN  OL
As indicated in the previous   "UTANOL
sections, the enhanced performance
of Katalco Apico leads to benefits in
many aspects of plant operation. 
These are summarised in Table 1.
The total value of these benefits is
over $25 million in additional meth- 
anol sales (assumed methanol price
$200/t). Additional savings related
to manning and material costs of

shutdown may also be realised.

Case study: Katalco Performance



A methanol plant in Asia using a +ATALCO 3 +ATALCO!PICO
combination of Katalco catalysts in
the tubular reformer contracted JM
Catalysts to do a specialist reformer
survey to check the performance Figure 11 Selectivity improvement with Katalco Apico
and make recommendations for any
possible optimisation. The furnace
was found to be in need of balanc­
ing and it was also shown that

the process gas temperature could
be safely increased without compro-
mising tube life. Detailed 

recommendations on balancing the


)NSITURADIALCRUSHSTRENGTH

furnace were implemented on the 


plant, with a resultant increase of
60 tpd of methanol make worth 
over $3 million/y.


Katalcojm and Apicojm are marks of the


Johnson Matthey Group of Companies. 


+ATALCO!PICO +ATALCO 3
Terry Fitzpatrick is Methanol Technology
Manager within the GTL group of Johnson
Matthey Catalysts. His work has a particular
emphasis on the development of technology Figure 12 In-situ radial pellet strength of Katalco Apico
and catalyst applications for methanol
chemistry. He joined the catalyst business
(then ICI) over 20 years ago and has a Operating benefits of Katalco Apico
bachelor’s degree in chemical engineering from
Cambridge University, UK.
Tom Hicks is a consultant with Johnson Feature Benefit to operator Extra MeOH, tonnes
Pre-reduced catalyst Faster start-up 3000
Matthey Catalysts. He has worked on steam
Higher, stable activity Additional MeOH made 55 000
reforming catalysts and technology, shift and Doubled catalyst life Eliminated change-out 23 750
methanol catalysts, acetylene hydrogenation 50% lower by-products Increased efficiency 45 000
catalysts and ammonia synthesis technology Total 126 750
for ICI and now Johnson Matthey, and has a
bachelor’s degree in chemistry from Durham
University, UK. Table 1

