PTQ Q1 2013

SPECIAL FEATURES
OUTLOOK 2013
AUTOMATION & CONTROL
ENVIRONMENTAL
REFINING
GAS PROCESSING
PETROCHEMICALS
PETROLEUM TECHNOLOGY QUARTERLY
ptq
Q1 2013
cover and spine copy 14.indd 1 7/12/12 11:26:25
Are you looking to step up plant perform ance?
Topses next generation B R IM catalysts offer refiners the opportunity to increase
perform ance through an increase in catalyst activity.
U sing the original B R IM technology Topse has developed several new catalysts, resulting
in higher activity at low er lling densities.
The next generation B R IM catalysts display
- high dispersion
- high porosity
- high activity
W e look forw ard to stepping up your perform ance!
W W W .T O P S O E .C O M
Stepping up perform ance
next generation B R IM

technology
topsoe.indd 1 6/12/12 12:55:24
2013. The entire content of this publication is protected by copyright full details of which are available from the publishers. All rights
reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic,
mechanical, photocopying, recording or otherwise without the prior permission of the copyright owner.
The opinions and views expressed by the authors in this publication are not necessarily those of the editor or publisher and while every care
has been taken in the preparation of all material included in Petroleum Technology Quarterly the publisher cannot be held responsible for any
statements, opinions or views or for any inaccuracies.
3 A 2020 vision of biofuels
Chris Cunningham

5 Outlook 2013

15 ptq&a

19 Refnery design in China
Sun Lili Sinopec Engineering Incorporation
31 Maximising ethane in liquids crackers
Muhammad Imran Technip Stone & Webster Process Technology
43 Tracking performance degradation in a debutaniser column
Ali Abdulal, Vinod Ramaseshan and Stanley Gustas III Saudi Aramco
Horia Orenstein, Mohammad Kurdi and Andrey Gaskov SAS Middle East
53 Operator training simulator for a multi-product fuels pipeline
Santosh Alladwar, Cordell Chapman, Billy Mayo, Surajit Dasgupta and
Javier Vazquez-esparragoza KBR
Daniel Vogt Krohne Oil and Gas
Keone Jackman Petroleum Company of Trinidad and Tobago Limited
63 Refnery operations gain from advances in digital automation
Tim Olsen Emerson Process Management
69 Controlling flm temperature in fred heaters
Jinyu Jiao, Yuriy Morayko, Morten Theilgaard and Michael Ho WorleyParsons Canada
73 High-capacity tray for debottlenecking a crude distillation unit
Kaushik Majumder Shell Projects & Technology
Giuseppe Mosca Sulzer Chemtech
Kent Mahon Refning NZ
79 Steam reformers for hydrogen and synthesis gas
Norm Pelletier and Goutam Shahani
Selas Fluid Processing Corporation, a division of Linde Engineering
85 VDU vacuum system revamp
Ahmet Bebek Turkish Petroleum Refneries Corporation
89 Reducing carbon emissions with antifoulants
India Nagi-hanspal, Mahesh Subramaniyam, Parag Shah and James Noland
Dorf Ketal Chemicals
95 Towards a zero (net) carbon refnery
Michael Stockle Foster Wheeler
101 Modular design of smaller-scale GTL plants
Kenneth Roberts Velocys
105 Starting up hydrotreating units with cracked feeds
Randy Alexander, Reinder Van Bruggen and Pierre Dufresne Eurecat US
111 Optimising a process scheme for platforming heaters
Adil Rehman, Suman Pachal, Shyam K Choudhary, Ugrasen Yadav and M K E Prasad
Technip KT India
119 Improved hydrogen yield in catalytic reforming
Roberto Amadei Chemical & Energy Development
127 Product News
Totals Central Germany refnery at Leuna, where Bilfnger Berger Industrial Services provides maintenance services.
Photo BIS Group
Q1 (J an, Feb, Mar) 2013
www.eptq.com
p
t
q
Y L R E T R A U Q Y G O L O N H C E T M U E L O R T E P

contents/ed com copy 17.indt 1 11/12/12 16:04:37
STRATEG IC
SO LU TIO N S
O PERATIN G
SO LU TIO N S
O RG A N ISATIO N A L
SO LU TIO N S
EN VIRO N M EN TA L
SO LU TIO N S
C A PITA L
SO LU TIO N S
SO FTW A RE
SO LU TIO N S
W hat is
Next Gen Per f ormance
SM
?
N extG en Perform ance is optim ising each and every opportunity. Its
addressing challenges before they becom e problem s and applying
solutions before others have fully considered them .
Its the result of tw o parties you and KBC w orking in a side-by-side
partnership, com bining our know ledge and experiences.
Why partner with KBC?
KBC is a leading independent consultancy to the energy sector
w orldw ide. W ith our proven m ethodologies and industry-leading
softw are, our experienced consultants consistently dem onstrate
innovative and creative thinking, fexibility, and a penchant for identi-
fying and im plem enting im provem ents that result in tangible value.
W e are dedicated to continually im proving our consultancy and
softw are solutions by being responsive to your requests, keeping
abreast of new technologies, and anticipating the needs of an
ever-changing industry.
W e m easure our success in term s of your success. Let us help
you achieve N extG en Perform ance.
Americas
+1 281 293 8200
Europe, Middle East , and Africa
+44 (0)1932 242424
Asia
+65 6735 5488
answers@kbcat.com w w w.kbcat.com
2011 KBC A dvanced Technologies plc. A ll rights reserved.
G iven our global experience in
w ide-ranging practice areas and
client sectors, KBC is uniquely
positioned to provide a fully
integrated response to your
m ost signifcant challenges.
kbc.indd 1 7/3/12 13:58:03
T
he European Union has arguably
been the global leader in biodiesel
production and use, with overall
biodiesel production increasing from 1.9
million tonnes in 2004 to nearly 10.3 million
tonnes in 2007. Biodiesel production in the
US has also increased dramatically in the
past few years from 2 million gallons in
2000 to approximately 450 million gallons
in 2007. According to the National Biodiesel
Board, 171 companies own biodiesel
manufacturing plants and are actively
marketing biodiesel.1. The global biodiesel
market is estimated to reach 37 billion
gallons by 2016, with an average annual
growth rate of 42%. Europe will continue to
be the major biodiesel market for the next
decade, followed closely by the US market.
Although high energy prices,
increasing global demand, drought
and other factors are the primary
drivers for higher food prices, food
competitive feedstocks have long
been and will continue to be a major
concern for the development of biofu-
els. To compete, the industry has
responded by developing methods to
increase process effciency, utilise or
upgrade by-products and operate
with lower quality lipids as
feedstocks.
Feedstocks
Biodiesel refers to a diesel-equivalent
fuel consisting of short-chain alkyl
(methyl or ethyl) esters, made by the
transesterifcation of triglycerides,
commonly known as vegetable oils or
animal fats. The most common form
uses methanol, the cheapest alcohol
available, to produce methyl esters.
The molecules in biodiesel are pri-
marily fatty acid methyl esters
(FAME), usually created by trans-
esterifcation between fats and metha-
nol. Currently, biodiesel is produced
from various vegetable and plant oils.
First-generation food-based feedstocks
are straight vegetable oils such as
soybean oil and animal fats such as
tallow, lard, yellow grease, chicken fat
and the by-products of the production
of Omega-3 fatty acids from fsh oil.
Soybean oil and rapeseeds oil are the
common source for biodiesel produc-
tion in the US and Europe in quanti-
ties that can produce enough biodie-
sel to be used in a commercial market
with currently applicable
PTQ Q1 2013 3
Editor
Chris Cunningham
editor@petroleumtechnology.com
Production Editor
Rachel Storry
production@petroleumtechnology.com
Graphics Editor
Rob Fris
graphics@petroleumtechnology.com
Editorial
tel +44 844 5888 773
fax +44 844 5888 667
Business Development Director
Paul Mason
sales@petroleumtechnology.com

Advertising Sales
Bob Aldridge
sales@petroleumtechnology.com
Advertising Sales Offce
tel +44 844 5888 771
fax +44 844 5888 662
Publisher
Nic Allen
publisher@petroleumtechnology.com
Circulation
Jacki Watts
circulation@petroleumtechnology.com
Crambeth Allen Publishing Ltd
Hopesay, Craven Arms SY7 8HD, UK
tel +44 844 5888 776
fax +44 844 5888 667

ptq (petroleum technology quarterly) (ISSN
No: 1632-363X, USPS No: 014-781) is published
quarterly plus annual Catalysis edition by Crambeth
Allen Publishing Ltd and is distributed in the US
by SP/Asendia, 17B South Middlesex Avenue,
Monroe NJ 08831. Periodicals postage paid at New
Brunswick, NJ. Postmaster: send address changes to
ptq (petroleum technology quarterly), 17B South
Middlesex Avenue, Monroe NJ 08831.
Back numbers available from the Publisher
at $30 per copy inc postage.
Vol 18 No 1
Q1 (J an, Feb, Mar) 2013
A 2020 vision of
biofuels
I
t is not just another new year; it is a year closer to 2020, when the worlds
commitment to the Kyoto protocol will be put to the test. Although
governmental talking shops on global strategy continue to make little
progress, the steady march of the decade seems to be concentrating refning
minds on meeting carbon limits. And a big benefciary looks set to be the
biofuels industry.
During November, the US Senate voted to cancel restrictions in its annual
defence policy bill that would have severely limited the militarys efforts to
develop biofuels for its jet aircraft. The National Defense Authorization Act
usually attracts most attention for policy changes on matters such as deten-
tion of terrorists, but the 2013 version brought biofuels to the fore by amend-
ing a rule that would have barred the military from buying biofuels if they
cost more than petroleum-derived jet fuel.
There remained the need to reconcile the Senate vote with the version yet
to be approved by the House of Representatives, which included limits on
the military use of biofuels, before the bill was ready for presidential signa-
ture. However, the latest move by US lawmakers is a major boost for the
alternative fuels industry. Production of relatively small batches of biofuels
inevitably costs far more than conventional, petroleum-based production of
jet fuel. As long as the price comparison test remained, there would be lim-
ited opportunity for a signifcant contribution of biofuels to the military
requirement of distillates. Since military approval generally leads the estab-
lishment of developing technologies, this was signifcant progress.
The US Air Force has been testing small but expensive batches of alterna-
tive fuels in its aircraft to prove they can be reliably used once prices become
competitive with petroleum fuels. For its part, the US Navy has been the big-
ger spender on biofuels for both warships and aircraft as part of its green
feet programme. With a 2020 deadline in mind, the Navy has set that year
to meet a target of 50% biofuels in its non-nuclear fuel pool by that date.
While opponents of biofuels in the US argue that freeing the use of cur-
rently expensive fuels is poor timing during a period of cutbacks, supporters
say that home-grown fuels support the domestic farming industry and create
jobs while protecting the military from price shocks in the international
petroleum markets. Bioethanol is, of course, big business for Midwest corn
growers, but it is not clear which US farming communities will be meeting
the vastly increased demand for vegetable oils to make jet fuel or biodiesel.
There is not much open mention of climate change amidst that controversy,
but there is unreserved acknowledgement of the Kyoto protocol and its
regional interpretation in other developments in favour of renewables. Enis
plan for wholesale conversion of its Venice refnery to biofuels production
amidst plant closures in Europe is based on environmental targets set by the
European Union. The Venice plan contrasts with news elsewhere in Italy,
where several refneries have already closed or will close temporarily in 2013.
In India, a leading refner has been reviewing the options for introducing
bio-jet fuel production, and we aim to report in PTQ the outcome of those
deliberations later in the year. After its public relations disasters of a few
years ago, it appears that the approach of 2020 may be turning the tide in
favour of the biofuels industry.
CHRIS CUNNINGHAM
p
t
q
Y L R E T R A U Q Y G O L O N H C E T M U E L O R T E P
contents/ed com copy 17.indt 2 11/12/12 14:54:39
Theres only one way to stay ahead of
the competition: always be evolving.
Meet Max Ovchinnikov:
Industry Game-Changer.
W hen youre leading the w orld in catalyst developm ent, your w ork is never done and thats just ne
by M ax O vchinnikov, C RITERIO N s Senior Research C hem ist and co-developer of such advanced
catalyst platform s as SEN TRY and C EN TERA
. H e, along w ith fellow scientists from Shells three R&D

centers, has m ade a career developing high throughput techniques that lead to the continual im provem ent
of catalysts. This m eans as M ax is evolving the com panys process and products, and launching a new
catalyst every three or four years, C RITERIO N is alw ays in front w hen it com es to innovation.
Leading minds. Advanced technologies.
www.CRITERIONCatalysts.com
criterion.indd 1 10/12/12 10:33:13
Maria van der Hoeven
Executive Director
International Energy Agency
W
ith relatively rapid shifts
on both the supply and
demand sides, the global
oil map is changing. That is the
message from two major recently
released IEA reports: the World
Energy Outlook 2012, which looks
out to 2035, and the Medium-Term
Oil Market Report 2012, which looks
out to 2017. Those shifts are happening along the entire
value chain, from upstream to downstream. But one
area where they are particularly stark is in refning.
Recent years have seen a steady shift from OECD
countries towards newer and often more advanced
refneries in emerging economies. Some 1.2 million b/d
of crude distillation capacity was shed from the OECD
in 2012, following 0.8 million b/d in 2011. Since the
economic downturn of 2008, 4.0 million b/d of primary
distillation capacity has been shut or scheduled to shut
in developed countries.
And yet, in 2012, the refning industry witnessed an
impressive recovery from end of 2011 lows, even in the
OECD. Europe, whose refnery industry had been
especially pressured by declining demand and surplus
capacity, saw margins and, in turn, refnery
throughputs rebound more strongly than other refning
centres, albeit from a low starting point. Regional
industry rationalisation has been especially severe in
Europe. In September, Brent cracking margins reached
their highest levels since at least 2006 in Northwest
Europe. Even simple hydroskimming margins
performed relatively well, with Northwest Europe
Brent attaining positive levels for a ffth consecutive
month, resulting in high regional throughput rates.
Other major refning centres also recovered, as product
markets are becoming increasingly globalised and
product trade is surging. The US refnery industry
continued to be supported by discounted feedstock
and refnery fuel, spurred by the impressive surge in
domestic oil and gas production.
The recovery in refnery proftability is expected to
be short-lived, however, with margins falling
back from current highs in most refning centres, as
localised product tightness eases. After three years
of shrinking spare capacity, global refnery
P
additions are set to exceed demand growth from 2013
onwards.
The Medium-Term Oil Market Report 2012 highlights the
fact that global refnery expansion plans are projected to
add 7.0 million b/d of crude distillation (CDU) capacity
in the fve years after 2012, reaching 100.5 million b/d in
2017. Non-OECD growth is partially offset by an
acceleration of refnery closures in the OECD, which are
now seen shedding more than 1.1 million b/d of
capacity in the same period. Among non-OECD
countries, Asia accounts for over 50% of additions. It is
led by China, which is still expected to expand its
distillation capacity by some 2.9 million
b/d, despite a more subdued outlook for domestic
demand growth. The rest of Asia, dominated by India,
is set to add 1.4 million b/d of capacity in the period.
The additions are concentrated in the earlier years of the
forecast period; indeed, a large portion of the new
capacity has already been commissioned earlier in 2012.
Major expansions are also taking place in the Middle
East, with at least two mega-projects in Saudi Arabia
and one in the United Arab Emirates contributing to 1.9
million b/d of aggregate regional incremental capacity.
Global net additions have been signifcantly lower
than previously seen, thanks to the acceleration of
closures in the OECD and some project delays in the
non-OECD over the past year. But they still exceed
projected oil product demand growth of 6.7 million
b/d over 2012-2017. All this demand growth is coming
from developing countries, whereas structural decline
is set to continue in the OECD. Furthermore, an
increasing share of oil product demand will be met by
supplies bypassing the refnery system. Those include
biofuels, NGLs, crude burned directly at power stations
and so forth. For refnery proftability to be maintained,
further closures or project delays/cancellations are
needed, or the industry will again see a prolonged
period of depressed margins worldwide. In that case,
the OECD will again bear the brunt, with further
structural effects.
Indeed, looking out over the longer term, the tectonic
shifts look set to continue. Even in an environment of
low margins, newer non-OECD plants will beneft from
their proximity to future crude production, as well as
the need to meet more stringent fuel quality standards
and increase light product yields (particularly low-
sulphur diesel). In some cases, whole facilities will be
bought up and shipped east. And there will be a
longer-term need worldwide for more hydrotreating
and upgrading of capacity to accommodate more
www.eptq.com PTQ Q1 2013 5
Outlook for 2013
What are the important trends affecting the downstream processing industry this year?
Executives and experts forecast challenges and prospects that could affect proftability
outlook copy.indd 1 7/12/12 14:47:54
heavy oil feedstock, as well as capacity to accommodate
more NGLs.
The trends we have observed over the past several
years are part of a fundamental rebalancing in the
global oil economy and they are here to stay, despite a
recent respite for European refners.
Eric Benazzi
Marketing Director
Axens
W
hile writing these lines,
forecasts indicate that
global GDP growth
should be about 3.6% for 2013,
showing that the recovery
continues even if it has weakened.
In fact, over the past few months,
global growth has slowed down
to its weakest pace since the 2008-
2009 recession. The latest forecasts for the second half
of 2012 indicate that, in the US, GDP is projected to
expand by about 1.5% (on an annualised basis), Japan
will probably stagnate and the GDP of the troubled
Euro zone is predicted to decline by 0.7%.
The most troublesome aspect of the current situation
is that the slowdown is also affecting emerging
countries. Among the four BRIC countries, Brazils
growth will probably be limited to 1.5% in 2012.
Chinas GDP, which until recently was in double digits,
has slowed down to a value of around 7%, indicating
that Chinas authorities have not succeeded in shifting
from an export-driven economy to one that relies on
domestic consumption. India will remain just below
5%, down from 7% last year, mainly because
investment has continuously slowed down.
Projected global GPD growth, 3.6% for 2013, will be
reached only if there is suffcient policy action in order
to prevent the so-called fscal cliff in the US and to
ease fnancial conditions in periphery economies of the
Euro zone. Regarding China, policy action will be
required to stimulate the shift of the Chinese economy
towards more domestic consumption. Otherwise, the
world economy could slow down even further.
As a result, the growth of demand for refnery
products could be limited to 0.8 million b/d, leading to
a global oil demand of about 90.5 million b/d in 2013.
In North America and probably in other regions, the
exploitation of unconventional gas and oil resources
will continue to be a game changer. In spite of that,
we have to remember that, in the long term, the crude
oil market will remain structurally tight and the era of
cheap crude oil is over.
Thus, in 2013, we can predict:
Tightening fuel quality specifcations worldwide will
call for additional hydrotreating capacities and even
higher performing catalysts. Impulse, our latest
hydroprocessing catalyst technology, is the result of a
long-term commitment to clean fuels and our answer
to this trend
Integration between refning and petrochemicals will
allow the recovery of valuable products from low-
value heavy feeds. High-propylene FCC, HS-FCC and
resid to propylene technologies enable the production
of propylene with a yield ranging from 8 to 20 wt%
from atmospheric or vacuum residue
Squeezing the bottom of the barrel will be necessary
to maximise motor fuel yields. In regions such as North
America, where unconventional gas is available at low
cost, cheap H
2
can be produced. Axens studies along
with clients feedback have shown that, in the case of a
refnery fed with bitumen and syncrude oil, the
addition of a residue hydrocracker ahead of an existing
delayed coker enables the debottlenecking of the
refnery and so improves its proftability
In mature markets such as Europe, increased
fexibility through revamping will contribute to reduce
the gasoline/diesel imbalance as well as the PolyFuel
technology, which transforms olefns contained in light
FCC cuts into good-quality middle distillates via
oligomerisation reactions
Process energy effciency improvement will continue
to be a must to reduce opex and improve asset
proftability.
Rajeev Gautam
President & CEO
UOP LLC, A Honeywell Company
T
aking a moment to look back,
2012 was a challenging and
exciting year for the refning,
petrochemicals and gas processing
industries. Continued shifts in
product mix to increase diesel
production affected strategies
globally, the use of heavy
crude as a feedstock remained
strong, and the value of refning and petrochemical
integration grew. The burgeoning natural gas
market and an expanding interest in renewable fuel
sources presented new challenges. Innovative
technology solutions to address these trends will
position the industry for great success in 2013 and
beyond.
While many challenges faced in recent years will
continue into 2013, the stakes will be higher. Continued
population growth and expansion of the middle class
will result in higher demand on energy sources such as
coal, oil and natural gas. Government regulations will
dictate tighter fuel specifcations and higher renewable
fuel content. Economic fux will force higher yields at
lower costs. The effects will vary by region, but these
challenges will be experienced worldwide. Across the
globe, technologies will provide the effcient,
economical and sustainable solutions to meet these
challenges head on and to build better business models
for stability.
Refners will need to fnd ways not only to maximise
fnished products from each barrel of oil, but to
6 PTQ Q1 2013 www.eptq.com
cbi.indd 1 6/12/12 13:24:13
upgrade their product slates for higher value. UOPs
recently commercialised Unifex process provides a
signifcant advantage by delivering an impressive 90%
conversion of low-quality residue streams to high-
quality transportation fuels, while minimising
byproducts.
In addition to maximising the yield from every
barrel, refners globally will be looking for ways to
improve the overall value of their product slate by
increasing petrochemicals production. Technology to
produce high-yield, high-reliability propylene, such as
UOPs Olefex technology, will remain an attractive
option as more economical LPG production comes on
stream.
In areas with limited availability to conventional
olefn feedstocks, the utilisation of cheaper, more
abundant feedstocks, such as coal, will be necessary.
Flexible process technology, such as the UOP/Hydro
Methanol-to-Olefns (MTO) process, will be vital for
helping producers address demands in unconventional
ways.
A short time ago, natural gas was, in many locations,
a byproduct of producing crude oil and it was often
fared off at the well site. Today, natural gas has
become a way towards high-value petrochemical
feedstocks, and it is on track towards becoming one of
the worlds top energy sources.
Global unconventional gas resources are estimated
to be enough to provide 250 years of supply,
providing a long-term opportunity for participants in
this market. Effcient contaminant removal,
purifcation and liquids recovery will be among the
essential steps to monetise natural gas resources. The
new UOP Russell product line offers packaged plants
that enable shale and conventional natural gas
producers to remove contaminants and recover high-
value natural gas liquids used for petrochemicals and
fuels.
Renewable fuels also have an enormous potential to
contribute to our energy supply. The frst UOP
Ecofning unit for renewable diesel production is
constructed and should be online shortly, with more
units in the fnal stages of funding. The technology is
preferred by refners due to the drop-in nature of the
fuel product leveraging existing infrastructure. This
year, our Rapid Thermal Processing (RTP) technology
will be used to upgrade renewable fuel oil into
transportation fuels. As customers and governments
continue to realise the importance of alternative fuel
options, and sustainable feedstocks become more
readily available, the momentum for renewable fuels
will continue to grow.
In 2013 and beyond, success will be achieved
through technology solutions. Whether trying to get
more value from every barrel processed, meeting
rising energy and petrochemical demands, or further
exploring how renewable fuels ft into business
strategies, technology will provide the answers that
will allow the industry to meet current and future
challenges, successfully providing energy globally for
years to come.
Ujjal Mukherjee
Vice President, Technology
Chevron Lummus Global
C
hevron Lummus Global
(CLG) has seen a marked
increase in the volume of
inquiries for hydrocracking, residue
upgrading and lubricant base oil
manufacturing, while Chevron
and Lummus Technology, the
parent companies, are very busy
addressing technology requests
for gas-to-liquids, ethylene, propylene, aromatics
and butadiene. From peaks in 2006-2007, equipment
pricing, especially those for high-pressure equipment,
has dropped signifcantly, leading to better project
economics.
The megaprojects are being driven either by large
national oil companies such as Rosneft, Aramco,
Petrobras and Petrochina or by large private companies
such as Reliance and SK. Despite the torrent of bad
news on the economic front from Europe and slight
slowdown in the Chinese, Indian and Brazilian
economies, crude prices remain high and most
projections peg crude to be in the $120-$140 range by
2020. At these prices, upgrading of opportunity crudes
remains attractive and some megaprojects may
materialise in Venezuela and Colombia. These projects
will involve consortia and complex project fnancing.
In the US, the refneries in the Gulf Coast retroftted
early to handle the diffcult Mayan crudes and
synthetic crude oil from Canada and Venezuela. These
refners have the option of continuing to import the
opportunity crudes or to process easier shale oils from
Eagleford or Bakken formations. In the Eastern US,
refners who have not upgraded to handle opportunity
crudes will continue to see depressed margins,
although even these refners may see an improvement
in margins driven by cheaper hydrogen costs from
shale gas and the availability of Bakken oil.
The availability of shale gas has also made the
production of ethylene proftable, and several large
projects were recently announced in the US Gulf Coast.
The product mix in the US is shifting signifcantly from
gasoline towards a mix of diesel and gasoline, with
many refners evaluating catalytic and process
solutions to make more diesel barrels.
The large new export refnery projects in the Middle
East, in Jubail, Yanbu and Jazan, all target Euro V
diesel. Most new Russian projects are targeting the
production of diesel and, in some instances, high-value
lube oil base stocks. In South East Asia, all
hydrocrackers target diesel and, occasionally, diesel
and lube oil base stocks. Although India is in a fuel
surplus situation today, with the completion of the
super highways across the country, the truck traffc
will grow exponentially, leading to a surging demand
for diesel. In China, the demand for gasoline and
diesel both remain strong. The increased gasoline and
petchem demand has spurred demand for residue
The Emerson logo is a trademark and a service mark of Emerson Electric Co. 2012 Emerson Electric Co.
See more, do more and be more profitable with the most trusted partner in wireless Emerson.
Emerson is your proven partner with Smart Wireless in more customer sites and with more operating
hours than anyone else in the process industry. Smart Wireless has the widest range of technologies
to expand your vision into more places across your operations. And its self-organising mesh network
delivers the highest reliability available. It is simply the most intelligent, secure and cost-effective
operation-wide wireless option available. See how Smart Wireless can empower your bottom line
at www.EmersonProcess.com/SmartWireless
Using wireless here and there is one thing.
But using it across my entire operation?
Theres no one I could trust to do that.
ITS
IMPOSSIBLE
115802 ins17_itsPossClimbers_PTQ.indd 1 31/08/2012 16:10
emerson.indd 1 11/9/12 11:41:27
desulphurisation projects with the bottoms feeding
RFCC units.
CLG has witnessed a growing global interest in
residue hydrocracking technology as standalone
solutions or in conjunction with existing delayed coking
units. Demand for residue hydrocracking is driven by
uncertainties about the future of high-sulphur residual
fuel oils and poor demand for petcoke.
For sustained proftability, a close integration of
refning and petrochemical complexes is essential. The
bottomless refnery would produce diesel, lube oil base
stocks and reformate for conversion to aromatics, in
particular paraxylene. Light naphtha from the refnery
would be converted to ethylene and propylene.
Butadiene production would remain a very attractive
option. Biofuels, driven by European airlines and the
US military, will become a necessary part of refnery
planning.
Umberto della Sala
President & Chief Operating Offcer
Foster Wheeler
W
e are certainly seeing
robust activity in all of
the hydrocarbon-related
business sectors in which Foster
Wheelers Global Engineering
and Construction Group operates:
onshore and offshore upstream oil
and gas, midstream/LNG, refning,
chemicals and power, as well as
in pharmaceuticals, environmental and metals and
mining, where we also operate.
Certainly, we have a good prospects pipeline covering
all of our business lines, including a number of projects
that are going through the fnal investment decision-
making process and for which we believe we are well
positioned. And we are seeing new opportunities
continuing to emerge, particularly in Asia, the Middle
East and the Americas.
There are some key themes that we are seeing today,
which look set to continue into 2013 and beyond. First
is local service delivery. This has always been important
to us and is becoming an even stronger area of focus
for us and for our clients. We have made further strides
forward this year, for example in establishing or
expanding operations in key locations such as Saudi
Arabia, Canada, Colombia, Indonesia and Malaysia. In
some cases, we are developing our own resources and
in others we are working together with local or regional
clients and/or contractors to enable us to deliver the
global Foster Wheeler brand locally and competitively
to our clients, for the long-term, in line with local
content requirements and our clients preferences.
Second is the emphasis on upstream. Many clients
with upstream and downstream businesses are
allocating more and more of their planned capital
spend to the upstream sector (many of them include
LNG in this category). The offshore and onshore
upstream sector remains a strategically important
market for us and one in which we are further
developing our skills, service portfolio and geographic
presence. We have expanded our upstream operations,
opening a new offshore operation in Woking, UK, and
expanding our upstream operations in Malaysia and
Trinidad and Tobago.
The third theme relates to the size and complexity of
projects. The scope and scale of some of todays
planned investments in the Middle East, the Americas
and Asia are truly impressive. These are projects where
we can leverage our size, technical expertise, track
record and innovative execution strategies, such as
employing a modular design and build approach in
areas where resources are constrained. A region to
highlight where large projects are planned is North
America, with signifcant investments in development
in chemicals, LNG liquefaction and gas-to-liquids,
based on monetising low-cost shale gas. We are
working with clients on the early phases of a number
of these planned investments.
Finally, we see more and more clients looking to
develop long-term strategic relationships with a
limited number of E&C contractors. This aligns well
with our approach, to seek to develop long-term
relationships with clients, working with them wherever
in the world they invest. We have signed agreements
with clients including Shell and Dow this year and
have a number of agreements already in place.
Competition remains strong everywhere. We are
focusing on those opportunities where we believe we
have differentiators, such as our technologies, our
know-how, our client relationships, our global presence
and our ability to work with clients from the very
earliest phases of projects to help them shape their
investment, and our proven track record of safely
delivering technically complex and very large projects.
Charles T Drevna
President
American Fuel & Petrochemical Manufacturers
T
he US has within its power
the ability to change the
course of not only our
energy future, but also that of
most components of the US
economy, balance of trade and
national security. For better or
worse, the course we chart will
depend upon decisions made and
framework implemented in 2013 and over the next
four years. Specifcally, the outlook for the US refning
and petrochemical industry in the years ahead is
inexorably tied and subject to these decisions, most
notably actions by President Barack Obama, his
Environmental Protection Agency and the newly
elected Congress.
During his run for re-election, President Obama
stressed energy independence for the nation. If the
president truly meant what he said and his words were
more than political rhetoric, we could see an end to the
anti-fossil fuel environment that has persisted
throughout his frst term in offce. I am hopeful that in
2013 this administration will work to advance a true
all-of-the-above energy strategy that recognises the
importance of all of our domestic energy resources and
fuel and petrochemical manufacturers in rebuilding
our nations economy.
The US, in combination with our ally and friend
Canada, has an abundance of natural resources capable
of bringing the nation to economic prosperity and
North American energy independence within the next
10 years. This can be done by expanding shale
development on federal lands, stopping the attacks on
hydraulic fracturing that threaten affordable feedstocks
necessary for all manufacturing, and increasing
offshore drilling permits.
Further, by approving the Keystone XL Pipeline, the
president could send a strong signal to the nation that
he is serious about creating thousands of domestic jobs
and improving our economy. More importantly, the
president would put the world on notice that the US is
charting a future to include energy independence and
security for the country, with a goal of ultimately
becoming a global energy provider.
Sadly, without a signifcant practical and political
course correction, growth is not possible for fuel and
petrochemical manufacturers in the current regulatory
environment. In January 2011, President Obama
ordered a review of federal regulations to be
eliminated because they hinder economic growth and
job creation. Two years later we are still waiting. A
commitment by the administration is needed to fx
what continues to be a regulatory nightmare for the
refning and petrochemical industry and an
unnecessary, costly burden for the American public.
For example, federal regulations require fuel and
petrochemical manufacturers to spend billions of
dollars to reduce greenhouse gas (GHG) emissions,
even though the EPA has acknowledged the
reductions would bring little or no environmental
beneft. These same regulations increase energy costs,
result in job loss, and harm the US economic and
national security.
AFPM will continue to support sensible and
benefcial environmental regulation, but we believe
that Americas national interest would best be served
by comprehensive and objective cost-beneft analyses
of regulations to determine which make sense and
which do more harm than good.
Today, it is clear that the vast majority of Americans
want to develop our own natural resources and
promote manufacturing jobs. Our nation is blessed
with an abundance of energy resources that could
revitalise job growth and our economy, enhance our
national security, and ensure a strong fuel and
petrochemical manufacturing industry. The decisions
made by the Obama administration in 2013 will
determine the future of the industry I represent and,
more importantly, the entire nation.
Dmitry Balandin
Chief Financial Offcer
Gazprom Neftekhim Salavat
W
e are living through
diffcult economic times.
Despite this, the market
for oil and gas production in
Russia remains stable. The main
source of this stability is high
oil prices, which allow OJSC
Gazprom Neftekhim Salavat and
other major market participants
to achieve positive margins. The domestic market still
tends to be infuenced by the activities of government
and the regulatory authorities, in particular the
measures government takes to reduce domestic fuel
prices. During the second half of 2011, the government
implemented a new export duty system (so-called 60-
66), which benefts the upstream industry, but is less
advantageous for companies within the downstream
sector. Thanks to our broad market appeal and
high-quality offerings, Gazprom Neftekhim Salavat
is fnding the export and domestic markets to be
proftable.
There are challenges that Russian process industry
companies operating in the oil and gas market need to
overcome in the next 12 months. One key challenge is
the high volatility of oil prices and crack spreads for oil
nitrogen removal. The extent of FCC yield improve-
ments were often a function of desired operational
cycle life and available hydrogen for the pretreat units.
Hydroprocessing catalyst systems were developed util-
ising cobalt molybdenum (CoMo) and nickel
molybdenum (NiMo), depending on these objectives
and constraints.
In todays clean fuel operations, much investment
has been made in ULSD and FCC naphtha HDS, with
few refners now achieving environmental compliance
via previously designed pretreat units. Additional
results of the global drive towards clean fuels are
continued advances in catalyst technology that have
provided signifcant gains in both HDS and HDN
performance. These technology gains are being utilised
to drive new FCC pretreat designs to very high levels
of performance and have provided refners with the
option of revisiting how best to maximise the value of
existing FCC pretreat units. This has resulted in many
units shifting catalyst system designs in order to
provide higher levels of nitrogen removal and aromatic
saturation by using more high-activity NiMo catalysts,
maximising FCC conversion capability.
If distillate maximisation is desired, many FCC
pretreat units can be revamped to effectively operate in
a MHC mode. This more severe operation is performed
with higher reactor temperatures and often by modify-
ing the catalyst system to include a more active
conversion catalyst such as an amorphous silica-
alumina (ASA) or zeolite.
Depending on the conversion and distillate selectiv-
ity required, all alumina, alumina/ASA or alumina/
zeolite stacked systems can be considered. Higher
conversions can be achieved by using alumina/ASA
stacks and even higher by using alumina/zeolite stacks
compared to a total alumina system. In specifying a
MHC catalyst system, the balance of hydrotreating
versus cracking catalyst and the potential addition of
reactor volume is largely infuenced by feed qualities
and the desired level of conversion. As many of the
feeds processed are high in contaminant metals,
sulphur and nitrogen, the pretreat section is required
to remove these contaminates to ensure a suffcient
cycle life can be maintained while both meeting any
product targets and minimising nitrogen slip into the
cracking section of the reactor. Feed quality and the
reactor and catalyst system specifed determine the
ultimate sulphur and nitrogen removal capability for a
given cycle life. HDS functionality can remain an
important criteria for some MHC units depending on
existing product specifcations, which are dependent
on site refnery constraints and capabilities. However,
HDN capability is often more important, as it infu-
ences cracking catalyst selection and performance due
to the remaining nitrogen heteroatoms, which reduce
cracking reactions. As mentioned, zeolite-containing
products can provide the highest levels of conversion,
but they tend to be the most sensitive to nitrogen slip,
reducing their long-term effectiveness in such cases.
Amorphous silica alumina (ASA) cracking catalysts
provide increased levels of nitrogen tolerance with a
lower level of conversion capability and, for units with
limited HDN capability, conventional pretreat catalyst
can be operated in a MHC mode but with a reduced
conversion capability.
Q
Can we use desa|ter water for coke cutting or are there
better recyc|ing options for the task?
A
I1w Marce||o Ierrara, Chairman, I1w, mferrara
itwtechno|ogies.com
When using desalter water for coke cutting, you
should at least consider some major problems related
to water composition: entrained salts; entrained caus-
tic; and entrained sludge and sediments. Entrained
inorganics might have an impact on the metallurgies of
the cutting tool, coke drum and lines, together with
potential salts precipitation in the cutting tools
nozzles. Some mitigation of such phenomena might be
found in acidifcation and/or anti-scalant injection, but
all of that needs to be carefully evaluated.
Also consider that caustic, oil carry-under and chem-
icals might act as emulsifers and can potentially
stabilise the frothing of coke particles, preventing/
limiting their precipitation in the water recovery
system and creating coke particles carry-over in the
cutting tool, which, in turns, creates plugging and
possible erosion/corrosion.
6 1Q Q4 2012 www.eptq.com
MINIVAP ON- LINE
P h o n e +4 3 1 2 8 2 1 6 2 7 -0 | F a x +4 3 1 2 8 0 7 3 3 4
i n fo . g ra b n e r-i n s tru m e n t s @ a m e te k . a t | w w w. g ra b n e r-i n s tru m e n t s . co m
VAPOR PRESSURE
PROCESS ANALYZER
VP o f G a so li n e , C ru d e O i l a n d L P G
Acco rd i n g to S ta n d a rd s
R VP , D VP , T VP a n d V/L = 2 0
H i g h e st P re ci si o n
U p to 2 S a m p le S tre a m s
Ad ju sta b le Te m p e ra tu re
R a n g e : 2 0 -6 0 C
Ad ju sta b le V/L R a ti o
Au to m a ti c C a li b ra ti o n
F a st a n d E a s y M a i n te n a n ce
Au to m a ti c L u b ri ca ti o n
Q&A copy 12.indd 2 11/9/12 13:23:03
products. At the same time, the whole sector is under
pressure from the necessity to perform major capital
expenditure programmes while coping with tight time
schedules for making downstream investments. Also,
there is still a need to comply with government-
implemented high technological standards to the
quality of oil products and refnery assets inside
Russia.
Energy effciency is important in relation to what
Gazprom would like to achieve domestically and
internationally. However, by modernising the
downstream oil and gas sector, and by implementing
new technologies, Russian operators are getting more
than one evident beneft. First of all, new technology
allows energy costs to be reduced and, as a result,
margins to be increased. Also, the quality and
ecological properties of products are improving.
Therefore, to consider energy effciency in isolation
without considering the synergetic effect on margins
would be wrong.
International markets are a priority for Russian
operators. Currently, Europe is the major market for
export-oriented refneries and petrochemical
companies. Oil products are also exported to other
regions including South America, the US and Africa.
At the same time, the Russian market becomes more
and more important and demanding for both oil and
petrochemical products.
Process optimisation of operations features in the
discussion of Russian oil and gas companies today.
Professionals working in an optimisation role in
Russian oil refneries and petrochemicals majors
communicate on an informal basis in spite of the
borders of their companies. However, informal
discussions with colleagues and peers together with
various industry forums organised by AspenTech, a
leading provider of process industry software, are of a
high value to the professional community and help us
to exchange views about the role of technology and
modern software in optimising processes in the
Russian oil refnery and petrochemical industry. We see
this role and business function increasing as more oil
refnery and petrochemical companies realise that
without this kind of technology optimisation
application they would not solve the problems raised
by global markets today.
The adoption of process software technology is
occurring across Russian companies in the oil and gas
sector. Russian companies are trying to achieve
innovation by investing in leading-edge software
solutions. Typically, as soon as they build new
processing units, they start to use modern software to
optimise their operations. It is also quite cheap to use
this kind of software in a modern facility. However, it
is relatively expensive to implement new innovative
software into an ageing facility that might be 20, 30 or
even 40 years old, for example. Sometimes, it is even
relatively cheaper to build a new facility than to
implement the new innovative software connected to
ageing technology.
The kinds of software used in the process industries
have changed signifcantly over the past 30 years. Just
in the last 10 years alone, these changes have been
particularly dramatic. Simply look at the way the IT
industry has evolved during that time period. We
could not have even imagined a decade ago the
possibility of developing an iPad device, for example.
My colleagues who are using the AspenTech software
say that the applications are changing and improving
continuously. They commented that, four or fve years
ago, the software was signifcantly different. Today, it
is of higher quality and more fexible. Within another
fve years, it is likely to change dramatically once
again.
Harry Hauptmeijer
President
Yokogawa Europe & Africa
A
ll the signs are there: the
European refnery sector is
facing the most challenging
time in its history. The global
recession, which is affecting refnery
utilisation, and the European
fnancial crisis are prolonging this
situation.
At the same time, changing
product demand is increasing imports of diesel and jet
fuel. Energy prices are rising, and new (lower) bunker
fuel sulphur specifcations are being introduced.
Couple all of these challenges with the forthcoming
European Union Emissions Trading Scheme, and the
whole future of the European refnery sector is under
considerable pressure.
In Europe, we have already seen refneries closing
and operating companies failing. However, there are
refners who are succeeding in a survival of the fttest
scenario.
These successful refners are those who have
managed to adapt to the changing environment, being
able and willing to share their vision and working
close together with select supplying partners. These
partners are characterised by a shared vision in
technologies, solutions and services that will allow the
refners to meet their newly set business targets and
achieve positive results.
As a leader in industrial automation and control,
Yokogawa develops leading-edge products, delivers
excellent services and commits to providing best ft
solutions to customers. With this strategy, Yokogawa
helps to mitigate the concerns of refners about the
optimal utilisation of their vital business resources:
people, assets and technologies.
For any refner working with Yokogawa, there are
four major concerns: safety, human reliability,
unplanned outage and effciency. Yokogawa provides
industry solutions to alleviate these concerns with the
following deliverable solutions:
Increased safety and reliability, for both process and
human resources
Empowering of human talent
Maximisation of asset value
Minimal costs for energy and utilisation of materials.
However, to achieve the maximum positive impact of
these solutions, a critical success factor comes into
account: the availability of skilled operators and
engineers. Unfortunately, the European workforce has
been adversely affected by demographic challenges as
these experienced and skilled operators and engineers
are retiring.
According to a study done by J&H Marsh and
McLennan, the largest single contributor to
unscheduled downtime is human error. As a result,
Yokogawa has developed an innovative industry
solution called MPA: Modular Procedure Automation.
Using this approach, refnery execution processes are
transformed from manual standard operation
procedures (SOPs) and best practices to an interactive
capability that guides operators consistently through
each and every procedure.
By capturing existing best practices from
experienced operators and then automating and
integrating those best practices into the control system
infrastructure, mistakes can be drastically reduced by
providing operators with guidance in real time. This
means that increased safety margins, process
effciency and staff effectiveness will be achieved
successfully.
Standardisation and automation of procedural
operations based on the best practices of experienced
operators results in safer operation and improved
proftability. In addition to a tool that allows skilled
operators to directly implement their knowledge into a
control system, an implementation methodology is
essential. Yokogawa caters for six sigma
implementation methodology together with a unique
consulting service. This ensures consistent and
optimum plant operations and guarantees skill transfer
to the next generation of operators.
Industry solutions embedded in our process control
and information systems form the basis for all the
information required by the end user, allowing the
refner to set clear and aligned goals for the business,
led by management and implemented through all
levels of staff. Of equal importance, this process will
motivate stronger use of business improvement
initiatives, appropriate technology and innovative
ways of working.
The refner can now continue to improve the
translation of market opportunities into integrated
demand-driven supply, manufacturing and distribution
plans.
It is essential that businesses create a positive spiral
of activities by addressing the challenges customers
are facing. They then need to meet their expectations
for a supplier in a timely manner and correctly, and
then to refect them in the development and
improvement of the next generation of products. We
believe that these interactive activities are the only
way to become a real partner of their customers and
to solve issues together.
www.eptq.com
SAMSON AG MESS- UND REGELTECHNIK
Weismllerstrae 3
60314 Frankfurt amMain Germany
Phone: +49 69 4009-0 Fax: +49 69 4009-1507
E-mail: samson@samson.de
Internet: www.samson.de
SAMSON GROUP www.samsongroup.net
A
0
1
1
2
0
E
N
Partner with
the Best
With over 50 independent subsidiar-
ies and more than 220 engineering
and sales ofces spread across the
world, SAMSON ensures the safety
and environmental compatibility of
your plants on any continent.
To offer the full range of high-quality
control equipment used in industrial
processes, SAMSON has brought
together highly specialized compa-
nies to formthe SAMSON GROUP.
pcs myth.indd 1 8/3/12 09:55:39
Q
When natural gas feedstock prices for steam crackers are
low and export routes for surplus gasoline are well established,
is there a sound economic case for optimised FCC propylene
production?
A
Raul Arriaga, Global FCC Applications Technology
Specialist, Albemarle, raul.arriaga@albemarle.com
Steam crackers (SC) have traditionally been used for
ethylene (C
2
=) production as an intermediate for poly-
ethylene and other products. They have also been used
to meet more than half the global demand for propyl-
ene (C
3
=).
Although SC main feed has traditionally been
comprised of naphtha, lower natural gas prices have
prompted petrochemical companies to shift their atten-
tion towards shale gas as a cheaper source of ethylene
production. The amount of propylene produced in an
ethane-consuming SC is about 10 times less than what
is produced when consuming naphtha-range material.
Therefore, the shift towards shale gas would result in
constrained propylene supply for petrochemical
consumption and a potential price increase.
Furthermore, the recent gasoline oversupply coupled
with declining demand has forced some companies to
shut down several refneries, resulting in even lower
production of propylene from FCC units. Finally,
according to market analysis experts, the global
demand for propylene is expected to increase between
3% and 6% over the next three years, with most of the
growth coming from developing regions.
Given the trends discussed above, alternative propyl-
ene-producing processes (ie, FCC unit processes and
their associated catalytic technologies) are becoming
increasingly well positioned to meet the growing
demand for propylene. For example, Albemarles AFX
line of catalysts has been specifcally designed for
maximum propylene applications. These catalysts
include features such as proprietary zeolite technology
(ADZT-150) with low UCS and high accessibility matri-
ces to reduce the rate of hydrogen transfer reactions
while maximising propylene production. AFX technol-
ogy has been used successfully for several years in
many units around the world.
Albemarle is also working closely with FCC technol-
ogy licensors to continue developing and tailoring new
catalyst and additive technologies for advanced FCC
processes. These new technologies are based on
features such as high reaction temperatures, secondary
risers, downfow reactors, olefns oligomerisation, high
catalyst-to-oils ratios, low hydrocarbon partial pres-
sures and long residence times, among others.
It is worth mentioning that new maximum propyl-
ene FCC units would have to compete for C
3
= market
share with other available on-purpose (non-FCC)
technologies, such as methanol to olefns (MTO),
metathesis, advanced off-gas recovery and propane
dehydrogenation (PDH). However, the overall contri-
bution of these new technologies is expected to be less
than one-third of the global production by 2020. For
these reasons, there is a sound economic case for opti-
mised propylene production from existing and new
FCC units until a new market equilibrium is achieved
and processing margins are restored to more competi-
tive levels.
Q
What standards for functional safety management should
I be considering in the installation of safety instrumented
systems?
A
Graeme Ellis, Senior Safety Consultant, ABB,
graeme.ellis@gb.abb.com
The generally accepted international standard for
safety instrumented systems (SIS), including trips and
interlocks, is published by the International
Electrotechnical Committee. Standard IEC 61508
applies to all industrial sectors, where failure of a SIS
to operate on demand could result in harm to people
or the environment. In the process industry, including
sectors such as chemicals, oil and gas, and pharmaceu-
ticals, the sector standard IEC 61511 should be used to
provide guidance on how to apply IEC 61508. This
second standard has terminology that is more readily
understood by people working in the process industry
and includes assessment tools that have been based on
industry best practices. It should be noted that these
standards are diffcult to interpret, require calibration
against company risk criteria before being used, and
assistance should be sought from a competent function
safety engineer, such as people approved by TV.
These standards cover the complete lifecycle, from
identifcation of hazards during the design phase of
projects through to maintaining the integrity of SIS
during the operational phase. Hazard studies such as
HAZOP are used to identify credible hazardous
events with serious consequences that require a SIS as
a layer of protection. The safety integrity level (SIL)
for all SIS needs to be determined using an appropri-
ate risk assessment technique, and IEC 61511 provides
various uncalibrated methods. Industry is currently
using suitably calibrated risk graphs for screening
low-severity events, with layer of protection analysis
ptq&a
Additional Q&A can be found at www.eptq.com/QandA
Q&A copy 13.indd 1 12/12/12 11:08:50
(LOPA) being used as a more refned approach for
major accident hazards. While the majority of SIS on
a site are likely to be ungraded, any that are
assessed with a SIL rating must be designed to
comply with the detailed requirements of the stan-
dards, including checks on the design architecture
and setting an appropriate proof test frequency. SIL 1
and SIL 2 systems are relatively straightforward to
design and operate. SIL 3 or SIL 4 systems will be
very complicated and should be avoided, generally
by providing other layers of protection based on a
different technology.
Q
Our crude slate has an increasing bitumen component.
Are there special precautions we need to take with regard to
potential poisoning of hydroprocessing catalyst?
A
Hiroshi Toshima, Global Application Specialist - HPC,
Albemarle, hiroshi.toshima@albemarle.com
Special precautions are needed for crude slates with
elevated levels of bitumen to properly protect
hydrotreaters and to prevent contamination from arse-
nic and silicon.
In practice, we have been seeing increasing amounts
of crudes with heavier bitumen components. The
bitumen-derived components inherently have higher
concentrations of contaminants, including sulphur,
nitrogen, aromatics and metals (eg, nickel, vanadium,
iron and sodium). The bitumen also exhibits higher
levels of indigenous species of arsenic, including
As
2
O
3
, Bu
3
As
5
and PhAsO
3
H
2
, among others. Arsenic
is particularly prevalent in the bitumen crudes from
Canadian Oil Sands, Venezuela Orinoco tar, West
Coast California, Wyoming Rocky Mountains and
Russian Urals. It is almost evenly distributed over the
entire boiling range, with concentrations ranging from
20 to 400 ppb in the resulting naphtha, distillate and
VGO.
Another contaminant of concern in the bitumen
component is silicon. Silicon contamination is primar-
ily derived from coker additives used as anti-foaming
agents when upgrading bitumen heavy feedstocks. In
these applications, the coker additives consist of
organo-siloxanes such as polydimethylsiloxane (PDMS)
that are applied at concentrations ranging from 5 to 10
ppm. The PDMS is thermally decomposed in the coker
unit, resulting in the accumulation of silicon deriva-
tives in the downstream coker naphtha and distillate
boiling ranges. The silicon concentration in these
streams depends on the species of silicon and the
addition rate of the anti-foaming agent in the coker
unit. In general, the ranges of silicon concentrations in
the coker naphtha and distillate streams are 5 to 10
ppm and 2 to 3 ppm, respectively.
Both arsenic and silicon contamination signifcantly
inhibit HDS activity in the downstream hydrotreaters.
For example, in diesel hydrotreating applications, a
contaminant deposition of 0.5 wt% arsenic on the cata-
lyst can result in a loss in HDS activity of
approximately 30%. Deposition of 10 to 20 wt% silicon
reduces activity by 30 to 50%. Due to this, the main-
bed catalyst must be protected with guard-bed catalyst
technology when hydrotreating bitumen crudes.
To address these issues, Albemarle has developed a
series of arsenic-trapping guard-bed catalysts. One
such catalyst, KG 6, has an arsenic-trapping capacity of
at least 2.0 wt%. To date, it has been used in over 50
commercial hydrotreaters worldwide to process naph-
tha, kerosene, distillate and VGO. Albemarle later
introduced KG 8, which provides an increased arsenic-
trapping capacity of 3.0 wt% (or higher) for
hydrotreaters that were strongly hindered by arsenic
contamination.
For silicon contamination, Albemarle introduced
Ketjenfne 647 (KF 647). KF 647 is a guard-bed catalyst
delivering a silicon trapping capacity of 20 wt% or
higher in the coker naphtha and distillate applications.
It has been utilised in more than 250 commercial units.
The catalyst can also remove arsenic, making it an
appealing choice for units suffering from both silicon
and arsenic contamination.
KF 859, a speciality NiMo main-bed catalyst, has an
exceptional tolerance for arsenic and arsenic contami-
nation in coker naphtha, kerosene and distillate
hydrotreating applications. The catalyst shows high
arsenic and arsenic