www.eptq.com CATALYSIS 2010 53

jm.indd 6 2/2/10 13:49:27


Catalytic steam generation

Exploration of an unconventional fuel mixture has resulted in a novel


catalyst-based technology

Franck Letellier and Dave Wardle


Oxford Catalyst Group

A
spin-off from the Department exhibit conversion efficiencies in the This catalyst has been optimised for
of Chemistry at the range of 70% per pass and are use in FT microchannel reactors, to
University of Oxford in 2004, designed for economical production enable the small-scale distributed
Oxford Catalysts works on the on a small scale. A single micro- production of next-generation
development of metal carbide cata- channel reactor block might biofuels from a wide range of waste
lysts for the generation of clean produce up to 40 barrels (bbls) of feedstocks. Demonstrations of FT
fuels from both conventional fossil liquid fuel per day. In contrast, microchannel reactor technology, to
fuels and sustainable, renewable conventional FT plants are designed produce liquid fuels from biomass,
sources such as biomass waste. to work at minimum capacities of are due to take place in 2010. The
These include improved hydrodes- 5000 bbl/day, and function well technology is also being adapted to
ulphurisation (HDS), chemisorption, and economically at capacities of create small-scale gas-to-liquid
steam methane reforming (SMR) 30 000 bbl/day or higher. They (GTL) facilities for use offshore to
and Fischer-Tropsch (FT) catalysts. typically exhibit conversion efficien- capture the energy from flare gas.
The group, as a whole, focuses on cies of 50% or less per pass.
developing new and improved Instant steam
catalysts and catalyst-based tech- Although the group’s main focus is
nologies to create cleaner fuels for Micro­channel on technology for cleaner and
the future. Within the group, greener liquid fuels, it also has a
Velocys concentrates on the devel- processing enables novel catalyst and fuel combination
opment of microchannel reactor that produces instant steam. This
technology, while catalyst develop- chemical reactions development is based partly on
ment is carried out by Oxford work originally carried out by Dr
Catalysts.
to occur 10 to 1000 Tiancun Xiao, one of the founders
Microchannel reactors are com- and Senior Scientific Advisor at
pact reactors that have channels
times faster than in Oxford Catalysts, as part of his
with millimetre-sized diameters. conventional systems search for a catalyst-based reaction
Small-diameter channels dissipate to create a portable hydrogen
heat more quickly than conven- generator for powering fuel cells.
tional reactors with larger-diameter Xiao’s starting point was an uncon-
channels in the range of 2.5–10cm Oxford Catalysts has also devel- ventional fuel mixture of an organic
(1–4 inch), so more active catalysts oped and patented a catalyst compound (methanol), with hydro-
can be used. Mass and heat transfer preparation method, known as gen peroxide passed over a
limitations reduce the efficiency of organic matrix combustion (OMX), platinum catalyst to produce hydro-
large, conventional high-pressure which makes it possible to achieve gen at room temperature via an
reactors used for hydroprocessing. high metal loadings while reducing exothermic reaction:
Micro­channel processing enables the need for precious metal promot-
chemical reactions to occur 10 to ers and enabling precise control of H2O2 (hydrogen peroxide) + CH3OH (methanol)
1000 times faster than in conven- crystallite sizes. OMX played a key ==> 2 H2 + CO2 + H2O
tional systems. role in the development of these
Microchannel FT reactors, devel- improved HDS catalysts, which are The large amounts of heat and
oped by Velocys and using a new, designed to enable refineries to steam produced during this reac-
highly active metal carbide FT cata- meet the requirements for cleaner tion also made it possible to
lyst developed by Oxford Catalysts, fuels from less than ideal feed- reformulate the fuel to produce
will soon be trialled for the small- stocks. OMX also lies behind the high-temperature steam at room
scale distributed production of development of the new, highly temperature and pressure via the
biofuels from waste. These reactors active metal carbide FT catalyst. reaction:

54 CATALYSIS 2010 www.eptq.com

oxford.indd 1 1/2/10 11:28:16


3 H2O2 + CH3OH ==> CO2 + 5 H2O additional challenge involved
designing a suitable reactor —
This work attracted the a capsule to contain the
seed funding from Oxford catalyst — to ensure the fuel
University that led to the was presented and distributed
establishment of Oxford over the catalyst in the most
Catalysts. It also formed the effective way. It also required
basis for the company’s first determination of the right
patents. particle size and diluent for
Although initially the idea the catalyst; finding a way to
was to develop the reaction for manage the heat to ensure the
hydrogen production, in the catalyst bed did not overheat
short term the commercial or become too cold; and find-
uses for high-temperature ing ways for the reactor to
steam were more obvious. As tolerate changes in the flow of
a result, the focus soon shifted fuel.
to developing the reaction as a The team then had to
way of generating portable consider how to adjust the
steam. The fact that the reac- fuel, catalyst and reactor
tion starts at room temperature parameters for the different
in just a few seconds sug­- applications envisaged for the
gests a range of possible technology. These fall broadly
applications. into two main groups —
industrial and consumer
Demonstrations goods — and both have very
The prototype demonstrations different requirements. In
of portable, instant steam were Figure 1 A prototype instant steam reactor industrial applications, the
impressive. In one demonstra- steam has to be extremely hot
tion, an ordinary plastic spray fuel mixtures to produce steam of and the catalyst very durable and
bottle was used to hold the fuel, different temperatures, as well as able to remain active for many
and the nozzle was adapted to hold identifying a catalyst support mate- hours. For consumer applications,
the catalyst and serve as the reac- rial capable of surviving the hot the requirement is to provide the
tor. When the trigger was squeezed, oxidising and water-wet atmos- right amount of catalyst to last just
high-temperature steam emerged phere produced by the reaction. the right amount of time to do the
from the nozzle (see Figure 1). Since water produced by the reac- job.
Another prototype reactor, the size tion also inhibits the reaction, the
of a sugar cube, was able to pump steam team had to discover the Conclusion
steam at a rate of 7 l/min at up to right balance between fuel composi- The effort expended has provided a
800°C from room temperature in foundation for evaluating this new
just 1–2 minutes, while a third technology. Although the commer-
prototype, just 2cm high, proved The technology has cial value of the instant steam
capable of producing 70 l/min of technology has still to be fully eval-
650°C steam. The technology has been developed uated, the group is in discussions
been developed so that it can now about industrial applications where
produce steam at 100–800°C.
so that it can now it is an advantage to be able to
The steam team had to develop a
new platinum-based catalyst espe-
produce steam at create superheated steam instanta-
neously and in difficult and
cially for the reaction. Moving 100–800°C unusual places.
towards commercialisation of this
new technology presented addi-
tional challenges. The overall goals tion and catalyst activity to enable
were to reformulate the fuel to the reaction to proceed efficiently. Franck Letellier is a Senior Research
enhance safety and to reduce Developing new methods to test Scientist and leader of the Steam Team at
Oxford Catalysts. He has an MSc and PhD in
costs, and to enhance catalyst variables, including catalyst life-
heterogeneous catalysis applied to petroleum
performance. time, catalyst activity, steam
processes from IFP and Marie Curie University,
Literature searches offered some temperatures, flow rates, space France.
clues as to the best way to tackle velocity through the reactor and the Dave Wardle is Business Development
outstanding problems. These composition of effluents, presented Director, Instant Steam, at Oxford Catalysts.
ranged from increasing the life and further difficulties. Since steam is He has a BSc from the Camborne School of
activity of the catalyst while reduc- generated by passing the fuel Mines and was a senior manager at the BOC
ing catalyst cost, to formulating safe mixture over the catalyst, an Group before joining Oxford Catalysts in 2007.