trapping capacities of 1.5 and 20
wt%, respectively, while maintaining high HDS and
HDN activities.
This robust catalytic portfolio allows bitumen refner-
ies to maximise the activities and cycle length of their
hydrotreaters units that are afficted by varying degrees
of arsenic and silicon contamination.
Q
Are there any hydrotreating catalysts or activation
techniques that allow the introduction of cracked feeds on
the frst day of the cycle?
A
Frederic Jardin, Vice President Sales & Marketing, Eurecat
US, fjardin@eurecat.com
Eurecat offers the Totsucat ex-situ activation and
sulphiding process combined with a carbonisation step
that allows the introduction of cracked feeds during
the frst day of a catalyst cycle. The carbonisation step
is referred to as cracked feed protection (CFP). Totsucat
CFP has been utilised in 15 hydrotreating units around
the world, including units treating coker naphtha,
distillates, FCC feed and LCO. Run data show that the
deactivation rate may also be reduced when processing
large amounts of cracked stocks with Totsucat CFP
treated catalysts.
Both arsenic and silicon
contamination signifcantly
inhibit HDS activity in the
downstream hydrotreaters
Q&A copy 13.indd 2 11/12/12 16:41:03
pcs 1.indd 1 8/3/12 09:50:20
1
2
0
7
_
e
w w w. a irliq u id e . c o m
Because our genes lead us to an irrepressible passion for new solutions.
Th e o n ly t h in g lo n g e r t h an o u r c o m p an y h is t o ry
is t h e lis t o f o u r in n o vat io n s .
When it comes to providing innovative
plant and system solutions, Air Liquide
Global E&C Solutions is your partner of
choice worldwide. No surprise, since
we innovate relentlessly in order to bring
sustainable and cost-efective solutions
to society, levera ging partnerships with
customers, suppliers, academics and
communities. This is not only demonstrated
by thousands of patents, but by pionee-
ring reference projects around the globe
as well. Yours could be the next one.
AIR2185025_AZ_TechnoDNA_10/12_RZ.indd 1 16.10.12 14:18
air liquide.indd 1 10/12/12 11:42:00
Refnery design in China
A
ccording to statistics, refning
capacity in mainland China
reached 575.27 million tons in
2010, 1.77 times the 2000 fgure.
Sinopec processed 144 crude oils in
2010, with a total processing capac-
ity of 212.972 million tons. Classifed
according to crude oil properties,
high-sulphur and high-TAN crude
oils account for 49.58% of the crude
oil processed, representing a year-
on-year increase of 0.64%, and sour,
acid, high-sulphur and high-TAN
crude oils account for 80.66% of the
crude oil processed, representing a
year-on-year increase of 0.68%.
In compliance with the
increasingly strict environmental
regulations of countries worldwide,
standards for gasoline and diesel
oil are continuously being
upgraded. The European Union,
Japan and Korea limit the sulphur
content of motor gasoline and diesel
to below 10 ppm, and North
America and Australia specify a
sulphur content of 30-50 ppm. It is
predicted that the production of
gasoline and diesel oil with a
sulphur content of less than 10 ppm
will exceed 60% of world gasoline
and diesel production this year. The
main trend of development in the
future is for ultra-low sulphur gaso-
line and diesel oil.
Since 2010, China has limited the
sulphur content of gasoline to
below 150 ppm, but still has a long
way to go compared with devel-
oped countries. Also, China has
limited the benzene content of gaso-
line to no more than 1%, which is
equivalent to the fgure specifed by
Europe, America, Japan and Korea.
In 2011, a new standard for motor
diesel was implemented in China,
Poor-quality crude oils and strict environmental regulations promote the use of
technologies that favour energy savings and environmental protection
SUN LILI
Sinopec Engineering Incorporation
which signifcantly lowered the
sulphur content of diesel oil.
The corresponding national
standard for integrated wastewater
discharge is continuously upgraded.
Chemical oxygen demand (COD) in
wastewater from refneries built
after 1998 should be less than
60 mg/l. Currently, most refneries
are following Grade II or Grade III
wastewater discharge standards.
The corresponding national stand-
ard for the integrated emission of air
pollutants is continuously upgraded.
SO
2
and NOx in waste gas emissions
from refneries built after January
1997 should be less than 960 mg/l
and 240 mg/g, respectively.
Currently, disparity exists
between actual emission of pollut-
ants and the requirements of the
new standard, and special attention
should be paid to NOx control.
Gap analysis between refneries
Waste of petroleum resources
There are quite a lot of small-scale
refneries in China, with a total
processing capacity above 50 million
t/y in 2010. Small-scale refneries
are characterised by simple process
technologies, poor-quality products,
high energy consumption and low
environmental protection levels.
According to statistics, light oil yield
in a small-scale refnery is 10%
lower than that in a large-scale
refnery, and annual light oil
production in a small-scale refnery
is fve million tons less than that in
a large-scale refnery.
Low use of petroleum resources
The average capacity of Chinese
refneries is lower than the world
average, and most refneries select
decarbonisation processes including
coking for heavy oil processing,
resulting in high energy consump-
tion, high material consumption
and low utilisation of petroleum
resources.
According to the 2010 statistics of
refneries in Sinopec and
PetroChina, the average light oil
yield is 76.41%, while the worlds
advanced level is 80.8%. The aver-
age comprehensive self-suffciency
rate is 5.11%, while the worlds
advanced level is below 4%. The
comprehensive commodity rate is
94.33%, while the worlds advanced
level is above 96%, and the process-
ing loss rate is 0.59%, while the
worlds advanced level is below
0.45%.
Currently, the consumption of
transportation oil and chemical oil
constitutes a relatively low propor-
tion of petroleum consumption.
Petroleum resources are not utilised
effectively and rationally.
Process technology gap leads to
high energy consumption
The performance in energy
consumption of most refneries in
China lags considerably behind the
worlds advanced level. In 2010, the
Energy Intensity Index (EII) of
refneries in Sinopec and PetroChina
was 87, while the worlds advanced
level was below 78 and at the high-
est level was below 70.
Currently, oil product quality in
China still lags behind that of
developed countries. To meet corre-
sponding quality standards, new
units should be installed, leading to
higher energy consumption and
energy loss.
The energy consumption of a
sinopec.indd 1 10/12/12 13:00:48
refnery accounts for 6-10% of the
crude oil processed, depending on
the complexity and quality upgrad-
ing level of the refnery. The fgure
is even higher in some refneries.
Treatment of waste gas, wastewater
and solid wastes to be strengthened
Flue gas from refnery furnaces and
boilers contains relatively high
levels of harmful substances,
including SO
2
and NOx, and the
treatment of wastewater and solid
wastes needs to be further strength-
ened. The average SO
2
in fue gas is
550 mg/m
3
, while the worlds
advanced level is below 400 mg/m
3

and the worlds leading level is
below 200 mg/m
3
. The average
COD in wastewater is 100 mg/l,
while the worlds advanced level is
below 60 mg/l.
Developed countries have real-
ised automatic monitoring of
environmental pollution and have
integrated the environmental moni-
toring system of refneries into local
environmental monitoring networks
for social and public supervision.
Chinese refneries should further
strengthen their environmental
monitoring measures.
Refnery design in China to keep
abreast of low carbon development
Diversifcation strategy for crude oil
processing
In refneries, the cost of crude
oil accounts for a relatively high
percentage, generally 70-80%, of
production costs. Due to the uneven
distribution of the worlds crude
oil resources, and the massive
exploitation and use of conventional
petroleum resources, the availability
of conventional petroleum resources
continuously decreases, while poor-
quality, high-sulphur, high-TAN
heavy oil resources are abundant.
Poor-quality, heavy oil will become
an important energy source in the
21st century. Refneries should
carry out a diversifcation strategy
for crude oil processing to increase
processing adaptability for changes
in crude oil properties (see Figure
1).
Deep processing strategy for higher
production of transportation fuel
Integrated and innovative residue
processing technologies should be
selected. Heavy oil upgrading is not
only the focus of refnery process
fow optimisation, but also the key
to cleaner production and one of
the major sources of beneft to
refneries.
It is found, in process engineering
research and industrial practices,
that neither a single advanced proc-
ess technology nor the conventional
combination of these advanced
technologies can perfectly solve the
problems brought by poor-quality
crude oils. Integrated application of
heavy oil processing technologies
should be further optimised, and
the feed conditions of the FCC unit
should be improved to increase the
liquid yield of the unit and reduce
the production of slurry and forma-
tion of coke, thus reducing carbon
emissions.
DC-RHT combined process
A small-scale delayed coking (DC)
unit is installed to improve the
properties of the residue hydrot-
reating (RHT) feed (see Figure 2).
The properties of crude oil are
getting worse. If the total content of
nickel and vanadium in the feed to
the RHT unit is higher than 150
ppm, measures for maximum utili-
sation of resources and light oil
yield are:
Add a small-scale DC unit to the
process fow of residue processing
Use deep-cut vacuum distillation
technology to condense vacuum
residue. Most of the vacuum resi-
due is sent to the RHT unit together
with vacuum gas oil and coker gas
oil
Send a small portion of the
condensed VR to the DC unit.
Petroleum coke product from the
DC unit is used as the feed to a
gasifcation unit to produce hydro-
gen for hydrogenation units.
30
40
20
10
-10
t
s
a
E

e
|
d
d
i
M
a
c
i
r
f
A

t
s
e
W
a
c
i
r
f
A

h
t
r
o
N
a
i
s
A

t
s
a
E

h
t
u
o
S
e
p
o
r
u
E
a
c
i
r
e
m
A

h
t
u
o
S
a
c
i
r
e
m
A

h
t
r
o
N
a
i
n
a
e
c
O

,
n
o
i
t
c
u
d
o
r
p

n
i

e
g
n
a
h
C
0
60
80
2
.
1
1
.
2
-
1
.
0
-
Year on year
2010
2009
1
.
5
9
.
1 2
.
1
-
0
.
5
0
.
2
7
.
0
-
9
.
2
8
.
4
1
.
9 7
.
0
7
.
0
0
.
0
8
.
0
5
.
0
4
.
0
2
.
4
6
9
.
6
6
4
.
6
2
4
.
4
2
4
.
9
2
.
9
Figure 1 Procurement ratio of imported crude oil by region
ADU RHT
VDU DCU
RFCC
C ru d e
o il
R e sid u e
H yd ro tre a te d
re sid u e
A R
VG O C G O
VR
C o k e
Figure 2 Process fow of DC-RHT combined process
sinopec.indd 2 11/12/12 16:42:44
According to preliminary esti-
mates, if a 1.2 million t/y DC unit
is installed in a 12.0 million t/y
renery that processes poor-quality
crude oil, the impurities content of
the crude oil can be even higher.
Sulphur content may be 8.8%
higher, the total content of nickel
and vanadium may be 18.1%
higher, and carbon residue content
may be 17.3% higher.
It is predicted, based on a pricing
system of $60 per barrel, that the
internal rate of return may increase
by 0.75% due to lower crude oil
costs and the inuence of compre-
hensive factors.
RHT-RFCC RICP
Heavy cycle oil (HCO) from the
RFCC unit is cycled to the RHT unit
to improve the properties of RHT
feed. HCO contains a large amount
of aromatic components that can
dissolve resin and asphaltene in
the residue, so as to improve RHT
reactions, reduce carbon deposits
on catalyst, prolong the service life
of catalyst, reduce gas yield, and
improve product distribution in the
RHT and RFCC units.
Both the light oil yield and liquid
yield of the RFCC unit in a typical
integrally combined process (RICP)
are 1.5-2 wt% higher than those of
a conventional process. Since less
slurry and coke is produced in the
RFCC unit, the energy consumption
of the unit is reduced.
According to preliminary
calculations, the overall energy
consumption of a RICP is 2 kg oe/t
of feed lower than that of a conven-
tional process.
S-RHT and IRCC-FR
Industrial application results also
show that due to the constraints of
operating conditions and catalyst in
the RHT unit, there is still a large
amount of non-cracking compo-
nents in the RFCC feed, which will
increase coke formation and gas
production during RFCC reactions.
The latest research shows that
coke formation and gas production
may be reduced and the light oil
yield of the RFCC unit may be
increased by reducing the operating
severity of the unit, increasing the
amount of cycle oil and routing the
cycle oil to a separately installed
RHT unit, where a hydro-saturation
reaction takes place in the presence
of a specic catalyst to deeply
improve the properties of RFCC
feed and convert non-cracking
components into cracking compo-
nents (see Figure 3).
The RFCC combined process
(IRCC-FR) is used to further
increase light oil yield. Preliminary
research results show that both
the light oil yield and liquid yield
of the RFCC unit increases by 6
wt%.
Further comprehensive research
is under way. The successful imple-
mentation of this research will
change traditional ideas of heavy
oil processing and provide more
exibility for poor-quality heavy oil
processing.
C ru d e
o il
ADU RHT
VDU HC HF
RFCC
Figure 3 Process ow of IRCC-FR
300
1000
800
200
100
m
p
p

,
d
e
e
f

e
h
t

n
i

V

+

i
N
0
Ebu||ated bed
S|urry bed
Net VR conversion,
50 60 70 80 30 40 90 100
Fixed bed
Figure 4 Adaptability of RHC technology
ADU
Gas oi|
HT
VDU
SDA
S|urry bed
RHC
FCC
CCR
Crude
oi|
Diese|
Aspha|t
Naphtha
Gas oi|
VR
Unconverted
oi| 5
Diese|
Gaso|ine
Gaso|ine
S<1000 ppm S<100 ppm
Figure 5 Process ow of the SDA-RHC-FCC combined process
sinopec.indd 3 10/12/12 13:01:12
SDA-RHC-FCC combined process
Hydrogen consumption for the
three processes is roughly the same,
but light oil yield in the slurry bed
process is remarkably higher than
that of the other two processes (see
Figures 4 and 5).
Since poor-quality heavy crude
oil is processed, deep-cut technol-
ogy and SDA technology are used
to treat heavier components, which
are difcult to process, in a slurry
bed hydrocracking unit, so as to
optimise the operating conditions
in the gas oil hydrotreating unit
and especially in the FCC unit.
Therefore, the renery-wide yield
of light oil in the slurry bed process
is the highest among the three
processes.
The problems of slurry bed tech-
nology concerning engineering
scale-up and operation stability
have been solved. It should be
commercialised at the appropriate
time so as to open up a new proc-
ess route for optimised processing
of poor-quality residue.
Cleaner production and clean
product strategy
It is reported that, by the end of 2010,
motor gasoline meeting National III
or National IV emission standards
accounted for over 96% of the
gasoline produced by Sinopec, and
the proportion of gasoline product
meeting National IV or higher
emission standards was only 12%.
Therefore, for the purposes of
gasoline quality upgrading, espe-
cially for the production of gasoline
meeting National IV or National V
emission standards, the sulphur
content of gasoline will decrease
from 150 ppm to 50 ppm or even
below 10 ppm.
The key is to solve the problem of
high sulphur content in FCC gaso-
line. Inuenced by chemical
equilibrium, the difculties of gaso-
line desulphurisation will further
increase, resulting in higher costs.
There are three ways to reduce
the sulphur content of gasoline
product: FCC feed pretreatment to
reduce the sulphur content of FCC
gasoline by improving the proper-
ties of FCC feed; desulphurisation
of FCC gasoline; and a proper
increase in the contribution of the
Gas oil
FCC
RFCC
G a s
o il
R e sid u e
F C C d ie se l
F C C g a so lin e
S < 1 0 0 0 p p m
S < 1 0 0 -1 5 0 p p m
S < 3 0 0 0 p p m
Gas oil
HT
RHT
Figure 6 Process ow of front-end hydrogenation of gasoline
Feedstock Slurry bed Ebullated bed Fixed bed
hydrocracking hydrocracking hydrocracking
Kuwait atmospheric residue 650.00 650.00 650.00
Hydrogen 15.27 14.17 13.18
Total 665.27 664.17 663.18
Output of main products
LPG 55.82 31.21 70.54
Naphtha 78.48 72.91 57.71
Gasoline 125.75 61.48 148.12
Kerosene 75.96 75.96 64.27
Diesel 249.11 198.93 223.97
Unconverted oil 11.07 81.55
FCC slurry 0.00 12.01
Total of light oil products 529.30 409.29 494.07
Case analysis: comparison of several combined processes
Table 1
Item Conventional Continous liquid Ultra-low
hydrogenation phase HDS sulphur HDS
Scope of Straight-run diesel No limit >90% >90%
application Coker diesel or No limit <10% <10%
FCC diesel
Technical Flow characteristics Downow, Upow, continous Two reaction
characteristics conventional trickle liquid phase bed zones,
bed technology technology conventional
trickle bed
technology
Reaction zone 1 1 2
Recycle hydrogen Recycle hydrogen Cycle oil Recycle hydrogen
or cycle oil
Reaction Hydrogen partial 6.4 6.4 6.4
conditions pressure, MPa(g)
Hydrogen-to-oil ratio 300-500:1 1.5-2.0 300:1
(liquid phase
circulation)
Reaction 360-390 360-390 360-390/280
temperature, C
Space velocity, h
-1
1.5-2.0 1.5-2.0 3.0/6.0
Product Sulphur content, ppm >350/50 >50 >10
specication Emission standard National III/IV diesel National IV National V
Investment Investment, % Basis 10-15 lower Equivalent
and energy Energy Basis 15-20 lower 10 lower
consumption consumption, %
Comparison of new diesel hydrogenation technologies with conventional technology

Table 2
sinopec.indd 4 11/12/12 16:52:27
limiting factors love
limitless possibilities
Overcome limiting factors affecting ref nery capacity and
operating f exibility with BASF innovative FCC products,
services and solutions. Our products deliver value to
enhance sustainability and performance.
At BASF, we create chemistry for a sustainable future.
Catalysts
Co-Catalysts
Additives
Services and Solutions

www.cat alyst s.basf.com/re ning
basf.indd 1 6/12/12 13:23:11
CCR unit to further reduce the
sulphur content of gasoline by
increasing the reformate yield.
Selection of FCC feed pretreatment
process
According to statistics, the sulphur
content of FCC gasoline is about 5-
10% of the sulphur content of the
FCC feed, and the sulphur content
of FCC diesel is about 80% of the
sulphur content of the FCC feed.
The best way to reduce the
sulphur content of FCC gasoline is
to pretreat FCC feed (see Figure 6).
The sulphur content of gasoline and
diesel can be signifcantly reduced,
and the light oil yield can be
improved by 2-3%. The olefn
content of gasoline and the aromatic
content of diesel can be reduced,
the cetane number of diesel can be
increased, and the operating diff-
culties of the unit and the
consumption of catalyst can be
lowered. Meanwhile, the sulphur
and nitrogen content of fue gas can
be greatly reduced, thus meeting
the requirements of fue gas
emissions.

Application of FCC gasoline
post-treatment process
Industrial application results show
that the olefn content of gasoline is
mainly reduced by using olefn
reduction technology for FCC gaso-
line. According to statistics, by
using olefn reduction technology
in the FCC unit, the olefn content
of FCC gasoline decreases by
10-22% and RON decreases by
0.2-2.2. The desulphurisation rate of
a S-Zorb unit is about 84-97%, while
RON decreases by 0.5-1.4 and olefn
content reduces by 4-6%. The
desulphurisation rate of a gasoline-
selective desulphurisation unit is
78-93%, and RON decreases by
0.2-3. At the same desulphurisation
rate, a higher olefn content of FCC
feed results in a higher degree of
olefn saturation.
Therefore, using olefn reduction
technology for FCC gasoline to
maintain a relatively low olefn
content of FCC gasoline can help
control olefn saturation during
reactions and will reduce RON
loss.
Domestic technologies and
catalysts for clean diesel production
are well developed. In particular,
hydrotreating technology and
capacity have been greatly
improved. Measures for diesel qual-
ity upgrading are as follows:
Unit confguration in the refnery
should be optimised to increase the
production of high-quality diesel by
properly building a hydrocracking
unit. A FCC feed pretreatment unit
should be installed to improve the
properties of diesel product by
improving FCC feed
The development and application
of new technologies should be
expedited, and the application level
of technology should be improved
to further reduce the cost of diesel
quality upgrading (see Table 2).
Continuous liquid-phase HDS
technology
Due to the high solubility of hydro-
gen in oil, the recycle hydrogen
system is removed in case chemical
hydrogen consumption is less than
the dissolved hydrogen in the
reactor.
A liquid-phase circulating pump
is installed in the reactor to remove
reaction heat by circulating the
appropriate amount of liquid phase,
so as to maintain a relatively small
rise in reaction temperature. A rela-
tively constant reaction temperature
is favourable to the desulphurisa-
tion and aromatics saturation
reactions. Meanwhile, circulating a
proper amount of liquid phase may
further make up the dissolved
hydrogen consumed during reac-
tions and dilute the impurity
concentration, which affects reac-
tion depth.
To avoid the precipitation of trace
quantities of excess dissolved
hydrogen in the reactor, upfow
reactor and proprietary internals
are developed for the continuous
liquid-phase HDS reactor.
Ultra-low sulphur HDS technology
Due to the differences between their
reaction kinetics and thermodynam-
ics, different reactions take place in
two specifc reaction zones. The
frst reactor is operated at high
temperature and high space
velocity. Desulphurisation and
denitrogenation reactions take place
at a pressure of 6.4 MPa, an aver-
age temperature of 360-390C and a
space velocity of 3.0 h
-1
, to remove
most of the sulphur and almost all
of the nitrogen. The second reactor
is operated at low temperature and
high space velocity.
The hydro-saturation reaction of
aromatics takes place at a pressure
of 6.3 MPa, an average temperature
of 280C and a space velocity of 6.0
h
-1
, to completely remove the
remaining sulphur. With a water-
white colour, the diesel product has
a sulphur content below 10 ppm
and a polycyclic aromatic hydrocar-
bon content below 11%.
Stable operation of the two reac-
tors in series at different
temperatures can be realised by
optimising the process fow and
using advanced control and heat
exchange; the heat exchange eff-
ciency can be improved by
enhancing heat transfer.
In normal operation, temperature
requirements at the inlet of the
reactor can be met by enhancing
heat transfer, so a reaction feed
furnace is put into service only
during startup.
This technology can not only
achieve ultra-deep desulphurisation
of diesel oil, but also realise innova-
tion in process technology and
process fows, thus further reducing
the investment level and energy
consumption of the unit.
Need to install a hydrocracker
According to estimates, if a 1.6
million t/y hydrocracking unit is
installed in a 10.0 million t/y fuel-
type refnery, overall hydrogen
consumption will increase, but
For the purposes
of gasoline quality
upgrading, the
sulphur content of
gasoline will decrease
from 150 ppm to 50
ppm or even below
10 ppm
sinopec.indd 5 11/12/12 16:43:23
Reliable SwiSS Quality
API 618
Rod load up to 1'500 kN/335'000 Ibs
Power up to 31'000 kW/42'100 HP
Full Range:
You get moRe than Just a
PRocess gas comPRessoR
Lubricated up to 1'000 bar, non-
lubricated up to 300 bar
For highest availability: We recom-
mend our own designed, in-house
engineered compressor valves and
key compressor components
Designed for easy maintenance
We are the competent partner
with the full range of services
worldwide
www.recip.com/api618
YouR BeneFit:
lowest liFe cYcle costs
burckhardt.indd 1 7/3/12 21:39:05
hydrogen production will reduce by
8% due to the increase in CCR
capacity. The feed to the hydrogen
plant will reduce by 13%, thus
lowering the CO
2
emissions of the
renery. By installing a hydroc-
racker, FCC decoking will reduce by
30%. The production of low freezing
point diesel will increase by 19%,
and kerosene and diesel production
will rise by 10%. Also, the light oil
yield will increase from 81% to 85%.
Building a large-scale renery and
production unit for scaled operation
To enhance competitiveness, ration-
alisation of reneries worldwide
continues to increase and renery
capacity is improving year by year.
The number of reneries worldwide
fell from 742 in 2000 to 662 in 2010,
a fall of 80 in 10 years. The average
capacity of reneries worldwide
gradually increased from 5.14
million t/y in 1992 to 5.63 million
t/y in 2000 and to 6.95 million t/y
in 2010. There are 21 reneries with
a capacity of over 20.0 million t/y,
while only two are in mainland
China (see Table 3).
The pros of a large-scale renery
are as follows:
Rational utilisation and optimum
allocation of petroleum resources
Higher processing depth for
comprehensive utilisation of
materials
Lower investment, smaller oor
area, lower operational cost and
lower energy consumption
Higher product yield and quality
Recovery of pollutants for
economic treatment.
The cons of a large-scale renery
are:
Poor adaptability to diversied
crude oils and a wide market for
products
Low load operation due to
insufcient feedstock supply
will result in higher energy
consumption
High maintenance levels and the
cost of large-scale units
Relatively high losses of product
during startup and shutdown and
high shutdown loss.
The preferred single-train capac-
ity of a future renery is no more
than 20 million t/y.
Renery and chemical integration
for lower crude oil consumption
Optimum utilisation of resources
A renery can supply reliable
chemical feedstocks for a petro-
chemical plant. Dry gas, saturated
hydrocarbon and coker gasoline
produced in a renery are high-
quality ethylene cracking feedstocks
with high added value. Pyrolysis
gasoline is hydrogenated and
returned to the renery as feedstock
for the production of high-octane
gasoline and aromatics. Methane
produced in large quantities in an
ethylene plant can only be used as
fuel in ethylene production, but it
is a good feedstock for the hydro-
gen plant in a renery, and naphtha
is a high-quality ethylene cracking
feedstock. It is obvious that renery
and chemical integration can not
only solve the feedstock supply
problems of an ethylene plant, but
can also realise optimum utilisation
of resources.
Optimisation of ethylene cracking
feedstock
Ethylene cracking feedstock includ-
ing saturated hydrocarbons (C
2
, C
3
and C
4
) from the renery, straight-
run naphtha and rafnate, coker
gasoline, as well as hydrocracking
light naphtha and tail oil should be
collected to the maximum. Ethylene
and propylene should be separated
from FCC dry gas and LPG.
A renery-chemical complex aims
to achieve crude oil consumption of
150-200 000 tons per 10 000 tons of
ethylene produced; the ratio of
ethylene production to crude
oil consumption is 1:15-20.
For a renery-chemical-aromatics
complex, this gure should be
better than 1:20.
Optimisation of chemical feedstock
Separation technology of group
composition should be applied. The
MaxEne process not only improves
ethylene yield, but also increases
reformate production and the quan-
tity of high-octane gasoline
components (see Figure 7). The
ethylene yield of a naphtha crack-
ing unit increases by 30%, while the
propylene yield remains unchanged.
The MaxEne process is applicable
to a large-scale renery-ethylene-
Pre-
fractionation
Pre-
hydrotreating
Adsorptive
separation
Ethylene
reforming
CCR
N a p h th a
i-p a ra ffin
n -p a ra ffin
n -p a ra ffin c o n te n t
> 9 5 %
i-p a ra ffin c o n te n t
> 8 5 %
C 5
-
Figure 7 Process ow of MaxEne process
Ranking 2000 2010
Country Capacity Country Capacity
1 USA 848.54 USA 870.54
2 China 324.64 China 575.27
3 Japan 259.14 Russia 280.38
4 Russia 257.45 Japan 218.55
5 Korea 133.18 India 200.70
6 Germany 120.40 Korea 136.18
7 India 117.37 Germany 118.60
8 Italy 115.05 Italy 114.80
9 France 98.56 Brazil 101.30
10 Canada 98.11 Canada 96.91
Total 2372.44 2713.23
Capacity of reneries worldwide in 2000 and 2010 (million t/y)
Table 3
sinopec.indd 6 11/12/12 16:52:51
The art of recovering maximum value from
your precious metal bearing materials.
The sc ienc e of recovering maximum value from
your precious metal bearing materials begins with
the most techni cal l y advanced anal yti cal ,
research, and sampling equipment, systems, and
procedures. The art of recovering maximum value is
a different story; thats based on seven decades of
experience and success serving thousands of organi-
zations around the world.
At Sabin Metal we combine science and art to
maximize your returns, speed process turnaround
times, and comply with environmental regulations
for safety, security, and peace of mind. If these
advantages are important to you, we'd like to
explain how we fit into your precious metals asset
recovery program. Let us hear from you today.
At Sabin Metal
weve turned science into art:
Learn more at sabinmetal.com
Sabin-ScienceToArt.indd 1 2/27/12 1:50 PM sabin.indd 1 1/6/12 12:12:48
aromatics complex. Raw material
consumption per unit of product
may be signifcantly reduced,
and resource allocation may be
optimised to achieve refnery-ethyl-
ene-aromatics integration.

Integrated application of advanced
and energy-saving technologies
Current energy consumption
High refnery-wide energy
consumption is attributed to the
processing of poor-quality crude
oils and the quality upgrading of
oil products. The sulphur content of
crude oil processed in a refnery is
mostly 1.0-2.7%. The steam
consumption of the desulphurisa-
tion unit, solvent regeneration unit
and sour water stripping unit
accounts for 5-9% of energy
consumption. The hydrotreating
capacity of a refnery accounts for
more than 95% of the crude oil
processed, not only increasing
hydrogen consumption but also
increasing the proportion of its
energy consumption in refnery-
wide energy consumption to above
20%.
Additionally, the production of
high value-added products results
in higher refnery-wide energy
consumption. Increased propylene
production not only leads to higher
decoking and energy consumption
in the FCC unit, but also increases
the energy consumption of the gas
fractionation unit by 20%. The
production of aromatics including
benzene and mixed xylene increases
the energy consumption of the CCR
unit by 6.7 kg oe/t of feed.
The energy consumption of proc-
ess units accounts for 85-90% of
refnery-wide energy consumption.
The proportion of large-scale units
in a refnery is high, and there is
heat integration between units and
between units and systems. The
energy consumption of utilities and
facilities accounts for 10-15% of
refnery-wide energy consumption.
Thermal systems, water supply and
drainage systems, as well as storage
and transportation systems account
for 51%, 28% and 16% of the energy
consumption of utilities and facili-
ties, respectively.
FCC, CCR and crude/vacuum
distillation units are the focus of
refnery energy savings. In a refn-
ery with a large-scale FCC unit, the
capacity of the unit accounts for
35% of the crude oil processed. As
the biggest energy consumer among
all the process units, its energy
consumption accounts for 28% of
that of the process units.
The capacity of the CCR unit
accounts for 15% of the crude oil
processed, and its energy consump-
tion accounts for 23% of that of the
process units. The energy consump-
tion of the crude/vacuum
distillation units accounts for 12%
of that of the process units.
In a refnery with a delayed coker,
the capacity of the CCR unit
accounts for 20% of the crude oil
processed, and its energy consump-
tion accounts for 24% of that of the
process units. The capacity of the
FCC unit accounts for 29% of the
crude oil processed, and its energy
consumption accounts for 20% of
that of the process units.
The energy consumption of the
crude/vacuum distillation units
accounts for 18% of that of the
process units (see Table 4).
Energy-saving measures
The overall process fow should be
optimised to improve feed proper-
ties. Suitable advanced catalysts
need to be used, appropriate oper-
ating conditions should be selected
for units, FCC feed needs to be
pretreated, and centralised treat-
ment of amine, centralised recovery
of light ends and centralised treat-
ment of wastewater should be
carried out.
The operating conditions of units
should be optimised to reduce proc-
ess energy demand. Pinch
technology needs to be used to
optimise the heat exchange
network, while a mixed vacuum
pumping system should be utilised
in the vacuum system to reduce
energy consumption by 0.5 kg oe/t
of feed. A pressurised operation
should be used in the xylene sepa-
ration unit to recover overhead
condenser heat. The separation
accuracy of the fractionator must be
optimised to reduce the energy
demand for separation, and the
recycle ratio of the delayed coker
needs to be reduced from 0.3 to 0.1
to lower the energy consumption
by 4 kg oe/t of feed.
Hot feed and heat integration
between units should be realised.
Hot feed between units must be
achieved and suitable division
temperatures should be selected for
energy optimisation. Heat integra-
tion between process units should
be realised to achieve double-effect
energy savings. The fractionator
pumparound must be designed to
achieve rational heat distribution
and utilisation. Low-temperature
heat should be recovered for use in
steam or hot water power genera-
tion besides process use. Pressure
equalisation design needs to be
achieved between units and
between unit and system to avoid
energy waste. Heavy oil storage
tanks should be designed as hot oil
tanks to reduce energy loss during
oil storage and transportation.
Advanced automatic control
systems should be used to minimise
interference between upstream
units and downstream units, so as
to realise direct hot feeding.
Advanced energy-saving equip-
ment and materials need to be
utilised. High-effciency heat
exchangers should be employed to
enhance heat transfer and improve
heat utilisation effciency. High-
Percentage (%) Improving measures
Fuel 30-40 To improve the thermal effciency of process furnace, and to
optimise process conditions and reduce reaction temperature
FCC decoking 10-40 To optimise and improve feed properties, and to optimise the
overall process fow
Power 20-30 To improve pump effciency, and to promote the application
of frequency conversion motor
Steam 10-20 To optimise steam network for cascade utilisation of steam

Analysis of energy consumption
Table 4
sinopec.indd 7 10/12/12 13:02:02
effciency trays must be used to reduce refux ratio
and lower the overhead cooling load. Flue gas
turbines, energy recovery hydraulic turbines, energy-
saving motors, frequency conversion motors and air
fow regulation systems of compressors should be
employed to recover pressure energy and reduce elec-
tric energy consumption. High-effciency intensifed
burners need to be used in furnaces to improve eff-
ciency. The exhaust temperature of furnaces should be
reduced to improve thermal effciency by 2-3%. New
insulation material needs to be employed to reduce
the heat loss of equipment and pipelines.
Low-temperature heat should be utilised by apply-
ing low-pressure steam generation technology,
low-temperature Organic Rankine Cycle (ORC)
systems and Kalina Cycle systems.
Energy optimisation of the area should be carried
out. Heat integration between the refnery and local
co-generation power plant needs to be realised. A
large quantity of low-temperature heat in a refnery
cannot be recovered because a heat sink is not availa-
ble, and this portion of low-temperature heat may be
used as a heat source for demineralised water and
boiler feed water in a cogeneration power plant. Also,
integration with the local chemical plant is important
to achieve material exchange and optimisation, as well
as optimum energy use. Integrated energy optimisa-
tion of the area should be carried out rather than
energy optimisation of a single refnery.
Conclusions
The priority of refning technology development and
the confguration of refning units in China have
distinct characteristics. During a long period in the
future, the FCC unit will still be the main secondary
conversion unit for gasoline and diesel production.
Therefore, improving the quality of FCC gasoline
and diesel is very important for oil product quality
upgrading technology in the future.
With the trend towards poor-quality crude oils,
future refneries should further optimise process fows
and change unit confguration so as to improve the
comprehensive utilisation rate of resources, and to
meet the requirements of energy savings and emis-
sions reduction.
More detailed classifcation of crude oil fractions
and the consideration of various narrow-cut process-
ing technologies will actively promote the low carbon
emission of refneries. Narrow-cut processing will
become the development trend in the overall process
fow optimisation of refneries in the 21st century.
Sun Lili is Vice President of Sinopec Engineering Incorporation in
charge of processing solution studies, engineering, construction and
startup of grassroots refneries and refnery modifcation projects. She
has over 20 years experience in engineering design and startup of
hydroprocessing units, has won several Prizes of National Scientifc
and Technological Progress, and holds a BS in petroleum refning from
China University of Petroleum.
Email: sunlili.sei@sinopec.com
www.eptq.com
www.dpharp.com
High Performance
Unmatched precision and real world
accuracy
Unconditional long term stability with lowest
Total Cost of Ownership (TCOj
Multi-sensing functionality
Safety
lnherently fail-safe sensor with no
undiscovered failure modes.
Certifed to lEC61508 as standard for single
use SlL2 and dual use SlL3
Robustness
Rugged, yet compact construction
Multiple material options for harsh industrial
environments
ln-built overpressure mechanism for
increased reliability
A New Standard
in Pressure
Measurement
EJA-E-270x86.indd 1 07/09/12 13:50
sinopec.indd 8 10/12/12 13:02:15
*ad copy 2.indt 1 1/6/12 14:52:43
Maximising ethane in liquids crackers
R
ecent advances in the recovery
of gas from shale reserves in
the US have resulted in a shift
in the economics of US ethylene
plant feedstock. The American
Chemistry Council estimates that US
shale deposits contain 100 years of
natural gas supply, a game
changer that could rejuvenate
Americas chemical industry. Strong
ethane supplies are positioning the
US as the most competitive, low-cost
ethylene producer, resulting in
increased investments in ethane
recovery and pipelines. As a result,
several companies have already
announced their plans for major
investments in the US ethylene
sector.
At present, many ethylene produc-
ers currently cracking liquid
feedstocks such as naphtha or gas
oil are either maximising or consid-
ering maximising the cracking of
lighter feeds such as ethane.
Producers who designed plants to
crack ethane and/or propane and
butane feeds have inherent advan-
tages since their plants require
minimal and/or no modifcations.
However, plants that were confg-
ured for heavier liquid feedstocks
such as naphtha and gas oil will be
forced to consider the impact on
their units if they are to process the
ethane feed. With shale gas advances
in other parts of the world, this
trend may also propagate in other
regions in the coming years.
The following factors need to be
considered before deciding to shift
to ethane feedstock.
Ethane availability and price
Shale gas recovery by horizontal
drilling and the use of fracking
Economics for energy and feedstock supply favour ethane feed cracking, but plant
constraints must be examined rigorously before maximum ethane feeds are pursued
MUHAMMAD IMRAN
Technip Stone & Webster Process Technology
technology has resulted in the US
having an abundance of natural gas.
Once natural gas is available, it is
fractionated to separate ethane from
the rest of the natural gas. The sepa-
rated ethane is then fed into the
pipeline. The US has a good pipeline
infrastructure to supply ethane from
the source to the ethylene producers.
In addition, new pipelines are being
constructed to meet the increased
demand of consumers. Depending
on the availability of shale gas reser-
voirs and the availability of fracking
technology, ethane cracking may
also become economically attractive
in other parts of the world in the
coming years. The desire to develop
shale gas reserves outside of the US
is strong. As per the International
Energy Agencys World Energy
Outlook 2011,
2
China has already
auctioned shale gas exploration
rights but with participation limited
to Chinese companies. International
companies willing to participate
have therefore sought to enter into
partnership with Chinese compa-
nies. Other countries becoming
active or considering becoming
active in shale gas exploration
include India, Poland, Germany,
Spain, the UK and Ukraine.
Liquids cracker confgurations
Ethylene plants are typically
designed with the demethaniser,
deethaniser or depropaniser tower
sequenced at the front end of the
recovery section. Each scheme will
present unique challenges for
maximising ethane feed fexibility.
Depending on the plant confgura-
tion, the cracked effuents fow rate
and composition will vary in each
section of the plant. Various
sections of the ethylene plant may
thus be more than adequate, with
other sections requiring possible
modifcations.
Capacity increase in addition to
fexibility
Some ethylene producers may well
target increasing plant capacity in
parallel to achieving maximum
ethane cracking fexibility. Capacity
increases will require additional
capital spending, particularly in
areas where existing equipment is
already tight and fow rates and
compositions are changing.
Available equipment margins in
current operation
The extent of modifcations required
will depend on whether the exist-
ing equipment has any remaining
margin. For maximum ethane
cracking, fexibility will depend on
the available margins in the existing
equipment.
Products shift and market
economics
In addition to ethylene, a
typical liquids cracker produces
Shale gas recovery by
horizontal drilling and
the use of fracking
technology has
resulted in the US
having an abundance
of natural gas
technip s&w.indd 1 10/12/12 13:11:37
dupont.indd 1 23/2/12 15:04:26
byproducts including hydrogen,
fuel gas, propylene, butadiene,
pyrolysis gasoline, fuel oil, C
4
s and
C
5
s. With ethane cracking, there is
a signifcant shift in the overall
product slate. Ethane cracking
yields for ethylene are higher than
those from liquid feed cracking;
however, being a lighter feedstock,
its cracking yields are signifcantly
lower for the heavier products.
Often, due to the small make, it
becomes uneconomical to separate
the heavier products formed by
ethane cracking. This product shift
can have an impact on the local
market and can result in shortages,
price hikes and possibly even
result in the import of heavier
products. There are technologies
available for converting lower-
chain olefns to higher-chain
olefns, but these processes require
additional capital, equipment and
utility consumption. These proc-
esses can be used to offset the
shortage of heavier co-products
due to ethane feed cracking.
Details of these processes are
outside the scope of this article.
Ability to maintain liquids cracking
capability
Ethane pricing and ethane cracking
economics may change in the future.
It is therefore important to maintain
fexibility so that, where possible,
owners can cost effectively shift back
to liquids cracking when desirable.
Ethane cracking and products
distribution
Figures 1, 2 and 3 explain the
change in product make when
gradually shifting the feedstock
from 100% naphtha to 100% ethane.
How much ethane can be cracked
in a particular naphtha or gas oil
liquids cracker depends on many
factors: available design margins on
equipment/piping, plant confgura-
tions, compressor performance
curves and quench system design.
The overall extent of the required
modifcations will have to be deter-
mined on a case-by-case basis. It
should be noted that product yields
are strongly dependent on cracking
severity and ethane conversion per
pass. Actual product yields will be
dependent on the cracking condi-
tions in a particular plant. Figures
1, 2 and 3 should be used only for
explaining the effect of increasing
ethane cracking on product distri-
bution and should not be used for
design work.
As Figure 1 shows, ethane
cracking results in higher ethylene
yields with minimum production of
byproducts. Typical once-through
ethylene yields from ethane crack-
ing are greater than 50% (depending
on ethane once-through per cent
conversion). Based on recycle crack-
ing of ethane to extinction, an
ultimate yield of ~80% ethylene is
achievable. For overall comparison
purposes, naphtha cracking gives a
25% to 36% ethylene yield depend-
ing on paraffn content and cracking
severity.
Due to higher ethylene yields,
ethane cracking requires a signif-
cantly lower feed rate (tons/ton of
ethylene) as compared to other
feedstocks. Table 1 shows feedstock
tons required per ton of ethylene.
As was explained above, shale
gas recovery in the US has resulted
in reduced pricing of ethane. The
combination of a relatively low
ethane price and better ethylene
yields can result in improved feed/
product margins by shifting from
naphtha to maximum possible
ethane cracking plus naphtha.
Shifting from heavier to lighter
feeds, however, substantially
changes plant traffc patterns,
decreasing plant loads for systems
handling C
3
+ components, while
burdening systems handling the
lighter C
2
-/C
2
fractions. The quench
area will probably be most affected;
a number of modifcations will
most likely be required, but the
changes will be a function of the
overall feedstock before and after
the addition of shale gas.
Figure 1 shows yield patterns for
recycle ethane and propylene. There
is a signifcant increase from 5 wt%
recycle ethane at 100% naphtha
cracking to 35 wt% recycle ethane
at 100% ethane cracking. Actual
recycle ethane within the system
will depend on how much fresh
ethane feed can be cracked in a
given liquids cracker. Recycle
ethane is typically fashed to a pres-
sure to achieve the equivalent of
the coldest propylene refrigeration
credit (-35F, -37C). Thus, a higher
recycle ethane yield has the poten-
tial for additional refrigeration
credit at the coldest propylene level.
This additional refrigeration credit
Feedstock T/T of ethylene
Ethane 1.25
Propane 2.3
Naphtha 3.0
Gas oil 3.85
Feedstock (tons/ton of ethylene) based
on ultimate yield
Table 1
6 0
5 0
3 0
3 0
2 0
1 0
0
%

e
t
h
a
n
e
1
0
0
%

n
a
p
h
t
h
a
2
0
%

e
t
h
a
n
e
8
0
%

n
a
p
h
t
h
a
4
0
%

e
t
h
a
n
e
6
0
%

n
a
p
h
t
h
a
6
0
%

e
t
h
a
n
e
4
0
%

n
a
p
h
t
h
a
8
0
%

e
t
h
a
n
e
2
0
%

n
a
p
h
t
h
a
1
0
0
%

e
t
h
a
n
e
0
%

n
a
p
h
t
h
a
F
u
r
n
a
c
e

e
f
f
l
u
e
n
t
s
,

w
t
%
0
P ro p yle n e
E th a n e
E th yle n e
2 8 % e th yle n e
4 3 % e th yle n e
5 0 % e th yle n e
E th yle n e yie ld will b e slig h tly h ig h e r d e p e n d in g o n e th a n e % c o n ve rsio n
E th yle n e yie ld will b e h ig h e r d e p e n d in g o n sa tu ra te s % a n d c ra c k in g se ve rity
*
**
Figure 1 Products shift: ethylene and propylene
technip s&w.indd 2 10/12/12 13:11:50
helps to reduce the required
horsepower of a propylene refriger-
ation compressor. Propylene
production, however, drops signif-
cantly when operating with
maximum ethane cracking. As can
be seen in Figure 1, the propylene
yield reduces from 14 wt% to 2
wt% when shifting from 100%
naphtha to 100% ethane feed.
Ethane cracking also has a signif-
cant impact on other, heavier
products. As shown in Figure 2,
with ethane cracking, 1,3 butadiene
(13BUTD), butenes, C
5
-200, BTX
and fuel oil production all are
reduced. Due to a very low produc-
tion of C
3
+ material, in most of the
cases it is not economical to sepa-
rate the C
3
and heavier products.
Since plants designed for liquids
cracking have C
3
and heavier prod-
uct fractionation systems installed,
many operators may decide to keep
these systems in operation by a
combination of external make-up
and internal recycles. Quench
system operation also requires care-
ful attention with possible
modifcations.
Figure 3 shows the shift in hydro-
gen and methane production with
ethane cracking. Hydrogen produc-
tion increases from 1 wt% to 4 wt%
and methane production drops
from 13 wt% to 5 wt%. Higher
hydrogen, lower methane and
higher C
2
s will lower the molecular
weight of cracked gas, impacting
the performance of cracked gas
compressors and other equipment
capabilities. A higher percentage of
hydrogen also results in a high
pressure drop in systems handling
the separation of hydrogen and
methane from other components.
Cracking furnace and utilities
considerations
Ethane cracking results in higher
ethylene yields than do other feeds;
therefore, a similar ethylene capac-
ity can be achieved with a lower
throughput of ethane feed. If a coil
designed for liquids cracking is
employed, run length and opera-
tional fexibility will be reduced
unless other adjustments are made.
The number of furnaces that can be
economically shifted from liquids
to ethane cracking will depend on
the design of the downstream
recovery section and its capability
to handle lighter cracked gas. If
operators want to crack more
ethane, additional capital spending
will be required in the recovery
section to achieve more fexibility.
A cracking furnace designed for
liquids cracking may have some
limitations when cracking ethane;
for example:
Reduced run length
Higher than normal steam-to-HC
ratio required to minimise coking
rate
Capacity limitations in radiant
coil
Burners and induced draft fan
limitations
Possible metallurgy issues
Convection section high pressure
drop due to a higher volumetric
fow rate
Primary quench exchanger
coking, resulting in a higher outlet
temperature of cracked gas.
Ethylene producers with the
objective of minimal capital spend-
ing have an option of using a
liquids cracking furnace (see Figure
4) on ethane cracking with reduced
operational fexibility. The existing
furnace liquids feed header and
feed control valves, however, will
need to be replaced to work effec-
tively with ethane feed. For greater
operational fexibility and longer
run lengths, an extensive furnace
revamp may be required. Ethylene
producers can opt for a feed fexible
radiant coil such as Technips Ultra
Selective Conversion (USC) W coil
furnace, which will more effectively
handle the shift from liquids crack-
ing to ethane cracking. In cases
where higher capacity and ethane
only cracking are the objectives,
1 2
1 0
1 6
1 4
8
6
4
2
0
%

e
t
h
a
n
e
1
0
0
%

n
a
p
h
t
h
a
2
0
%

e
t
h
a
n
e
8
0
%

n
a
p
h
t
h
a
4
0
%

e
t
h
a
n
e
6
0
%

n
a
p
h
t
h
a
6
0
%

e
t
h
a
n
e
4
0
%

n
a
p
h
t
h
a
8
0
%

e
t
h
a
n
e
2
0
%

n
a
p
h
t
h
a
1
0
0
%

e
t
h
a
n
e
0
%

n
a
p
h
t
h
a
F
u
r
n
a
c
e

e
f
f
l
u
e
n
t
s
,

w
t
%
0
B u te n e s
C 5 -2 0 0 e x B T X)
B T X
1 3 B U T D
F u e l o il
Figure 2 Heavy products shift
1 0
1 4
1 2
8
6
4
2
F
u
r
n
a
c
e

e
f
f
l
u
e
n
t
s
,

w
t
0
H yd ro g e n
M e th a n e
0
%

e
t
h
a
n
e
1
0
0
%

n
a
p
h
t
h
a
2
0
%

e
t
h
a
n
e
8
0
%

n
a
p
h
t
h
a
4
0
%

e
t
h
a
n
e
6
0
%

n
a
p
h
t
h
a
6
0
%

e
t
h
a
n
e
4
0
%

n
a
p
h
t
h
a
8
0
%

e
t
h
a
n
e
2
0
%

n
a
p
h
t
h
a
1
0
0
%

e
t
h
a
n
e
0
%

n
a
p
h
t
h
a
Figure 3 Light products shift
technip s&w.indd 3 10/12/12 13:12:00
What you can do
with a
touch of blue.
Improve your renery protability by maximizing the production of clean
transportation fuels with our leading residue upgrading technologies.
Deliver the best in renery hydrogen production while reducing
your operating costs with our unique Terrace Wall reformer design.
Enhance the efciency of your overall sulfur recovery to achieve peak
operating and environmental performance with our SRU technology.
You can
And these are just the technology options. There is so much more
you can do with a touch of blue. Visit www.fwc.com/touchofblue
fw.indd 1 7/3/12 13:55:37
Quench system considerations
For ethylene producers accus-
tomed to cracking naphtha or
heavier feed, one of the most
challenging systems to main-
tain effcient operation when
introducing ethane feed is the
quench oil system. As ethylene
producers consider cracking
various feedstocks, they
attribute a certain amount of
fuel oil make to both naphtha
and gas feeds. The net fuel oil
make from naphtha cracking is
typically within 4 to 6% of the
total furnace effuent. Ethane
cracking, at the other extreme
of the spectrum, makes up less
than 0.2%. Therefore, one of
the consequences of transition-
ing to a lighter feed slate is a
signifcant reduction in fuel oil
make from the furnaces, thus
the need for additional external fuel
oil or fux oil to ensure the quench
oil circulation system is operational.
The quench oil system thrives on
having an abundance of heavy and
middle distillate oils. The shortage
of these components, which is
common when cracking primarily
ethane, leads to ineffciencies and
potential plant reliability problems.
With ethane cracking, light and
middle boiling components vapor-
ise in the quench oil tower, leaving
behind the heavy components. A
lower yield of heavy components
results in an increase in the resi-
dence time in the system, causing
the formation of polynuclear
aromatics (PNA) and tar due to
agglomeration. Viscosity therefore
increases, causing a drop in the
heat transfer coeffcient in the heat
exchangers. With time, the concen-
tration of PNA and tar increases
due to polymerisation, resulting in
a further increase in viscosity. A
stage is reached when it is no
longer feasible to maintain quench
oil circulation at the required
temperature and the system has to
be shut down for cleaning.
To maintain effcient and reliable
operation, producers shifting to
majority ethane feedstocks need to
properly evaluate the quench
system and may need to consider
importing external oil (fux oil) to
maintain adequate quench oil
Technips USC M coil is a
better choice. Revamp of the
convection section, burners,
induced draft fan and primary
quench exchangers may also
be required to achieve the
required capacity and run
length while providing greater
operational fexibility.
Another important consider-
ation when converting a
liquids furnace for ethane
cracking is the addition of the
secondary quench exchangers
downstream of the primary
quench exchangers. Typically,
gas crackers are designed to
have a lower cracked effuent
temperature to the quench
section as compared to liquids
crackers. In liquids crackers,
the temperature is kept high to
avoid liquid condensation and
possible quench exchanger and
transfer line fouling. For reference,
for gas
cracking, the cracked effuent
temperature to quench section is
~350F (177C) compared to ~700F
(370C) for liquids crackers. A
secondary quench exchanger can
therefore be added to produce
additional high-pressure steam
during ethane cracking. Bypass
provision can be included to mini-
mise the duty on the secondary
quench exchanger during naphtha
or gas oil cracking.
With a secondary quench
exchanger in service, the combined
furnace effuent temperature from
liquids and gas cracking furnaces
will be reduced. This will lower the
bottom temperature of the quench
oil tower and may reduce the
amount of usable waste heat.
Typically, in liquids crackers,
quench oil is used for dilution
steam generation. A lower cracked
gas effuent temperature will have
a signifcant impact on dilution
steam production. Additional
medium-pressure steam import
may be required to cover for
reduced dilution steam.
Typically, liquids crackers have
hydrogen recovery and purifcation
sections. As was explained above,
ethane cracking generates addi-
tional hydrogen, which can cause
limitations in the hydrogen recov-
ery area. There is a possibility of
increased hydrogen content in the
residue gas, but cracking furnace
burners must be designed to oper-
ate within a certain percentage
range of hydrogen in residue gas.
The fuel gas balance must be
checked with consideration of natu-
ral gas import to maintain a
reasonable residue gas composition.
Modifcations in the hydrogen
recovery area may be required to
maximise hydrogen recovery and
limit hydrogen content in the resi-
due gas. In the extreme case,
furnace burner modifcations or
replacement may be required. With
no modifcations or replacement,
existing furnace burners can poten-
tially set the limit on how much
ethane can be cracked in a liquids
cracker.
Residue gas will have higher
hydrogen and lower methane
contents because of a high ethylene
yield and a signifcant drop in
methane yields with ethane
cracking. Typically, liquids crackers
have a hydrogen recovery system
installed as part of the process
scheme. There will be a reduction
in the residue gas make if hydrogen
recovery is still a requirement
during majority ethane cracking.
An increased import of natural gas
therefore may be required while
transitioning from liquids to major-
ity ethane cracking.
Figure 4 Cracking furnaces
technip s&w.indd 4 10/12/12 13:12:11
Only Grace Catalysts Technologies offers refiners a full portfolio of high technology, state-of-the-art
FCC catalysts and additives. Driven by our world-class research and development in the U.S. and
Europe, only Grace Catalysts Technologies can customize products designed specifically to optimize
your operation.
Our industry-leading technical service provides support, including unit monitoring and troubleshooting,
backed by our high level of analytical services including Ecat, spent catalysts and fines analysis.
We have been the leader in FCC catalysts since inventing the first FCC catalyst over 70 years ago
and we are committed to the future of the refining industry.
Have Grace Catalysts Technologies design your optimal catalyst. Contact us.
Propylene Maximization
Resid Conversion
Gasoline Maximization
LCO Maximization
Coke Selectivity
RepLaCeR
Solutions
grace.indd 1 10/12/12 10:30:41
quality. Proper evaluation and
selection of fux oil is required to
ensure compatibility with the fuel
oil produced in the system.
Incompatible fux oil can lead to
precipitation of asphaltenes in the
system. Flux oil characterisation
should be done to confrm compati-
bility with the system. The
following tests are recommended:
Aromatic carbon
Unsubstituted aromatic carbon
Substituted aromatic carbon
Average chain length.
Most favourable fux oils are
heavy catalytic naphtha (HCN) and
heavy cycle oil (HCO). These are
produced in refneries. However,
these are not readily available, as
refners prefer to produce diesel.
The second best choice for fux oil
is light cycle oil (LCO) and it is
generally available from refneries.
There must be extensive changes
in the quench system in case of
limitations in fux oil availability.
Depending on how much ethane is
cracked, the quench oil tower may
need to be bypassed, with possible
modifcations to the quench water
tower to provide additional heat
transfer capability. Additional
equipment may need to be provided
to handle tar removal. A beneft of
bypassing the quench oil tower is
reduced pressure drop between
furnaces and the cracked gas
compressor. A higher cracked gas
compressor suction pressure can
therefore be employed.
Existing quench oil and quench
water towers may be limiting if
capacity increase is another objec-
tive in addition to feedstock
fexibility. Technips Ripple trays
have been used successfully for
numerous revamp applications for
quench oil and quench water
towers (see Figure 5).

Cracked gas system
Cracked gas compressor perform-
ance needs thorough evaluation.
With a higher ethane feed diet,
molecular weight reduces from ~30
to ~19 with an increase in volumet-
ric fow rate. One of the important
parameters to consider when shift-
ing from liquids cracking to
majority ethane cracking is poly-
tropic head. Polytropic head is
inversely proportional to molecular
weight. As Equation 1 shows, a
lower molecular weight gas will
require a higher polytropic head for
the same compression ratio:
[1]
Where
H
P
Polytropic head, ft
MW Molecular weight
Z
AVG
Average compressibility
T1 Suction temperature, deg R
n Compression coeffcient
P1 Suction pressure, psia
P2 Discharge pressure, psia
Also, as Equation 2 shows,
compressor horsepower is directly
proportional to mass fow rate and
polytropic head:

[2]
Where
SHP Shaft horsepower
M Mass fow rate
H
P
Polytropic head
n
P
Polytropic effciency
1.02 2% gear losses
From the above equations, it is
clear that, for the same mass fow
rate and compression ratio, a lighter
cracked gas from majority ethane
cracking will require a higher shaft
horsepower as compared to heavier
cracked gas. However, as Figure 1
shows, majority ethane cracking
results in a higher ethylene yield
and therefore a lower mass fow of
cracked gas will be required to
produce the same quantity of ethyl-
ene as compared to liquids cracking.
An important variable to overcome
potentially higher horsepower
requirements is to increase frst-
stage suction pressure. As Equation
1 indicates, increased compressor
suction pressure will reduce the
compression ratio and polytropic
head requirements. On the negative
side, a higher frst-stage suction
pressure results in a higher radiant
coil outlet pressure. Since a higher
pressure at the radiant coil outlet
degrades the ethylene yield, increas-
ing the frst-stage suction pressure
tends to lower ethylene make. On
the positive side, the liquids crack-
ing ethylene yield is a stronger
function of the coil outlet pressure
as compared to gas cracking. With
a small sacrifce in ethylene yields,
gas cracking furnaces can therefore
be operated at a higher coil outlet
pressure. There will be a reduced
impact on coil outlet pressure in
case the quench oil tower is
bypassed.
A higher coil outlet pressure also
lowers the volumetric fow rate,
which is helpful in avoiding poten-
tial casing or suction nozzle
limitations. If feed fexibility is the
only target, with no increase in
ethylene production, it is sometimes
possible that a cracked gas compres-
sor driver designed for liquids
cracking furnace effuents can work
for cases with majority ethane
cracking effuents. In some cases,
especially where both feed fexibil-
ity and an increase in ethylene
production are targets, it may not
be possible to operate by just
increasing the frst-stage suction
pressure. Possible solutions for this
situation are:
Adding a booster stage
Adding a parallel stage with a
dedicated driver
Adding a new driver for one of
the stages to avoid limitations on
the existing driver
Adding wheels on stages
|
|
|
|
|
|
|
|
|
|
j
j
)
|
|
\
[
j
)
|
\
[
=
j
)
|
\
[
1
1
1545
1
1
2
1
n
n
AVG P
P
P
n
n
T
MW
Z H
02 . 1 ] 3300 ) [(
) (
n
H
P
P
m
SHP =
Figure 5 Quench oil and quench water
towers
technip s&w.indd 5 10/12/12 13:12:22
requiring higher polytropic head.
In case the quench oil tower is
bypassed to address quench area
issues, there will be a reduced
pressure drop between furnaces
and the cracked gas compressor.
For the same coil outlet pressure, it
will therefore be possible to
increase the compressor suction
pressure.
Another important consideration
for the cracked gas compressor is to
confrm that a suffcient operating
margin is kept between operating
point and surge limit. This operat-
ing margin tends to be narrowed
due to a higher operating speed
and polytropic head required for
lighter cracked gas operation.
Again, a combination of increased
suction pressure and a lower mass
fow required for lighter cracked
gas operation can help to stay away
from the surge limit. Compressor
performance curves need to be
studied carefully by use of a proc-
ess simulator to confrm compressor
performance for lighter cracked gas
effuent.
Chilling train and
expander/recompressor
As Figure 3 shows, switching from
naphtha cracking to majority ethane
cracking results in an increase in
hydrogen production from ~1 wt%
to ~4 wt% and a lower methane
production from ~13 wt% to
~5 wt%. Higher hydrogen produc-
tion may result in a higher
volumetric fow to the chilling train
(depending on composition) and
an increased pressure drop.
Replacement of some piping and
exchangers in the hydrogen system
may help to lower the pressure
drop. In the case of no modifca-
tions, this additional pressure drop
has to be compensated by addi-
tional discharge pressure from the
cracked gas compressor. This is not
always possible due to potential
limitations in cracked gas compres-
sors designed for operating with
heavier cracked gas.
Lower methane and C
3
+ produc-
tion may reduce the demethaniser
warmer level feeds, but can add a
burden to colder level feeds. If
hydrogen is being recovered in the
original liquids cracking scheme,
with no slip stream to the expander,
there will be an increase in the
hydrogen stream fow to the hydro-
gen recovery core exchanger, with
potential hydraulic and surface area
limitations. There will also be
reduced fow to the expander from
the demethaniser overhead due to a
lower methane content in the
cracked gas. A reduced fow to the
expander will lower the expander
effciency. The combined effect of
lower effciency and a reduced fow
is a reduction in refrigeration avail-
able to the chilling train from the
residue gas expander. The expander
feed system can potentially be
modifed to add load and recover
more refrigeration (see Figure 7).
Expander internals or the expander
itself may need to be replaced to
make it suitable for the new set of
operating fow rate and composi-
tion. Additional expander stages
are sometimes needed to process
the lighter molecular weight gas.
Ethylene losses from the demeth-
aniser overhead will also increase
due to the lower methane-to-
hydrogen ratio. Modifcations can
be made to the overhead system to
reduce ethylene losses. These modi-
fcations should be fexible enough
to allow reverting back to naphtha
or gas oil cracking, if required.
Another important consideration
is the percentage of hydrogen in the
residue gas. An ethane cracker
results in lower molecular weight
residue gas due to the increase in
hydrogen content. Since residue gas
is used as a regeneration gas for
dryers, there is the potential for an
increased pressure drop in the
regeneration gas system. It is,
however, possible to reduce the
regeneration gas fow and therefore
reduce the pressure drop due to the
improved specifc heat of lighter
regeneration gas.
Ethylene splitter, C
2
hydrogenation
and deethaniser systems
Shifting from naphtha or gas
oil cracking to majority ethane
cracking results in a higher ethane
per cent to ethylene splitter feed,
that is a lower purity feed to the
Figure 6 Ethylene unit Credit:Technip
R e -c o m p re sso r E x p a n d e r
Figure 7 Simplifed sketch of expander/recompressor
technip s&w.indd 6 11/12/12 16:49:42
ethylene splitter. This will require a
higher refux rate for the same
production capacity. If additional
capacity is another objective, the
required tower internal traffc will
be further increased. The existing
ethylene tower trays would need to
be checked for the extra capacity.
Possible solutions to overcome limi-
tations in tray hydraulics are the
replacement of existing trays with
high-capacity trays, or the addition
of another tower in series. The
deethaniser and C
2
hydrogenation
reactor systems also need to be
evaluated for hydraulic constraints
and any possible negative impact
due to the potential higher fow
rates.
Ethylene and propylene refrigeration
compressors
Depending on composition, availa-
ble credits and mass fow rate, a
higher ethylene refrigeration duty
may be required for the chilling
train during maximum ethane crack-
ing. An ethylene refrigeration
compressor revamp might be
required in this case. Scope of the
revamp can be extensive if the objec-
tive is to both increase plant capacity
and add feed fexibility. Overall, the
ethylene refrigeration compressor
may need a higher hydraulic capac-
ity and horsepower to enable
increased ethane cracking.
Propylene refrigeration load will
increase with two major users
during maximum ethane cracking.
First, as was explained above, the
ethylene splitter will require a
comparatively higher refux rate for
majority ethane cracking as
compared to naphtha or gas oil
cracking. For non-heat pumped
systems, a higher refux rate will
require a higher duty on the ethyl-
ene splitter condenser, which is the
largest user of the coldest propyl-
ene refrigerant. For low-pressure
heat pump systems integrated with
the ethylene compressor, there will
be an increased load on the heat
pumped ethylene compressor.
Second, the ethylene refrigeration
condenser duty will increase due to
a higher demand for ethylene
refrigeration. The ethylene refriger-
ation condenser is normally the
second largest user of the coldest
propylene refrigerant.
On the positive side, there will
be two important refrigeration
credits to propylene refrigeration
with increased ethane cracking.
These credits can be used to reduce
the load on the propylene refriger-
ation compressor. The frst credit
is the increased recycle ethane
production from ~5 wt% to
~35 wt%. Recycle ethane can
potentially be used for the coldest
level refrigeration credit. A higher
recycle ethane refrigeration credit
can help to lower signifcantly the
propylene refrigeration compressor
horsepower. An additional credit
is the fresh liquid ethane feed,
which can be fashed to furnace
feed pressure to achieve additional
refrigeration credit at the coldest
level. Ethane feed, however, may
need to be dried before fashing to
a lower pressure and temperature
to avoid the formation of hydrates.
Additional investment in dryers
and regeneration systems may
need to be considered. Higher
duties on the ethylene splitter and
ethylene refrigeration condensers
can be balanced by these additional
credits during periods of high
ethane cracking. The propylene
refrigeration compressor must be
checked on a case-by-case basis
before shifting to higher ethane
cracking.
Conclusion
While current economics for energy
and feedstock supply favour ethane
feed cracking, plant owners must
rigorously examine operational and
capacity constraints before transi-
tioning to maximum ethane feeds.
Detailed feasibility studies need to
be undertaken to evaluate the
impact of maximising ethane crack-
ing on plants originally designed
for naphtha or gas oil cracking. A
well-planned feasibility study can
help improve production margins.
Each ethylene plant is unique and
needs dedicated investigation for
suitability in feedstock transition.
References
1 American Chemistry Council, Shale Gas
and New Petrochemicals Investment: Benefts
for the Economy, Jobs, and US Manufacturing,
March 2011.
2 International Energy Agency, World Energy
Outlook 2011.
3 Bernard A, Pickett T M, Manek B M, Frye D K,
Flux oil stream import to quench system risk
and impacts, paper number 33e, AIChE 2011
Spring National Meeting, Chicago, Illinois, 14
March 2011.
Muhammad Imran is a Lead Process Engineer
for Technip Stone & Webster Process
Technology, Houston, Texas. He has 16 years
of experience with petrochemical, refnery and
gas processing projects, with the past seven
years focused on ethylene. He holds a bachelors
degree in chemical engineering from University
of Engineering and Technology Lahore and
a masters in chemical engineering from the
University of Houston.
Figure 8 Grassroots ethylene plant
technip s&w.indd 7 10/12/12 13:13:00
roMax process slmulauon soware by 8ryan 8esearch & Lnglneerlng, lnc.
oqloeetloq 5olouoos fot tbe Oll, Cos, kefoloq & cbemlcol loJosttles.
sales[bre.com www.bre.com 979 776-3220 uS 800 776-3220
Process
Comparing Physical
Solvents for Acid Gas
Removal
Insight:
Physical solvents such as DEPG, NMP, Methanol, and Propylene Carbonate
are often used to treat sour gas. These physical solvents differ from chemical
solvents such as ethanolamines and hot potassium carbonate in a number of
ways. The regeneration of chemical solvents is achieved by the application of
heat whereas physical solvents can often be stripped of impurities by simply
reducing the pressure. Physical solvents tend to be favored over chemical
solvents when the concentration of acid gases or other impurities is very high
and the operating pressure is high. Unlike chemical solvents, physical solvents
are non-corrosive, requiring only carbon steel construction. Aphysical solvents
capac|ly lor aosoro|rg ac|d gases |rcreases s|gr|lcarl|y as lre lerperalure
decreases, resulting in reduced circulation rate and associated operating costs.
Typical Physical Solvent Process
DEPG (Dimethyl Ether of Polyethylene Glycol)
DEPG is a mixture of dimethyl ethers of polyethylene glycol.
Solvents containing DEPG are marketed by several companies including
Coastal Chemical Company (as Coastal AGR
), Dow (Selexol), and UOP

(Selexol). DEPG can be used for selective H
2
3 rerova| ard car oe corlgured
to yield both a rich H
2
S feed to the Claus unit as well as bulk CO
2
removal.
DEPG is suitable for operation at temperatures up to 347F (175C). The
minimum operating temperature is usually 0F (-18C).
MeOH (Methanol)
The most common Methanol processes for acid gas removal are
the Rectisol process (by Lurgi AG) and Ifpexol
process (by Prosernat). The

ra|r app||cal|or lor lre Recl|so| process |s pur|lcal|or ol syrlres|s gases
der|ved lror lre gas|lcal|or ol reavy o|| ard coa| ralrer lrar ralura| gas
treating applications. The two-stage Ifpexol process can be used for natural gas
applications. Methanol has a relatively high vapor pressure at normal process
conditions, so deep refrigeration or special recovery methods are required to
prevent high solvent losses. The process usually operates between -40F and
-80F (-40C and -62C).
NMP (N-Methyl-2-Pyrrolidone)
The Purisol Process uses NMP
and is marketed by Lurgi AG.

Tre loW screres used lor lr|s so|verl are s|r||ar lo lrose lor 0EP0. Tre
process can be operated either at ambient temperature or with refrigeration
down to about 5F (-15C). The Purisol process is particularly well suited to the
pur|lcal|or ol r|gr-pressure, r|gr C0
2
synthesis gas for gas turbine integrated
gas|lcal|or coro|red cyc|e (l0CC) syslers oecause ol lre r|gr se|ecl|v|ly lor
H
2
S.
PC (Propylene Carbonate)
The Fluor Solvent process uses JEFFSOL
PC and is by Fluor
Daniel, Inc. The light hydrocarbons in natural gas and hydrogen in synthesis
gas are less soluble in PC than in the other solvents. PC cannot be used for
selective H
2
S treating because it is unstable at the high temperature required
to completely strip H
2
S from the rich solvent. The FLUOR Solvent process is
generally limited to treating feed gases containing less than 20 ppmv; however,
|rproved slr|pp|rg W|lr red|ur pressure lasr gas |r a vacuur slr|pper a||oWs
treatment to 4 ppmv for gases containing up to 200 ppmv H
2
S. The operating
temperature for PC is limited to a minimum of 0F (-18C) and a maximum of
149F (65C).
Gas Solubilities in Physical Solvents
All of these physical solvents are more selective for acid gas than
for the main constituent of the gas. Relative solubilities of some selected gases
in solvents relative to carbon dioxide are presented in the following table.
The solubility of hydrocarbons in physical solvents increases with
the molecular weight of the hydrocarbon. Since heavy hydrocarbons tend
to accumulate in the solvent, physical solvent processes are generally not
economical for the treatment of hydrocarbon streams that contain a substantial
amount of pentane-plus unless a stripping column with a reboiler is used.
Gas Component
DEPG
at 25C
PC
at 25C
NMP
at 25C
MeOH
at
-25C
H
2
0.013 0.0078 0.0064 0.0054
Methane 0.066 0.038 0.072 0.051
Ethane 0.42 0.17 0.38 0.42
CO
2
1.0 1.0 1.0 1.0
Propane 1.01 0.51 1.07 2.35
n-Butane 2.37 1.75 3.48 -
COS 2.30 1.88 2.72 3.92
H
2
S 8.82 3.29 10.2 7.06
n-Hexane 11.0 13.5 42.7 -
Methyl Mercaptan 22.4 27.2 34.0 -
Choosing the Best Alternative
Adetailed analysis must be performed to determine the most economical
choice of solvent based on the product requirements. Feed gas composition, minor
components present, and limitations of the individual physical solvent processes are
all important factors in the selection process. Engineers can easily investigate the
ava||ao|e a|lerral|ves us|rg a ver|led process s|ru|alor sucr as ProVax
which has
oeer ver|led W|lr p|arl operal|rg dala.
For additional information about this topic, view the technical
article A Comparison of Physical Solvents for Acid Gas Removal at
http://www.bre.com/tabid/147/Default.aspx. For more information about ProMax,
contact Bryan Research & Engineering or visit www.bre.com.
bre.indd 1 7/3/12 10:45:54
For m ore inform ation, visit www.uop.com/uniex.
2012 H oneyw ell International, Inc. A ll rights reserved.
refinem ent redefined
UOP Unihox rosiduo-upgrading toonnology yiolds 25% moro
oloan uol, turning "bottom o tno barrol" into "top o tno lino."
Got tno most rom ovory barrol. Tno UOP Unihox Proooss oan doublo tno diosol yiold
o otnor rosiduo-upgrading toonnologios, turning tno "bottom o tno barrol" into moro
blaok on your bottom lino. Tno Unihox Proooss is a nign-oonvorsion, oommoroially
provon toonnology tnat prooossos low-quality rosiduo stroams liko vaouum rosiduo into
nign-quality distillato produots loading to rohnory margin inoroasos o up to 100%.
Simply put, witn tno Unihox Proooss rom UOP, you'll maximizo produotion and prohts
rom ovory barrol.
SPM-UOP-30 Uniflex Ad_Comp_v6.indd 1 3/8/12 2:35 PM
uop.indd 1 9/3/12 12:31:44
Tracking performance degradation in a
debutaniser column
C
omplex, capital-intensive
industries such as rening and
petrochemicals face a pressing
demand to minimise unplanned
downtime and achieve higher asset
uptime and improved Health Safety
& Environment (HSE) targets. To
improve the reliability and availabil-
ity of facilities and assets, Saudi
Aramco has been investigating the
application of SAS Predictive
Analytics and Data Mining in opera-
tions and maintenance. In this
regard, a pilot project was conducted
on the ability to deliver leading indi-
cators to the problem of performance
degradation in the Ras Tanura ren-
erys hydrocracking unit debutaniser
column. This article deals with this
A project applying statistical and data mining techniques to historical operating
data identied the root causes of a debutanisers poor performance
ALI ABDULAL, VINOD RAMASESHAN and STANLEY GUSTAS III Saudi Aramco
HORIA ORENSTEIN, MOHAMMAD KURDI and ANDREY GASKOV SAS Middle East
pilot project; understanding of the
use of a business analytical tool,
resulting in successfully achieving
the goals that deliver prediction of
performance degradation; the ag-
ging of leading indicators; and the
development of a dashboard for
operating staff, with recommenda-
tions for revision to the multivariable
control system (MVC) system to
alleviate the problem.
Hydrocracking unit
Saudi Aramcos Ras Tanura ren-
ery operates a 44 000 b/d
Unicracking process unit (hydroc-
racking unit) designed by UOP and
commissioned in 1999. This unit is
a two-stage maximum conversion
unit (>97%), where the primary
feed vacuum gas oil (SC-7 and SC-
8) is treated and cracked over three
reactors (C-100 and C-101 in the
rst stage, and C-200 in the second
stage) to naphtha and distillate
products.
Since the unit was originally
conceived to operate in both maxi-
mum distillate and naphtha mode,
the fractionation section is what in
hydrocracking parlance is termed a
debutaniser rst fraction scheme.
The reactor efuent rst comes to
the debutaniser, where lighter ends
are removed before being routed to
the main fractionation section.
Consequently, the debutaniser acts
both as a conventional stabiliser
Reaction section
Recycle compressor
Second stage cooler
First stage cooler
J 8 0 -D -1 3 0
Debutaniser Fractionator
Fractionation section
Feed/effluent
exchangers
VG O fe e d
E -3 1 5 E -3 2 4
J 8 0 -F -3 1 0
J 8 0 -F -3 2 0
J 8 0 -G -3 1 1
Figure 1 Ras Tanura hydrocracking unit
aramco.indd 1 10/12/12 13:18:43
and as a hydrogen sulphide (H
2
S)
rejection column. To act as a H
2
S
rejection system, it is imperative
that the vapour-to-liquid ratio at
the bottom of this column is high
(>0.5 on mole basis). If it is lower,
H
2
S rejection is poor, leading to
corrosion in the downstream frac-
tionation column and pressure drop
in the second-stage reactor (C-200)
on account of particulate carry-over
with the recycled oil, over and
above possible leaks in the piping
on account of high temperature and
wet H
2
S corrosion.
Subsequently, since the feed to
the debutaniser essentially comes
from the reaction section, the
vapour-to-liquid ratio is strongly
inuenced by the product slate. Ras
Tanuras hydrocracking unit, since
start-up, has predominantly oper-
ated in distillate mode (>95% of the
time) and with catalyst systems
changed to low-zeolite catalyst in
the rst stage and amorphous cata-
lysts in the second (as opposed to
high-zeolite catalyst in the past).
The product slate has predomi-
nantly been heavier (a atter true
boiling point, TBP, curve), leading
to a higher than design requirement
in the reboiler duty (based on a
preferential once-through reboiler
[red heater] design), with no
improvement in the vaporisation
rate, leading to H
2
S slippage and
associated corrosion and reliability
problems. To mitigate this issue, a
multivariable controller (MVC)-
based control system for the
second-stage conversion levels was
developed by changing the C-200
inlet temperature and therefore the
weighted average bed temperature
(WABT). This was done because the
distillate selectivity per pass conver-
sion on the second stage is lower
when compared to the rst stage
for the same catalyst (the effect of
ammonia comes into the picture).
Consequently, the light end make
would increase and thereby main-
tain the vapour-to-liquid ratio in
the debutaniser. However, this
procedure leads to a debit in distil-
late make (the product of choice for
the renery).
A pilot project with SAS Institute,
using predictive analytics and data
mining techniques, was applied to
this problem, to ascertain if other
leading indicators could be tracked
to alleviate the situation and main-
tain the distillate make from the
unit. While the renery will conduct
other engineering modications to
the unit in the future, data mining
was considered to be a good
concept to deal with the current
situation, alleviate the problem, if
possible, and therefore optimise
any additional engineering solution.
Figure 1 shows a high-level over-
view of the hydrocracking unit in
question.
Technique and analytic procedure
The objectives of the project, called
the Predictive Performance
Degradation (PPD) project, can be
summarised as follows:
Perform data mining techniques
to understand inter-dependencies
among different debutaniser opera-
tional parameters that include:
Hot and cold feed tempera-
tures, ows and compositions
Internal reux
Tray 23 temperature: the
seventh tray from the top in the
rectication section
Reboiler duty and bottom
temperatures
Identify the cause and effect rela-
tionships among these operating
parameters and the following debu-
taniser performance-related issues:
Debutaniser instability
Light hydrocarbon slippage
and Reid vapour pressure
deviations
H
2
S slippage and corrosion
problems at the bottom of the debu-
taniser and the fractionator column
Proactively identify the critical
event and root causes that contrib-
ute the most to the above
performance issues
Generate proactive dashboards,
reports and other user interface
screens that provide information
about events, symptoms, causes
and effects of the performance
degradation of the debutaniser,
7 0
8 0
6 0
5 0
4 0
3 0
2 0
2 0 /4 /2 0 1 0 2 2 /4 /2 0 1 0 2 4 /4 /2 0 1 0
1 0
C A S E 1 0

R
e
f
l
u
x

f
r
o
m

G
-
3
1
0
A
/
B

f
l
o
w
Figure 2 Case 18, reux ow rate vs time
Analysis description SAS module used
H
2
S slippage event and root causes sequence analysis SAS Enterprise Guide
SAS Enterprise Miner
Visual discovery analysis for 24 H
2
S indirect indicators SAS/BASE
SAS/CONNECT
SAS JMP
Correlation and regression analysis for major operation parameters SAS/JMP
SAS/ETS
Non-linear regression analysis for cold-hot feed composition SAS/ETS
Neural network modelling for rst-stage efuent density SAS/BASE
SAS JMP
Visual discovery analysis for reux loss root causes SAS/BASE
SAS JMP
SAS Predictive Asset Maintenance modules used
Table 1
aramco.indd 2 12/12/12 10:52:14
thereby helping engineers and
operators to react in time and
minimise the debutanisers
instability.
The performed predictive analysis
is based on historical process infor-
mation data. It applies statistical
and data mining techniques to this
data to nd the factors and root
causes that affect the debutanisers
performance. Analytical capability
is the highlighted outcome of the
project. It has been determined that
there is a huge potential in benets
from the exploitation of historical
data. It has also been determined
that exploitation of such data in a
similar way has not been fully
applied in the past. Therefore, this
project was designed to utilise the
power of analytics to unleash the
potential of such historical data and
to make operations more stable and
protable.
Although the performance of the
debutaniser with regard to H
2
S slip-
page and its stability has shown
signicant improvement after the
MVC systems implementation, the
PPD project has shown that there is
still room for further improvements.
To facilitate the analysis, apart
from mechanical data, process and
control data were also analysed.
More than 300 tags for three years
data were processed (data taken
from before the implementation of
the MVC and also after), which
resulted in over 750 million records
being analysed and correlated to
nd the root cause of the debutan-
iser columns performance
degradation. The required modules
and components from the analytical
bundle SAS Predictive Asset
Maintenance were used. Table 1
lists the tools used.
Since H
2
S slippage from the debu-
taniser column bottom is primarily
due to slumping of column oper-
ations, the analysis was started with
this in focus; however, measure-
ment of H
2
S, in particular transient
high values, in the laboratory and
in the oil stream proved to be a
great challenge. This resulted in
unreliable deterministic analysis of
H
2
S. To overcome this obstacle, 24
H
2
S slippage indirect indicators
were considered, based on opera-
tions and engineering opinion along
5 8
5 9
5 7
5 6
5 5
5 4
5 3
2 0 /4 /2 0 1 0 2 2 /4 /2 0 1 0 2 4 /4 /2 0 1 0
5 2
C A S E 1 8
2 7
2 8
2 6
2 5
2 4
2 3
2 2
2 0 /4 /2 0 1 0 2 2 /4 /2 0 1 0 2 4 /4 /2 0 1 0
2 1
C A S E 1 8

D
-
1
4
0

c
o
l
d

f
l
a
s
h

d
r
u
m

l
i
q
u
i
d

f
l
o
w

D
-
1
2
0

b
o
t
t
o
m
s

t
o

C
-
1
3
0

f
l
o
w
Figure 3 Case 18, cold and hot feed rates
4 8 0
4 6 0
4 4 0
4 2 0
4 0 0
3 8 0
2 0 /4 /2 0 1 0 2 2 /4 /2 0 1 0 2 4 /4 /2 0 1 0
3 6 0
C A S E 1 8
C
-
3
1
0

t
r
a
y

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

C
3 . 4
3 . 0
2 . 6
2 . 2
1 . 8
1 . 4
2 0 /4 /2 0 1 0 2 2 /4 /2 0 1 0 2 4 /4 /2 0 1 0
1 . 0
C A S E 1 8

R
e
b
o
i
l
e
r

r
e
t
u
r
n

t
o

C
-
3
1
0

p
r
e
s
s
u
r
e

d
i
f
f
e
r
e
n
c
e

Figure 4 Case 18 Tray 4 temperature and eccentric orice pressure drop
aramco.indd 3 10/12/12 13:19:04
debutaniser column; manipulated
variables and process disturbances
based on changes to feed quality;
and/or other disturbances to the
system.
Based on the inuence map,
multiple cases were analysed.
Figure 2 shows the trend of reux
ow changes for Case 18.
It is evident that initially the
reux rate is stable and then drops,
steadies itself, nally plunging to
15 000 b/d within 24 hours. When
this time period was superimposed
with changes in the hot and cold
feed rate from the hot and cold
ash drums to the debutaniser
column (the reactor efuents), it is
evident that the hot feed rate
increases when the cold rate corre-
spondingly goes down (see Figure
3). The colder feed is essential
liquid ashed from the hot separa-
tor/ash drum and, as such,
constitutes the lighter components
of the reactor efuent.
The reboiler duty is constant, but
jumps to a higher value as reux
suddenly drops. (The reboiler duty
is maintained constant, as the outlet
temperature is close to the equip-
ments maximum temperature)
There is a surge in tray 4s tempera-
ture and, despite the increase in
reboiler outlet temperature, the net
vaporisation has decreased on
account of a at TBP curve for
the reactor efuent. Figure 4 shows
the complete scenario.
It is important to note that the
reboiler outlet temperature is high,
but boil-up rates are low. This can
only happen if the debutaniser
bottoms liquid has a high density,
indicating a heavier hydrocarbon
composition.
To prove this hypothesis, the
liquid density at the inlet of the
reboiler (F-310) is required.
Unfortunately, there are no
sampling points in this area to
develop the corresponding distilla-
tion curve. Therefore, inferential
density was calculated, taking into
account the fractionators product
ow rates. The idea behind this
estimation is as follows. The liquid
composition coming into the F-310
reboiler is the same as the liquid
composition owing into the main
fractionation column. Nearly all
with their hypotheses. Using SAS
Institute tools for statistical analysis
and data mining techniques, all of
these indicators and hypotheses
were carefully examined. The
discovery at this stage was that the
most signicant problem is loss of
reux ow in the debutaniser.
During reux losses following
signicant H
2
S slippage, leading
indicators were affected; namely, a
reduction in tray 23s temperature;
surges in the hot feed component
rate, with a corresponding dip in
the cold feed component rate;
surges in tray 4s (four trays above
the reboiler return line) tempera-
ture; a reduction in reboiler duty;
and a drop in the reboiler return
eccentric orice pressure drop,
along with a reduction in the
condenser outlet temperature.
Consequently, 22 cases of signi-
cant reux loss during a three-year
period from 2008-2010 were
analysed. Eleven of these were not
considered because of extremely
unstable debutaniser column opera-
tion. The other 11 cases of
signicant reux loss during the
period 2009-2010 were analysed in
greater detail.
Based on the above, a reux prob-
lem inuence map was developed,
which basically analyses the nd-
ings and shows inter-dependencies
between the various parameters:
control variables around the
0 . 8 2
0 . 8 3
0 . 8 1
1
7
/
0
4
/
2
0
1
0
0
8
:
0
0
2
1
/
0
4
/
2
0
1
0
0
:
0
0
2
3
/
0
4
/
2
0
1
0
1
7
:
0
0
2
6
/
0
4
/
2
0
1
0
0
9
:
0
0
2
9
/
0
4
/
2
0
1
0
0
:
0
0
D
e
n
s
i
t
y

a
v
e
r
a
g
e
0 . 8 0
C A S E 1 8
2 0
3 0
1 0
1
7
/
0
4
/
2
0
1
0
0
8
:
0
0
2
1
/
0
4
/
2
0
1
0
0
:
0
0
2
3
/
0
4
/
2
0
1
0
1
7
:
0
0
2
6
/
0
4
/
2
0
1
0
0
9
:
0
0
2
9
/
0
4
/
2
0
1
0
0
:
0
0
R
e
f
l
u
x

f
r
o
m

G
-
3
1
0
A
/
B

f
l
o
w
0
6 0
7 0
8 0
9 0
5 0
4 0
C A S E 1 8
Figure 5 Case 18, density and reux ow rates
High reboiler duty
but low vaporisation
and hence a sudden
reux drop
aramco.indd 4 11/12/12 17:05:45
combustion and environmental solutions. pure and simple.
Emissions regulations continue to tighten worldwide.
Stay ahead of the regulatory curve and protect your
bottom line by choosing innovative combustion solutions.
From our smokeless flares to our Next Generation Ultra-Low NO
x
GLSF Free-Jet burners, Zeeco pushes combustion technology
forward. Driven by the challenges our customers face, we design
flares, burners, and thermal oxidisers that simply work better and
last longer. Global offices, local expertise, and smart design.
Zeeco is combustion innovation... pure and simple.
burners flares thermal oxidisers aftermarket products & services
Zeeco, Inc. 22151 E 91st St., Broken Arrow, OK 74014 USA
+1-918-258-8551 sales@ zeeco.com zeeco.com
ZEECO
is a registered trademark of Zeeco, Inc. in the U.S.