www.eptq.com C ATALYSIS 2010 55


oxford.indd 2 1/2/10 11:28:29


Alphabetical list of advertisers
Albermarle Catalyst 9 Haldor Topsøe IFC
www.albemarle.com www.topsoe.com

Amistco Separation Products 30, 32 W C Heraeus 49


www.amistco.com www.heraeus.com

Axens OBC Intercat IBC
www.axens.net www.intercatinc.com

BASF Catalyst 7 Johnson Matthey Catalysts 20


www.basf-catalysts.com www.jmcatalysts.com

Buchen-ICS 19 MBI Global 46


www.buchen-ics.com www.blastresistantbuildings.com

CB&I 14 Mourik International 29


www.cbi.com www.mourik.com

Chevron Lummus Global 13 Sabin Metal Corporation 34


www.chevronlummus.com www.sabinmetal.com

CRI Catalyst Company 45 Saint-Gobain NorPro 36


www.cricatalyst.com www.norpro.saint-gobain.com

Criterion Catalyst 4 Süd Chemie 10


www.criterioncatalysts.com www.sud-chemie.com

ExxonMobil Research and Engineering 27 Tricat Catalyst Service 24


www.exxonmobil.com/refiningtechnologies www.tricatgroup.de

Grace Davison 2 Zeochem 41


www.gracedavison.com www.zeochem.com

For more information on these advertisers, go to www.ptqenquiry.com

56 CATALYSIS 2010 www.eptq.com

ad index copy 5.indd 1 2/2/10 13:44:36


www.ptqenquiry.com
for further information

intercat.indd 1 29/1/10 13:00:10


Our advanced
catalytic engineering
has powerful attraction
Axens’ HR series ACE™ technology catalysts deliver sustained, on-spec ultra-
low-sulfur diesel. The powerful desulfurization activity of this new series is
particularly enhanced in combination with EquiFlow™ reactor internals and
Catapac™ dense loading technology.

With its submicron-level control, ACE technology provides dual activity for
superior levels of sulfur and nitrogen removal in all applications. Easy to load
and activate, HR series catalysts also offer excellent regeneration properties.

Axens’ quality products are based on reference technology – commercially


proven, dependable, reliable and cost-effective.

The performance improvement specialists


www.axens.net

For more information www.ptqenquiry.com


Paris +33 1 47 14 25 14 Houston +1 713 840 1133 Email Catalysts@axens.net for further information

HR_ptq.indd 1 28/11/08 12:44:39


axens.indd 1 29/1/10 13:02:15