U.S. Patent Nos. 6,499,990, 6,626,661
Zeeco, Inc. 2012
Zeeco_PTQ_full_Dec.indd 1 12/10/12 2:18 PM
zeeco.indd 1 11/12/12 18:52:00
Among these innovative products is a line of internals
used in down fow processes to retain the often-costly
media and to provide a collection area for the process
fow across the entire vessel diameter or length.
hydrotreaters
desulfurizers
hydrocrackers
dryers
sand flters
gas sweeteners
other absorption systems
EUROPE - MIDDLE EAST - AFRICA
TEL: +33 5 4902 1600
saleseurope@johnsonscreens.com
AUSTRALIA - ASIA PACIFIC
TEL: +61 7 3867 5555
salesasiapacifc@johnsonscreens.com
www.johnsonscreens.com
JAPAN
TEL: +81 55 997 8511
sales.jpn@johnsonscreens.co.jp
NORTH, SOUTH & CENTRAL AMERICA
TEL: +1 651 636 3900
hpisalesnsa@johnsonscreens.com
CONTACT US!
Because of their strength,
durability & fow characteristics,
Johnson Screens support grid
and distributor tray systems are
widely used in:
Johnson Screens

designs and
manufactures a wide range of vessel
internals for media retention.
j screens.indd 1 07/01/2013 17:27
hydrocarbons owing into the frac-
tionation column are separated into
light naphtha, heavy naphtha, kero-
sene, diesel and fractionator
bottoms (consisting of drag oil and
feed to the second stage). By meas-
uring the densities for these
streams, one can calculate the aver-
age fractionator products density.
While this formula is only an
approximation (as it does not take
into account volume changes at
different temperature and other
properties), it is considered suf-
cient to ascertain trends and
abnormalities during normal opera-
tions. The laboratory measures
distillation curves; API gravities for
all of these streams, in most cases,
are available. A plot of the distribu-
tions of streams densities was then
developed, based on the following:
Calculated densities for streams
for a near-three-year period
Removed observations that are
connected with transient intervals.
Based on the data distribution,
the most probable densities (median
value) for all ve streams were
considered and the inferential
density was calculated.
Figure 5 shows density behaviour
during the corresponding reux
loss case. The red arrows indicate
the beginning of the reux loss
problem. The intervals that are
shaded show unexpected and unde-
sirable jumps in density.
In Figure 5, the jump in density
shows the sudden drop in reux
rate. In the second instance, the
reux drop is not pronounced on
account of operator intervention.
It is evident that density change
has a big impact on reux drop.
However, for the sake of argu-
ment, even if a density measurement
was provided to alert the operator,
the value and hence the trend would
still be a lagging indicator. It is
evident that density change occurs
on account of changes to the cold
and hot incoming liquid stream rates
from the reactor and therefore origi-
nates in the reaction section.
Therefore, to obtain a leading
indicator, a quality inuence map
was developed for the reactor/
debutaniser along with the reactor
feed. The map developed the rela-
tionship between:
The vacuum column operating
parameters, along with ambient
conditions and their inuence on
the quality and quantity of LVGO
and HVGO (SC-7 and SC-8), which
are the components of fresh feed to
the hydrocracking unit
The inuence of the feed on
C-100, C-101 temperature proles,
and the quality and quantity of
efuent
The inuence of the recycle oil
feed on C-200 temperature proles,
and the quality and quantity of
reactor efuent
The reactor efuents inuence on
debutaniser operating variables and
their consequent effect.
Feed quality, along with tempera-
ture rises in the individual beds of
C-100, C-101 and C-200, were all
plotted, as was the total tempera-
ture rise of the treating and cracking
catalyst beds. Indicators such as the
effect of ambient temperature on
vacuum column pressure, and
therefore its effect on the feed qual-
ity to the hydrocracking unit, was
also investigated. It became evident
that as the feed to the hydrocrack-
ing unit becomes heavier (and
pressure in the vacuum column
uctuates on account of uctuations
in ambient temperature), the total
temperature rise over the rst reac-
tor changes.
An analysis of the temperature
prole of C-100 and C-101 Bed 1
was conducted over a two-month
period. The 50% distillation point
was taken as a measure of density
of the combined feed. It was
observed that as the density rose,
the overall total temperature rise in
6 9
8 5 0
8 6 0
8 4 0
1
2 3
4
8 3 0
8 2 0
6 8
6 7
6 6
6 5
6 4
1
/
7
/
2
0
1
0
1
5
/
7
/
2
0
1
0
2
2
/
7
/
2
0
1
0
2
9
/
7
/
2
0
1
0
5
/
8
/
2
0
1
0
1
2
/
8
/
2
0
1
0
1
9
/
8
/
2
0
1
0
2
6
/
8
/
2
0
1
0
8
/
7
/
2
0
1
0
C
-
1
0
0
6 3
Figure 6 Fluctuation in density and changes in total temperature rise for C-100
8 5 0
8 6 0
8 4 0
1
2 3
4
8 3 0
8 2 0
2 0
1 9
1 8
1 7
1 6
1
/
7
/
2
0
1
0
1
5
/
7
/
2
0
1
0
2
2
/
7
/
2
0
1
0
2
9
/
7
/
2
0
1
0
5
/
8
/
2
0
1
0
1
2
/
8
/
2
0
1
0
1
9
/
8
/
2
0
1
0
2
6
/
8
/
2
0
1
0
8
/
7
/
2
0
1
0
C
-
1
0
1

b
e
d

t
r
a
y

1

t
e
m
p
e
r
a
t
u
r
e

r
i
s
e
1 5
2 0
1 9
1 8
1 7
1 6
C
-
1
0
1

b
e
d

t
r
a
y

1

t
e
m
p
e
r
a
t
u
r
e

r
i
s
e
1 5
7 2 0
7 1 8
7 1 6
7 1 4
7 1 2
C
-
1
0
1

i
n
l
e
t
7 1 0
C -1 0 1 in le t
C -1 0 1 b e d tra y 1
Figure 7 Fluctuation in density and changes in C-101 Bed 1s inlet temperature and
temperature rise
aramco.indd 5 10/12/12 13:19:24
C-100 decreased. Various periods
(shown as points 1-4 in Figure 6)
follow the trend of drop in total
temperature rise vs change in
density, whenever operator interfer-
ence with change in inlet
temperature is not superimposed.
Figures 6 and 7 show the uctua-
tion of the total temperature rise at
C-100 and C-101 Bed 1. It was also
observed, and interesting to note,
that the performances of C-101 Bed
2 and C-101 Bed 3 follow opposite
directions when inlet temperatures
to these beds were maintained at
nearly constant levels (in fact, there
was hardly any change in Bed 2
delta Ts).
So, having established the fact
that feed changes affect the rst-
stage performance, the reactor
efuent quality (composition) was
then plotted against the same
period of time, as this is essentially
the feed to the debutaniser. Since
distillation and density measure-
ments for reactor efuent are not
available for the same period of
time (high-pressure product
sampling is restricted for safety
reasons), inferential density over
the period and the amount of 640F
minus material (reboiler return
temperature is limited to 640F) in
the efuent (based on available
distillation data) were plotted
against feed distillation and C-
100/C-101 Bed 1 performance. It
was evident that a loss in reux
usually followed a loss in tempera-
ture rise in C-101 Bed 1. Further,
growth in inferential density and a
loss in temperature rise over the
rst-stage C-100 and C-101 Bed 1
temperature occur well in advance
of the reux collapse. Figure 8
shows a graphical view of this for
a given case.
Inferences and recommendations
The analysis reveals that degrada-
tion in the performance of the
debutaniser is affected by debutan-
iser feed composition. This scenario
takes place when C-100 and C-101
Bed 1 (hydrocracker pretreat)
performance deteriorates. The main
reason for deterioration of these
reactors performance during stable
operation is heavy feed, more
specically the feed from the
vacuum distillation column.
Deterioration in the performance of
C-101 Bed 1 is usually more severe
than deterioration in the perform-
ance of C-100. Due to an undersized
C-101 inlet quench valve, the MVC
system and the operators have very
limited control over the perform-
ance of C-101 Bed 1.
The analysis also revealed that
the VDU column pressure signi-
cantly impacts the hydrocracker
feed quality and causes day/night
performance uctuations. The
temperature rise in C-101 Bed 1 is
a leading indicator and can be
used as a control in the MVC to
achieve stable debutaniser
operations.
A dashboard and reporting
system has been proposed and is
being developed, which will allow
monitoring and identication of the
reactors performance deterioration
at the early stages, while agging
the factors and root causes affecting
performance and destabilising
operations.
Additional recommendations, both
from an engineering and operational
point of view, have also been devel-
oped, such as ensuring sufcient H
2

availability (valve sizing) at the inlet
of C-101, which will help stabilise
the C-101 Bed 1 inlet temperature
and increase control. Prioritisation
18
20
19
22
21
CASE 18
17
16
15
14
13
0
1
0
2
/
4
/
4
1
0
1
0
2
/
4
/
8
1
0
1
0
2
/
4
/
2
2
0
1
0
2
/
4
/
6
2
0
1
0
2
/
4
/
0
3
0
1
0
2
/
5
/
4
c
e
n
a
m
r
o
f
r
e
p

1

d
e
b

1
0
1
-
C
12
C-101 bed 1 performance drop
|eading indicator.
8 days to dangerous situation
71
75
73
CASE 18
69
67
65
63
61
0
1
0
2
/
4
/
4
1
0
1
0
2
/
4
/
8
1
0
1
0
2
/
4
/
2
2
0
1
0
2
/
4
/
6
2
0
1
0
2
/
4
/
0
3
0
1
0
2
/
5
/
4
c
e
n
a
m
r
o
f
r
e
p

0
0
1
-
C
59
C-100 performance drop
|eading indicator.
7 days to reux |osses,
14 days to dangerous situation
Figure 8 C-101 Bed 1 performance and inferential density growth
The analysis reveals
that degradation
in the performance
of the debutaniser
is affected by
debutaniser feed
composition
aramco.indd 6 11/12/12 17:02:26
between C-100 and C-101 Bed 1 total
temperature rises and WABT
controls to maintain constant
performance of the debutaniser and
control the C-100 Bed 3 outlet
temperature to prevent a low inlet
temperature into the C-101 Bed 1
(once sufcient H
2
is available at C-
101 top bed) have also been
recommended and are being
implemented.
On the whole, this pilot project
has shown that if the data set to be
analysed is very large, data mining
techniques, along with statistical
and mathematical methods,
provide quantitative results, which
can be translated into a meaningful
control strategy when coupled
with experience and empirical-
based qualitative logic. The project
has also shown the capability to
detect early development of
performance degradation, utilising
data mining techniques and actual
plant history, while employing a
systematic and largely automated
methodology.
Unicracking is a trademark of UOP, a Honeywell
company.
Acknowledgment
The authors would like to acknowledge all
members of Saudi Aramco Central Engineering
group, along with the management for their
uninching support during this pilot project
exercise.
Ali H Abdulal is an Engineering Consultant
with Saudi Aramco Process & Control Systems
Department. He has over 24 years experience
commissioning, operating, troubleshooting
and optimising renery and gas plant units
within Saudi Aramco. He holds a BSc degree in
chemical engineering from Tri State University
and has one patent under his name.
Vinod Ramaseshan is a Process Engineer with
Saudi Aramco Process & Control Systems
Department, with over 19 years experience
commissioning, operating, troubleshooting
and optimising new/revamped hydrocracking
and hydrotreating units in the EMEA and India
region. He holds a masters degree in chemical
engineering from the Indian Institute of
Technology, Mumbai and is a Chartered
Engineer in the UK.
Stanley Gustas III is an Operations Engineer
at Saudi Aramco Ras Tanura renerys
Engineering Department. He has over 22
years experience designing, commissioning,
operating, troubleshooting and optimising
new/revamped hydrocracking and
hydrotreating units worldwide. He holds a
bachelors degree in chemical engineering
from Rice University, Houston, Texas, and is a
Registered Professional Engineer in Texas.
Horio Orenstein is the Director of Advisory
in the Global Oil & Gas Business Unit of
SAS Institute. His 20 years experience in the
process industry includes the development,
standardisation and implementation of
business concepts and solutions in renery
operations optimisation, integrated asset
management, integrated supply chain
optimisation, integrated planning, facility
integrity and reliability.
Mohammad Kurdi is an Advisor with the Oil
& Gas Business Unit of SAS Institute. With 10
years experience in analytical intelligence,
he holds a BSc in electrical engineering from
Jordan University of Science & Technology.
Andrey Gaskov is a Statistical Analyst
with SAS Middle East FZ-LLC. With over
10 years experience providing consulting
services, analytics and data mining for oil
and gas industry companies, upstream and
downstream, he holds a masters degree in
information technologies from Novosibirsk
State University, Russia.
ellciency lor induslry
One sinple slep lo added produclivily
Kurila Europe CnbH
.
nduslriering 43
.
4!75! Viersen
.
Cernany
.
Phone:+49 2!62 9530!!0
.
www.kurila.de
Troublelree operalion is lhe only lhing lhal guaranlees your
syslen will work al naxinun ellciency during produclion.
Thals why al Kurila, weve developed innovalive producls lor
processes in relneries and lhe pelrochenical induslry. These
producls will help you averl produclion bolllenecks caused
by corrosion, scaling and louling. And our capable engineers
will review whal you need and cone up wilh a concepl lhals
jusl righl. vhen you choose Kurila, youre naking a choice lor
producls wilh sound lechnology - and services lhal range lron
consullalion lo inplenenlalion and nainlenance.
See whal a dillerence grealer ellciency and lruly worrylree
operalion nake!
aramco.indd 7 10/12/12 13:19:45
We put tested ref ning technologiesand processes
to work in your world. From hydroprocessing to
f uid catalytic cracking and Veba combi-cracking to
resid processing, we of er proven know-how. So
you can improve productivity and lives. KBR
Technology licenses deliver for greenf eld and
existing ref neries of virtually every type and size.
See HOW we can help you meet mission-critical
goals. Visit refning.kbr.com to learn more.
KNOW-HOWDELIVERED
REFI NI NG
C
M
Y
CM
MY
CY
CMY
K
advert_PTQ copy.pdf 1 12/6/2012 4:50:19 PM
kbr.indd 1 7/12/12 11:21:57
Operator training simulator for a
multi-product fuels pipeline
A
s part of a larger project to
design and build a new
multi-product fuels pipeline
system for Trinidad and Tobago, an
innovative operator training simu-
lator (OTS) was developed that
allows highly realistic training of
the pipeline operators in all phases
of operations prior to actual startup
of the pipeline and for continuous
training after startup. The pipeline,
owned by a consortium of
Trinidadian entities, including the
Government of Trinidad and
Tobago, the National Gas Company
of Trinidad and Tobago (NGC), the
Trinidad and Tobago National
Petroleum Marketing Company
(TTNPMC) and Petroleum
Company of Trinidad and Tobago
(Petrotrin), is designed to deliver
up to 36 000 b/d of multiple fuel
products (gasoline, diesel and avia-
tion fuels) at peak by a
state-of-the-art automated pipeline
system roughly 47 km in length
from the islands Petrotrin West
refnery to a new distribution termi-
nal in Caroni County in the central
part of Trinidad. Also included in
the distribution system is a new
11 km pipeline to deliver aviation
fuel from the Caroni terminal to
Trinidads Piarco Airport (see
Figure 1).
The pipeline system is being
built to ensure secure and on-time
delivery of fuel products to key
points on the island, which other-
wise relies on delivery by trucks.
To ensure the quickest delivery
times, the pipeline also employs a
new pigless delivery system that
loads one liquid product immedi-
ately behind another in a
continuous, seamless fashion. The
An operator training simulator for a fuel products pipeline system was developed
and delivered before the pipeline was completed to train personnel in operations
SANTOSH ALLADWAR, CORDELL CHAPMAN, BILLY MAYO, SURAJIT DASGUPTA and JAVIER VAZQUEZ-ESPARRAGOZA KBR
DANIEL VOGT Krohne Oil and Gas
KEONE JACKMAN Petroleum Company of Trinidad and Tobago Limited
proper operation of the pipeline in
this pigless mode is ensured by a
leak detection system (LDS) and
batch management system (BMS)
from Krohne Oil and Gas, a
company specialising in metering
skid solutions and computational
pipeline monitoring (CPM) pipe-
line leak detection. The LDS
and BMS were integrated into the
OTS, which would be used in
tandem with the pipeline control
system (PCS) and the terminal
automation system (TAS) to plan,
schedule and track product
batches, keep track of the feed and
product tanks, as well as provide
alarms and/or advice to operators
on abnormal conditions, and
continuously detect for potential
pipeline leaks. As some of
these systems are new and to
some extent custom built for this
project, it was agreed and planned
in advance to have an OTS to
aid in the rapid familiarisation
and training of operations
personnel.

Process overview
The batch of refned products from
Petrotrin West Refnery storage
tanks consisting of unleaded
RON92 gasoline, unleaded RON95
gasoline, diesel and Jet A-1 fuel
will be metered and transferred by
pipeline pump using individual
Figure 1 Pipeline path on Trinidad and Tobago map
The pipeline system is
being built to ensure
secure and on-time
delivery of fuel
products to key points
on the island, which
otherwise relies on
delivery by trucks
Piarco Airport
kbr.indd 1 12/12/12 11:09:51
AS DESIGNED AS BUILT AS OPERATING
Operations Integrity Management
Building Reputations
In an increasingly complex and regulated environment, Owner Operators must meet
demanding levels of operations integrity.
AVEVAs information management solutions help to meet this challenge by giving you
the information you need to make the right decisions - maximising asset and project
performance. FromFEEDthrough to decommissioning, an AVEVA solution improves
productivity, increases reliability and reduces risk.
The leader in design, engineering and information management software for the process
plant, power and marine industries, AVEVA invests in our customers success through a
global sales and support network in more than 40 countries.
AVEVA building solid reputations for 45 years
www.aveva.com/oim
Playing it Safe in
Plant Operations
Download our new Process
Safety Management
Business Paper
www.aveva.com/qr/process-safety
aveva.indd 1 10/12/12 12:52:12
booster pumps for each product.
The preliminary batch sequence
would be as follows:
Jet A-1 Diesel RON 92 RON 95
RON 92 Jet A-1
This batch sequence will be trans-
ferred to the Caroni facility, the
distribution centre, using a pipeline
pump via an 8in multi-fuels pipe-
line. Before transferring to the
Caroni facilitys storage tanks, all
the fuels will be metered at a meter-
ing skid entering the pipeline. The
meter will be used to control the
mainline pumps, while the tank
gauging systems will be used for
custody transfer. The general opera-
tion is designed to run with no
pigs; however, an additional pig
launcher was built to launch an
intelligent/batch pig in cases when
this operation was needed.
As soon as a batch arrives at the
Caroni facility, it will be diverted to
the appropriate destination tank by
the operator based on the interface
detection. For batch interface detec-
tion, density is the primary
parameter used. Two densitometers
are used to detect the interface of
the multi-fuels. The rst interface
detector is installed 4.4 km from the
Caroni facility to alert the operator
about the product coming in the
pipeline. The operator may use a
second interface detector, located at
the Caroni centre, to decide the
correct time to switch valves to
transfer the product into an appro-
priate tank.
The liquid fuels interface is one
of the most important parameters
to measure at the Caroni facility in
order to transfer the product into
an appropriate tank. When the
multi-fuels are transferred using
one pipeline, an interface mixture
is generated. The interface
mixture/off-spec product can be
diverted to slop tanks or to the
lower-quality product tank. Two
slop tanks are provided to accom-
modate the interface a high
ash slop tank and a low ash
slop tank. The interface/off-spec
product from the slop tanks is
trucked back to the Petrotrin West
renery for reprocessing.
Fuel from the individual storage
tanks is transferred to (truck) load-
ing bays using individual transfer
pumps for each product.
For the batch control of two
grades of gasoline RON 92
gasoline and RON 95 gasoline
their physical properties would be
too close to reliably differentiate
the products based on density
alone. Optical interface detectors
(OIDs) are also installed near to
the densitometers. The two gaso-
line grades are dyed for taxation
purposes, resulting in distinct
optical signatures (primarily
colour), which the OID can distin-
guish and signal to the operator.
Additionally, the ow meter count
could be utilised to guide the
operator
An additional dedicated pipeline
8in diameter and 11.2 km length is
installed to transfer the Jet A-1 from
Caroni to the Piarco facility. There
are two isolation valves installed on
the pipeline because of a river
crossing, one valve at each side of
the river. A 600-gallon-per-minute
capacity pump can deliver jet fuel
from the Caroni distribution centre
to Piarco Airport.
Simulator development
The OTS was designed and deliv-
ered by KBRs Advanced Chemical
Engineering group with support
from the Pipeline and Operations
groups. Overall, the OTS consists of
four components:
A custom dynamic model of the
pipeline system (resembling actual
pipeline operation) and an instruc-
tor station to train the operator,
with various features such as oper-
Printer
Ethernet Switch(LAN)
Operator TraineeConsole
(I-Station)
SimulationPC &Instructor
Station
LeakDetectionand
BatchTracking Console
(BMS/LDS)
D yn a m ic m o d e l o f T a n k s a n d p ip e lin e
S im u la te s a ll p h ys ic a l e q u ip m e n t a n d
in s tru m e n ts
I n c lu d e s m a lfu n c tio n s & tra in in g
s c e n a rio s
S e s s io n in itia liza tio n c o n tro l lo g ic &
file s
D is p la ys a ll c o n tro ls g ra p h ic s
a n d o p e ra to r d is p la ys
A ll h a rd w a re c o n n e c te d
th ro u g h E th e rn e t sw itc h
EngineeringandAppStation
D e d ic a te d E n g in e e rin g S e rve r
h o s tin g b o th th e P ip e lin e &
T e rm in a l S C A D A d a ta b a s e s
BMS Simulation connection to OTS
through OPC link
Internet Gateway
Switch
Figure 2 Architecture of the operator training simulator
kbr.indd 2 11/12/12 14:51:56
ating training on feld devices. The
safety instrumented system (SIS)
logic was also implemented in the
model. The SIS hardware/software
was not replicated because it was
possible to model its functionality
with the pipeline software. A
portion of the pipeline system, the
TAS, handling loading on logistics
operations, was not included in the
OTS system. Only some of the main
operations screens were simulated
A BMS and LDS identical to the
real pipelines BMS and LDS
systems
A dedicated Emerson DeltaV
distributed control system (DCS)
identical to the real pipelines DCS
DeltaV Operator consoles identi-
cal to the actual pipelines consoles.
These OTS components are inter-
faced through a standalone local
area Ethernet network using
industry-standard OLE for Process
Control (OPC) protocol for commu-
nications. Figure 2 shows the
architecture of the OTS.
The OTS also includes auxiliary
equipment to ensure the proper
working of the integrated system as
a standalone OTS. As such, the OTS
had no physical connections to
actual pipelines PCS or other pipe-
line systems.
Software
On reviewing the details of the
process and requirements for the
modelling of various process upsets,
it was decided to use the Stoner
Pipeline Simulator from GL Noble
Denton to provide the required
rigour and accuracy for the pipeline
model.
The Instructor graphics were
developed using Iconics Genesis32
that comes with the Stoner
software.
The BMS, was developed by
Krohne Oil and Gas, and it is used
to manage and track the batching
operations. It was fully integrated
with Krohnes LDS, PipePatrol,
used for the detection of leaks along
the length of the pipeline.
A dedicated Emerson DeltaV DCS
identical to the actual pipeline Delta
V hardware with the complete PCS
and TAS database provides a highly
realistic training environment.
Actual pipeline and associated
equipment modelling
The model included all physical
equipment (pipes, valves, pumps,
and so on) in the pipeline from the
feed tanks to fnal destination tanks
and includes complexities such as
truck loading racks and loaders. In
addition to the physical equipment
available in the software library,
programming was done to defne
operation of special equipment like
Yarway valves, truck loading, pig
status, and so on. The model also
included the complete safety instru-
ment system logic resembling the
actual SIS system. Figure 3 shows a
screen capture of a Stoner Pipeline
Software (SPS) scheme.
Instructor station
Overall control of the OTS is
through a special graphical display
system called the instructor inter-
face, which provides such
functionality as the run/pause of
the simulation, load/save of initial
conditions, touch points for failing
critical equipment, and standard-
ised scenarios for operator training.
Figure 4 shows a screen capture of
the Instructor interface.
Leak detection and batch tracking
management system
The Krohne leak detection and
batch tracking/management system
functionalities are intended to
manage the batching process, to
track the current position of a batch
in the pipeline and to schedule the
next batch. Also, operators can be
Figure 3 Model development screen
kbr.indd 3 11/12/12 14:52:05
creeping leak is a leak that does
not cause signifcant pressure
drops or pressure waves. For false
alarm-free operation, these param-
eters have been fne-tuned based
on recorded data to fnd the best
compromise between minimum
detectable leak rate and detection
time. Next to the leak detection,
which determines the volume of a
leak, this system has two different
algorithms based on pipeline
informed of the likely location of a
leak in the pipeline if it occurs.
Krohne was required to create
and test two new functionalities
for the BMS pause/start and
speed-up/slow-down in order
to complete the communication
and integration of the OTS system.
Many rounds of further testing
had to be done to ensure proper
two-way communication between
the BMS and the DCS. Figure 5
shows a screen capture of the
batch management system. Figure
6 shows a screen detail of the
status of pipeline batches and
alarms.
The LDS is a real-time transient
model (RTTM) type. It has special
parameters that determine its
sensitivity during pumping and
standstill conditions. Also, the
parameters that detect a creeping
leak can be set in the system. A
Figure 4 Instructor graphics showing feed and product tanks
Figure 5 Batch management system typical display
kbr.indd 4 11/12/12 14:52:15
pressure and fow to determine the
leaks location.
Modelling and integration
The overall modelling effort for the
OTS was rigorous and detailed to
include the complexity of the
modelling of some equipment, such
as the Yarway valves downstream
of the pumps and, in particular, the
prover circuits, the implementation
of the logics, the interlock system
and the communications among the
OTS components.
Some new functionality and
custom features had to be incorpo-
rated into the SPS software in order
to properly implement OTS func-
tionality, such as pause/resume
and save/restore, which had to be
transferred to the DCS and the
instructor interface systems. In
order to save a run or a fle for the
initial condition of a scenario, the
OTS system saves a fle in the
DeltaV system and a fle in the
simulation system simultaneously,
and both fles are loaded at the
same time when the instructor
restores a saved run. The BMS is a
real-time system that was originally
not intended to perform as an OTS,
but the feature of pausing and
restoring the simulation was added
successfully; pausing of the signal
to the BMS is available via OPC
command but not necessary because
the fow totaliser automatically
stops counting in the pause mode
of the Stoner simulator.
The DCS database is complex.
Many extra logic points were
required to be developed in the
simulation and connected to model
variables to allow a complete range
of operator actions from the DCS.
The DeltaV system is capable of
working in simulation mode when
it is not connected to the real and
physical system.
Figure 7 shows a distance plot of
both of the pipelines, from Petrotrin
Refnery to Caroni Distribution
Centre, and from Caroni to Piarco
Airport.
Reverse fow operation had to be
defned and confgured within the
operator training system. This oper-
ation occurs when a substantial
amount of off-spec product is not
accepted at the distribution centre
and has to be sent back to the refn-
ery via the pipeline. This procedure
is a manual operation and consists
of opening a set of valves, in the
right sequence and at the right time,
from the tank with the off-spec fuel
to the reverse fow pipeline at the
Caroni Centre. Then, operators
have to start the dedicated reverse
pump and continue with the open-
ing of the downstream valves to
direct the fow to the main pipeline
and let the rejected product travel
in reverse motion to the refnery.
Training for this operation proce-
dure is important because it will be
done very seldom at the actual
pipeline facilities; therefore, the
only means by which the operators
can be trained for this eventuality is
the OTS system. By doing this, the
operator will be ready when a
reverse operation is needed in the
real pipeline.
A particular limitation of the
model was the OID, which was
modelled as though it is driven by
density due to the absence of an
optical characteristic model in the
software.
Communication using OPC
connectivity
OPC Mirror, a server-to-server/
client communication software util-
ity from the DCS vendor, was used
to connect OPC servers (pipeline
model to DCS and BMS/LDS) on
the OTS system and to enable two-
way data communication from one
system to another. In the OTS, the
linkage was between the DCS, OPC
server and the Instructor server,
StOPC. Additionally, the Instructor
server and the Iconics graphics
interface communicate with the
simulation model variables. The
OPC communication was server-
server between the DCS and the
Iconics graphical interface, as a
client-server between Iconics
graphics and the simulation model,
and as a client-server between the
Wa rn in g s, if va lve
switc h e s a re d u e so o n
Wa rn in g s, if a n y ta n k
c o m b in a tio n is
in c o n siste n t
Wa rn in g s, if ta n k le ve ls
a re o ff h ig h a n d lo w)
Wa rn in g s d u e to
re d u n d a n c y sta te
L D S wa rn in g s
C u rre n tly
lo g g e d -o n o p e ra to r
Figure 6 Overview screen of pipeline, batches and alarm status
The overall modelling
effort for the OTS
was rigorous and
detailed to include
the complexity of the
modelling of some
equipment
kbr.indd 5 11/12/12 14:52:24
OPC mirror and BMS system. The
development team had to go
through additional technical devel-
opment on the DCS, the graphical
interface and the BMS to be able to
engineer a fully working system.
Figure 8 shows the general OPC
connectivity architecture within the
OPC system components.
Operations support
Additional support was required
from the Commissioning and
Startup Services group of KBR and
the client operations personnel to
help defne and test the training
scenarios and to defne the details
of the standard operation proce-
dures. In order to have detailed
procedures on the standard opera-
tions of the pipeline, batch startup,
batch changing, batch receiving,
interface disposal and so on, client
personnel and KBR Commissioning
and Startup Services personnel had
several meetings defning and
simulating the procedures and the
exercises to prepare the training
program for the operators. Some of
those procedures were also taken as
exercises for the fnal site accept-
ance test of the system.
Training
Table 1 shows a set of the different
scenarios that were prepared for
training the trainer initially and for
further training of operators. Some
of them are related to the OTS as an
integrated system tool and others
are related to operations procedures.
Benefts
While the primary purpose of the
OTS is operator training, several
benefts accrued to the project as a
result of the early development of
the simulator. It was also benefcial
to project staff who came to the
project late as a conceptual tool in
understanding the scope and objec-
tives of the pipeline system.
One of the key benefts was that
it was possible to completely check
the designs of both the DCS and
BMS systems prior to actual imple-
mentation. Several missing details
and incomplete functions were
found and corrected, particularly
those relating to the travelling and
volume of the interface along the
length of the pipeline and its prop-
erties shown on the systems in real
time; for instance, the distance from
the pumps and length of the inter-
Figure 7 Pipeline distance plot, from the simulation program and displayed on the Instructor station
One of the key
benefts was that
it was possible to
completely check the
designs of both the
DCS and BMS systems
prior to actual
implementation
kbr.indd 6 11/12/12 14:52:32
face between two consecutive
batches.
Further, it provided an invaluable
platform for completely testing and
validating the operating procedures
for the pipeline. It was possible for
KBR operations and training experts
to run through each procedure
many times to ensure that all
aspects were thoroughly tested. As
part of the testing process, NGC
has also already reaped signicant
benets through early familiarisa-
tion and training of their operations
supervisors.
By using the OTS in a repeated
fashion, it is envisaged that opera-
tors will quickly become familiar
with and learn how to control
daily activities in the pipeline, such
as starting a batch of a specied
product, modifying batches,
receiving a batch at the distribu-
tion centre, performing operations
such as starting and shutting
down the pipeline system, and
responding to upsets. The company
plans to complete this OTS-based
training before the actual
startup of the pipeline so that
operators are completely familiar-
ised with actual operations well in
advance.
Server
D e lta V
E n g in e e rin g S ta tio n
P S C
O P C M irro r
Server
S P S S to n e r
S im u la tio n
S tO P C
Client
K ro h n e
B M S
L D S
O P C lin k
P P _se rve r
Client
D e lta V
O p e ra to r S ta tio n
P C S
Client
I c o n ic s g ra p h ic s
d isp la y
Client
d isp la y
Client
d isp la y
D e lta V L A N O P C M irro r
S tO P C
O P C M irro r
S tO P C
P C S E n g in e e rin g S ta tio n syste m is a n O P C se rve r.
S to n e r S im u la tio n is a n O P C se rve r.
P C S D e lta V O p e ra to r S ta tio n is a n O P C c lie n t.
B M S /L D S syste m is a n O P C c lie n t.
E a c h I c o n ic s g ra p h ic in sta n c e is a n O P C c lie n t.
O P C M irro r is a D e lta V O P C c o m m u n ic a tio n to o l.
S tO P C is a S P S O P C c o m m u n ic a tio n to o l.
Figure 8 Software/OPC communications
Scenario Description Modelling preparation Process results
Speed up the OTS Batching speed is increased for Speed factor is increased and then Batches will travel faster through the pipeline
training purposes decreased
Downstream valve A valve fails closed A selected valve is closed Upstream pressure will increase until valve is xed
failure and open again (or interlock system works against
the failure)
Truck loading A RON95 truck is requesting a A valve in the model at the loading The amount of liquid requested is delivered in the
certain volume of product at racks for RON95 will open and release determined time
loading rack the amount of uid requested

Leak scenario A leak in the pipeline section A leak ow is dened in that section The material ow through the pipeline will
between PaP and Caroni will occur of pipeline decrease and the LDS will create an alarm
Filter swap A jet fuel lter is plugged Jet fuel lter E206A is plugged, causing Upstream pressure will increase and material ow
upstream pressure to increase amount will decrease until the operator switches to
E206B lter
Equipment and Densitometer, tank level or Selected equipment fails and stops PCS will not receive the selected equipment signal
instrument failure other equipment show a failure communications with PCS
RON 92 gasoline Batching of RON92 from Prepare and start a RON95 batch, A Batch of RON95 is dispatched to the Caroni
batching renery to Caroni performed including starting of auxiliary and main Centre
as per procedures document and pumps and opening of corresponding
BMS guide valves
Batching Jet A-1 to Batching of Jet A-1 from Caroni Prepare and start a jet fuel batch, A batch of jet fuel is dispatched to the airport tanks
Piarco Airport to Piarco performed as per including starting of auxiliary and main
procedures document pumps and opening of corresponding valves
Receiving RON95 Receiving a batch of RON95 at Prepare tank and valves at Caroni Centre A batch of RON95 is received, metered and directed
Caroni performed as per for receiving a batch of RON95 to the RON95 tank at the Caroni Centre, BMS
procedures document and BMS system is used to determine the time to manage
guide valves and pipeline interface
Scenarios for training operators
Table 1
kbr.indd 7 11/12/12 14:52:45
Conclusion
This was the frst integrated OTS
system for pipeline operations
developed by KBR. Close working
relationships between various
groups and support from the client
were critical to the successful
completion of the OTS. Acceptance
by the client has been high based
on early assessment. Experience
showed that such pipeline systems
can derive signifcant beneft from a
BMS and LDS as standard
equipment.
The authors wish to thank the following
for their help in developing the OTS: Ernest
Williams, Kendal Lindsay, Terrence N Ramsundar
and Vishard Birusingh of the National Gas
Company of Trinidad and Tobago, Ian Ramoutar
of Trinidad and Tobago National Petroleum
Marketing Company Limited, Reiner Jorewitz
of Krohne Oil and Gas, Travis Stripling, David
Smith, Celine Thomerson and David H. Jones
of KBR. The technical support from GL Noble
Denton is also appreciated.
Santosh Alladwar is a Senior Technical
Professional with KBR Technology in New
Delhi, India. He has seven years experience
Close working
relationships between
various groups and
support from the
client were critical
to the successful
completion of the OTS
in petrochemical, refning and olefns
technologies, and holds bachelors and masters
degrees in chemical engineering.
Cordell Chapman was the Engineering
Manager with KBR for the design of the liquid
fuels pipeline and terminal facilities, and wrote
the performance specifcation for the real-
time surveillance software and supervised its
development and testing. He holds a BS in
geological engineering from Colorado Mines
and a MS in engineering management from
University of Alaska-Anchorage.
Billy Mayo has worked for KBR for 36
years in maintenance and construction for
petrochemical, plastics and speciality chemical
facilities. For the last fve years he has worked
in the KBR Commissioning and Start-up Service
group, preparing operating and maintenance
manuals for a variety of projects.
Surajit Dasgupta is the Senior Technology
Manager for Advanced Simulation Applications
& Services in KBR and was the Technical
Manager for this OTS. He has almost 35 years
experience in advanced simulation techniques
and applications.
Javier Vazquez-Esparragoza is an Engineer in
the Advanced Chemical Engineering group at
KBR. He is a registered professional engineer
in Texas, has worked in Process Automation for
the last 17 years, and holds a PhD in chemical
engineering from Texas A&M University.
Daniel Vogt works for Krohne in Duisburg,
Germany, on leak detection systems and
supervised the development of real-time
surveillance software for this project. He
graduated from the University of Applied
Sciences in Gelsenkirchen with a Dipl.Ing. (FH)
EE and MScMT.
Keone Jackman is a Senior Engineer with
Petroleum Company of Trinidad and Tobago.
He is a Registered Electrical Engineer with more
than 12 years experience in instrumentation
and control systems, is a Senior Member of
the International Society of Automation (ISA)
and former President, former Webmaster and
Executive of the San Fernando, ISA Trinidad
Section.
Operate safely and optimize
your HF alkylation process.
Measurement made easy?
Certainly.
ABB Inc.
Anal yt i cal Measurement s
Phone: +1 418- 877- 2944
1 800 858- 3847 (Nor t h Americ a)
Email: f t ir@c a.abb.com
The FTPA2000- HP20 is a mult i- channel fibre- opt ic FT- NIR analyser especially
dedicated for HF lkylat ion nit operat ions wit h a specifically engineered field
acid sample system for HF cid ecycle purit y monitoring. The analyzer is capable
of monitoring up to eight remote sample cells. resides in a locat ion
out side t he plant bat ter y limit s and means t hat t he dangers of lengt hy sample
t ranspor t are eliminated. Learn more at www.abb.com/analyt ical
kbr.indd 8 11/12/12 14:52:56
000_zymeflow.indd 1 6/12/12 13:29:04
Refnery operations gain from advances in
digital automation
A
signifcant number of experi-
enced refnery personnel will
retire in the next decade.
With new personnel being younger
and less experienced than those
they are replacing, will refneries
continue to operate safely, reliably
and effciently? The short answer is:
Yes, by utilising the recent
advances in automation such as
smart devices and accompanying
embedded human centre designed
systems. The younger generations
are more conditioned to current
digital technology, embracing all of
the real-time, online information
and capabilities available from
these smart instruments and
systems. Smart devices replace the
ear of a seasoned feld operator or
technician. With information about
asset health and performance more
readily available, refners can
achieve safer and more reliable
operations than previously experi-
enced. Asset health information can
be used by maintenance for long-
term condition-based maintenance
on an as-needed basis, and opera-
tions can be alerted only on
short-term imminent failure or
abnormal operation.
The simplest analogy to represent
technology advances is to look at
the cellular phone. In the late 1980s,
The Brick phone was available
for about $4000. The main and only
function was to make a phone call.
Today, there is the Apple iPhone
and other similar smart phones that
are a fraction of the cost of a Brick,
but with much more functionality
beyond making phone calls, includ-
ing many real-time applications.
Many refners still have DCS
systems from the late 1980s and
Advances in digital automation can better enable the next generation of refnery
workers to operate more effciently, reliably and safely
TIM OLSEN
Emerson Process Management
1990s operating their plants. These
systems are robust and control the
process, but without additional
features (no apps). Todays
modern systems include additional
features not found in legacy
systems, such as embedded
advanced process control, statistical
monitoring, smart device monitor-
ing, asset health monitoring and
more. Like the iPhone, the console
operator is presented with more
analysed information to make better
and more informed decisions, with
time to take corrective action. If the
console operator continues to use
the modern system in exactly the
same way as the older replaced
system, the beneft of additional
functionality and information is lost
(like people using an iPhone only
for the phone call capability).
Refners challenges
The challenge of getting skilled and
experienced resources to replace
seasoned personnel who are retir-
ing is not limited just to the refning
industry. Many industries are
facing the same challenge with a
noticeable gap in the availability of
experienced staff. There are also
fewer people wanting to enter the
refning industry. This combination
has led to changes in the way refn-
ers operate to ensure reliable and
safe operations.
In the past, a single train refnery
had about 20 000 I/Os, whereas
today that same refnery design can
have 50 000 I/Os. This means there
are greater challenges for older,
well-established facilities in the US
and Europe when competing with
newer, modern assets in the Middle
and Far East. The refning industry
is faced with both competition and
opportunity from the globalisation
of fuels and petrochemicals distri-
bution across regions and countries.
For example, when demand is
down in one region, opportunities
arise to fnd consumers in other
markets where production and
distribution costs make this
economically attractive. Basically,
this means that modern refners
with lower operating costs can look
for additional opportunity markets
beyond their own regional markets.
This leads to the need to control
costs. Most people consider gaso-
line, diesel and other fuel products
as commodities, making it diffcult
to extract premium pricing and
additional margins from a differen-
tiated product. The reality is that
additional margins come from oper-
ating the refnery more effciently
and more reliably than other
refners.
While there are niche market
opportunities to extract higher
margins, overall the refning indus-
try is a commodity market where
an individual refnery has limited
control over raw material costs and
fnished product values. Refnery
net margins have historically aver-
aged below $2/bbl. Refners have a
signifcant investment in fxed
assets, which are expensive to
maintain and subject to on-going
environmental, health and safety
regulations. These realities give rise
to an industry focused on cost
reduction without jeopardising
plant reliability or safety. Top quar-
tile refners with high availability
also tend to operate more safely.
For the refnery operator, safety
and environmental compliance are
emerson.indd 1 10/12/12 13:48:14
bete.indd 1 07/01/2013 17:27
at the top of every refners list of
important concerns. Safety and envi-
ronmental programmes represent a
large, on-going cost to the refnery,
and incidents can be extremely
expensive. With hundreds of
employees, many hazardous areas,
and with the presence of explosive,
carcinogenic and poisonous materi-
als, refners are extremely interested
in systems that can provide early
warning of pending failures and
prevent unsafe or environmentally
unfriendly conditions.
Unplanned shutdowns incur costs
beyond lost production. Emergency
shutdowns and subsequent startups
put abnormal wear-and-tear
demands on equipment, increase
catalyst and chemical losses,
produce off-spec products and
require additional manpower, over-
time and expedited 24-hour service.
Just one unplanned shutdown in a
major FCC unit can cost a large
refnery over $1 million per day.
Similar costs can be incurred by
shutdowns in any of the refnerys
key process units. Based on the
Marsh Report, a majority of severe
incidents occur during startup and
shutdown (transition) and piping:
Losses in the refnery industry
have continued to increase over the
last few years and the causes high-
light the ageing facilities in this
category. A signifcant number of
larger losses (over $10 000 000) have
been caused by piping failures or
piping leaks, leading to fres and/
or explosions. Several large losses
due to piping failures were due to
corrosion issues or using the wrong
metallurgy. ...Incidents occurring
during startup or shutdown
continue to cause signifcant dollar
losses. Marsh Report - The 100
Largest Losses 1972-2001, Large
Property Damage Losses in the
Hydrocarbon-Chemical Industries.
Utilising advances from digital
automation
Operations and engineering person-
nel represent a large fxed-cost
component in a refnerys operating
budget. Any opportunity to
improve the effciency and effec-
tiveness of refnery staff can be
translated into economic benefts by
increasing the value of the activities
they perform. Timely information
that helps identify sub-optimal
process control on critical loops is
the key to augmenting the perform-
ance of operations and engineering.
Modern systems with statistical
monitoring can report on control
loops with high variability or exces-
sive alarms. What used to go
unnoticed in the past is now front
and centre to get attention.
Easy availability of process and
historical data through the continu-
ous historian, integrated with
Microsoft Offce and Internet access
tools, saves engineers and manag-
ers time and effort. Troubleshooting
operating problems can be
completed quickly, and more timely
decisions can be made that save the
company money. Internet tools
allow specialists located anywhere
in the world access to plant data
and displays, to aid in problem
solving and analysis. Key perform-
ance indicators, equipment alerts,
status information and quality data
are available in real-time for display
and alerting the appropriate person-
nel of impending problems before
they happen. These tools allow
refneries to accomplish more with
less by maximising their entire
staffs productivity.
In addition, modern systems with
simulation capabilities allow
frequent training on infrequent
events like startup, shutdown and
emergency situations. This ensures
operators are familiar with the
required actions to take should an
emergency situation occur. This
also helps familiarise new employ-
ees with operations prior to taking
on a live role in the refnery.
Utilising modern digital automa-
tion can address the challenges
and opportunities for cost reduc-
tion. For example, feld instruments
that are equipped with smart diag-
nostics can alert key maintenance,
operations or safety personnel
about asset problems in real-time
before they fail or cause abnormal
operation. The key beneft is time
to act on the information since the
asset health information detects
problems long before the process
sees the effects. The diagnostics
can identify the root cause of the
troubled asset, so maintenance can
effectively correct the problem. In
addition, this same root cause
predictive diagnostic capability
can be utilised to identify and
plan for a smarter turnaround by
repairing only the assets requiring
work.
Effcient use of maintenance
resources, both for on-going main-
tenance and turnarounds, is an area
that can often be improved through
the use of modern technologies.
Process and control equipment
deteriorates over time and eventu-
ally needs attention. Heat
exchangers, compressors and
turbines foul over time, becoming
less effcient and increasing produc-
tion costs. Without maintenance,
instruments tend to drift, control
valves wear and process dynamics
change, decreasing the overall effec-
tiveness of the control system. The
question of when to pull a piece of
equipment for repair is a complex
issue that should be based on its
performance history, the operation
performance cost of degradation,
the cost of repair and the risk and
consequence of a failure. In a cost-
cutting environment, the appeal of
delaying maintenance activities
must be weighed against the
increased risk of unplanned
shutdowns and the cost of operat-
ing below peak equipment
performance.
Maintenance costs can be reduced
by improving the effciency of the
maintenance staff, reducing the
frequency of unplanned shutdowns
and reducing unnecessary mainte-
nance work. Studies have shown
that more than 60% of a typical
instrument technicians trips to the
feld result in either no action or a
minor confguration adjustment that
Utilising modern
digital automation
can address the
challenges and
opportunities for
cost reduction
emerson.indd 2 10/12/12 13:48:30
could be done from their offce with
smart instruments (see Figure 1).
Refner feedback estimated a
manpower saving of 1.6 hours per
instrument, per year, by automating
instrumentation work practices and
reducing trips to the feld. For a
large refnery with over 10 000 proc-
ess inputs, a digital solution could
save 16 000 hours per year, for
instrument and valve maintenance.
This provides the opportunity to
create a predictive maintenance
culture that acts on alerts and root
cause information in real-time,
rather than reactive. At the 2010
NPRA Reliability and Maintenance
Conference, a Gulf Coast refner
indicated that reactive maintenance
cost their facility 50% more than
planned maintenance. Their focus
was to reduce the number of inci-
dences of reactive maintenance.
Embedded APC
In the 1990s, many refners took
advantage of advanced process
control (APC). However, APC
tended to be expensive and complex
to set up, understand and maintain.
The end result was many refners
turned off APC within two years of
implementation.
Advanced control is still a desired
functionality for refners because it
helps address energy cost savings
by improving the effciency of fred
heaters and boilers (supply) and
distillations columns (demand).
While turned on, it also has the
added beneft of consistent, high-
performance operation across
operating shifts, no matter who the
console operator is.
Reducing process variability
directly impacts a refnerys bottom
line in terms of energy effciency,
capacity and quality. Modern
digital automation can automati-
cally identify high variability in key
control loops to ensure visibility to
both operators and operations engi-
neers. Properly tuned and
performing control loops can extend
catalyst life and improve yield and
quality, which directly beneft the
bottom-line revenues.
APC applications reduce energy
costs by further stabilising the proc-
ess and pushing closer to actual
equipment constraints and quality
specifcations. Modern, embedded
APC applications are built with
these standard tools that allow fast
execution so distillation columns
can be operated at minimum refux,
compressor recycles lowered, heater
excess air reduced, catalyst losses
minimised, heat recovery maxim-
ised and so on, all impacting the
total energy consumed by the refn-
ery. Using APC technology, refners
can be assured that the control
system only utilises the right
amount of energy needed to
economically recover the most valu-
able products.
Manual checked measurements
As was mentioned early in this arti-
cle, older refnery designs had a
limited I/O count, only that which
was required to operate the refnery
safely. However, many process and
asset health measurements that are
not online are captured through
manual measurements on a daily,
weekly, monthly or longer period
of time basis, depending on the
expected mean time between fail-
ures. Only those assets deemed
critical enough to have online moni-
toring provided continuous
information to operations and/or
maintenance.
Manual measurements do not
always prevent asset failure and
they can also create a potential
safety hazard. Some examples
where additional danger is intro-
duced are:
Manually dipping a tank to
confrm the level
Collecting pump data from a
pump within an alkylation unit
Taking temperature measure-
ments on the crude unit preheat
exchanger train
Climbing ladders while holding
onto a clipboard.
On the example of the crude unit
heat exchangers, many bundles
have empty thermowells in between
tube bundles. For someone to
manually check the temperature,
they would need to get on a ladder
and take the measurement around
hot exchangers. Many people also
use a heat gun, but this can create
errors in the data collected, poten-
tially leading to incorrect decisions,
such as to pull and clean for the
upcoming turnaround.
One recent solution has been the
use of wireless to capture missing
process and asset health measure-
ments. On average, a refner has
hundreds of process pumps with
only about 10% monitoring online.
The remaining pumps are checked
manually, typically once per month.
This has been found to be insuff-
cient in preventing pump failures
and the potential safety and envi-
ronmental incident associated with
unscheduled shutdowns. For exam-
ple, one refner sends a technician
out each morning to look for prob-
lems; a better approach would be to
have the information available in the
morning that identifed the pumps
requiring attention and then to send
maintenance to service them.
Wireless asset health measurements
enable planned maintenance that
results in higher reliability.
Online vibration monitoring
determines whether a pump or
motor needs maintenance. These
3 0
4 0
3 5
3 5 %
2 0 %
6 % 6 %
2 5
2 0
1 5
1 0
5
R o u tin e
c h e c k
N o
p ro b le m s
C a lib ra tio n
sh ift
Z e ro o ff P lu g g e d
lin e s
F a ile d
in stru m e n t
P
r
o
p
o
r
t
i
o
n
,

%
0
2 8 %
4 %
Figure 1 Advanced process control optimises distillation columns, enabling signifcant
energy savings
emerson.indd 3 10/12/12 13:48:40
Conclusion
Advances in digital automation can
augment the next generation of
refnery staff to operate more eff-
ciently, reliably and safely. The
younger generations are more condi-
tioned to current digital technology,
thus embracing all of the real-time,
online information and capabilities
from smart instruments and modern
systems. Additional measurements,
and the beneft of time to act on the
early detection of wear or abnormal
operation, are the keys to ensuring
refneries can operate even better
than they did in the past.
Tim Olsen is a Refning Consultant within the
PlantWeb global refning industry solutions
group of Emerson Process Management, where
he supports Emersons technical and business
strategy. He is currently the Division Chair for
the AIChE Fuels & Petrochemicals Division,
holds a BS in chemical engineering from Iowa
State University with emphasis in industrial
engineering and process control, and an MBA
from the University of Iowa.
Email: Tim.Olsen@Emerson.com
tools have demonstrated how
modern maintenance tools and
methodologies can be used in a
predictive manner to save both time
and money during a turnaround
without sacrifcing reliability or
safety.
Online monitoring tools and auto-
matic alerts can help prevent
unplanned shutdowns and their
associated costs (see Figure 2).
Certainly, not all of the causes of
unplanned shutdowns can be
addressed by process control and
automation systems, but faulty
measurements, malfunctioning
instruments, operator error and
equipment failure are some of the
common problems that smart feld
devices address. The online moni-
toring tools provide the opportunity
for a refner to create a more predic-
tive maintenance culture that acts
on information in real-time to
prevent shutdowns or safety inci-
dents, rather than the traditional
reactive or preventive maintenance
culture.
Figure 2 Online monitoring tools help prevent unplanned shutdowns and reduce
time-consuming operator rounds
Det erminat ion of Sulf ur and ot her element s
at -line and in t he laborat ory
El ement al Anal ysi s
of Fuel s
To keep pace w i t h t he demandi ng qual i t y requi rement s of
modern f uel s, advanced, preci se and easy t o use anal yt i cal
t echnol ogy i s requi red. Wi t h a compl et e range of XRF and
ICP spect romet ers, SPECTROs uni que sol ut i ons f or at -l i ne
and l aborat or y el ement al anal ysi s are capabl e of meet i ng
t he most demandi ng product speci f i cat i on t est i ng
requi rement s.
Measure your fuels
- at sub 10-15 ppm levels of sulfur t o ensure federal
government agency compliance
- at t race ppm levels for met al element s such as Cu, Ca, Mg,
Na, K, and P, t o ensure low engine emissions as well as
t rouble-free mot oring
- wit h reliabilit y, speed and accuracy for any part icular applicat ion
- manually or fully aut omat ically
Discover more exciting details at
w w w.spect ro.com/f uels
w w w.di gi t al ref i ni ng.com/ 105.
Visi t us at
Pi t t con 2013,
17-21 March,
Phi l adel phi a, PA,
USA
emerson.indd 4 10/12/12 13:48:52
cw.indd 1 11/12/12 17:10:49
Controlling flm temperature in fred
heaters
D
irect-fred heaters have been
widely used in the oil refn-
ing and chemical process
industries to heat the crude oil
contained in tubular coils by the
combustion of fuel within an inter-
nally insulated enclosure. A
successful fred heater design relies
on many factors. Film temperature
control is one of the key factors that
play a crucial role in fred heater
design, particularly for units
processing heavy feedstocks that
are thermally unstable, such
as Canadian oil sands-based
feedstocks.
Film temperature determines the
susceptibility of a process fuid
towards coking. Bulk oil tempera-
ture plus a temperature rise across
the oil flm sets the flm tempera-
ture. In most applications, it is the
oil flm temperature, not the bulk
oil temperature, that limits the
heater duty and the oil life.
1
Film
temperature is an important factor
in fred heater design for many
reasons. Firstly, oil degradation
starts in the fuid flm, since this is
the hottest place for the bulk oil.
Fluid life is shortened because of
degradation, which can lead to a
costly result. Secondly, if the flm
temperature exceeds the limitation,
the stationary fuid flm on the
inside tube surfaces is subject to
thermal decomposition, which
results in coke deposition at that
location. Coke deposits increase
resistance to heat transfer and raise
the tube metals temperature. Once
the tube wall temperature reaches
the design temperature, the heater
must be shut down for decoking
to avoid coil damage. Thirdly,
overheating of the fuid flm
Film temperature control is critical to the sucessful design of fred heaters,
especially for heaters employed in upgrading heavy feedstocks
JINYU JIAO, YURIY MORAYKO, MORTEN THEILGAARD and MICHAEL HO
WorleyParsons Canada
accelerates the fouling rate. Fouling
requires more heat input and a
hotter tube metal temperature to
maintain the same heater outlet
temperature. These factors cause
heaters to shut down much more
frequently and eventually reduce
the whole plants proftability.
Due to the importance of the flm
temperature, its control has become
a hot topic for fred heater designs,
especially for crude heaters,
vacuum heaters and coker heaters.
In this article, some feasible meth-
ods of controlling the flm
temperature for fred heater design
are presented. Methods including
steam injection, reducing tube size,
using double fring and lowering
the average heat fux are discussed,
and some methods are presented
with accompanying examples.
Steam injection
Steam injection is one of the best
options for lowering the flm
temperature, as long as there are
no unintended consequences
downstream of the units. Steam
reduces the oil residence time by
increasing the fuid velocity. High
fuid velocity improves heat trans-
fer in the flm layer, which lowers
the differential temperature
between the tube wall and the bulk
fuid. Figure 1 shows the effect of
steam injection on flm tempera-
ture for one crude heater. The
horizontal axis represents the radi-
ant coil growth from inlet to outlet.
It can be seen that the flm temper-
ature drops around 20C after
injecting 1 wt% of steam into the
fuid. The results shown
in Figure 1 arise from the simula-
tion of a crude heater at a design
duty of 63 MW with diluted bitu-
men as the process feed. The
simulation was performed with a
commercial fred heater rating
program, FRNC-5PC, which has
been widely used and has been
proven to be reliable in predicting
heater performance.
3 6 0
3 8 0
3 4 0
3 2 0
3 0 0
2 8 0
2 6 0
F
i
l
m

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

C
From i nl et t o out l et
2 4 0
With o u t ste a m in je c tio n
With ste a m in je c tio n
Figure 1 Effect of steam injection on flm temperatures
worley.indd 1 11/12/12 17:16:39
gains radiant heat from the refrac-
tory. The portion of the tube facing
the burners has a higher local heat
fux, while the side facing the
refractory is much lower. For a
given fred heater with nominal two
diameter tube spacing and a very
uniform longitudinal heat fux
distribution, the local peak heat fux
(q
m
) is approximately 1.8 times the
average heat fux (q
a
) for single-
fred heating. In contrast, the
double-fred heater has radiant heat
on both sides of the tubes, which
greatly reduces the peak fux to
about 1.2 times the average heat
fux.
3
The correlations mentioned
above for single and double fred
can be simply represented in the
following equation:
q
m
= Xq
a
(1)
where X represents the time factor,
which is approximately equal to 1.8
and 1.2 for single and double fred,
respectively.
The local flm temperature can be
calculated by the following
equations:
4

T
f
= T
b
+ T
f
(2)

T
f
=
q
m

(
D
o

)
(3)
K
f
D
i
where T
f
and T
b
are flm tempera-
ture and oil bulk temperature,
respectively. T
f
is the flm temper-
ature rise and K
f
is a flm heat
transfer coeffcient.
From the equations, it can be seen
that it is the localised heat fux, not
the average heat fux, that directly
governs the flm temperature. For a
heater with a given average heat
fux, a double-fred heater has a
lower localised heat fux distribution
than a single-fred heater. A lower
localised heat fux reduces the flm
temperature at that location. Figure
3 shows a comparison of the flm
temperature between single and
double fring for the same heater
with the same average heat fux. The
results are from the simulation of a
vacuum heater with a design duty
of 50 MW. It can be seen that using
double fred can greatly reduce the
flm temperature of the radiant coil
for the heater.
Selecting the correct location and
amount of steam injection is critical.
It must be injected upstream of the
heater tubes with the highest flm
temperature, yet far enough down-
stream in the radiant section to
minimise incremental pressure drop
to ensure charge pump capacity is
not exceeded.
2

Steam injection can also change
the fow regime for two-phase fow.
The problem associated with slug
fow can be mitigated by steam
injection. However, it is not the
intention of this article to discuss
the effect of steam injection on the
fow regime because it is not related
to the topic. Caution is advised in
selecting a suitable steam condition
to match the process fuid condi-
tion, to make sure that no
condensation occurs after steam
injection into the process fuid.
Reducing tube size
Depending on the allowable pres-
sure drop, fred heater coils are
usually divided into multiple passes
to accommodate the total fow and
to meet pressure drop require-
ments. For a given fow rate and
the number of fow passes, oil mass
velocity increases with reducing
tube size. A higher oil mass veloc-
ity reduces the oil residence time
and increases the flm heat transfer
coeffcient, therefore lowering the
flm temperature. Figure 2 shows
the effect of tube size on flm
temperatures. The horizontal axis
represents the heater coil growth
from radiant inlet to radiant outlet.
The results come from the simula-
tion of a small heater with a design
duty of 15 MW. The heater is a
vertical cylindrical type with four-
pass fow. To demonstrate the effect
of tube size on the flm tempera-
ture, the radiant tube size is
changed from 6.625in to 4.5in, with
no change in the size of the convec-
tion tubes. More radiant tubes have
been added to the radiant section to
compensate for surface loss caused
by the reduction in tube size, to
keep the same heat fux. Case 1 is
for the heater with all tubes sized at
6.625in, while case 2 is for the same
heater with tube sizes changed
from 6.625in 4.5in, for the radiant
coils only. It can be seen that the
flm temperature in the radiant
section has been decreased because
of the reduction in tube size.
Reducing the tube size increases
the heater pressure drop, which
requires a much higher pump head
upstream of the heater. Since flm
temperature control is basically
intended to control the peak flm
temperature, it is advisable to
reduce tube sizes for tubes with
peak flm temperature only, to
minimise the increase in pressure
drop caused by reducing tube sizes.
Double fred vs single fred
A fred heater can be single or
double fred. The heat fux on the
tubes circumferential surface is not
uniform because of the shading of
radiant heat. The single-fred heater
receives radiant heat on one side of
the process tubes (directly from the
burner fame), while the other side
of the tubes, facing the heater wall,
4 1 0
4 2 0
4 0 0
3 9 0
3 8 0
3 7 0
3 6 0
F
i
l
m

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

C
3 5 0
4 . 5 in tu b e s
6 . 6 2 5 in tu b e s
Figure 2 Effect of tube size on flm temperatures
worley.indd 2 11/12/12 17:16:50
Lowering the average heat fux
The frst step in designing a fred
heater is to set up the allowable
average radiant heat fux. For a
given heater, either single or double
fred, it is helpful to control the flm
temperature by lowering the aver-
age heat fux. From Equation 1, the
localised heat fux reduces with
lower average heat fux, no matter
whether it is single or double fred.
Oil flm temperature depends on
the heat fux and oil mass velocity.
Decreasing the heat fux reduces
the oil flm temperature at a fxed
mass velocity.
5

The average radiant section heat
fux is defned as the total radiant
section absorbed duty divided by
the total radiant section tube surface
area. For a given radiant duty of a
fred heater, the only way to lower
the average heat fux is to increase
the radiant sections surface area. It
may be reasonable to assume a rela-
tively low average heat fux to
design a fred heater with very tight
flm temperature control. However,
it is also noted that oil residence
time increases as the surface area
increases, which may partially
counteract the beneft of decreasing
the flm temperature by lowering
the average heat fux. A lower aver-
age heat fux means more capital
cost in the heater coils, which is
another drawback in controlling the
flm temperature by reducing the
average heat fux. Thus, the effect
on heater design of lowering the
average heat fux should be care-
fully evaluated before a decision
is made.
Other design considerations
There are other design considera-
tions that should not be neglected
when designing a fred heater with
better flm temperature control. The
radiant section heat fux at any
point in the heater is controlled by
the temperature difference between
the hot fue gas and oil in the tube.
The heat transfer rate increases with
the temperature difference between
the hot fue gas and the cold oil.
6
In
a vertical up-fred heater, it is not
rare to see that heat fux is low at
the foor and gradually increases
along the length of the fame. It
is highest at the point where
maximum combustion takes place
in the fame, then reduces at the top
of the fre box. Thus, real fred heat-
ers may encounter more or less
signifcant heat fux imbalances.
This heat fux imbalance can cause
high flm temperatures and high
rates of fouling formation. Efforts
should be made to optimise the
design parameters to minimise the
heat fux imbalance. These design
parameters include radiant section
height to width ratio, burner to
tube distance, number of burners,
fame shape and dimensions, and
radiant section tube layout.
7
Flame impingement can cause
extremely high localised heat fux,
which results in a higher flm
temperature and rapid coke forma-
tion. Flame impingement occurs
when a fame actually touches or
engulfs the tubes. Vinayagam has
discussed the causes of fame
impingement for a fred heater.
8

Some precautions need to be
considered in the design of fred
heaters to prevent fame impinge-
ment occurring; for instance, an
adequate fre box to contain the
fame, more and equally spaced
burners, the correct type of burners,
and improved distribution of
combustion air fow.
Summary
Film temperature control is critical
to the successful design of fred
heaters, especially for those heaters
employed in upgrading heavy feed-
stocks. This article has discussed
several ways to control the flm
temperature in the design of fred
heaters. These methods have been
discussed in detail and have proved
to be effective means of controlling
flm temperature.
References
1 Pelini R G, Heat fux and flm temperature
in fred thermal-fuid heaters, Chemical
Engineering, Dec 2008.
2 Hanson D, Martin M, Low capital revamp
increases vacuum gas oil yield, March, Oil &
Gas Journal, 2002.
3 Romero S, Delayed coker fred heater
design and operations, Rio Oil & Gas Expo and
Conference, 2010.
4 API 530, Calculation of heater-tube thickness
in petroleum refneries, 6th ed, Sept 2008.
5 Golden S W, Barletta T, Designing vacuum
units, Separations, Apr 2006.
6 Martin G R, Heat-fux imbalance in fred
heaters cause operating problems, Hydrocarbon
Processing, May 1998.
7 Nogay R, Prasad A, Better design method
for fred heaters, Hydrocarbon Processing, Nov
1985.
8 Vinayagam K, Minimizing fame
impingements in fred heaters, Chemical
Engineering, May 2007.
Jinyu Jiao is a Mechanical Engineer with
WorleyParsons in Calgary, Canada. He holds an
MS degree in mechanical engineering as well as
a PhD in chemical engineering.
Yuriy Morayko is a Mechanical Engineer with
WorleyParsons in Calgary, Canada. He holds a
BSc degree in mechanical engineering.
Morten Theilgaard is a Mechanical Department
Manager for WorleyParsons in Calgary, Canada.
He holds a BSc in mechanical engineering.
Michael Ho has eight years experience in the
oil and gas industry in an EPCM environment.
He holds masters and bachelors degrees in
mechanical engineering.
3 2 0
3 4 0
3 6 0
3 8 0
3 0 0
2 8 0
2 6 0
2 4 0
2 2 0
F
i
l
m

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

C
2 0 0
S in g le fire d
D o u b le fire d
Figure 3 Comparison of the flm temperature between single and double fring
worley.indd 3 11/12/12 17:17:21
*ad copy 3.indt 1 10/12/12 16:40:47
High-capacity tray for debottlenecking a
crude distillation unit
T
he main fractionator of the
crude distillation unit (CDU-1)
in the Whangarei refnery of
The New Zealand Refning
Company (NZRC) was retroftted
with high-capacity internals to
increase the unit throughput from
8500 t/d to 13 000 t/d. Ultra-high-
capacity Shell ConSep* trays were
applied in the most capacity-
constrained HGO pumparound
(mid circulating refux) section of
the column, as no other frst-
generation, high-capacity tray was
found adequate to debottleneck this
section. By the application of these
trays, capex savings of the order of
$5.5-6 million were achieved
Use of ultra-high-capacity trays in the most constrained section of a CDU enabled
over a 50% increase in throughput at less cost than other debottlenecking options
KAUSHIK MAJUMDER Shell Projects & Technology GIUSEPPE MOSCA Sulzer Chemtech
KENT MAHON Refning NZ
compared to other conventional
debottlenecking options. This was
the frst application of ConSep trays
in a CDU main fractionator, and the
post-revamp test run established
realisation of the expected
performance.
Whangarei refnery targeted
expanding its refning capacity
through the Point Forward Project.
1

The project involved increasing the
throughput of the CDU-1 from 8500
t/d to 13 000 t/d, thereby increas-
ing the distillate component to
downstream processing and gener-
ating additional long residue to
replace imported long residue for
loading the vacuum distiller. Figure
1 shows a simplifed process fow
diagram of CDU-1.
Shell Global Solutions International
(SGSi) carried out the feasibility
study for the expansion of CDU-1.
Several options were studied to
debottleneck the main fractionator:
Replacement of the existing
column internals with high-capacity
internals including the ConSep tray
for the most capacity-constrained
HGO pumparound section
Installation of a new crude pre-
fractionator column to separate off
light naphtha and reduce the load
to the main fractionator. The capex
for this option was found to be $6
million higher than for option 1
Crude
desalter
Furnace-2
Furnace-1
Preheat
train-1
Preheat
train-2
Preheat
train-3
C ru d e
c h a rg e
Preflash
vessel
Main
fractionator
(C-150)
U n sta b ilise d
n a p h th a
K e ro se n e
H G O
L G O
L o n g re sid u e
S trip p in g
ste a m
S trip p in g
ste a m
S trip p in g
ste a m
Figure 1 Simplifed process fow diagram of CDU-1
sulzer.indd 1 10/12/12 13:57:21
Installation of a new heavy end
column to recover the heavy gas oil
(HGO) dropped into the long resi-
due to ofoad the main fractionator.
The capex for this option was found
to be $5.5 million higher than for
option 1.
On the basis of a comparison of
the revamp options, NZRC decided
to proceed with the ConSep tray
alternative owing to this options
lowest capex and most favourable
economics.
Shell ConSep tray technology
The ConSep tray utilises the princi-
ple of de-entrainment by centrifugal
forces to remove the gravitational
limitation of jet ood. Separation of
the entrained liquid before entering
the next tray allows very high
vapour velocities to be achieved in
the column. The tray combines the
features of a contacting deck and a
separator deck in a single tray. The
basic features of the tray are shown
in Figure 2. The functioning of the
contacting deck, which in fact is a
normal tray, is limited by three
hydraulic mechanisms: jet ooding,
downcomer choking and down-
comer backup. The use of a
separator deck inuences all three
mechanisms:
2
The jet ooding limit is extended
as the entrained liquid is efciently
separated from the vapour to
prevent carry-over of liquid to the
tray above
The liquid entering the main
downcomer is largely coming from
the separator deck, where it is well
degassed. As a result, the down-
comer liquid handling capacity is
substantially increased
To eliminate downcomer backup
limitation, the separator deck is
designed with low-pressure drop
swirl tubes. The contacting deck is
also designed with a relatively high
open area.
Figure 3 shows the expected capac-
ity gain of the ConSep tray over
conventional trays and packing.
3
The ow parameter () is dened
as:
=
L
P
v
V P
L

where
L
v
represents the liquid-to-
vapour mass ow ratio and
Pv
P
L

Swirl
tubes
Contacting
deck
Hi-fi
downcomer
Swirl deck
downcomer
Separator
deck
Figure 2 Schematic of ConSep tray
1 . 2
2 . 0
1 . 8
1 . 6
1 . 4
1 . 0
0 . 8
0 . 6
0 . 4
0 . 2
0 . 0 1 0 . 1 0 1 . 0 0
R
e
l
a
t
i
v
e

c
a
p
a
c
i
t
y
Fl ow paramet er, ()
0 . 0
C o n S e p tra y
S tru c tu ra l p a c k a g in g
C o n v. d o wn c o m e r tra y
H i-fi c a lm in g se c tio n tra y
Figure 3 Comparison of column internal capacities
Location Diameter Application Installed Capacity increase, %
UK 1.9 NGL debutaniser 1995 22
1
Australia 1.9 FCCU debutaniser 1996 30
2
Germany 2.2 HCU main fractionator 1999 50
Australia 1.7 NGL debutaniser 1999 15
1
Singapore 2.5 FCCU debutaniser 2000 20
1
Japan 2.1 FCC debutaniser 2006 10
Sweden 2.0 FCC debutaniser 2006 20
Sweden 1.0 C
3
/C
4
splitter 2007 50
USA 2.3 C
3
/C
4
splitter 2006 12
Canada 1.0 Depropaniser 2007 20
New Zealand 4.6 Crude distillation 2009 22
2
Singapore 1.8 PO drying column 2011 20
2
China 3.2 Ethylene fractionator 2011 Target 25%
3
Japan 2.5 FCC debutaniser 2013 (scheduled) Target 15%
3
1 Limited by reboiler capacity 2 Limited by feed to column 3 No data yet
Overview of ConSep tray applications
Table 1
sulzer.indd 2 11/12/12 18:05:10
represents the ratio of vapour-to-
liquid density.
Typically, the ConSep tray is capa-
ble of offering a 40-50% capacity
advantage over a wide range of frst-
generation, high-capacity trays. In
most revamps, the column retroft-
ted with the ConSep tray becomes
limited by other factors such as
availability of feed and/or
constraints on auxiliary equipment,
including reboiler, condenser,
pumps and so on, even before the
full potential of the ConSep tray is
realised. Table 1 shows some appli-
cations of this tray along with
benefts achieved and constraints
faced.
Modifcations of main fractionator
(C-150)
During the feasibility study, the
HGO pumparound section of the
column was found to be severely
limiting for the targeted throughput
of 13 000 t/d. This section was
already ftted with Shell Calming
Section (CS)* trays. Since the frst
generation of high-capacity inter-
nals was found inadequate to
debottleneck this section, ultra-
high-capacity ConSep trays were
selected. The trays were designed
to achieve 33% more capacity
compared to the CS tray.
The HGO pumparound section
consisted of three contacting trays
with a tray spacing of 500mm. A
one-for-one tray replacement with
ConSep trays was selected. Figure 4
shows a schematic drawing of the
HGO pumparound section ftted
with these trays. For the remaining
sections of the column, the follow-
ing internals were suggested:
Stripping section: Shell HiFi*
trays
Wash section: MellapakPlus
252Y** packing
All other sections: Shell CS trays.
As this was the frst application
of ConSep trays in this service, a
detailed study was carried out to
address the risks associated with
this application and the mitigations
were applied in the design. The
trays were manufactured by Sulzer
Chemtech. To ensure proper
performance of the trays in a rela-
tively new application, rigorous
quality control steps were followed
at the manufacturing site and a
detailed mock-up assembly of tray
components was carried out at the
refnery site prior to installation in
the column.
Post-revamp performance
From conception to implementa-
tion, the project took four years,
and the revamp was implemented
during a normal planned shutdown
in October 2009 without affecting
the units availability. The
performance test run of CDU-1 was
conducted in September 2010.
Crude throughput was maintained
at 13 000 t/d, although the crude
blend used during the test run was
marginally heavier than the one
considered for the revamp design.
H G O +
in te r re flu x
Tray 14
Tray 13
Tray 12
I n te r
re flu x
L G O
H G O
va p o u r
Figure 4 Schematic of HGO pumparound section of C-150
5 0 0
8 0 0
7 0 0
6 0 0
4 0 0
3 0 0
2 0 0
1 0 0
0
0 2 0 4 0 6 0 8 0 1 0 0
T
B
P

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

C
Wei ght , %
1 0 0
D e sig n b le n d
Te st ru n b le n d
Figure 5 Comparison of design and test run crude TBP
sulzer.indd 3 11/12/12 18:05:29
run conditions to evaluate hydrau-
lic loading of the ConSep trays. In
Table 3, the key performance
indicators for these trays operating
under test run conditions are
compared with the design condi-
tions. During the test run, the trays
were operating 10-15% lower than
design capacity, even at a design
intake of 13 000 t/d, due to:
Heavier feed than in the design
case
Less preheat recovery of the order
of 10-12C due to limitations in the
crude preheat train.
Based on a review of operating
experience since startup and the
Comparison of the design crude
blend and crude blend used during
the test run is shown in Figure 5.
Key design conditions and test run
operating conditions of C-150 are
shown in Table 2. The product qual-
ity is compared in Figure 6 (a-e).
C-150 was simulated for the test
Parameters Design Test run Parameters Design Test run
Crude intake, T/D 13 000 13 077 Flash Zone press, barg 1.90 2.24
Naphtha, T/D 4405.0 3516.4 Feed temp, C 361.5 345.0
Kerosene, T/D 1498.8 2169.3 Top temp, C 180.8 175.7
Light gas oil, T/D 1517.0 1116.1 Kero draw temp, C 216.5 223.4
Heavy gas oil, T/D 2014.5 1731.9 LGO draw temp, C 244.2 250.1
Long residue, T/D 3573.3 4271.1 HGO draw temp, C 280.7 278.3
Strip steam, T/D 106.9 149.5 Flash zone temp, C 343.9 327.3
Top press, barg 1.65 1.95 Bottom temp, C 333.3 322.061

Key design and test run operating conditions of C-150
Table 2
3 0 0
4 0 0
2 0 0
6 0 0
7 0 0
5 0 0
1 0 0
0 2 0 4 0 6 0 8 0 1 0 0
T
e
m
p
e
r
a
t
u
r
e
,

C
Wt , %
0
Long resi due ASTM D86
1 5 0
4 0 0
1 0 0
5 0
3 0 0
3 5 0
2 5 0
2 0 0
0 2 0 4 0 6 0 8 0 1 0 0
T
e
m
p
e
r
a
t
u
r
e
,

C
LV, %
0
Heavy gas oi l ASTM D86
1 5 0
2 0 0
1 0 0
3 0 0
3 5 0
2 5 0
5 0
0 2 0 4 0 6 0 8 0 1 0 0
T
e
m
p
e
r
a
t
u
r
e
,

C
LV, %
0
Li ght gas oi l ASTM D86
D e sig n b le n d
Te st ru n b le n d
1 5 0
2 0 0
1 0 0
5 0
0 2 0 4 0 6 0 8 0 1 0 0
T
e
m
p
e
r
a
t
u
r
e
,

C
LV, %
0
Napht ha ASTM D86
3 0 0
2 0 0
1 0 0
2 5 0
1 5 0
5 0
0 2 0 4 0 6 0 8 0 1 0 0
T
e
m
p
e
r
a
t
u
r
e
,

C
LV, %
0
Kerosene ASTM D86
Figure 6(a-e) Comparison of design and test run product quality
sulzer.indd 4 10/12/12 13:57:54
test runs results, it could be
concluded that the revamp targets
for the CDU-1 main fractionator
(C-150) were achieved. No hydrau-
lic constraint was experienced in
achieving the design intake of
13 000 t/d and the required prod-
uct quality was achieved.
Conclusions
The performance of Shell ConSep
trays in the HGO pumparound
section of the CDU-1 main fractiona-
tor met the target of capacity
enhancement without any drawback
compared to the pre-revamp condi-
tions. During the test run, the trays
were operating at 10-15% lower than
the design capacity even at the
design intake of 13 000 t/d due to
heavier crude feed and lower feed
temperature. However, the built-in
capacity margin enabled stable oper-
ation for the trays at much above
the capacity limit of the frst genera-
tion of high-capacity trays.
The options to debottleneck
columns already equipped with the
frst generation of high-capacity
trays are limited. ConSep trays
provide an attractive solution for
such cases. In this revamp project,
use of only three of these trays in
the most capacity-constrained
section of the column made it possi-
ble to retroft the existing column
and made the capex option more
attractive over the other debottle-
necking options.
* Shell ConSep, Shell CS and Shell HiFi are Shell
trademarks. ** Mellapak Plus 252Y is a Sulzer
Chemtech trademark.
References
1 Refnery expansion means NZ more self
reliant, media release by NZRC, 16 July 2010.
2 Wilkinson P M, De Villiers W E, Mosca
G, Tonon L, Achieve challenging targets in
propylene yield using ultra system fractionation
trays, ERTC 2006.
3 De Villiers W E, Bravo J L, Wilkinson P M,
Summers D R, Further advances in light
hydrocarbon fractionation, PTQ Q3 2004.
Kaushik Majumder is Distillation Team Lead of
Shell Projects & Technology in Bangalore, India.
He holds a bachelors degree from Jadavpur
University, India, and a masters and doctorate
from Indian Institute of Technology, Delhi.
Email: Kaushik.Majumder@shell.com
Giuseppe Mosca is the Global Refnery
Technology Manager of Sulzer Chemtech.
He holds BS and MS degrees in chemical
engineering from the University La Sapienza
Rome, Italy.
Email: Giuseppe.Mosca@sulzer.com
Kent Mahon is a Process Engineer at Refning
NZ. He was the Senior Process Engineer and
Commissioning Process Engineer during the
Point Forward Project.
Email: Kent.Mahon@refningnz.com
Parameters Design Test run
Froth backup/CS height, % 68 60
Tray pressure drop, mbar 12.3 9.2
Tube food , % 73 60
Flow parameter 0.17 0.19
Overall column load factor, m/s 0.12 0.10
Flooding (CS tray), % 133 112
Key performance indicators for ConSep trays
Table 3
sulzer.indd 5 11/12/12 18:05:58
va|ue acc|ng cata|ysts, abscrbents, acc|t|ves anc
prccess technc|cgy fcr c|| refn|ng prccesses.
www.jmcata|ysts.com|rehner|es
UK
Te| +44 (0)1642 553601
USA
Te| +1 630 268 6300
Te| +1 732 223 4644
Naphtha HDS
D|ese| HDS
Hydrctreating cata|ysts
vacuum tcwer
Hydrcgen cata|ysts
FCC un|t
FCC additives
SOx NOx remcva|
l|ght c|efn prccuct|cn
Bcttcms ccnvers|cn
Veta|s traps
/ct|v|ty bccsters
CO cx|cat|cn
6as | |iquid
purifcaticn
S, Hg, C| remcva|
Cruce un|t
vCO HDS
Hycrcgen p|ant
Prccess
diagncstics
Untitled-1 1 31/07/2012 17:18
j matthey.indd 1 6/12/12 13:25:56
Steam reformers for hydrogen and
synthesis gas
T
he popular press and scien-
tic literature has been rife
with references to the discov-
ery of vast quantities of shale gas in
North America. This resource has
become economically recoverable
due to advances in horizontal drill-
ing and well fracturing or fracking.
The advent of shale gas has been
referred to as a game-changer,
which has the potential to make
North America energy independent
and the rening and petrochemical
sectors more cost competitive.
Many renery operations and
petrochemical processes require
large quantities of hydrogen (H
2
)
and synthesis gas (carbon monox-
ide [CO] and H
2
, also known as
syngas). The dominant method of
producing H
2
and syngas is steam
reforming of light and medium
hydrocarbons. This technology will
become even more popular in the
future with the advent of relatively
cheap natural gas.
Shale gas
The historic and future supply,
demand and pricing for US natural
gas in a recent Department of
Energy study is shown in Figure 1.
This graph shows total natural gas
production in the US, where shale
gas production will increase from 5
trillion cu ft in 2010 to 13.6 trillion
cu ft in 2035, representing an annual
increase of 4.1% per annum. This
growth in natural gas production
will make the US a net exporter in
the early part of next decade. The
price of natural gas is expected to
increase to ~$7.4/MMbtu (2010
dollars) in 2035 based on supply,
demand and incremental cost of
production. It is expected that natu-
Modern steam methane reformer design is a complex process that needs to be
carefully optimised in terms of capital and operating costs
NORM PELLETIER and GOUTAM SHAHANI
Selas Fluid Processing Corporation, a division of Linde Engineering
ral gas will be relatively cheap
compared to other fuels on an
energy equivalent basis. Based on
these projections, it is expected that
the dominant method of producing
H
2
and syngas is set to remain
steam methane reforming.
Petrochemical and rening
The rening and petrochemical
sectors consume large quantities of
H
2
, CO and syngas, which are
collectively known as HYCO.
Hydrogen is used for rening crude
oil, which is increasingly becoming
heavier and more sour. Also, ever
more stringent environmental regu-
lations will drive H
2
demand. In
addition, H
2
, CO and syngas are
used in various chemical synthesis
reactions. A summary of the major
industrial gases applications in
rening and petrochemicals is
shown in Table 1.
As noted previously, the domi-
nant method of producing H
2
, CO
and syngas is steam reforming of
natural gas. H
2
, CO and syngas
plants require a hydrocarbon feed-
stock such as natural gas or naphtha
and utilities such as water, electric-
ity and nitrogen. The outputs are
H
2
, CO, syngas and varying
amounts of steam. Given that these
plants are designed to last decades,
it is very important to understand
the current and future costs of feed-
stocks. This is essential to make the
appropriate trade-off between
capital cost and feedstock
consumption.
In particular, consumers of
tonnage quantities of industrial
gases have to make informed
investment decisions in an uncer-
tain economic environment. Plant
owners/operators need to examine
carefully their HYCO needs in
order to develop the most cost-
effective plant conguration for a
steam reformer. This can be done
by partnering with an experienced
engineering company that owns a
complete technology portfolio for
2 5
3 0
2 0
1 5
1 9 9 0 2 0 0 0 2 0 1 0 2 0 2 0 2 0 3 5
V
o
l
u
m
e
,

t
r
i
l
l
i
o
n
s

c
u
b
i
c

f
e
e
t
H isto ry P ro je c tio n s
1 0
2 0 1 0 2 0 3 5 1 9 9 0
0
5
1 0
H e n ry H u b sp o t m a rk e t
n a tu ra l g a s p ric e s
2 0 1 0 d o lla rs p e r m illio n B T U )
N e t e x p o rts, 2 0 3 5 = 5 %
N e t im p o rts, 2 0 1 0 = 1 1 %
C o n su m p tio n
D o m e stic p ro d u c tio n
Figure 1 Total US natural gas production, consumption and net imports
Source: Annual Energy Outlook 2013, Energy Information Administration, June 2012
selas.indd 1 10/12/12 14:05:47
producing H
2
, CO and syngas.
Ideally, the engineering and manu-
facturing companies have to work
together as a single team to identify
the most effcient and economic
plant design. This includes an in-
depth assessment of capital and
operating costs, taking plant relia-
bility and process safety into
account in order to deliver an opti-
mal solution. It is important to
consider all the industrial gases as
well as steam and utility needs at a
given manufacturing complex, in
conjunction with available feedstock
and wastes to develop the best
possible long-term solution. All of
these industrial gas requirements
have to be examined over a long-
term horizon in a holistic manner.
Reformer
Steam reforming of hydrocarbon
feedstock to produce H
2
, CO and
syngas has been the dominant proc-
ess for producing H
2
, CO and
syngas for over 80 years. A simpli-
fed block fow diagram of the
steam methane reforming process
showing how the reformer fts into
the overall plant is depicted in
Figure 2.
Conversion of steam/hydrocar-
bon mixtures is carried out in
catalyst-flled tubes heated from the
outside for the production of
syngases rich in H
2
and CO.
Feedstock can be methane, ethane,
LPG or naphtha. The basic design
criteria for reformers are burner
design and confguration, heat
recovery from the hot fue gases
and furnace control. A variety of
design options with burners
arranged in the ceiling or at the
sidewalls of the furnace can be
considered. The blue-shaded steam
reforming block in Figure 2 repre-
sents the furnaces radiant section,
where the tubes receive most of the
input heat by radiation. In the
convection section, heat is recov-
ered in both the process streams
and steam system by convective
heat transfer. Some other considera-
tions are:
Desired operating temperature
and pressure
Amount of export steam
Feed fexibility for natural gas,
LPG and heavy naphtha
Burners for low NOx generation
and low noise generation
Selective catalytic reduction for
NOx control, if required
Flue gas heat recovery by preheat-
ing feed, water and generating
steam
Fully automatic reformer using
advanced control technology.
For large installations, the top-
fred design combines both process
and physical benefts to minimise
the total cost of ownership. The
top-fred design allows either
modularised or stick-built construc-
tion, taking the relative differences
in shop assembly versus feld
construction costs into account.
Furthermore, this confguration
results in a compact frebox, leading
to fewer burners relative to the
number of tubes. The smaller
surface area per unit volume when
compared with sidewall-fred
reformers minimises heat loss and
reduces construction time. The
main advantages of a top-fred
design are.
Easy access to burners
Simplifed combustion control
with fewer burners
Heat recovery fexibility (omega,
horizontal, vertical orientations
available)
Maximum reforming effciency
Highest heat fux due to co-
current fow.
Waste heat
recovery
Feed
pretreatment
Steam
reforming
CO-shift
conversion
Pressure
swing
absorption
F e e d
F u e l
D e m in e ra lise d
w a te r
H yd ro g e n
E x p o rt
ste a m
H
2
re c yc le
S te a m p ro c e ss
Ta ilg a s
Figure 2 Steam reforming
Carbon monoxide (CO) Hydrogen (H
2
)
Acetic acid Hydrotreating
Formic acid Hydrocracking
Polycarbonates Ammonia
Toluene di-isocyanate Aniline
1,4 Butanediol
Syngas (CO + H
2
) Cyclohexane
Methanol Caprolactum
Oxo-alcohols 2-Ethylhexanol
Hexamethylenediame (HMDA)
Hydrogen peroxide
Isononyl alcohol
Toluene diamine (TDI)
Industrial gases in refning and
petrochemicals
Table 1
The engineering
and manufacturing
companies have
to work as a single
team to identify the
most effcient and
economic plant design
selas.indd 2 10/12/12 14:06:03
3D TRASAR For
Crude Overhead Systems
Nalco Downstream Energy
Services is the global leader in
Refning Process Treatments.
For more than 80 years, we have
worked to build an unrivaled level
of experience and expertise with
over 1,200 employees operating
in more than 350 refneries and
400 petrochemical plants across
six continents.
3D TRASAR for Crude Overhead
Systems is a patent pending new
technology that combines an
automated analyzer, Pathfnder
(suite of chemistries and predictive

simulation software), and experts to
deliver real time overhead
corrosion control. This innova-
tive new program allows the right
amount of chemistry to be delivered
at the right time by automating best
practices. Through technologies like
3D TRASAR for Crude Overhead
Systems, Nalco helps global refners
meet their corrosion control goals.
www.nalco.com
INNOVATION SQUARED
NALCO, the Logos and Tagline are trademarks of Nalco Company
Ecolab is a trademark of Ecolab USA, nc.
nalco.indd 1 11/9/12 11:54:21
Furthermore, optimisation of the
reformer with the waste heat recov-
ery section can be tailored to suit
multiple design cases for varying
feedstock, fuels and steam require-
ments. The use of computational
uid dynamic (CFD) modelling and
three-dimensional design programs
integrates process design with the
physical layout.
Design considerations
The cost of producing industrial
gases is a primary consideration for
most owner -operators. A schematic
depicting the important cost consid-
erations is presented in Table 2.
While some plant owner-operators
may focus on initial capital cost,
others pay equal attention to annual
operating cost. In reality, both
capital and operating costs are
important. For a rigorous assess-
ment, it is ultimately worth
examining the unit cost of the
industrial gas. This captures both
capital and operating costs in a
single metric that can be optimised
as a function of size, scope and
location.
It is usually very expensive to
achieve the last increment of
performance, whether it is
extremely high purity (99.999%) or
extremely high on-stream (99.9%).
This entails adding features such as
multiple instruments in critical
control loops, multiple rotating
equipment and backup power
supply. All of this adds cost and
may not provide value. By relaxing
overly stringent performance
criteria, it is possible to save time
and money. This can best be
achieved by an early dialog between
Stage IV
2 years
Project
execution
E n g in e e rin g
E q u ip m e n t
C o n stru c tio n
S ta rt-u p
Stage III
1 year
Project
development
R e q u e st fo r
q u o ta tio n R F Q )
E stim a te
P ro p o sa l
P e rm its
Stage III
1 year
Project
development
R e q u e st fo r
E stim a te
P ro p o sa l
P e rm its
Stage III
1 year
Project
development
R e q u e st fo r
E stim a te
P ro p o sa l
P e rm its
Stage II
1 year
Feasibility
study
P re lim in a ry
d e sig n
E c o n o m ic
e va lu a tio n
P ro je c t
ju stific a tio n
A u th o risa tio n
Stage III
1 year
Project
development
R e q u e st fo r
E stim a te
P ro p o sa l
P e rm its
Stage I
1 year
Pre-feasibility
M a rk e t re se a rc h
E n e rg y a n a lysis
P ro je c t sc o p e
S ite sc re e n in g
Engagement
Figure 3 Project development
Capital cost
Process selection
Size of plant
Infrastructure (cooling water, electrical)
Energy efciency
Operating cost
Purity
Pressure
Plant optimisation
Capital/energy trade-off
Local cost of energy
Utility
Water availability
Electricity
Steam
Operations
Reliability
Time to full production
Turndown
Design considerations
Table 2 Figure 4 Large reformer with required process optimisation
selas.indd 3 11/12/12 18:06:55
the plant owner-operator and plant
designer. A large, modern reformer
with the required process optimisa-
tion is shown in Figure 4.
Project development
It is also very important to consider
the co-products that can be
produced from a single process
plant. In general, producing multi-
ple products such as H
2
, CO, CO
2
,
syngas, steam and electricity from a
HYCO plant reduces the unit cost
of each product. Producing multi-
ple products from a single
production plant can be very cost
effective, since the fxed capital and
variable operating costs increase
only incrementally. Furthermore,
these increased costs can be allo-
cated to multiple products, thereby
reducing the unit cost of all prod-
ucts. In order to achieve this sweet
spot, it is essential to consider all
possible scenarios early in the
development of a project.
A high-level timeline of typical
project development is presented in
Figure 3. Typically, a large, complex
industrial gas plant can take up to
fve years from concept to startup.
During pre-feasibility (Stage I), it
is important to consider all possible
scenarios including co-products,
feedstock and capital versus energy
cost trade-offs. As the project gets
better defned, different process
cycles can be examined in the feasi-
bility stage (Stage II). Once project
development is authorised and
permits are applied for in project
development (Stage III), further
changes to scope should be resisted.
While some changes may appear to
be cost justifed for a small portion
of the project, usually these changes
add cost to the overall project and
delay the schedule. It is important
to maintain discipline during
project execution (Stage IV) as
equipment is ordered and feldwork
begins. Even the best-conceived
projects on paper can be disasters
in the feld if proper engineering
work practices are not followed.
Conclusion
Modern steam methane reformer
design is a complex process that
needs to be carefully optimised in
terms of capital and operating
costs, taking current and future
requirements into account. The
optimum solution is application
and site specifc and needs to be
custom engineered to a large extent.
In order to achieve the best overall
solution, early engagement with a
qualifed engineering company is
essential.
Norm Pelletier is Product Manager of Steam
Reformers at Selas Fluid Processing, a division
of Linde Engineering, Blue Bell, PA. He has
15 years process engineering and sales
experience across Lindes fred equipment
product lines lines, including refnery heaters,
LNG vapourisers and thermal oxidisers as
well as his current position in hydrogen and
syngas reforming. He holds a bachelors
degree in chemical engineering from Villanova
University.
Goutam Shahani is Business Development
Manager at Selas Fluid Processing, a division
of Linde Engineering, Blue Bell, PA. He has over
25 years industry experience specialising in
industrial gases for the energy, refning and
chemical industries. He holds bachelors and
masters degrees in chemical engineering and
an MBA.
Emission impossible
BCD Packing Ring -
leading operators and OEMs conrm:
Are you satised with the effectiveness of your compressor packings?
Are you happy with leaking gas?
Are you satised with the durability of your packing rings?
Are you comfortable with the impact your compressor may be having
on the environment?
The BCD packing ring has proven its outstanding
performance in numerous applications worldwide.
Many compressor operators have installed the
HOERBIGER BCD packing ring and leading
OEMs have conrmed its high lifetime
and superior sealing efciency. Their
compressors now run more efciently
and with less fugitive emissions.
For extended lifetime, reduced emissions and cost savings, ask
for the BCD packing ring: ringsandpacking@hoerbiger.com
www.hoerbiger.com
selas.indd 4 10/12/12 14:57:59
enersul.indd 1 6/12/12 13:25:03
VDU vacuum system revamp
T
he non-condensed gases from
a vacuum distillation unit
(VDU) are considered to be
waste gas. Since waste gas has some
H
2
S and H
2
O content, the disposal
of this waste gas poses serious safety
and environmental problems. In this
respect, engineers at Tpras
Kirikkale refnery redesigned the
Revamping a VDUs vacuum system delivered operational improvements in a
refnery, but also led to safer and more effcient handling of non-condensed gases
AHMET BEBEK
Turkish Petroleum Refneries Corporation
VDU tower top outlet downstream
section to treat the overhead gases
(non-condensables) leaving the
vacuum column. In place of an
underground atmospheric pit, the
installation of a barometric seal
system for a VDU vacuum system
was achieved. The existing vacuum
unit operation was not affected by
the erection, commissioning and
startup of the revamped vacuum
system. As a result of the installation
of a new barometric seal system,
gaseous releases to the atmosphere,
exposure of personnel to gaseous
releases and the risk of back-fow of
air into the vacuum unit have been
eliminated. The seal system also
H -2 A
H -2 B
H -3 A
H -3 B H -1 A
E -1 A E -1 A E -1 B
E -2 E -3 E -4
H -1 B
H -2 B
H -4 A
H -4 B
N o n -c o n d e n se d
va p o u r to
fu rn a c e
K n o c k -o u t
ve sse l
S W to S WS
fe e d d ru m
H C to slo p s
Va c u u m
u n it
va p o u r
Va c u u m
u n it
va p o u r
S te a m
P 1
Ve n t to
a tm o sp h e re
D ra in
P 1
P 2
A tm o sp h e ric p it
Figure 1 Vacuum system with atmospheric pit
tupras.indd 1 10/12/12 14:08:02
helped in the disengagement of
liquid and non-condensables, lead-
ing to improvements in reliability
(for instance, a lower level of plug-
ging in the non-condensables line
and improved vacuum in the tower).
Lastly, waste gas is burned by
a dedicated burner in the
vacuum heater furnace, which repre-
sents a signifcant saving in energy
costs.
Vacuum system before revamp
The VDU vacuum system is a
booster type (in four stages) with a
condensate leg for each condenser
(see Figure 1). Each stage has two
ejectors. The system achieves a
pressure of approximately 13.3
mbara top pressure and a fash
zone pressure of 66.6 mbara. The
ejectors are driven by medium-
pressure steam (12.75 kg/cm
2
and
192C). Condensers downstream of
the ejectors use water as a cooling
medium.
The non-condensed vapour from
the last-stage ejector was passed to
the VDUs furnace via a small knock-
out vessel. This stream had a high
wet H
2
S content. It was common
(once a week) for this line to become
plugged. When this occurred, the
vacuum level in the VDU decreased,
resulting in a signifcant disturbance
to operations. While the plugged line
was steamed out, the non-condensa-
bles from the last-stage ejector were
directed to atmosphere, giving rise
to offensive smells in the area.
The condensed steam and hydro-
carbons from each ejector condenser
were passed directly to an under-
ground atmospheric pit. Overfow
separation of hydrocarbons took
place in this pit. The liquid hydro-
carbon was sent to a slops system
and the aqueous phase was directed
for further treatment to the sour
water stripper unit.
Even when waste gas was burned
in the furnace, there was still
airborne H
2
S in the area surround-
ing the vacuum columns
underground atmospheric pit, caus-
ing offensive smells. This was due
to fact that the atmospheric pit was
H -2 A
H -2 B
H -3 A
H -3 B H -1 A
E -1 A E -1 A E -1 B
E -2 E -3
H -1 B
H -2 B
H -4 A
H -4 B
E -4
V3
V4
N o n -c o n d e n se d
va p o u r to
fu rn a c e
Ve n t to fla re
Ve n t to sa fe
lo c a tio n
S W to S WS
fe e d d ru m
M C to slo p s
Va c u u m
u n it
va p o u r
Va c u u m
u n it
va p o u r
S te a m
V1 V2
P 2
P 3
P 1
Figure 2 Vacuum system with barometric seal vessels
Waste gas is burned
by a dedicated
burner in the vacuum
heater furnace,
which represents a
signifcant saving in
energy costs
tupras.indd 2 12/12/12 11:11:36
not sealed and waste gas dissolved
in the condensate was being
released to the atmosphere. Clearly,
there had to be an improvement in
the management of health, safety
and environment in the area.
The approach
Tpras Kirikkale refnery constantly
invests to minimise the adverse
impact of refning activities on the
environment, employees and the
public, to offer the safest possible
working environment. In this
respect, the existing vacuum unit
was revamped to prevent waste gas
release to the atmosphere, which
was causing safety problems for
personnel. The other basis for this
work was to prevent the risk of air
backfow into the vacuum unit to
deal with the case of failure of ejec-
tors. Lastly, the revamp was
designed to eliminate operational
disturbances resulting from plug-
ging of the waste gas line, which
was leading to a decrease in unit
capacity and loss of operating
margin.
The major component of the
condensed stream originated from
the vacuum towers non-condensed
vapour and from medium-pressure
steam as a utility. The installation
of a barometric seal system for the
VDU vacuum system seemed the
best option to solve the problems
mentioned, especially the health,
safety and environmental issues.
The existing vacuum units opera-
tion should not be adversely
affected by the erection, commis-
sioning and startup of the revamped
vacuum system. The revamp should
not give rise to any negative impact
on the columns vacuum level.
New vacuum system
The underground atmospheric pit
was replaced with a barometric seal
system (see Figure 2). The purpose
of the barometric seal system was
to achieve good separation of non-
condensed vacuum tower top
outlets after they are condensed by
the ejectors. In addition to this
change, the fnal-stage ejector
nozzles were replaced in order to
provide a pressure high enough to
send waste gas to the furnace.
There was no other design change
to either the ejectors or the
condensers.
The confguration of the present
barometric seal system is described
below. The condensed hydrocar-
bons and water from the frst-,
second- and third-stage condensers
pass to the frst seal vessel via a
barometric seal leg. The condensed
hydrocarbons and water from the
fourth-stage condenser pass to the
second seal vessel. The non-
condensed vapour from the
fourth-stage condenser passes to a
waste gas seal vessel via a seal pipe
that is about 17m high, to prevent
ingress of air. The last vessel is the
knockout vessel. The primary
purpose of this vessel is to remove
entrained liquid droplets in the
offgas stream before it is sent for
disposal. The offgas from this vessel
then passes to three separate loca-
tions: the vacuum furnace (a
dedicated waste gas burner), the
fare (when there are problems
sending to furnace) and the safe
location (only during startup). The
routes to the furnace or safe loca-
tion are provided with fame
arresters.
Water in the water compartment
of the frst seal vessel is circulated
between the other vessels at all
times to prevent any breakthrough
of H
2
S and thus prevent build-up of
corrosive and fouling material.
Excess sour water is pumped to the
sour water stripper unit. Since sepa-
ration of water and hydrocarbons is
improved in comparison to the old
underground pit design (in which
hydrocarbon was sent to the slops
system), the hydrocarbon from the
seal vessels can be pumped to the
feed pool of the deep diesel desul-
phurisation unit.
Results
Tpras Kirikkale refnery rede-
signed the vacuum system of the
vacuum distillation unit. An atmos-
pheric pit was replaced with a
barometric seal system. This change
resulted in good separation of non-
condensed gases from the vacuum
towers top outlets after condensing
by ejectors. The achievements of
this revamp are:
Wet H
2
S release to atmosphere
from both the atmospheric pit and
the knock-out vessel has been
prevented
The possibility of exposure of
personnel to wet H
2
S release has
been avoided
The risk of a back-fow of air into
the vacuum unit has been
eliminated
The heating value of hydrocar-
bons in the waste gas can be used.
300 kg/h of waste gas is consumed
in a dedicated special burner in the
VDUs furnace. This results in a
saving in fuel of 1.4 Gcal/h
The reliability of the vacuum
system has been improved because
the seal system has helped in
disengagement of liquids and non-
condensables. There has been less
plugging of the non-condensables
line to the furnace and an improved
vacuum level in the tower. In this
way, a decrease in capacity leading
to margin loss has also been
prevented
Separation of hydrocarbons and
water has been improved. As a
result, hydrocarbons are being sent
to the diesel pool instead of the
slops system, thereby avoiding
reprocessing in the crude unit and
so decreasing operational costs.
Ahmet Bebek is an Operations Superintendent
in the Production Department of Tpras
Kirikkale Refnery, Turkey. He holds a BS
in chemical engineering from Middle East
Technical University, Ankara, Turkey, a MS in
process systems engineering from Technical
University of Dortmund, Germany, and a MS
in industrial engineering from Middle East
Technical University, Ankara.
Email: Ahmet.Bebek@tupras.com.tr
The existing vacuum
units operation
should not be
adversely affected
by the erection,
commissioning
and startup of the
revamped vacuum
system
tupras.indd 3 10/12/12 14:08:26
GET A RETURN ON INVESTMENT
WITHIN 24 HOURS
Using ITW Online Cleaning on a proactive way will help improve
TURNAROUND OPERATIONS and OPERATIONAL EXCELLENCE
An entire Process Unit can be ITW Online Cleaned in as low as 24 hours on an oil-to-oil basis. Our cleaning technology will
clean all of the Units equipment quickly, safely and environmental friendly.
To recover a Units performance, ITW Online Cleaning can be applied in a Turnaround during shutdown operations, to
reduce/eliminate the need for mechanical cleaning, or during the run.
Furthermore, use of our cleaning technology eliminates the need to open or enter hazardous process equipment.
ITW Online Cleaning is the only technology which has an immediate Return On Investment (ROI).
By using ITW Online Cleaning during the run, to recover Units performance, at Unit start-up (just after 24h, ITW
Technology has been applied), all the benefits are immediately monetized in terms of:
- improved throughput
- improved energy efficiency
- improved product quality
- reduced emission taxes
Regular application of ITW Online Cleaning will target an increased run length under clean conditions, with a far higher
ROI.
For Turnaround applications, a further ROI increase can be added by applying ITW Improved Degassing/Decontamination
to achieve quick and effective safe entry conditions.
EVALUATE ITW ONLINE CLEANING TODAY TO IMPROVE YOUR PLANTS PROFITABILITY !
For more informations contact : ITW S.r.l.- C.da S.Cusumano - 96011 Augusta - Italy
Tel. +39 (0931) 766011
E-mail: info@itwtechnologies.com
www.itwtechnologies.com
Join ITW Team worldwide and send your Curriculum Vitae to : jobs@itwtechnologies.com
ITW
Innovative
Technologies
Worldwide
Now hiring
Professionals
Worldwide
itw.indd 1 11/9/12 11:45:38
Reducing carbon emissions with
antifoulants
T
he UN Kyoto Protocol set the
frst binding targets for reduc-
ing greenhouse gas emissions
in 1997. Although the US and China
declined to participate, 37 industr-
ialised countries and the European
Union now regulate carbon emis-
sions, and the trend seems clear.
The EU Energy Pact targets a 20%
reduction in CO
2
emissions by 2020,
and a carbon pricing scheme takes
effect in Australia in 2013.
These are challenging times for
refneries. Sulphur levels must be
reduced in fnished fuels to meet
increasingly strict specifcations
driven by new emission control
technologies in motor vehicles.
Meanwhile, crude feedstocks are
becoming heavier, higher in sulphur
and more diffcult (and energy
intensive) to process.
Refneries are signifcant sources
of carbon emissions, much of it in
the form of CO
2
from burning fuels
to distill, crack and hydrotreat their
feedstocks. In spite of the caps and
fees imposed on carbon, demand
for refnery products continues to
grow, increasing refnery energy
use and emissions. Heavier, sourer
crude feedstock adds to the
problem.
The costs are substantial: energy
costs for a typical refnery are 50-
60% of total operating costs,
excluding feedstocks.
1
Effciency,
always a high priority in refnery
operations, has never been more
important or more diffcult to
attain.
Since crude cost is the single most
important determinant of a refn-
erys proftability,
2
price differentials
on challenging crudes have consid-
erable appeal, even though unit
Carbon emission regulations make refnery operation more costly, but appropriate
antifoulant treatment and monitoring can reduce these costs substantially
INDIA NAGI-HANSPAL, MAHESH SUBRAMANIYAM, PARAG SHAH and JAMES NOLAND
Dorf Ketal Chemicals
designs often limit feedstock fexi-
bility and heavy crudes can lead to
fouling problems both of which
increase carbon emissions that must
be factored into the refnery operat-
ing cost model.
Consider the current approximate
prices of Murban crude (0.6 wt%
asphaltenes) and Maya (10.0 wt%
asphaltenes). Since Maya is $ 13.61/
bbl cheaper than Murban, a refnery
processing 100 000 b/d can save up
to $ 1 361 000/day in feedstock
costs alone. As the small sample of
crudes in Figure 1 shows, Maya is
just one of many common high-
asphaltene crudes.
Processing these heavy, high-
n
a
i
r
o
d
a
u
c
E
a
y
a
M
h
s
o
o
r
o
S
o
p
a
N
e
t
n
e
i
r
O
6
1

y
e
r
e
M
n
a
r
d
u
B

s
a
R
b
i
r
a
h
G

s
a
R
z
o
o
r
w
o
N
e
n
e
c
o
E
h
e
i
d
e
u
o
S
a
n
a
r
f
a
a
Z
n
a
y
y
a
R

l
A
m
u
i
y
a
l
e
B
i
w
a
t
a
R
o
v
l
o
P
i
j
f
a
h
K
2
2

a
n
o
e
L
4
2

a
n
o
e
L
n
a
g
r
e
h
a
B
2
2

A
S
V
D
P
y
v
a
e
H

b
a
r
A
m
i
l
r
a
M
o
t
i
u
K
%

,
t
n
e
t
n
o
c

e
n
e
t
l
a
h
p
s
A
Figure 1 Asphaltene content of selected crudes

90
80
70
60
50
40
0 5 10 15 20 25 30 35 40

,
s
n
o
i
s
s
i
m
e

G
H
G

g
n
i
n
i
f
e
R
O
C

g
k
2
l
i
o

l
b
b
/
e
Crude gravity, API
30
22
20
18
16
14
12

,
s
n
o
i
s
s
i
m
e

G
H
G

g
n
i
n
i
f
e
R
O
C

g
2
B
O
B
R

J
M
/
e
10
g GHGs/MJ RBOB
kg/bb| crude
y = -0.77x + 86.39
R
2
= 0.86
y = -0.09x + 15.13
R
2
= 0.72
Figure 2 Greenhouse gas emissions decline with increase in API
3
dorf ketal.indd 1 10/12/12 14:19:03
sulphur crudes consumes more
energy and increases greenhouse
gas emissions (see Figure 2).
Calculations of actual costs can be
quite complex, in part because
different fuels are often used at vari-
ous stages within the refnery. Fuel
oil and fuel gas are common choices
in refnery furnaces, and their energy
content and emissions differ.
Although it typically takes less refn-
ery fuel oil to heat the feedstock to
the target temperature than would
be the case with refnery fuel gas
(refnery fuel oil has a calorifc value
of 8740 kcal/m
3
compared to 10 000
kcal/m
3
for fuel gas), fuel oil carbon
emissions are usually higher per
unit of fuel consumed.
Refnery feedstock is another
important consideration. Crude
production may generate enough
CO
2
emissions to make a given feed-
stock more costly overall. Oil shale
production, for example, has been
shown to contribute more heavily to
carbon emissions than the extraction
of other hydrocarbons.
These issues are increasingly
important because of the way
carbon emissions regulations work.
The Western Climate Initiative in
selected Canadian provinces and
California is running a cap-and-
trade scheme. In Europe, the EU
Energy Pact contains key targets for
the year 2020, in particular a 20%
reduction in CO
2
emissions (from
1990 levels) that is designed to
ensure that at least 20% of total
energy consumption comes from
renewable sources.
Emission Trading Scheme
To meet the new targets, many
countries will adopt Kyoto mecha-
nisms such as the Emission Trading
Scheme (ETS), through which coun-
tries can buy carbon credits known
as emission reduction units. These
can be bought from clean develop-
ment mechanism projects or carbon
emission reductions from joint
implementation projects.
European Union cap-and-trade
regulations are the largest such ETS
to date. Companies there are
granted emissions allowances that
they can buy, sell or trade with
each other, but at the end of the
year each company must have
enough emissions allowances to
cover their total emissions.
The EU ETS regulates 46% of the
EUs CO
2
emissions by capping the
amount of CO
2
that can be emitted
from factories and plants. Once
Phase III (2013-2020) of the scheme
is under way, more restrictive
controls on emissions can be
expected, along with more efforts to
reduce carbon credit consumption.
Support for these schemes is not
unanimous. Canada withdrew from
the Kyoto Protocol in December
2011 to avoid heavy fnes for failure
to meet emissions targets. China,
one of the worlds largest emitters of
greenhouse gases, has not signed the
Kyoto Protocol, but even there plans
are already under way to launch a
few pilot cap-and-trade markets and
to establish a fully operational
carbon market by 2015.
In spite of these regional differ-
ences, it is clear to refners
worldwide that carbon costs are
becoming signifcant variables in the
refnery cost equation, and many are
actively seeking opportunities to
reduce emissions by increasing eff-
ciency. Their frst targets are the
systems that consume fuel for
instance, furnaces and preheaters
where effciency depends on
feedstock, fuel source and combus-
tion effciency.
Effciency gained from feedstock
changes must be weighed against
the proft potential from lower-
priced crudes. Changing fuel types
can entail a signifcant investment
and can have a substantial impact
on operations. Combustion improv-
ers are a lower-cost option that may
help in some cases.
Carbon capture and storage (CCS)
is another alternative. As the name
implies, CCS limits the amount of
CO
2
released into the atmosphere
by capturing CO
2
and storing it in
geological formations underground.
This, too, is capital intensive. CCS
is the way of the future despite its
economic implications.
Antifoulants
Antifoulants offer another way to
improve effciency, a proven
approach that entails little or no
capital investment. Antifoulants can
also improve gross margins by
enhancing refnery feedstock fexibil-
ity, and costs are usually very low
in comparison to the alternatives.
Uncontrolled fouling decreases
heat transfer effciency and through-
put, increasing fuel consumption
and carbon emissions. Feedstock
fexibility is impaired and, if left
untreated, fouling reduces through-
put and can force units offine for
cleaning or repair.
Fouling is of two general types:
inorganic and organic. The former is
usually caused by elevated levels of
metals in refnery feedstocks, typi-
cally occurs between 150 and 360C,
and tends to increase the potential
for costly and dangerous corrosion.
Crudes produced from deep oceanic
locations often exhibit inorganic
fouling due to contaminants such as
salts, flterable solids, basic sedi-
ments and corrosion products.
Organic fouling generally occurs
above 250C in cracked streams,
often as a result of high asphaltene
content or incompatible blends of
asphaltenic and paraffnic crudes.
Whether the fouling is inorganic or
organic, success with antifoulants
depends on careful monitoring. Key
parameters include heat transfer
rates, heat exchanger duties,
approach enthalpies, feedstock
composition, CO
2
emissions and
fuel combustion effciency.
Selection of antifoulant is also
important, especially with todays
increasingly sour feedstocks.
Sulphidation is common with these
crudes, leading to iron sulphide-
promoted fouling. In most cases,
antifoulants must therefore be effec-
tive on asphaltenes and iron
sulphide.
Antifoulants work by stabilising
asphaltenes that would otherwise
become destabilised when heated.
This prevents deposition of polynu-
clear aromatics that, upon further
heating, can form coke. Left
unchecked, fouling reduces heat
transfer from the heating media to
the cold stream and increases the
furnace loading needed to achieve
the required coil outlet temperature.
Figures 3 and 4 illustrate antifou-
lant functionality by comparing
untreated feedstock with treated
samples. Asphaltenes that agglom-
erate and settle out in minutes
dorf ketal.indd 2 12/12/12 12:48:32
safe & sound
Integrating people, process and knowledge for better performance.
Experion
PKS O rion provides the only truly integrated safety and

security solution, supporting com plete operational integration through
a single dashboard. Additionally, a new U niversal Safety I/O m odule
supporting control in the eld, the U niversal Safety Logic Solver,
safeguards the process even in the event com m unication w ith the
Safety M anager is interrupted. This advance reduces cost w hile
increasing availability and ef ciency.
2012 Honeywell International, Inc. All right reserved.
w w w .honeyw ellprocess.com
honeywell.indd 1 11/9/12 11:44:18
without antifoulant treatment
remained stabilised for an hour or
more in the test.
Experience indicates that antifou-
lants can increase furnace inlet
temperature by 5 to 15C in fouled
systems. It is possible to do even
better with periodic cleaning. The
way the antifoulant is applied has
considerable inuence on the
results, and choosing the correct
injection point is especially impor-
tant. A suction pump upstream of
the main fouling exchangers is
ideal.
Case study
The following case study illustrates
the potential benets of antifoulants
on fuel cost and CO
2
emissions.
Renery X was running at an aver-
age throughput of >300 000 b/d.
The average coil outlet temperature
(COT) when the feedstock was
treated with antifoulants met the
renery standard required to
produce target yields of down-
stream nished products. Without
antifoulant, the target COT was
often impossible to achieve, and
considerably more energy was
required (see Figure 5).
Antifoulant treatment signi-
cantly reduced the fuel consumption
required to maintain target COT,
lowering specic fuel costs by
nearly 4% (see Figure 6). This saved
the renery approximately 41 000
per month on fuel alone.
At an average carbon credit value
of 16 per ton of CO
2
, renery
carbon cost savings would total
59 000, raising the overall nan-
cial impact of antifoulant treatment
to approximately 100 000 per
month (see Figure 7).
Conclusion
Carbon emission regulations make
renery operation more costly, but
appropriate antifoulant treatment
and monitoring have been shown
to reduce these costs substantially.
Antifoulants also allow reners to
enhance gross rening margins by
exploiting lower-cost feedstocks.
They reduce the fuel consumption
required to maintain coil outlet
temperatures for target throughput
rates. As a whole, antifoulants
are an environmentally friendly
choice with attractive economic
benets.
Without additive With additive
Figure 3 Asphaltene dispersion studies with and without antifoulant
A g g lo m e ra te d
a sp h a lte n e s
D isp e rse d
a sp h a lte n e s
1 1 : 4 1 a m 3 : 5 0 p m
Figure 4 Testing under a microscope: agglomerated and dispersed asphaltenes
1 . 9 0
1 . 9 2
1 . 8 8
1 . 8 6
1 . 8 4
1 . 8 2
1 . 8 0
N o a n tifo u la n t With a n tifo u la n t
T
h
r
o
u
g
h
p
u
t
,

/
m
3
1 . 7 8
Figure 5 Specic fuel cost with and without antifoulant
dorf ketal.indd 3 10/12/12 14:19:23
References
1 Based on a natural gas price of about $6/MM
Btu for a typical 100 KBPSD renery that emits
1.2-1.5 MM t/yr of CO
2
.
2 Stockle M, Carter D, Jones L, Optimising
Renery CO
2
Emissions, Foster Wheeler
Technical Paper www.fwc.com/publications/
tech_papers/les/ERTC%20CO2%20paper%2
0Nov07.pdf
3 Brandt A R, Unnasch S, Energy intensity and
greenhouse gas emissions from California
thermal enhanced oil recovery, Energy & Fuels
2010: Keesom W, Unnasch S, Moretta J, Life cycle
assessment comparison of North American
and imported crudes. Technical report, Jacobs
Consultancy and Life Cycle Associates for
Alberta Energy Resources Institute, 2009.
India Nagi-Hanspal is Lead Renery Engineer,
Technical Services with Dorf Ketal Chemicals,
Mumbai, India. She holds a MEng degree in
chemical engineering from Imperial College
London. Email: indianagi@dorfketal.com
Mahesh Subramaniyam is Director of Research
& Development with Dorf Ketal Chemicals. He
7 . 9 6
7 . 9 8
7 . 9 4
7 . 9 2
7 . 9 0
7 . 8 8
7 . 8 6
N o a n tifo u la n t With a n tifo u la n t
S
p
e
c
i
f
i
c

f
u
e
l

c
o
n
s
u
m
p
t
i
o
n
,

k
g
/
m
3
7 . 8 4
Figure 6 Specic fuel consumption with and without antifoulant
4 1 , 0 0 0 in
fu e l sa vin g s
5 9 , 0 0 0 in
c a rb o n c re d its
Figure 7 Financial benet of antifoulant
treatment
holds a PhD in chemistry from Indian Institute
of Technology, Mumbai.
Email: drmaheshs@dorfketal.co.in
Parag Shah works in Global Renery Technical
Services with Dorf Ketal Chemicals in software
development for desalter adequacy testing and
monitoring fouling in preheat exchanger trains.
He holds a BEng in chemical engineering from
Mumbai University.
Email: paragshah@dorfketal.com
James Noland is Senior Director of the Process
Chemicals Division of Dorf Ketal USA, LLC. He
holds a BEng in chemical engineering from
Mississippi State University, USA.
Email: james.noland@dorfketal.com
dorf ketal.indd 4 12/12/12 10:55:27
www.metso.com/oilandgas
Focusing on your safety
W e of er peace of m ind w hen it com es to your process safety and productivity.
M etso is a single source for all your process control needs, from autom ation system s
to intelligent f ow control solutions that are proven in the toughest conditions.
O ur 30,000 professionals based in over 50 countries deliver sustainability and
prof tability to custom ers w orldw ide.
metso.indd 1 11/9/12 11:42:58
Towards a zero (net) carbon refnery
T
he well-to-wheel emissions
for petrol and diesel can be
split into three main parts: oil
production and transportation, oil
processing and transportation, and
emissions from the consuming
engines. In order to reduce emis-
sions and meet targets for carbon
Review of options for refners to reduce the carbon emissions from their refneries.
The technologies already exist to develop a refnery that has zero net emissions
MICHAEL STOCKLE
Foster Wheeler
reduction, it is likely that all three
parts of the chain will need to make
some contribution.
While the processing of oil in the
refnery contributes around
only 5-10% of the total well-to-
wheel emissions, the opportunities
for reducing these are signifcant
due to the nature of the processes
used, the fuels available and the
fact that these are large-scale fxed
location processes making applica-
tions such as carbon capture and
storage (CCS) more practical than
they would be on individual
vehicles.
H
2 S F G
VDU
FCC
FCC Nap
HDT
Visibreaker
Amine
treatment
Fuel gas
system
Hydrogen
plant
SRU
L P G
L C O
N a p h th a
L C O /D C O L C O /D C O
C ru d e
H
2
L P G
H
2
H
2
S F G
H e a vy
N a p h th a
L ig h t
N a p h th a
S F G
H
2
S F G
S F G H
2
S F G WN
F G
L P G
CDU
LPG
treater
Naphtha
HDT
Naphtha
splitter
Reformate
splitter
Kero
HDT
CCR
ISOM
Diesel
HDT
N a p h th a
K e ro se n e
D ie se l
VG O
Va c u u m re sid u a ls
H
2
S
G a s o il
D ie se l
K e ro se n e
R e sid u e
A tm o sp h e ric
re sid u a ls
L P G
G a so lin e
H S F O
N a p h th a
J e t
D ie se l
S u lp h u r
H
2
H
2
R e fo rm a te
I so -M
C
4
s
H e a vy
re fo rm a te
R e fo rm a te
Tre a te d g a so lin e
A tm o sp h e ric re sid u a ls
F G
L P G
S F G
Figure 1 Refnery confguration
fw.indd 1 12/12/12 12:13:12
This article will look at the main
sources of carbon emissions from a
refnery and consider a range of
options for reducing the carbon
impact of the refnery, starting with
fuel substitution and energy eff-
ciency, then moving on to how
technologies such as CCS and
renewable power generation could
be integrated into the refnery.
The article will look at the impact
of making these changes on refnery
economics and show one possible
solution that, at a suffciently high
carbon price, could see a zero net
carbon refnery (that is, one where
no fossil fuel-derived carbon is
emitted).
Sources of CO
2
in the refnery
Sources of carbon dioxide (CO
2
) in
the refnery can be categorised into
four main groups:
Fuel to process units
Steam and power production
Hydrogen production
FCC coke (where applicable).
The emissions of an individual
refnery depend on a number of
factors. The confguration of the
refnery is a key factor, as complex
upgrading refneries produce more
CO
2
than simple hydroskimming
confgurations, but they tend to
produce more of the fuels society
demands (gasoline and diesel) per
barrel of oil processed. The fuels
used in the refnery also have an
impact, as do the crudes processed.
Heavy, sour crudes require more
energy to process than do light,
sweet crudes.
In order to quantify the relative
sizes of the emissions, we can look
at a typical refnery confguration
and the CO
2
emitted from it. In this
article, we have considered an FCC-
based European refnery processing
150 000 b/d of Ekofsk Crude that
generates all power and steam on-
site from refnery fuel gas and fuel
oil. Figure 1 shows the main confg-
uration of this refnery. The
emissions from this refnery are
shown in Figure 2.
Options for reducing refnery carbon
emissions
There are a number of ways of
reducing the carbon emissions of a
refnery, ranging from relatively
simple, low-cost options to complex,
capital-cost-intensive options. These
options can be grouped into a
number of areas and we will
consider each area in turn.
Energy effciency
The frst area of focus is energy eff-
ciency. The other options we pursue
will tend to be more expensive than
the current energy sources in the
refnery (if they were cheaper, they
would already be used) and so
making the most of the energy we
do use will become even more
important.
Even refneries that consider
themselves good performers in
terms of energy effciency can do
more, and this is illustrated by a
recent study Foster Wheeler
completed for a top-quartile refn-
ery in northern Europe. This study
resulted in operational improve-
ment and investments being
identifed that could save around
10% of the fuel used in the refnery,
and the potential projects had an
estimated payback of less than 18
months. In our example, we will
assume our base case 150 000 b/d
refnery has already achieved a
high level of effciency and
5 0 0 0
4 0 0 0
3 0 0 0
2 0 0 0
1 0 0 0
T
P
D

o
f

C
O
2
Base sources
0
U tilitie s
H P U
Visb re a k e r
F C C
D H T
N a p h th a
C D U
S R U
5 0 0 0
4 0 0 0
3 0 0 0
2 0 0 0
1 0 0 0
T
P
D

o
f

C
O
2
Base Nat . gas f i ri ng
0
U tilitie s
H P U S R U
Visb re a k e r
F C C
D H T
N a p h th a
C D U
U tilitie s
H P U
S R U
Visb re a k e r
F C C
D H T
N a p h th a
C D U
Figure 2 Base case carbon emissions
Figure 3 Comparison of impact of natural gas fring
Fuel Calorifc value, kJ/kg CO
2
tonnes/tonne CO
2
tonnes/FOE* tonne
*Fuel oil equivalent
Methane 50 000 2.75 2.20
Ethane 47 500 2.93 2.47
Propane 46 300 3.00 2.59
Ethylene 47 200 3.14 2.67
Fuel oil 40 000 3.21 3.21
Hydrogen 119 900 0 0
CO
2
emissions for typical fuels
Table 1
Hydrocarbon fuels
contain very different
levels of carbon and
result in signifcant
carbon emissions
for the same level of
energy requirement
fw.indd 2 12/12/12 12:13:23
will focus on other options to
reduce the emissions from the
refnery.
Fuel substitution
Hydrocarbon fuels contain very
different levels of carbon and result
in signifcant carbon emissions for
the same level of energy require-
ment. Table 1 compares the impact
of different fuel sources on carbon
emissions.
Figure 3 shows the total refnery
CO
2
emissions from our 150 000
b/d refnery, assuming all heat and
power is generated on-site,
compared with likely emissions if
natural gas was imported into the
refnery for power generation.
We can see that moving from
fuel oil fring to natural gas fring
reduces emissions from the refn-
ery by about 13%. Of course, the
economics of this option are highly
dependent on the relative prices of
natural gas and fuel oil. In the US,
where the differential between
natural gas and crude price is
wide, natural gas fring will be
increasingly attractive. In Europe,
the incentive will be lower, but
will still generally favour natural
gas fring where available. Figure 4
shows the natural gas price
required to break even on energy
cost versus fuel oil at varying
carbon prices.
Green heat and power
In our example 150 000 b/d refn-
ery, we have assumed that all of
the refnery's power requirements
are generated on-site. One way to
reduce the refnery's emissions
would be to generate this power
from low-carbon sources, either on-
site or by purchasing green power
over the fence.
If we replace all of the on-site
power generation with imported
low-carbon power, emissions drop,
as shown in Figure 5.
This option reduces on-site emis-
sions by 28%, but is dependent on
a reliable supply of green power. If
power was required to be gener-
ated on-site, options such as wind
and solar could be evaluated, but
again the dependability of these
sources may be an issue, and
investment is likely to be required
3 0
2 5
2 0
1 5
1 0
5
5 0 6 0 7 0 8 0 9 0 1 0 0 1 1 0 1 2 0
N
a
t
u
r
a
l

g
a
s

p
r
i
c
e
,

$
/
M
M
B
T
U
Fuel oi l pri ce, $/ bbl
0
0
$ /to n n e C O
2
5 0
2 5
5 0 0 0
4 0 0 0
3 0 0 0
2 0 0 0
1 0 0 0
T
P
D

o
f

C
O
2
Base Low carbon
el ect ri ci t y purchase
0
U tilitie s
H P U S R U
Visb re a k e r
F C C
D H T
N a p h th a
C D U
U tilitie s
Visb re a k e r
F C C
D H T
N a p h th a
C D U
H P U S R U
3 , 0 0 0 , 0 0 0
2 , 5 0 0 , 0 0 0
2 , 0 0 0 , 0 0 0
1 , 5 0 0 , 0 0 0
1 , 0 0 0 , 0 0 0
5 0 0 , 0 0 0
B a se F u ll p o st-c o m b u stio n
c a p tu re
P re -c o m b u stio n
c a p tu re e x c e p t F C C )
T
P
A

C
O
2
0
E m itte d
C a p tu re d
in backup systems (which will not
offer a carbon saving) or energy
storage.
If we combined natural gas fring
and low-carbon power purchase,
we can achieve an emissions reduc-
tion by almost 40% from the
original base case. To make further
reductions in emissions, other, less
conventional options need to be
investigated. These could include:
Renewable sources of heat, such
as heat from concentrated solar
power (CSP) or electrical heating
from green power
Carbon capture
Figure 4 Natural gas breakeven price at varying carbon price
Figure 5 Impact on carbon emissions of electrical power purchase
Figure 6 Impact of carbon capture
fw.indd 3 12/12/12 12:13:35
costs of this are high and signifcant
levels of power are required.
Therefore, this is not considered
further here.
A number of biomass-based fuels
could potentially be used in the
refnery, with everything from
wood chips to algae being potential
fuels for the utilities boilers. For the
process units, converting to solid
fuels would be more challenging:
so, in order to keep the assump-
tions simple, we will look at an
example where some of the
purchased natural gas is replaced
with bio-derived methane. This
could be produced from anaerobic
digestion or by methanation of bio-
derived syngas.
The production of the biogas uses
energy and, because this is not all
carbon-free, there are still some
emissions associated with using this
gas. In our example where we
combine power import, natural gas
import and full post-combustion
capture, our refnery still emits
around 140 000 t/y of CO
2
(about
8% of the original emissions). To
get to a zero net carbon refnery,
this amount would need to be
equivalent to the bio-derived
content of the biogas.
Figure 7 shows the emissions on
the zero net refnery. Table 2 shows
the level of biogas fring required at
different levels of carbon reduction
achieved by the biogas.
From Table 2, we can see that at
high rates of replacement of natural
gas with biogas, it is possible to
actually achieve negative overall
emissions from the refnery and, in
effect, decarbonise some of the
product, although the impact on
products is small, with even the
most optimistic case reducing emis-
sions from the products by about
only 2.5%.
Conclusions
We have seen that a number of
options exist for refners to reduce
the carbon emissions from their
refneries. The technologies already
exist to develop a refnery that has
zero net emissions. The challenge in
achieving this is not technical, but
commercial, with refners requiring
suffcient incentive and capital to
invest, and suffcient protection
Use of biomass-derived or other
low-net carbon fuels.
There are several potential
renewable sources of heat that
could be used in a refnery. Green
power could be used in electrical
heaters, and CSP or geothermal
power could be used as sources of
heat to generate steam. These
options have not been considered
in detail in this article, but could
offer opportunities for refners
with easy access to abundant and
cost-effective sources of these types
of energy.
Carbon capture
There are a number of ways that
carbon capture can be integrated
into the refnery. All three of the
commonly proposed capture
schemes (pre-combustion, post-
combustion and oxyfuel capture)
can be used in the refnery, with all
three schemes technically viable in
some of the refnery processes.
Carbon capture can typically
recover up to 90% of the carbon in
the fuel and can enable a refnery to
get much closer to a zero carbon
target. We will look at two options.
The frst one sees us change all of
the fred heaters in the refnery to
fre on hydrogen and capture the
CO
2
produced in making hydrogen
(pre-combustion capture). For the
second, post-combustion capture
can be added on all sources of
carbon from the refnery.
Carbon capture can deliver a
large reduction in emissions, but it
still cannot achieve 100% carbon
emission-free processing in the
refnery. Even combining capture
with the other options considered,
using natural gas as fuel, importing
power and implementing CCS,
around 8% of the carbon emitted in
the base case is still emitted. Of
course, consideration also needs to
be given to where the CO
2
is stored,
but this is outside the scope of this
article.
To achieve a net emission of no
fossil-derived carbon, we need
alternative fuel sources that are not
fossil-fuel based. The obvious
choice here is biomass-based fuels.
The carbon from these sources has
been captured from the atmosphere
and will be captured again as plants
grow, absorbing the emissions. An
alternative option could be to use
hydrogen produced from electroly-
sis using green electricity, but the
2 0 0 , 0 0 0
4 0 0 , 0 0 0
6 0 0 , 0 0 0
8 0 0 , 0 0 0
1 , 0 0 0 , 0 0 0
1 , 2 0 0 , 0 0 0
1 , 4 0 0 , 0 0 0
C a rb o n
e m issio n s
T
P
A

C
O
2
0
C a p tu re d
B io -d e rive d
Figure 7 Carbon emissions from the zero
net carbon refnery
Net emissions (TPA) % natural gas from biomass
when using biogas 10% 20% 50% 70% 100%
Biogas emissions as 0% 68 830 886 -202 944 -338 830 -542 660
% of natural gas 20% 82 418 28 064 -135 000 -243 710 -406 774
50% 102 801 68 830 -33 085 -101 029 -202 944
80% 123 184 109 595 68 830 41 652 886
Impact of bio-derived gas and carbon capture on emissions
Table 2
Carbon capture can
typically recover up
to 90% of the carbon
in the fuel and can
enable a refnery to
get much closer to a
zero carbon target
fw.indd 4 12/12/12 12:13:49
against the impacts that lower
carbon emissions could have on
their competitive position in their
domestic and export markets.
Current high oil prices mean that
the returns on energy efciency are
good and most reneries are likely
to nd some attractive investments
in this area. Where natural gas is
available, changing from fuel oil
ring to natural gas ring is also
likely to be attractive. Steps beyond
these options are generally less
attractive nancially unless there is
a sufciently high carbon price or
appropriate level of government
subsidy.
Previous Foster Wheeler studies
have shown a carbon price of
around $60/t CO
2
is required to
justify the investment in carbon
capture. The higher cost of
biomass-derived fuels means that
similar or higher levels of subsidy/
carbon tax are required to
make this option economically
attractive.
This results in a hierarchy of
actions for any rener looking to
reduce carbon emissions:
Maximise design efciency and
operate efciently
Minimise the carbon in the fossils
fuels you use
Capture the carbon you cannot
avoid and/or move to low-carbon
fuels and power.
The right solution will, of course,
depend upon the specic circum-
stances and business objectives of a
particular renery, and the full
range of options need to be studied
in order to develop a robust techni-
cal solution and optimised
investment plan.
Mike Stockle is Chief Engineer of Rening
Technology with Foster Wheeler and currently
works in the Business Solutions Group in
Reading, UK. During his time at Foster Wheeler,
he has worked on a number of rening projects,
ranging from a grassroots renery conguration
studies and FEEDs, through major renery
revamps, to a number of small studies. He is an
experienced LP modeller and has undertaken
a number of feasibility and front-end studies
looking at the impacts of changing markets
and legislation on reneries across the globe.
He holds an MEng in chemical engineering
from Nottingham University and is a Chartered
Engineer and a Fellow of the IChemE.
Email: michael_stockle@fwuk.fwc.com
Current high oil prices
mean that the returns
on energy efciency
are good and most
reneries are likely to
nd some attractive
investments in
this area
Tripl e eccentric
Butterfl y Val ves
Check Val ves
Doubl e Bl ock and Bl eed
ESD Val ves
www.zwick-armaturen.de
fw.indd 5 12/12/12 12:13:59
C O M P R E S S O R S T U R B I N E S G L O B A L S E R V I C E
EBARA CORPORATION
w w w .elliott-turbo.com
Cust omer:
Ethylene plant, Saudi Arabia.
Chal l enge:
O verhaul and repair four critical steam turbines in 12 days.
Resul t :
Elliott coordinated six m onths of pre-planning
and com pleted the project four days early.
They turned to Elliott
for planning and execution.
The custom er turned to Elliott because they understood that planning is everything
and that few com panies can m atch Elliott in turnaround planning and execution.
W ho w ill you turn to?
The world turns to Elliott.
elliott.indd 1 1/6/12 21:23:46
Modular design of smaller-scale GTL
plants
M
odularisation is a construc-
tion method that involves
designing and building
refnery and gas processing facili-
ties as series of unitised (or
modularised) process and utility
systems. The units are designed
and fabricated in a controlled shop
environment, then shipped to the
project site, where they are installed
on foundations and connected
together. Compared to traditional
feld construction methods, where
all materials are delivered to the
project site and then built on loca-
tion, modularisation is proving to
be an effective solution for achiev-
ing fast-track project schedules, and
for building refning and gas
processing facilities in remote
locations. For example, Ventech
Engineers International LLC
(Ventech), based in Pasadena,
Texas, is an engineering, procure-
ment and construction company
that specialises in the design and
fabrication of modularised facilities.
A project in Siberia illustrates
the capabilities of modularisation
when dealing with remote and
challenging environments. Ventech
designed, fabricated and shipped a
2000 b/d crude processing plant to
the remote Siberian town of Urai in
just six months after contract sign-
ing. Urais temperatures drop as
low as -45C (-50F), so the modules
were designed and fabricated to
accommodate these conditions. In
particular, the crude process
module was completely enclosed
and heated. Ventech personnel,
assisted by local labour, installed
the modules at the project site and
started refnery operations four
weeks after the modules delivery.
Modularisation is opening up opportunities for GTL plants to monetise small-scale
natural gas deposits both on- and offshore
KenneTh RobeRTs
Velocys
Ventech has applied modular
design and construction most
recently in the Kurdistan region of
northern Iraq. A large-scale expan-
sion to a refnery in Kurdistan
started during 2010, with Ventech
providing a 20 000 b/d modular-
ised crude processing plant that
was added to the existing 20 000
b/d refnery. Made up of
26 modules, the new crude unit
was designed and fabricated at
Ventechs Pasadena fabrication site.
The modules were shipped to the
port of Mersin, Turkey, then trans-
ported by truck to the project site
in Kurdistan, where the refnery
was erected in 16 weeks.
The refnery has continued to
expand in a modular fashion; once
completed in 2013, total capacity
will be over 185 000 b/d and it
will remain the sole producer of
unleaded gasoline in Iraq. In all,
Ventech has provided modularised
crude distillation units, naphtha
hydrotreaters, catalytic reformers,
isomerisation units, demercaptani-
sation systems, gas plants and
supporting utilities to this three-
phase project. Modularisation has
made the difference in this
geographically and politically chal-
lenging location, enabling expansion
work to be accomplished quickly.
Design concepts for modularisation
In modular plant design, the differ-
ent process units are contained in
separate modules. For ease of ship-
ment to any location, Ventech
fabricates its modules to a standard
size of 13.5ft wide x 12ft high x 40ft
long. Each of the modules and
its associated equipment has a
reference code, which helps
Figure 1 Fischer-Tropsch module process unit Courtesy of Ventech
velocys.indd 1 10/12/12 14:27:48
communicate where the module
will be installed on the plot plan.
Connections between the modules
are designed to be similar in confg-
uration so that construction is
relatively straightforward. Ventech
estimates that, with modularisation,
approximately 70% of a project is
already complete even before the
modules are shipped from their
facility. This greatly decreases feld
construction time to deliver an
operational facility (see Figure 1).
These methods also facilitate easy
disassembly and relocation, if
necessary, at some point in the
future. For example, a remotely
located gas processing facility could
be easily taken apart and moved to
a new natural gas source if an exist-
ing supply was depleted in its
current location.
Applying modularisation to refn-
ery construction has advantages
with regard to productivity, prod-
uct quality and ensuring the safety
of construction personnel. Since the
modules are built in a well-lit,
climate-controlled environment,
work can continue around the clock
regardless of weather conditions,
for greater productivity and easier
quality control. Since module height
is restricted, safety is enhanced, as
workers build at limited heights
within the fabrication facility.
Modularising GTL
The same advantages of modular
construction of refneries are being
applied to the construction of
distributed GTL plants. The GTL
process involves two operations: the
conversion of natural gas to a
mixture of carbon monoxide (CO)
and hydrogen (H
2
), known as
syngas, followed by a Fischer-
Tropsch (FT) process to convert the
syngas into paraffnic hydrocarbons
that can be further refned to
produce a wide range of
hydrocarbon-based products, includ-
ing clean-burning, sulphur-free
diesel and jet fuel. Speciality prod-
ucts including food-grade waxes,
solvents and lubricants can also be
produced from the paraffnic
hydrocarbons.
Large, commercial-scale GTL
plants, including the Sasol Oryx and
the Shell Pearl plants (both located
in Qatar), have been built at enor-
mous capital cost. The Oryx plant,
designed for production levels of
34 000 b/d, cost around $1.5 billion
to build. The Shell Pearl plant, with
an ultimate design capacity of 140
000 b/d of GTL products and 120
b/d of natural gas liquids, cost
around $18-19 billion. Conventional
GTL plant designs rely on econo-
mies of scale to drive positive
fnancial returns and are viable only
where there are large supplies of
low-priced natural gas.
However, another option being
developed smaller-sized and
distributed GTL plants shows
promise for deriving value from
smaller accumulations of unconven-
tional gas that would otherwise be
left underground, such as shale gas,
tight gas, coal bed methane and
stranded gas (gas felds located too
far from existing pipeline infra-
structure). A small, modularised
GTL plant has the fexibility to be
installed close to the trapped
resource and then used to process
that resource locally. Associated gas
(gas produced along with oil) is
another area of opportunity for
modularised GTL plants. This gas
is typically disposed of either by re-
injection, at considerable expense,
back into the reservoir or by the
wasteful and environmentally
damaging practice of faring, which
is subject to increasing regulation.
Modularised GTL plants enable this
otherwise wasted gas to be
converted into additional revenue.
In the larger economic picture, a
modular GTL capability can be the
key factor that enables the construc-
tion of upstream projects that
would otherwise be cancelled
because of poor results derived
from economic models. For exam-
ple, some shale gas discoveries are
being hampered by high develop-
ment costs, which result in marginal
economics due to gas prices that
are often low. These projects can be
enhanced by converting the gas to
higher-value clean fuels produced
in the GTL process.
ITW
Innovative
Technologies
Worldwide
ITW is a fast growing Company, marketing and implementing unique and patented Production Units Online
Cleaning, Tank Cleaning, Decontamination and Reclamation technologies, along with Specialty Chemicals.
Given the considerable success, ITW is expanding its markets and activities and is looking for experienced
professionals worldwide. The candidates should have minimum a 5 years experience in at least one of these
fields:
refining/petrochemicals process specialty chemicals sales
refining/petrochemicals process technology, operations, maintenance, turnaround
and should be willing to travel nationwide and worldwide, along with being prone to hard working and sales.
Having a technical degree and good market knowledge is also required.
The available positions will cover: technical sales, implementation of ITW technologies on the field,
sales and operations management.
Interesting compensation plans will be given along with serious career possibilities. Please contact:
ITW S.r.l. S.Cusumano 96011 Augusta - Italy jobs@itwtechnologies.com www.itwtechnologies.com
HIRING EXPERIENCED PROFESSIONALS WORLDWIDE
velocys.indd 2 10/12/12 14:27:56
Distributed GTL plants, using
smaller Fisher-Tropsch (FT) reactors
and located near the natural gas
source, can be designed for use in
both offshore and remote on-shore
locations. The FT reactors devel-
oped by Velocys, the US-based
member of the Oxford Catalysts
Group, employ microchannel tech-
nology. This is a developing feld of
chemical processing that intensifes
chemical reactions and reduces the
dimensions of the reactor systems
by increasing the heat removal
capability to enable reactions to
occur at rates 10-1000 times faster
than in conventional systems. This
makes it possible to use more active
FT catalysts.
Velocys reactors are the only FT
reactors to use a highly active FT
catalyst developed by Oxford
Catalysts to accelerate FT reactions
by a factor of 10-15 times compared
to conventional reactors. Individual
reactors are designed to produce
125 b/d of FT product, and desired
plant capacity is reached by linking
together, or numbering up, multiple
reactors to meet the availability of
gas feed. For example, by linking
up 40 individual 125 b/d reactors,
a capacity of 5000 b/d can be
achieved with a natural gas feed
rate of approximately 50 million cu
ft/day. As a result, the distributed
GTL plant technology developed by
Velocys has the fexibility to scale
production to match the available
resource. Plants of this type can
also be used as a fexible and
economical way to expand the
capacity at existing petroleum refn-
eries when an economic gas supply
is available.
Velocys is currently working with
Haldor Topse and Ventech to
design a modular GTL plant using
its microchannel FT reactor (see
Figure 2). The modular GTL plant
will be comprised of standard-sized
13.5 x 12 x 40ft (4.1 x 3.65 x 12m)
modules that will be built in
Ventechs fabrication facility. In
these GTL plant designs, autother-
mal reforming (ATR) reactors from
Haldor Topse will be used to
produce the syngas feedstock for
the Velocys FT reactors. Haldor
Topse is also providing the hydro-
processing unit that takes raw
products from the FT reactor and
produces fnished fuels.
First commercial plant
The frst company to select the
modular GTL plant design is
Calumet Specialty Products Partners
L.P., an independent producer of
speciality hydrocarbon products in
North America. Calumet plans to
use the modular GTL plant, which
has a nominal capacity of 1000 bpd,
in the expansion of its Specialty
Products facility at Karns City,
Pennsylvania. The very pure,
paraffnic hydrocarbons produced
by the GTL plant will be used as
feedstock for some of Calumets
1500 products, including food-grade
waxes, solvents and lubricants.
Calumet was attracted to the idea
of modular GTL technology because
it offers the ability to use readily
available, low-cost natural gas to
reduce overall costs, increase the
security of feedstock supply and
improve product quality.
The GTL plants modules will be
fabricated in Ventechs Texas fabri-
cation facility, then transported by
truck to the Karns City site for
installation and integration with
Calumets existing refnery. Velocys
engineers are working with Ventech
and Haldor Topse on the plant
design, which is due be completed
by late 2012. Fabrication is expected
to begin during the frst half of 2013
and the modular GTL plant would
come on-stream during the second
half of 2014.
More projects
Meanwhile, both Velocys and
Ventech expect that the Calumet
project will be just the frst of many
small-scale modular GTL plants
announced, several of which are
already undergoing preliminary
design. Velocys continues to receive
numerous inquiries about the
microchannel technology, not only
for distributed GTL but also for
biomass-to-liquids (BTL) applica-
tions. For example, Solena has
recently chosen the Velocys FT
process for its GreenSky London
Project, which will produce renew-
able jet fuel in the UK.
Given the low natural gas prices
in the US, and the small and often
remote nature of many gas deposits
around the world, interest in modu-
lar distributed GTL plants
incorporating microchannel FT reac-
tors is expected to grow. Once the
Calumet GTL plant comes on
stream, Velocys, Ventech and Haldor
Topse are confdent that such
plants will prove to be the best way
forward for monetising stranded
and undervalued natural gas.
Kenneth Roberts is Senior Business
Development Director with Velocys, Houston,
Texas. With over 35 years experience in the
energy industry, he supports engineering
and business development activities in
collaboration with Ventech and Haldor Topse
for the modular GTL design. He holds a BS in
mechanical engineering and a MBA in strategic
planning from the University of Texas at
Austin.
H ig h h e a t flu x
1 0 tim e s h ig h e r th a n
c o n ve n tio n a l re a c to rs
S yn g a s
F T p ro d u c ts
0 . 2 5 to 1 0 m m
0 . 2 5 to 1 0 m m
Figure 2 Microchannel FT reactor internals Courtesy of Oxford Catalysts Group
velocys.indd 3 10/12/12 14:28:05
Commerci al l y Proven Route for the Producti on of Gasol i ne through
Methanol Synthesi s and Conversi on
ExxonMobil Research and Engineering Companys Methanol to Gasoline (MTG) Technology, was commercially operated in
New Zealand for over 10 years. Now we offer the second generation MTG technology which provides:
8ulphur-free low oenzene gasoline
A product full] compatiole with refner] gasoline
8imple, reliaole fxed oed reactor design
So Advanc ed,
Yet So Si mpl e.
3225 Gallows Road, Fairfax, virginia 22037-0001, USA
+1-703-846-2568 fax +1-703-846-3872
www.exxonmobil.com/refningtechnologies tsl.email@exxonmobil.com
exxon.indd 1 10/12/12 11:14:52
Starting up hydrotreating units with
cracked feeds
R
efners can realise signifcant
operating margins by process-
ing cracked feedstocks such as
coker naphtha and light cycle oil
(LCO). Unfortunately, after catalyst
change-outs on hydrotreating
(HDS) units, catalyst manufacturers
recommend that the refner only
process straight-run feeds for at
least the frst three days of the cata-
lyst cycle. This delay in processing
cracked feeds has several potential
costs including:
Lost profts from processing
cracked and heavy feeds
Storage costs for a three-day
supply of straight-run feed
Storage costs for three days of
cracked feed that must be held
aside for processing later
The purchase of additional
straight-run feed if a three-day
supply is not available.
Problem with cracked feeds
Cracked feeds such as coker naph-
tha contain signifcant quantities of
highly reactive olefns and diole-
fns, while cracked gas oils also
contain polynuclear aromatics. If
these components are exposed to
the hyperactive catalytic sites that
exist on freshly sulphided cata-
lysts, coke and gums will form that
readily deposit on the catalyst
surface. Such deposits tend to
obstruct the catalysts pores and
inhibit access to the active sites,
leading to a permanent loss of
catalyst activity and a reduction in
cycle length.
To avoid the adverse effects of
cracked feeds at startup, catalyst
manufacturers typically recommend
a break-in period of at least three
days, feeding only straight-run
A catalyst sulphiding process enables the introduction of cracked or heavy feeds
to HDS units without the recommended three-day delay following changeover
RANDY ALEXANDER, REiNDER VAN BRuggEN and PiERRE DufRESNE
Eurecat US
product to the unit. During this
three-day break-in period, a small
amount of soft coke will form on
the catalyst surface, gradually
tempering the hyperactivity of the
catalyst. Once catalyst activity has
stabilised, cracked feeds can be
introduced to the unit without the
risk of premature coke and gum
formation.
Eurecat has developed an ex-situ
treatment that allows the introduc-
tion of cracked or heavy feeds
without the recommended three-
day delay. This process, called
Totsucat CFP, combines the
Totsucat activation process with an
additional step to protect the cata-
lyst from the harmful effects of
cracked feed, ensuring long-term
reactor performance irrespective of
when cracked feeds are introduced
to the unit.
This is a carefully controlled
procedure that gently moderates
the hyperactivity that occurs imme-
diately after catalyst sulphiding.
Catalysts treated with this are
fully sulphided and activated prior
to loading. The activity will mimic a
catalyst that has already been condi-
tioned with straight-run feed for
three days. Cracked and heavy feed-
stocks can be gradually introduced
during the frst hours of a startup
without causing permanent damage
to the catalyst. Coke and gum
deposits will be minimised, since the
catalyst activity has been carefully
stabilised following activation.
Recent unit data also show that
this treatment can signifcantly
reduce the deactivation rate of a
hydrotreating unit, allowing the
refner to extend their cycle by as
much as 50%.
Totsucat catalyst activation process
Unlike with other pre-sulphurisa-
tion processes, Totsucat-treated
catalysts are delivered to the HDS
unit totally activated and ready to
work immediately. The uncertainty
of incomplete activation is elimi-
nated and startup time is greatly
reduced.
The advantages of the process
include:
Load-and-go reactor startups. The
catalyst is fully sulphided and acti-
vated and does not require any
additional sulphiding agents or
complex activation procedures
No sulphiding exotherms since
the sulphiding step is completed in
Eurecats equipment
No need for additional hydrogen
at startup
Minimal sour water formation
during startup
The hazards and odours associ-
ated with sulphiding chemicals are
avoided since spiking agents are no
longer needed at startup
No need to measure H
2
S levels in
the recycle gas
Catalyst
manufacturers
recommend a break-
in period of at least
three days, feeding
only straight-run
product to the unit
eurecat.indd 1 10/12/12 14:32:54
Negligible amounts of H
2
S are
released during reactor heat-up,
protecting sulphur-sensitive units
downstream and avoiding the need
to are undesirable byproducts
Circulating high concentrations of
wet H
2
S during in-situ sulphiding
can cause severe corrosion prob-
lems. Totsucat eliminates this issue
since only low concentrations of
dry H
2
S will be recirculating
through the system.
Starting a unit with Totsucat-
treated catalysts is similar to a
restart after an emergency shut-
down. Once the reactor is loaded,
it only requires heating up to the
start-of-run temperature. Time-
consuming drying steps, holding
steps and tests to determine H
2
S
concentrations are not required.
Startup time is reduced to a few
hours, while startup risks are mini-
mised. The catalyst bed will
achieve peak performance since
sulphiding and activation are
controlled at very precise
conditions.
The treatment is particularly
useful for:
Critical path units where lost
production time is costly
Units with temperature limits,
where effective sulphiding is dif-
cult, if not impossible, to achieve
Reneries with limited hydrogen
supplies
HDS units that run in parallel or
series, where only one unit is taken
down while the other is still
operating
Units that cannot tolerate H
2
S
breakthrough to downstream units
Units that utilise gas-phase star-
tup, increasing the risk of
catastrophic exotherms during in-
situ sulphiding
Units where the use of sulphiding
chemicals is undesirable or
uneconomical.
Case study 1
A diesel hydrotreater (DHT) located
on the US East Coast was loaded
with Totsucat CFP-activated cata-
lyst in the summer of 2011. This
moderate-pressure unit (800 psi
with a hydrogen partial pressure of
6 8 0
7 2 0
7 1 0
7 0 0
6 9 0
6 7 0
6 6 0
6 5 0
6 4 0
6 3 0
0 2 0 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0 2 6 0 2 8 0
T
e
m
p
e
r
a
t
u
r
e
,

F
6 2 0
C yc le 2 5 n o rm a lise d
C yc le 2 5 a c tu a l
C yc le 2 4 n o rm a lise d
y = 0 . 1 6 4 2 x +6 3 4 . 5 1
R
2
= 0 . 4 6 8 2
Figure 1 Reactor temperature (WABT) after 90 days on oil
7 2 0
7 0 0
6 8 0
6 6 0
6 4 0
6 2 0
0 3 0 6 0 9 0 1 2 0 1 5 0 1 8 0 2 1 0
T
e
m
p
e
r
a
t
u
r
e
,

F
Days on st ream
6 0 0
A c tu a l c yc le
N o rm a lise d c u rre n t c yc le T-re q .
N o rm a lise d p re vio u s c yc le T-re q .
D e a c tiva tio n = 3 . 7 F /m o n th
S O R WA B T = 6 4 5 F
P re vio u s d e a c tiva tio n = 8 . 9 F /m o n th
7 2 0
7 6 0
7 4 0
7 0 0
6 8 0
6 6 0
6 4 0
6 2 0
0 3 0 6 0 9 0 1 2 0 1 5 0 1 8 0 2 1 0 2 4 0 2 7 0 3 0 0 3 3 0
T
e
m
p
e
r
a
t
u
r
e
,

F
Days on st ream
6 0 0
A c tu a l c yc le
N o rm a lise d c u rre n t c yc le T-re q .
N o rm a lise d p re vio u s c yc le T-re q .
C u rre n t c yc le d e a c tiva tio n = 4 . 2 F /m o n th
S O R WA B T = 6 4 5 F
P re vio u s c yc le d e a c tiva tio n = 7 . 8 F /m o n th
Circulating high
concentrations
of wet H
2
S during
in-situ sulphiding
can cause severe
corrosion problems
eurecat.indd 2 10/12/12 14:33:07
A B B Lifecycle S ervices. N o one else know s your
system s like w e do.
W ith over 30 years of autom ation system heritage, no one else know s your A B B
control system s better than w e do! A nd w ith our outstanding lifecycle policies and
program s, system expertise, evolution options, value added services and determ ined
com m itm ent to our custom ers, w e can help you evolve to the latest in control system
technology w ith S ystem 800xA . S o, if you have an IN FI90, M O D 300, A dvant, D C I, or
other control system , w e are here to m ake sure your control system w ill continue to
m eet your business needs now and in the future. For m ore inform ation:
w w w. a b b . c o m /c o n tro ls y s te m s
PTQ_Sweden_Evol_A4.indd 1 08/09/2011 12:30 abb auto.indd 1 10/12/12 11:07:28
550-600 psi) processes up to 46 000
b/d of a mix of 35-40% straight-run
feed and 60-65% cracked stocks,
including 30% LCO. The reactor
holds 106 000 lb (48 tons) of CoMo
catalyst. Due to the severity of the
feeds being processed and the size
of the reactor, the rener typically
achieves cycle lengths of less than
12 months before the end-of-run
(EOR) temperature is reached, forc-
ing a unit shutdown and a catalyst
change.
The following data compare the
reactor weighted average bed
temperature (WABT) for two cycles:
one utilising Totsucat D-treated
catalyst (Cycle 24) and another
where Totsucat CFP was used to
activate and protect the catalyst
(Cycle 25). Figure 1 shows the
WABT for the unit after 90 days.
Cycle 24 was considered a success-
ful run since it was one of the
longest cycles achieved by this unit.
But, in Cycle 25, LCO was intro-
duced in signicant quantities on
Day 2. After 90 days on oil, Cycle
25 exhibited a lower deactivation
rate than Cycle 24.
Figure 2 conrms that the deacti-
vation rate of the catalyst remains
much lower with Cycle 25 at
180 days on oil. The rate was only
4F (2.2C) per month versus 9F
(5C) per month for the previous
cycle.
Figure 3 shows that the trend
continued past the normal EOR for
this unit. Unit run data show that
the lower deactivation rate with
Cycle 25 continued until EOR. This
allowed the rener to extend the
cycle to 420 days, which is approxi-
mately 60% longer than anticipated
for this catalyst ll.
In comparing the two cycles, the
rener explained that Cycle 25 was
actually more severe than Cycle 24.
For Cycle 25, an additional
325 000 barrels of LCO had been
accumulated offsite during the
turnaround and required treatment.
Therefore, in addition to the normal
amount of LCO processed, an addi-
tional 3000-5000 b/d was fed to the
reactor for the rst 90 days of Cycle
25. The rener also explained that
this unit had averaged a deactiva-
tion rate of 20F (11C) per month
over the last 10 cycles.
In the past, the catalyst manufac-
turer recommended that a gentle
and lengthy startup should be
employed for this unit in order to
avoid premature activity loss.
In this case, Totsucat CFP has
proven that a run can be signi-
cantly extended without the
need for a gentle startup, allowing
the rener to gain signicant
6 0 0
7 0 0
6 5 0
5 5 0
5 0 0
0 5 0 1 0 0 1 5 0 2 0 0 2 5 0 3 0 0 3 5 0
W
A
B
T
,

F
Days on oi l
4 5 0
To tsu c a t C F P R u n 2
To tsu c a t C F P R u n 1
I n situ with b re a k in
Figure 4 Reactor temperature (WABT)
1 2 0
1 4 0
1 0 0
8 0
6 0
4 0
2 0
0 5 0 1 0 0 1 5 0 2 0 0 2 5 0 3 0 0 3 5 0
T
e
m
p
e
r
a
t
u
r
e
,

F
Days on oi l
0
I n le t to b o tto m T To tsu c a t C F P R u n 2
I n le t to b o tto m T To tsu c a t C F P R u n 1
I n le t to b o tto m T in situ b re a k in
Figure 5 Inlet to bottom T
120
140
100
80
60
40
20
0 50 100 150 200 250 300 350

0
0

b
/
d
,
d
e
e
f

a
h
t
h
p
a
n

r
e
k
o
C
Days on oil
0
Coker naphtha - Totsucat CFP Run 2
Coker naphtha - Totsucat CFP Run 1
Coker naphtha - in situ break in
Figure 6 Coker naphtha feed rate
eurecat.indd 3 12/12/12 10:59:06
exhibited a greatly reduced deacti-
vation rate, extending the unit cycle
length by 60%. And in a coker
naphtha unit, it helped extend unit
cycle length by over 50%.
Randy Alexander is the Director of Sales and
Marketing for Eurecat US, Houston, Texas. He
has over 25 years experience in the areas of
hydrotreating and hydrogenation catalysts, is
responsible for sales and marketing of catalyst
sulphiding and regeneration services and for
directing new product development, and holds
a bachelors degree in chemical engineering
from the University of Texas and a MBA from
the University of Houston.
Email: ralexander@eurecat.com.
Reinder van Bruggen joined Eurecat France in
2003 as Area Sales Manager and is currently
Totsucat Product Manager. He holds an MSc in
chemical engineering from Twente University
of Technology, The Netherlands.
Pierre Dufresne is Vice President, Research and
Development, for Eurecat SA. He holds a degree
from the Chemistry School of Mulhouse and a
PhD from the University of Lille. He has co-
authored around 70 patents.
Email: p.dufresne@eurecat.fr
margins by processing much more
LCO.
Case study 2
A naphtha hydrotreater (NHT)
located in a Gulf Coast renery is
now on its eighth cycle using
Totsucat CFP to pre-activate the
catalyst. The unit processes 35 000
b/d of feed consisting of 25-50%
coker naphtha. Due to limited stor-
age capacity, the rener struggled
to delay the introduction of cracked
feeds and typically brought them in
prior to the recommended three-
day break-in period.
Figure 4 shows the reactor
temperature for a typical cycle as
compared to two cycles utilising
Totsucat CFP ex-situ activation.
The rst run, dened as in-situ
with break-in, utilised a catalyst
that was sulphided in-situ and
then fed only virgin feed for three
days. For the other two cycles, the
reactor was loaded with treated
catalysts and progressively fed
coker naphtha during the rst day
of the cycle. Both of these latter
runs showed steady performance.
Figure 5 denotes the temperature
difference across the reactor during
each cycle. On average, Totsucat
CFP provided a temperature gradi-
ent across the bed similar to the
in-situ cycle.
Figure 6 shows that the amount
of coker naphtha fed to the unit
during each run was comparable.
The reners experience shows
that using Totsucat CFP pre-
activated catalysts can extend run
length without the need to delay
cracked feeds for three days at the
beginning of the cycle.
Conclusion
The addition of CFP to the Totsucat
process enables the rener to begin
introducing cracked stocks to their
hydrotreater on the rst day of a
unit startup without the need for
the three-day break-in period utilis-
ing only straight-run feed. In a
moderate-pressure diesel HDS unit,
catalyst treated with this process
Taking the Guesswork Out of CO
2
Removal
ProTreat Simulation Software
Get Real
Dont Guess!
SYNGAS, LNG AND CARBON CAPTURE
ProTreat accurately predicts column performance.
100% Mass-Transfer-Rate based no ideal stages.
Grassroots? Revamp? Designing for Packing?
Using MDEA with piperazine? HotPot?
Build a virtual plant with the industrys
most advanced gas treating simulator.
www.ogt rt .com +1 281 970 2700 info@ogt rt .com
Learn how state-of-the-art simulation science can beneft you.
OGT-0117 PTQ Half Horizontal Ad.indd 1 11/13/12 10:40 PM
eurecat.indd 4 12/12/12 10:58:27
Don t mi ss out
on t he i ndust r y s
most i mpor t ant
event of 2013
Marriott Rivercenter
San Antonio, Texas
March 17 19, 2013
The AFPM Annual
Meeting brings
together the rening
industrys most
important players
each year. Dont miss
this opportunity to be
a part of the action.
The 2013 program has something
for everyone from process
and operations engineers to
senior management. Keynote
sessions feature top government
ofcials and corporate CEOs.
Breakout sessions cover nearly
every facet of rening technology,
as well as public policy
issues impacting the industry.
Dont miss the networking
opportunities at our receptions
and afliate- hosted events.
Be there and be sure to ex tend
your reach within the industry.
Register at www.afpm.org.
afpm.org
EXTEND
YOUR REACH
ATTEND THE 2013
AFPM ANNUAL MEETING
afpm.indd 1 6/12/12 12:54:27
Optimising a process scheme for
platforming heaters
C
oncern over safety issues,
especially in industries such
as rening, lead operating
personnel to be more and more
reluctant to operate their plants
near design or permissible limits,
thereby avoiding the risk of mishap
or industrial accident. On the other
hand, they are under pressure to
increase their prot margins.
CCR platforming heaters in an
operating oil renery have been in
operation for the last 30 years or so.
Due to the existing process scheme,
a thermal imbalance in all four of
the CCR platforming units heaters
had been observed in terms of
permissible limits for tube metal
temperatures and uneven bridge-
wall temperatures. The operator
decided to engage a consultant to
analyse the problem and to come
up with an optimised process
scheme, thereby avoiding operation
of the furnaces near the design
limits of the process coils.
The study was carried out based
on various document and data
provided for the existing system in
the form of DCS output to obtain
full details of the process.
Methodology of study
The feed property grid was gener-
ated using PRO II and a
thermodynamic method (Grayson
Streed). A reasonable process-side
fouling factor in the convection
zone has been considered, which is
quite likely to happen over a long
period of operation. The CCR plat-
forming heaters were studied using
commercial software (FRNC-5PC
Version 4.18 Mod 7.6) with the
following assumptions and data for
modelling: radiant heat loss of 2.5%;
Thermal imbalance resulting from an existing process scheme for CCR platforming
furnaces is addressed by process schemes for distributing heat load more uniformly
ADIL REHMAN, SUMAN PACHAL, SHYAM K CHOUDHARY, UGRASEN YADAV and M K E PRASAD
Technip KT India
ambient air temperature of 17C
and relative humidity of 95%, as
per the meteorological data
provided by the client; maintaining
the oxygen percentage at the outlet
of the convection section by keep-
ing the quantity of air the same as
in the existing operating case in
order to account for the ingress of
air, which might have taken place
over a long period of operation.
The algorithm followed for the
study is shown in Figure 1.
Denition of uid
Fluid in this case is dened and
identied as process feed entering
the reactor. Thus, process uid enter-
ing Reactor 1 is called Fluid 1,
process feed entering Reactor 2 is
Fluid 2 and process feed entering
Reactor 3 is Fluid 3. This denition
of uid is shown in Figure 2.
Comparison of existing scheme with
proposed scheme
The existing scheme according to
DCS outputs is summarised in the
process ow diagram shown in
Figure 3. The proposed schemes
results obtained from FRNC have
been summarised in the form of
S te p 1
E va lu a tio n a n d c h e c k in g th e
c o m p le te n e ss o f in p u t d a ta
p ro vid e d b y th e c lie n t
S te p 2
U tilise th e c o lle c te d d a ta to
d e ve lo p a b a se c a se
sim u la tio n m o d e l
S te p 3
R e vie w th e re su lts o f th e
sim u la tio n m o d e l fo r th e
a c tu a l c a se
S te p 4
D e ve lo p sim u la tio n m o d e ls
fo r p ro p o se d o p tio n s
S te p 5
R e vie w th e re su lts o f d iffe re n t
sim u la tio n m o d e ls fo r
p ro p o se d o p tio n s
S te p 6
C o m p a riso n o f va rio u s
o p tio n s a n d se le c tio n o f m o st
o p tim ise d o p tio n
S te p 7
P re p a ra tio n o f a stu d y re p o rt
a n d issu e th e re p o rt fo r c lie n t
re vie w a n d c o m m e n ts if a n y
Figure 1 Algorithm followed for the study
P a c k in o x
F lu id 1
F lu id 2 F lu id 3
C o n ve c tio n +
5 0 1 A a n d 5 0 1 B
R e a c to r 1
R e a c to r 2
R e a c to r 3
F u rn a c e 5 0 2
F u rn a c e 5 0 3
Figure 2 Denition of uid
technip.indd 1 10/12/12 14:37:10
process fow diagrams shown in
Figures 4, 5 and 6 for proposed
Options I, II and III, respectively.
Results and discussion
Mechanical aspects
Option I
This scheme is identical to the exist-
ing scheme with regard to the
process fow pass arrangement
inside the heaters. Thus, modifca-
tions inside the heater are minimal
with respect to the existing scheme.
However, the north-side intercon-
necting process header between
501A and 501B needs to be cut in
order to connect the process outlet
of 501A to the inlet of Reactor 1.
Similarly, the outlet of Reactor 2
needs to be connected to the inlet of
501`B. Also, the outlet of 502, which
goes to Reactor 2 in the existing
scheme, needs to be dismantled and
rerouted to the inlet of 503.
Ultimately, this scheme would
require a lot of external mechanical
modifcations and rerouting of hot
piping, cutting of large-sized process
headers and so on, and thus would
require a long shutdown time.
Option II
In this scheme, the interconnecting
north-side header between 501A
and 501B needs to be blinded/
dismantled and the outlet of 501B,
which goes to the inlet of Reactor
1 in the existing scheme, needs to
be routed to Reactor 3. Similarly,
the outlet of Reactor 2, which goes
to the inlet of 503 in the existing
scheme, needs to be rerouted to
the inlet of 501B, and the outlet of
502, which goes to Reactor 2 in the
existing scheme, needs to be
connected to the inlet of 503. The
interconnecting north-side header
between the inlet and the outlet of
502 needs to be dismantled/
blinded. This option would require
changes inside the heater process
fow pass arrangement (single pass
for 501A and 501B in Option II,
against double pass for 501A and
501B in the existing scheme). This
scheme would require a lot of
external rerouting of heater outlet
and inlet piping to and from the
reactors and thus would direction-
ally require a longer shutdown
time compared with Option I.
Option III
This scheme is similar to the exist-
ing scheme with regard to the
process fow arrangement inside
the heaters. The interconnecting
north-side header between 501A
and 501B need not be cut.
However, the Packinox outlet,
which goes to the inlet of 501A in
the existing scheme, now needs to
be routed to the inlet of 502. The
Reactor 1 outlet, which goes to the
inlet of 502 in the existing scheme,
needs to be routed to the inlet of
501A. Similarly, the outlet of 502,
which goes to the inlet of Reactor
2 in the existing scheme, needs to
be routed to the inlet of Reactor 1.
Apart from external rerouting,
there is no cutting/dismantling of
interconnecting headers and no
change in heater internal pass fow
arrangements, with respect to the
existing scheme, in Option III.
Thus, Option III would direction-
ally require minimum mechanical
modifcation and rerouting
compared with Options I and II,
and consequently would require
the least shutdown time.
A b s d u ty 3 . 4 8 M W
Convection
section
Platformer heater
B-501A
D o u b le -p a ss d e sig n )
A b s d u ty 0 . 1 2 M W
M a x . T M T 4 9 9 C
B rid g e wa ll te m p . 4 7 6 C
Platformer heater
B-502
A b s d u ty 9 . 1 M W
M a x . T M T 5 7 5 C
Platformer heater
B-503
S in g le -p a ss d e sig n )
A b s d u ty 3 . 0 5 M W
M a x . T M T 5 6 0 C
Platformer heater
B-501B
A b s d u ty 2 . 1 1 M W
M a x . T M T 5 3 2 C
R e c yc le g a s
3 2 . 2 t/h 8 0 t/h
M ix stre a m 8 0 t/h +3 2 . 2 t/h )
Packinox
C-516
Reactor 1
D-501
Reactor 3
D-503
Reactor 2
D-502
4 7 1 C
3 1 B a rg
5 1 5 C
3 0 . 3 3 B a rg
4 7 9 C
2 9 . 0 5 B a rg
4 3 2 C
2 9 . 6 3 B a rg
5 1 0 C
2 9 . 5 5 B a rg
4 9 8 C
3 0 . 4 6 B a rg
5 0 6 C
2 8 . 9 7 B a rg
4 9 7 C
3 0 . 5 9 B a rg
5 0 1 . 4 C
3 0 . 5 9 B a rg
Figure 3 Existing process scheme
technip.indd 2 10/12/12 14:37:21
A b s d u ty 2 . 0 7 M W
Convection
section
Platformer heater
B-501A
M a x . T M T 5 4 0 C
Platformer heater
B-502
A b s d u ty 7 . 1 4 M W
M a x . T M T 5 4 5 C
Platformer heater
B-503
A b s d u ty 1 . 9 6 M W
M a x . T M T 5 4 4 C
Platformer heater
B-501B
A b s d u ty 3 . 0 5 M W
M a x . T M T 5 3 1 C
R e c yc le g a s
3 2 . 2 t/h
8 0 t/h
M ix stre a m 8 0 t/h +3 2 . 2 t/h )
Packinox
C-516
Reactor 1
D-501
Reactor 3
D-503
Reactor 2
D-502
4 7 1 C
3 1 B a rg
5 1 5 C
3 0 . 4 6 B a rg
4 7 9 C
2 9 . 0 5 B a rg
4 3 2 C
2 9 . 7 6 B a rg
4 9 4 . 3 C
2 9 . 6 9 B a rg
5 1 0 C
2 9 . 6 1 B a rg
4 7 9 C
2 9 . 1 1 B a rg
5 0 6 C
2 8 . 9 7 B a rg
4 9 0 . 1 C
3 0 . 5 9 B a rg
4 9 0 . 1 C
3 0 . 5 9 B a rg
Figure 4 Proposed process scheme (Option I)
A b s d u ty 2 . 0 7 M W
Convection
section
Platformer heater
B-501A
A b s d u ty 3 . 1 1 M W
M a x . T M T 5 5 2 C
Platformer heater
B-502
A b s d u ty 7 . 1 4 M W
M a x . T M T 5 4 5 C
Platformer heater
B-503
A b s d u ty 1 . 9 6 M W
M a x . T M T 5 4 4 C
Platformer heater
B-501B
A b s d u ty 3 . 0 5 M W
M a x . T M T 5 4 2 C
R e c yc le g a s
3 2 . 2 t/h
M ix stre a m 8 0 t/h +3 2 . 2 t/h )
Packinox
C-516
Reactor 1
D-501
Reactor 3
D-503
Reactor 2
D-502
4 7 1 C
3 1 B a rg
5 1 5 C
3 0 . 5 7 B a rg
4 7 9 C
2 9 . 0 5 B a rg
4 3 2 C
2 9 . 8 7 B a rg
4 9 4 . 3 C
2 9 . 7 9 B a rg
5 1 0 C
2 9 . 7 1 B a rg
4 7 9 C
2 9 . 2 1 B a rg
5 0 6 C
2 9 . 1 9 B a rg
4 9 0 . 1 C
3 0 . 5 9 B a rg
4 9 0 . 1 C
3 0 . 5 9 B a rg
8 0 t/h
Figure 5 Proposed process scheme (Option II)
technip.indd 3 10/12/12 14:37:32
Maximum tube skin temperature
The maximum tube skin tempera-
ture encountered in Options I and
II is 545C, well within the permis-
sible tube metal temperature of
559.8C. A comparison of maximum
tube metal temperature is shown in
Figure 7 for various proposed proc-
ess schemes compared with that
encountered in the existing process
scheme.
Bridgewall temperature
Heating for Reactor 1: comparing
bridgewall temperature by fuid
number, we observe that, for Fluid
1, a maximum bridgewall tempera-
ture of 707C is encountered in
Options I and II, while Option III
results in a bridgewall temperature
of 652C
Options I and II, while Option III
results in a maximum bridgewall
temperature of 784C
Option III, while Options I and II
result in a bridgewall temperature
of 698C. A comparison of bridge-
wall temperatures is shown in
Figure 8 for the various proposed
process schemes compared with that
encountered in the existing scheme.
Imbalance in bridgewall
temperature
Imbalance in the bridgewall temper-
ature has been observed in the actual
scheme, with hot fue gases at 631C
for 501A to 822C for 503. This, as
such, is not good for intermediate
tube supports in the convection
section and would result in uneven
thermal expansion problems.
The problem of imbalance in the
bridgewall temperature has been
addressed in Options I and II,
where bridgewall temperatures are
almost in the same range, from
707C for 501A to 761C for 503,
thereby reducing uneven stresses
on heater hardware and minimising
fue gas maldistribution and associ-
ated hot fue gas mixing problems.
The problem of imbalance in the
bridgewall temperature has also
been addressed in Option III, where
bridgewall temperatures are almost
in the same range, from 765C for
501A to 871C for 503, thereby
reducing uneven stresses on heater
hardware and minimising fue gas
maldistribution and associated hot
fue gas mixing problems.
Suitability of furnace hardware
The existing case bridgewall
temperature for 501A is 631C. It
has been found that the bridgewall
temperature for 501A is 765C,
which is the maximum for Option
III, against 707C for Options I and
II. However, a bridgewall tempera-
ture of 765C is well below the
permissible limit of 950C. Thus,
the furnace hardware of 501A is
suitable for all options.
temperature for 501B is 669C. It
has been found that the bridgewall
temperature for 501B is a maximum
784C for Option III, against 698C
Convection
section
Platformer heater
B-501A
A b s d u ty 4 . 5 5 M W
M a x . T M T 5 0 9 C
Platformer heater
B-502
A b s d u ty 2 . 9 8 M W
M a x . T M T 5 3 6 C
Platformer heater
B-503
A b s d u ty 3 . 0 6 M W
M a x . T M T 5 5 9 . 8 C
Platformer heater
B-501B
A b s d u ty 4 . 5 6 M W
M a x . T M T 5 4 7 C
R e c yc le g a s
3 2 . 2 t/h
M ix stre a m 8 0 t/h +3 2 . 2 t/h )
Packinox
C-516
Reactor 1
D-501
Reactor 3
D-503
Reactor 2
D-502
4 7 1 C
3 1 B a rg
5 1 0 C
2 9 . 5 7 B a rg
4 7 9 C
2 9 . 0 7 B a rg
5 1 5 C
3 0 . 5 3 B a rg
4 7 1 C
2 9 . 7 B a rg
5 0 6 C
2 9 B a rg
4 9 1 . 1 C
3 0 . 5 9 B a rg
8 0 t/h
Figure 6 Proposed process scheme (Option III)
technip.indd 4 10/12/12 14:37:42
Beach Rotana Hotel, Abu Dhabi
March 24-28
2013
Organised by
Officially
Supported by
F
O
R

P
R
O
G
R
A
M
M
E
S

&

R
E
G
I
S
T
R
A
T
I
O
N
p
l
e
a
s
e

v
i
s
i
t
w
w
w
.
s
o
g
a
t
.
o
r
g
www.sogat.org
March 24
Workshops
Process Optimisation (Day 1)
Filtration and Separation (Day 1)
Acid Gas Injection
H2S Safety Detection and Survival Solutions
March 25
Workshops
Process Optimisation (Day 2)
Filtration and Separation (Day 2)
Reducing Liquid Flaring and Oil Wastage
during Sour Well Testing and Intervention
Claus andTail Gas Treatment
March 26-28
9
th
International
SOGAT Conference
March 25-28
SOGAT Exhibition
Official
Media Partner
Shaila DSa Dome Exhibitions PO Box 52641 Abu Dhabi, UAE Tel: +971 2 674 4040 Fax: +971 2 672 1217
C
M
Y
CM
MY
CY
CMY
K
SOGAT2013Ad1 BLEED216x303mm for PTQ.pdf 1 12/6/2012 4:23:00 PM
sogat.indd 1 6/12/12 13:30:14
for Options I and II. However, a
bridgewall temperature of 784C is
well below the permissible limit of
950C. Thus, the furnace hardware
of 501B is suitable for all options.
temperature for 502 is 893C. It has
been found that the bridgewall
temperature for 502 is a maximum
814C for Options I and II, against
652C for Option III. Option III
brings about relief to the heater 502
by bringing down the bridgewall
temperature to the lowest among
all three options and is well below
the permissible limit of 950C. Thus,
the furnace hardware of 502 is suit-
able for all options.
temperature for 503 is 822C. It has
been found that the bridgewall
temperature for 503 is a maximum
871C for Option III, against 761C
for Options I and II. However, a
bridgewall temperature of 871C is
well below the permissible limit of
950C. Thus, the furnace hardware
of 503 is suitable for all options.
Burner fexibility and utilisation
It has been found that, for the exist-
ing scheme, the burner for 501A is
operating at a very low load (27.6%
of design load). The total available
facility of 10 burners has not been
utilised and only one burner is put
into operation. Thus, burner fame
stability, fuel gas pressure control,
optimum excess air requirement
and control may be some issues of
concern.
In Options I and II, the burner
needs to be operated at around 62%
of the burner design load, which is
quite good from the point of view
of operations and burner utilisation,
while there is better utilisation of
the 501A burner in Option III,
where the burner load is 96.5% of
design load capacity.
It has been found that for the
existing scheme, the burner for
501B is operating at almost half
load (48.6% of the burner design
load). In all, 9 out of 10 burners are
put to use.
In Options I and II, the burner
needs to be operated at around 61%
of the burner design load, which is
quite good from the point of view
of operations and burner utilisation,
while there is better utilisation of
the 501B burner in Option III,
where the burner load is 98.5% of
design load capacity.
For the existing scheme, burner
502 is operating at around 88% of
design load, which is a good level
of utilisation. All 15 out of 15 burn-
ers are put to use.
In proposed Options I and II, the
burner needs to be operated at
around 64% of the burner design
load, which is quite a good load
from the point of view of opera-
tions and burner utilisation, while
the level of utilisation of the 502
burner in Option III is 23% of
burner design load capacity. Since
5 7 0
6 1 0
5 9 0
5 5 0
5 3 0
5 1 0
4 9 0
4 7 0
E x istin g sc h e m e
F R N C o u tp u t)
F u rn a c e 5 0 3
sin g le p a ss)
F u rn a c e 5 0 1 B
d o u b le p a ss)
sin g le p a ss)
F u rn a c e 5 0 3
sin g le p a ss)
5 6 0
5 3 1
5 4 2
5 5 9 . 8
P ro p o se d
sc h e m e I
P ro p o se d
sc h e m e I I
P ro p o se d
sc h e m e I I I
T
u
b
e

m
e
t
a
l

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

C
Case
4 5 0
5 7 0
6 1 0
5 9 0
5 5 0
5 3 0
5 1 0
4 9 0
4 7 0
F R N C o u tp u t)
F u rn a c e 5 0 2
d o u b le p a ss)
F u rn a c e 5 0 1 A /B
d o u b le p a ss)
5 0 2 d o u b le p a ss)
5 0 3 sin g le p a ss)
P ro p o se d
sc h e m e I
P ro p o se d
sc h e m e I I
P ro p o se d
sc h e m e I I I
T
u
b
e

m
e
t
a
l

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

C
Case
4 5 0
5 7 5
5 4 5 5 4 5
5 4 7
5 7 0
6 1 0
5 9 0
5 5 0
5 3 0
5 1 0
4 9 0
4 7 0
F R N C o u tp u t)
C o n ve c . +5 0 1 A /B
d o u b le p a ss)
5 3 2
5 4 0
5 5 2
5 3 6
C o n ve c . +5 0 1 A
d o u b le p a ss)
C o n ve c . +5 0 1 A
sin g le p a ss)
C o n ve c . +5 0 2
d o u b le p a ss)
P ro p o se d
sc h e m e I
P ro p o se d
sc h e m e I I
P ro p o se d
sc h e m e I I I
T
u
b
e

m
e
t
a
l

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

C
Case
4 5 0
Figure 7 Maximum tube metal temperatures for three process schemes
technip.indd 5 10/12/12 14:37:53
there are 15 burners in 502, a few
burners may be put out of opera-
tion in such a way that uniformity
of heat distribution is maintained
(maybe by using an alternate burner
as far as possible) and keeping the
heat load of the burner above the
turndown limit.
It has been found that, for the
existing scheme, the burner for 503
is operating at 94.5% of burner
design load. In the existing scheme,
fve out of fve burners are put to
use. In proposed Options I and II,
the burner needs to be operated at
around half load (50% of burner
design load). There is better utilisa-
tion of the 503 burner in Option III,
where the burner load is 87% of
burner design load capacity, which
is quite good from operation and
burner utilisation point of view.
Conclusion
Considering mechanical aspects,
including time required for carrying
out the modifcations, and process
aspects, such as maximum tube
metal temperature, thermal imbal-
ance and burner fexibility,
improvements would be best
addressed by carrying out modifca-
tions to the existing process scheme
recommended in proposed Option
III to achieve the desired process
optimisation within the least time.
Acknowledgement
Special thanks to Mr S Pachal for his
contribution to carry out the study and analysis
of the problem.
Adil Rehman is Principal Engineer, Syngas-
Thermal-Oil and Gas and Special Projects,
Process & Technology Department with
Technip KT India Ltd, New Delhi. He holds a
masters degree in petrochemical engineering
from Aligarh Muslim University, Aligarh, India.
Email: arehman@technip.com
Suman Pachal is Senior Engineer, Syngas-
Thermal-Oil and Gas and Special Projects,
Process and Technology Department with
Technip KT India Ltd. He holds a bachelors
degree in chemical engineering from University
of Calcutta, Kolkata, India.
Email: spachal@technip.com
Shyam K Choudhary is Principal Engineer,
Syngas-Thermal-Oil and Gas and Special
Projects, Process & Technology Department
with Technip KT India Ltd. He holds a bachelors
degree in chemical engineering from BIT, Sindri,
India. Email: skchoudhary@technip.com
Ugrasen Yadav is Deputy General Manager,
Refnery & Petrochemicals, Process &
Technology Department with Technip KT India
Ltd. He holds a masters degree in chemical
engineering from HBTI, Kanpur, India.
Email: uyadav@technip.com
M K E Prasad is Head of the Process and
Technology Department with Technip KT
India Ltd. He holds a BTech degree in chemical
engineering from Osmania University,
Hyderabad, India.
Email: mkeprasad@technip.com
7 0 0
8 0 0
7 5 0
9 0 0
8 5 0
6 5 0
6 0 0
5 5 0
5 0 0
4 5 0
F R N C o u tp u t)
7 6 1 7 6 1
P ro p o se d
sc h e m e I
P ro p o se d
sc h e m e I I
P ro p o se d
sc h e m e I I I
B
r
i
d
g
e
w
a
l
l

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

C
Case
4 0 0
F u rn a c e 5 0 2
d o u b le p a ss)
F u rn a c e 5 0 1 A /B
d o u b le p a ss)
8 8 9
7 8 4
7 0 0
8 0 0
7 5 0
9 0 0
8 5 0
6 5 0
6 0 0
5 5 0
5 0 0
4 5 0
F R N C o u tp u t)
C o n ve c . +5 0 1 A /B
d o u b le p a ss)
6 4 9
7 0 7 7 0 7
6 5 2
C o n ve c . +5 0 1 A
d o u b le p a ss)
C o n ve c . +5 0 1 A
sin g le p a ss)
C o n ve c . +5 0 2
d o u b le p a ss)
P ro p o se d
sc h e m e I
P ro p o se d
sc h e m e I I
P ro p o se d
sc h e m e I I I
B
r
i
d
g
e
w
a
l
l

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

C
Case
4 0 0
7 0 0
8 0 0
7 5 0
9 0 0
8 5 0
6 5 0
6 0 0
5 5 0
5 0 0
4 5 0
F R N C o u tp u t)
6 9 8 6 9 8
P ro p o se d
sc h e m e I
P ro p o se d
sc h e m e I I
P ro p o se d
sc h e m e I I I
B
r
i
d
g
e
w
a
l
l

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

C
Case
4 0 0
8 7 4 8 7 1
F u rn a c e 5 0 3
sin g le p a ss)
d o u b le p a ss)
sin g le p a ss)
F u rn a c e 5 0 3
sin g le p a ss)
Figure 8 Comparison of bridge wall temperatures for three process schemes
technip.indd 6 12/12/12 11:01:05
www.wraconferences.com/europeanfuels | +44 (0) 20 7384 7700 | Hannah.Wharrier@clarionevents.com
5 -7 March 2013
NH Grand Hotel Krasnapolsky
Amsterdam, T e Netherlands
www.wraconferences.com/europeanfuels
Find out how to capitalize on
the emerging market for biofuels
for transportation with case studies
from Shell and the EFOA
Engage with end-users from
across the transport fuels industry,
including Boeing, KLM, Daimler
and Renault
Gain insight into how the whole
fuels ecosystem is evolving
Hear how partnerships and
new operational efciencies
can help you with renery survival
and efciency focus sessions
Network with 200 senior
executives from across the
European fuels value chain
and benchmark your
operating strategies
Malcolm McDowell,
Head of Unit, DG Energy,
European Commission
Chris Beddoes,
Secretary General,
European Petroleum
Industry Association
Marco Schiavetti,
Director - Supply & Trading,
Saras
Atanas Ivanov,
Deputy Chief
Process Engineer,
Lukoil
Arthur Reijnhart,
General Manager
- Alternative Energy Fuel
Development Strategy, Shell
Aksel Skjervheim,
Head of Fuel Markets,
Gasnor
Ignaas Caryn,
Director Innovation and
Corporate Venturing and
Biofuels Strategy, KLM
Marek Herra, Production
Director, Grupa Lotos
3DAY EVENT
5-7 March 2013
GALA DINNER
5 March
Excellent forum:
engaged, competent
and honest participants
contributed to positioning
some of the bricks of this
very complex puzzle.
Ron Timmers,
Head of Industry Segment,
Linde AG
Very informative conference,
will denitely recommend it to
my colleagues.
Faisa Mohammed
Al Haddad,
Senior Analyst
- Projects & Special Studies,
Kuwait Petroleum
Corporation
Sponsored by: Media Partners: Supporting Associations:
THE EUROPEAN
FUELS CONFERENCE
14
th
Annual Meeting
UNITING THE EUROPEAN REFINING AND TRANSPORTATION
INDUSTRY IN NEW FUEL SOLUTIONS
KEY SPEAKERS:
KEY REASONS TO ATTEND:
efc.indd 1 10/12/12 16:42:25
Improved hydrogen yield in catalytic
reforming
T
he advanced fuel technol-
ogy is a patented invention
(US Patent 6207040, European
Patent 0914405) and states that the
higher-boiling C
6
hydrocarbons,
including naphthenes, benzene and
hexane, are more proftably sent to
the light tops isomerisation process
rather than to catalytic reforming.
The critical full-range naphtha-split-
ting mode to be constructed in the
refnery (see Figure 1) requires
negligible investment.
The economic gains of advanced
fuel technology
1
include:
A gain of more than 4% on the
recombined isomerate-reformate
blend volume yield per 100 tonnes
of isomerisation-reforming feed.
2
(This is mainly explained by the
good C
6
isomerate octane and, in
particular, by the even better C
6
blending isomerate octane, plus the
C
6
isomerate weight yield being
much higher and the C
6
isomerate
density being much lower than the
reformate weight yield and
density.)
A gain of around three octane
number points on the recombined
isomerate-reformate blend (result-
ing from 5-6 more points of the
reformate RON
1,3
acquired thanks
to the better reformability quality of
the reforming feed), exchangeable
for an additional gasoline yield.
These results are obtained by
managing the full-range naphtha
splitting unit in such a way as to
determine the two ranges of concen-
trations (0-4 vol% C
7
hydrocarbons
in the tops, 0-0.5 vol% C
6
hydrocar-
bons in the bottom, see Figure 1),
which defne the splitting
mode and are the necessary and
suffcient condition for performing
A process step that sends higher-boiling C
6
hydrocarbons to light tops isomerisation
delivers an increase in hydrogen production from naphtha catalytic reforming
ROBERTO AMADEI
Chemical & Energy Development
the advanced fuel technologys
functions.
Effcient separation by distillation
in the full-range naphtha splitting
unit can be achieved according to
different strategies, depending on
choices the refner has to make
between highly effcient equipment,
which is more capital intensive, and
highly effcient operation, which
consumes more energy. The devel-
opment in no way limits freedom
of choice in a compromise between
these options.
In practice, separation effciency
can be improved by modifying the
equipment (for instance, number
and type of distillation stages, inter-
nals type) and/or the refux ratio.
There is also the option to deter-
mine an optimum compromise
between improvements in equip-
ment and refux ratio.
Hydrogen
A further important feature of
advanced fuel technology is the
gain it offers in catalytic reforming
hydrogen production/availability.
The gain in production of catalytic
reforming hydrogen (net of
increased use in isomerisation) is
estimated in the range of 28% to
over 48% based on conventional
hydrogen production in catalytic
reforming. When removing all or
nearly all of the C
6
molecules (shift-
ing them to the isomerisation
process) from the catalytic reform-
ing feed, it has been demonstrated
in the refnery that the yield of cata-
lytic reforming hydrogen increases
signifcantly. This is in agreement
with the theory that the reforming
feed quality, after C
6
removal,
becomes much more favourable to
reformings dehydrogenation reac-
tions with hydrogen production
and much less favourable to reform-
ings hydrocracking reactions with
hydrogen consumption.
We will frst analyse the basis of
the technology before examining
the results of the refnery runs.
Reformate octane number
First, we can consider the aromati-
sation catalytic reforming reactions:
1 naphthene 1 aromatic + 3 H
2
- 48-55 Mcal/
kmol (million calories per thousand moles) (1)

1 paraffn 1 aromatic + 4 H
2
- 60-65 Mcal/
kmol (2)
It is well known that the delta
octane number [C
8
+ aromatics - C
8
+
(naphthenes + paraffns)] is much
higher (around four times) than for
[C
6
aromatics - C
6
naphthenes]. C
6

paraffns cannot be considered
because the C
6
paraffns that pass
through catalytic reforming do not,
for practical purposes, increase their
octane number. In fact, C
6
paraffns
either crack or pass through uncon-
verted (although an exception is
Isomerisation
unit
Catalytic
reforming
unit
N a p h th a
L ig h t to p s
0 -4 vo l% C
7
)
H e a vy n a p h th a
0 -0 . 5 vo l% C
6
)
Splitting
unit
Figure 1 The advanced fuel fow diagram
ced.indd 1 11/12/12 14:58:40
made for a 10% maximum quota of
normal hexane with a RON value
of about 26).
The heat absorbed in catalytic
reforming by a one-molecule
aromatisation reaction, generating
one aromatic molecule, is roughly
speaking about the same, irrespec-
tive of the number of carbon atoms.
This means that, with about the
same heat absorption, C
8
+ has a
delta octane from the aromatisation
of one molecule about four times
higher than for a C
6
molecule.
Regarding one molecule, with
approximately the same heat
absorption, C
8
+ has a delta octane
four times higher than C
6
s delta
octane for a gasoline quantity on
average about 40% higher (because
the relevant gasoline weight is
exactly proportional to the molecu-
lar weight, while the relevant
gasoline volume is approximately
proportional).
The hydrocarbons C
6
, C
7
, C
8
, C
9
and so on compete with one
another in order to make use of the
available heat. When C
6
is present,
it provides a modest octane increase
by making use of a portion of the
heat subtracted from the availabil-
ity of C
7
+. When C
6
is present, in
order to increase the available heat,
the only option is to increase the
catalytic bed temperature.
When C
6
is removed, the refor-
mate octane increases automatically
due to the strong octane upgrading
available for C
7
+, which is simply a
result of the absence of C
6
. The
presence of C
6
appears to be very
harmful to catalytic reforming
performance.
The highest boiling hydrocarbon
among all of the C
6
s is cyclohexane,
which in reforming has a fair
conversion rate (although lower
than the C
7
+ naphthenes), but a
relatively modest potential for
octane increase. As a consequence,
it is important to transfer as much
C
6
as possible from the reforming
feed to the isomerisation feed. The
more C
6
transferred, the better the
reforming performance and the
refnery economics.
The theory applies to any crude
oil and to any refnery confgura-
tion. This theory is in full agreement
with refnery experimental results.
1
The frst conclusion is that the
removal of C
6
hydrocarbons
increases signifcantly the reformate
octane number process variables
being equal and the catalytic bed
temperature, and consequently the
catalyst life, being equal. In order
to keep the reformate octane
number constant, it is necessary to
decrease signifcantly the catalytic
bed temperature.
Catalytic reforming hydrogen
production
Catalytic bed temperature decrease
When C
6
is removed from the cata-
lytic reforming process, the
reformate octane number increases
signifcantly, which means that a
large decrease in the catalytic bed
temperature is necessary to obtain a
determined reformate octane
number.
We will now discuss hydrocrack-
ing reactions on the premise that
the categories of hydrocracking
reactions include the dealkylation
reactions. Hydrocracking reactions
are rather slow and very sensitive
to temperature, hydrocracking rates
decreasing rapidly as temperature
decreases. Thus, as soon as the C
6
s
are removed from the catalytic
reforming process, hydrocracking
decreases sharply. As a result, the
following main consequences arise,
all favouring a net increase in
hydrogen production during cata-
lytic reforming:
A quantity of hydrogen molecules
is not consumed and is therefore
saved, the proportion of saved
hydrogen molecules being one
molecule for each hydrocarbon
molecule not cracked. In other
words, catalytic reforming hydrogen
production increases by one mole-
cule for each molecule not cracked
The decrease in the contribution
of paraffns hydrocracking to refor-
mate octane causes a compensating
increase in aromatisation, implying
a further increase in hydrogen
production
The decrease in paraffns and
naphthenes hydrocracking favours
dehydrogenation reactions and
produces hydrogen
By means of dealkylation, the
heavier aromatics are converted
into light aromatics such as toluene
pile caps to spread the ground-
bearing loads. The super lift crane
was scheduled to perform numer-
ous super lifts during a 38-day
outage period. It would be lifting a
total weight of more than 9.4
million lb (4250 tonnes).
Due to the risks associated with
lifts, let alone the risks associated
with super lifts, a detailed manual
was produced and a third-party
company was employed to help
provide a cold set of eyes not only
to the lift plans but also to the erec-
tion and disassembly of the crane.
Additionally, each crane foundation
was tested to 125% of expected load
and checked for settlement.
Other yard activities included:
Assembly of regenerator cyclones
and top head
Assembly of reactor cyclones and
top head
Final assembly of middle section
of regenerator
Final assembly of lower section of
regenerator with upper combustor
riser
Final assembly of lower reactor
and stripper vessel
Final assembly of lower section of
combustor riser
Final assembly of regenerator
internal riser.
Replacement of the FCC unit
cooler bundle had presented an
initial problem. But that was solved
when the development of a new
method to replace the catalyst
cooler bundle was added to the
work scope. The original plan was
to use a 1000-tonne crane ftted
with nearly 400ft (122m) of boom
due to the height and weight of the
bundle. But now that the unit had a
new confguration, access was
prohibited. To solve the problem,
the project team developed an inno-
vative approach for bundle
replacement utilising two 100-tonne
chain hoists. The new procedure
was both safer and more cost effec-
tive than the previously scheduled
method.
The project faced a delay in the
delivery of a large expansion joint
assembly that would have been
held up due to Canadian travel
restrictions. The team opted to
complete the expansion joints on
site, which allowed the three
smaller sections to be shipped
instead of the overall assembly that
weighed approximately 150 000 lb
(68 tonnes) and was 15ft (4.5m)
wide, 15ft (4.5m) tall and over 40ft
(12.2m) long. Further complicating
the fabrication were the tolerances
on the joint that included a 7-inch
cold offset and bellow welding that
had to be completed under feld
conditions.
Many areas were completed as
far as possible during the pre-
turnaround construction phase. The
project was completed within
schedule, on budget and with a
safety record of zero recordables.
Over $3 billion in upgrades have
made the plant one of the most eff-
cient in the US. Valero St Charles is
now producing a variety of quality
fuels, making extra energy available
to the area grid and employment to
many workers.
Some of the projects fnal statis-
tics included:
Over 200 000 man hours no
recordables
Under 0.5% weld reject rate
AltairStricklands portion of the
turnaround was completed in 46
scheduled days
4.5% under budget
Over 20 million lb (9000 tonnes)
of steel removed/replaced
Over 33 super lifts during TAR
with Deep Souths 36000
Over 40 major lifts during TAR
with CC2800
with 800T
with CC2400
Over 150 critical lifts (over
50 000 lb [23-tonne] lifts) overall.
whitney Strick|and ls vl ChleI MurkeLlng
OIIcer wlLh AlLulrSLrlckund und
AlLulrSLrlcklund Holdlngs, LLC. He munuges
boLh domesLlc und lnLernuLlonul sules Ior
AlLulrSLrlcklund und domesLlc sules Ior
AlLulrSLrlcklund Holdlngs. He uLLended
Lhe UnlverslLy oI 1exus uL AusLln und klce
UnlverslLy ln HousLon, 1exus, und ls MCSL
quullIed.
/ma||. wstt|c||aaJa|ta|tstt|c||aaJ.com
Chad 8eckemeyer ls currenLly u lro|ecL
Munuger Ior AlLulrSLrlcklund und hus been
wlLh Lhe compuny slnce 1998.
www.eptq.com kVAMS 2012 25
assess achievement of emission
reductions required under the
uated Linde in regard to how
accurately its products measured
, nitrogen oxide and carbon
dioxide. Lindes US production site,
at Alpha New Jersey, passed the
audit with zero failures and, in
2011, the company was granted
continued approval to produce EPA
protocol gas standards both at this
site in the US and at an additional
As in other parts of the world, the
European Union is introducing
sions from light road vehicles,
gen particulates and oxides. Since
2009, it has been mandatory to have
ultra-low-sulphur petrol and diesel;
that is, fuels containing less than 10
Analysis of sulphur compounds in
fuels has therefore not only become
a critical requirement, but so has the
need to measure for lower and lower
levels of sulphur compounds. It is
now vital to be able to detect
extremely low levels of sulphur,
down to parts per billion, and a
range of instrumentation techniques
and detectors are required to meet
cal methods used for the
determination of sulphur species in
fuel samples is GC separation
Heavy crude 0il
Atmospheric 0istillation
Vacuum 0istillation
0oker & Visbreaker Feed
Fluidized 0atalytic 0racker
Bitumen
Non-nlrusve low
measuremenl up lo /000
FeLd-Proven al Relneres
Trouble Iree operation at
extreme pipe temperatures
No clogging, no pressure losses
nstallation and maintenance
without process interruption
ndependent oI uid or pressure
Hazardous area approved

When lhe 0ong

gels I0T.
relnery.lexm.com
altair s.indd 3 17/9/12 12:36:32 ced.indd 2 11/12/12 14:59:06
JOIN MAJOR
INDUSTRY PLAYERS
AT ME-TECH 2 01 3
The Meeting Plac e for the
Mid d le East Downstream Ind ustry:
Conferenc e and Exhib ition.
Foc us on latest tec hnology d evelop ments
for Gas Proc essing, Oil Rening,
Resid ue Up grad ing & Petroc hemic als
ME- TECH 2 0 1 3
1 8 - 2 0 FEBRUARY
MADINAT JUMEIRAH, DUBAI
www. me- tec h. b iz
ofc e@europ etro- me. c om
event organised b y
sup p orted b y
epc.indd 1 6/12/12 13:26:47
and xylenes. As C
7
and C
8
paraffns
are unfavourable for conversion to
aromatics, dealkylation has an unfa-
vourable effect on aromatisation of
these light paraffns; hence, the
implied decrease in dealkylation
favours aromatising dehydrocycli-
sation reactions of light paraffns,
producing signifcant quantities of
hydrogen.
Replacement of C
6
naphthenes
aromatisation by C
7
+ paraffns
aromatisation
Aromatisation of C
6
naphthenes is
for the most part replaced by C
7
+
paraffns aromatisation rather than
by C
7
+ naphthenes aromatisation.
The reason for this is that, at the
stage when C
6
is present, the
concentration of C
7
+ naphthenes in
the reformate is low, while the C
7
+
paraffns concentration is high. As
the reaction rate depends on the
concentration of the reactants, when
the equilibrium of catalytic reform-
ing reactions is broken by the
removal of C
6
, the C
7
+ paraffns are
more likely to aromatise than are
the C
7
+ naphthenes. As we see from
reactions 1 and 2, the paraffn
aromatisation reaction yields 33%
more hydrogen than does the naph-
thene aromatisation reaction.
The second conclusion is that,
process variables being equal,
including the catalytic bed tempera-
ture and consequently the catalyst
life, the removal of C
6
hydrocarbons
increases signifcantly the hydrogen
production yield from catalytic
reforming.
Confrmation of increased hydrogen
production
Any [full-range naphtha splitting]
unit built after 1950...is either a
simple depentaniser or a rather
intense deisohexaniser. Oil & Gas
Journal, 21 March 1994, p52.
Typical splitter designs...only
deisohexanise the reforming unit
feed. Oil & Gas Journal, 11
September 2006, p40.
We decided to conservatively
consider the typical naphtha splitter
as a deisohexaniser. A naphtha
deisohexanising splitter generates a
bottom cut (the heavy naphtha),
feeding catalytic reforming, contain-
ing around 15 vol% of the total C
6
hydrocarbons. This is the base case
of our study.
That said, experimental confrma-
tion of the gain in catalytic
reforming hydrogen production
versus C
6
hydrocarbons removal
was examined. To this end, we
constructed a normalisation algo-
rithm in order to bring proper
refnery run data to constant proc-
ess variables.
We then considered normalised
catalytic reforming hydrogen
production versus the relevant vol%
content of feed C
6
. Our goal was to
describe the catalytic reforming
hydrogen production yield in two
situations:
A scarce concentration of C
6
naph-
thenes, in particular of cyclohexane
originating aromatisation reactions,
in the reforming feed
A fair concentration of C
6
naph-
thenes, in particular of cyclohexane,
in the reforming feed.
To this end, we constructed two
regression lines, corresponding to
these situations:
Catalytic reforming normalised
hydrogen production wt% on the
reforming feed versus feed total C
6

hydrocarbons content vol%, for a
feed total C
6
hydrocarbons content
from 15 vol% to 4 vol% (see
Figure 2)
Catalytic reforming normalised
hydrogen production wt% on the
reforming feed versus feed total C
6
hydrocarbons content vol%, for a
feed total C
6
hydrocarbons content
from 4 vol% to 0.3 vol% (see
Figure 3).
The plots gave us a percentage
increase of catalytic reforming
hydrogen production, when the
total C
6
hydrocarbons content of the
reforming feed decreases from 15
vol% to 0.3 vol%. The resulting
increase is 54.59 vol%, the reformate
octane number and all the process
variables being equal (see Figures 2
and 3).
3 . 0
4 . 5
4 . 0
3 . 5
2 . 5
2 . 0
1 . 5
1 . 0
0 . 5
0 5 1 0 1 5 2 0
R
e
f
o
r
m
i
n
g

h
y
d
r
o
g
e
n

n
o
r
m
a
l
i
s
e
d

y
i
e
l
d
,

w
t
%
Feed C
6
concent rat i on, vol %
0
Figure 2 Reforming hydrogen normalised production vs C
6
- vol% in feed
3 . 2
3 . 0
2 . 8
2 . 6
2 . 4
2 . 2
0 1 2 4 3 5
Feed C
6
concent rat i on, vol %
0
R
e
f
o
r
m
i
n
g

h
y
d
r
o
g
e
n

n
o
r
m
a
l
i
s
e
d

y
i
e
l
d
,

w
t
%
Figure 3 Reforming hydrogen normalised production vs C
6
- vol% in feed
ced.indd 3 11/12/12 16:27:29
Need A Catalyst Bed Support....NOW!
Why wait 10 ,15, 20 weeks for delivery. Ask AMACS
about our Fast Track delivery for our full line of wedge wire screen
products. With in-house engineering support and fabrication, we
can use your existing drawings or modify them to improve
your process.
Catalyst Bed Supports
Basket Strainers
Nozzles
Outlet/Inlet Baskets
Distributors
Hub and Header Laterals
DISTRIBUTORS & SUPPORTS
Manufactured to customer
specications or engineered to
meet performance requirements
COALESCERS
Oil water separations Haze removal
from fuels Removal of tower wet reux
Caustic treater applications
STRUCTURED PACKING
Woven, sheet metal, and knitted
structured packing. Built to spec
or performance requirement.
Mesh numerous alloys & plastics
Vane (Chevron)
Double Pocket Vane
Insertion MistFix
MIST ELIMINATORS RANDOM PACKING

Valve (numerous options)
Bafe Sieve or perforated
Dual ow Bubble cap trays
Cartridge trays
TRAYS
Wide variety of random packing
types, sizes and materials in
stock! Ask us how our patented
SuperBlend 2-Pac can increase
your capacity and efciency.
w w w .amac s.c om 24hr EM ERGENCY SERVI CE 281- 716- 1179
Visit our new website at
w w w.amacs.com
amacs.indd 1 12/12/12 09:50:42
fgures, we arrive at 2.64 - 0.08 =
2.56% on the base case reforming
feed rate, equivalent to 2.56/2 =
1.28 times the base case hydrogen
production, or a 28% net gain in
hydrogen production on the reform-
ing base case.
If the reforming feed rate is fully
replenished with good-quality feed,
we will have the full 46 wt% yield
increase and will retain a smaller
fraction of the gain in reformate
octane points obtained.
The result of the relevant calcula-
tion gives a net gain in reforming
hydrogen production of around
48% on base case hydrogen
production.
Economic and environmental
improvements
It is possible to exceed a 48% gain
in reforming hydrogen production
through the installation of addi-
tional catalytic reforming and
isomerisation capacities. This option
could be evaluated if we consider
that the current trend in crude oil
towards heavier and sourer quality,
as much as the trend towards
better-quality products, requires a
bigger and growing addition of
hydrogen (also keeping in mind
that on-purpose hydrogen produc-
tion processes consume enormous
energy quantities).
In addition to the gain in hydro-
gen, the advanced fuel technology
delivers:
A strong gain in gasoline yield
A strong gain in gasoline octane
number (exchangeable at will for an
additional strong gain in gasoline
yield)
Prevention of benzene formation
Accomplishment, with a margin,
of the most stringent specifcations
for gasoline benzene content (in
particular, for the US, accomplish-
ment of the US Environmental
Protection Agencys requirement of
0.62 vol% average benzene content
for the whole gasoline pool, result-
ing in saleable benzene requirement
credits)
Major energy savings
Reduction in pollutants, both
in gasoline production and
consumption
Higher capacity increase in gaso-
line octane production
We applied a safety margin of
15%, thus reducing the gain in
catalytic reforming hydrogen
production from 54.59 wt% to 46
wt% compared with hydrogen
production in the reforming base
case. The reforming feed of the base
case contains 15 vol% of the total C
6

hydrocarbons.
Should the removed C
6
not be
replaced at all that is, should
catalytic reforming not be replen-
ished at all with good-quality feed,
almost free of C
6
we would have
a decrease in the reforming feed
rate. This decrease would entail a
reduction in the gain of hydrogen
production from 46 wt% to 25 wt%
based on hydrogen production in
the reforming base case.
Trend in hydrogen production
increase
Generally speaking, C
6
paraffns
originate nearly zero hydrogen.
Line 2 in Figure 3 is steeper than
line 1 in Figure 2; this was expected
because, from 4 vol% to 0.3 vol% C
6

(see Figure 3), the cyclohexane
concentration in the C
6
of the
reforming feed grows much more
rapidly versus the total content of
C
6
hydrocarbons in the feed. In
such a situation, the benefcial
effects of C
6
transfer from reform-
ing to isomerisation, also enhancing
the hydrogen generation, grows
much more rapidly. That is to say,
in this range of feed total C
6
hydro-
carbons content, the hydrogen
production yield grows more
rapidly compared with the decrease
in feed total C
6
hydrocarbons
content. This also means that, in
this range of feed total C
6
hydrocar-
bons content, the improvement in
refnery economics, environmental
performance and energy savings
delivered by the technology
increases more rapidly with the
decrease in feed total C
6
hydrocar-
bons content.
As for line 1 in Figure 2, we can
also say that the increase in hydro-
gen production yield with the
decrease in feed total C
6
hydrocar-
bons content is largely due to a
reduction in dilution of the reform-
ing feed with a component (the C
6
paraffns) that effectively does not
produce hydrogen.
Hydrogen production increase
through partial exploitation of
reformate octane gain
We have seen that, through C
6
removal, there is an important gain
in reformate octane points, the cata-
lytic bed temperature being
constant. In order to keep constant
the octane number of the isomerate-
reformate blend which would
otherwise decrease as a conse-
quence of the shift of about 11 wt%
on full-range naphtha from reform-
ing to isomerisation
1
we must
retain a minor fraction of this gain;
that is, the reformate octane number
is increased above the base case
value. The increase in reformate
octane entails an increase in reform-
ing hydrogen production. Hydrogen
production will increase from 1.25
to around 1.32 times the base case
value.
If we keep the other gained
octane points in reformate, the
gain in hydrogen production yield
in catalytic reforming will further
increase. But we can change at will
these supplemental gained refor-
mate octane points into an
additional strong gain in volume
yield, on top of the gain resulting
directly from the C
6
shift from
reforming to isomerisation. The
reforming catalytic bed tempera-
ture will thus decrease, so further
increasing the reforming catalysts
life, which has already increased
due to the lower feed rate
(in the case of non-replacement of
C
6
).
Returning to reforming hydrogen
production: as the base case hydro-
gen production is approximately 2
tonnes per 100 tonnes of reforming
feed, reforming hydrogen produc-
tion after C
6
removal is 2*1.32
tonnes% (2.64 tonnes%) on the base
case reforming feed rate.
Hydrogen production gain net of
increased use for isomerisation
Isomerisation hydrogen use of the
shifted C
6
(11 tonnes per 78 tonnes
of reforming throughput
1
), evalu-
ated at 0.6 wt% of the isomerisation
throughput, is worth 11*0.6/
100*100/78 = 0.08 wt% on the base
case reforming feed rate.
By subtracting isomerisation use
from the hydrogen production
ced.indd 4 11/12/12 16:27:48
Improved engine operation and
maintenance.
1,4
Economic value
Calculations are provided with the
purpose of demonstrating the order
of magnitude of the economics
involved. We will conservatively
consider two of the approximately
three points of octane number
gained by the technology. These
two points of octane gain are
commercially worth, on the basis of
the revenue difference between
premium and regular grades,
about 0.02 * 2 = 0.04/litre of
isomerate-reformate blend, or about
0.05/kg of full-range naphtha
($0.07/kg or $70/tonne of full-
range naphtha).
We assume a gasoline revenue of
$3/gal, equivalent to $0.7926/litre.
On the basis of a conservative
3 vol% gain and proportioning from
earlier calculations,
1
where the
assumed price of gasoline was
$0.20/litre, we obtain a value of the
higher gasoline yield of $20.80/
tonne of full-range naphtha and a
lower byproducts value, to be
subtracted from the above fgure, of
$8.14/tonne of full-range naphtha.
Hence, the better gasoline yield
gives us a total net economic gain
of $(20.80 - 8.14) = $12.66/tonne of
full-range naphtha.
We assume a hydrogen cost of
$2150/tonne.
5
On this basis, as the
reforming feed heavy naphtha of
the base case represents about 78%
of the full-range naphtha,
1
and base
case hydrogen production is 2
tonnes per 100 tonnes of reforming
feed rate, the hydrogen gain is basi-
cally worth between 0.02 * 0.28 *
0.78 * 2150 = $16.10/tonne of full-
range naphtha and over 0.02 * 0.48
* 0.78 * 2150 = $16.10/tonne of full-
range naphtha.
Based on the fgures above, we
obtain a proft of $92.05-98.76/tonne
of full-range naphtha. In an average
refnery, with a full-range naphtha
splitting feed rate of 4000 tonnes/
day, total annual gain offered by
the invention considered results is
$134-144 million.
Conclusion
The technology described offers
gasoline yield gain, gasoline octane
gain, net gain in catalytic reforming
hydrogen production and abate-
ment of refnery benzene
production.
The transfer from catalytic
reforming to isomerisation of any
C
6
hydrocarbon species appears to
be benefcial. Among these species,
the most benefcial transfer concerns
the higher boiling ones, cyclohex-
ane transfer bringing the biggest
beneft.
The overall result of the technol-
ogy encompasses important
improvements in refnery econom-
ics and environmental performance,
climate change mitigation, energy
savings and improvement in gaso-
line technical quality.
Acknowledgment
Revised and updated from the Q&A Gasoline
Processes answers and interventions at the 2011
National Petrochemical & Refners Association
(now American Fuel & Petrochemical
Manufacturers) Q&A and Technology Forum,
9-12 October 2011, San Antonio, Texas.
References
1 Amadei R, A proposal for an economic and
environmental gasoline, PTQ, Summer 1999.
2 Amadei R, Opposition Appeal Patent
proprietor reply, European Patent Offce,
European Patent Register, Munich, 16 June
2005, https://register.epo.org/espacenet/applic
ation?documentId=EI57C1OY1001J11&numb
er=EP97937512&lng=en&npl=false (accessed
29 Sep 2012).
3 Westphalen D, Shethna H, Refnery wide
simulation, Hydrocarbon Engineering, March
2004.
4 Amadei R, in the Attachment to Industry
comment for the GHG Scoping Plan - 1st
Workshop: An economic and environmental
gasoline, California Environmental Protection
Agency, Air Resources Board, 30 July 2008,
www.arb.ca.gov/lispub/comm2/bccommlog.
php?listname=sp-industry-ws(#7) (accessed
29 Sep 2012).
5 Parihar P, et al, Optimise hydrogen
management for distillate production,
Hydrocarbon Processing, March 2012.
Roberto Amadei is Managing Director of
Chemical & Energy Development, Genoa. Italy.
He has 40 years experience in oil refning and
energy technology, planning and optimization,
and holds a PhD in chemical engineering from
the University of Genoa.
Email: ramadei1@alice.it
Our new clean-fuels
plant is straining the
auxiliary units!
How can we boost their capacity
without major construction?
Read more on this topic at
www.amacs.com
TO MEET CLEAN FUELS
requirements for gasoline, a medium-
sized Midwestern refnery added a low-
sulfur fuels technology plant. Gasoline
throughput was unchanged. However,
the sharp increase in sulfur removal
required more hydrogen from the
hydrogen unit and sent more sulfur
gases to the amine treaters and
downstream sulfur units. These
auxiliary units became bottlenecks,
overdriven at the cost of product
purity and amine consumption.
On studying the hydrogen, amine,
and sulfur units, AMACS found many
opportunities for improving separation
effciency and capacity. The problems
were solved without major construction
by applying modern technology to mist
eliminators, liquid-liquid separators,
and tower trays and packing. Results
included haze-free product and
reduced amine consumption. Now a
diesel clean-fuels plant is being added.
Prore: Z13-131-0931 Fax: Z13-133-201
amacs@amacs.com
21-rr Erergercy 3erv|ce:
1 -8 0 0 -2 3 1 -0 0 7 7
Visit our new website at
www.amacs.com
ced.indd 5 12/12/12 09:52:37
Globally, refineries face the problem of losing and wasting millions of dollars every year due to
plant shutdowns and energy inefficiency. Equipment failures and weather are both major con-
tributors to refinery downtime.
This three-day event addresses technology solutions and business strategies for
reducing power disruptions and improving energy efficiency.
Attendees will gain knowledge, experience, and insight from leading experts, who will present
on topics ranging from low-hanging fruit approaches for energy improvement, power failure
prevention and protection, and swift unit restarts, to cutting-edge technologies that ensure the
utmost in energy security, high operating availability, fast payback on investments, and mini-
mized carbon footprint.
Don't miss out on this opportunity for networking, idea-sharing, and strategizing!
REGISTER TODAY.
Hosted by:
Energy Efficiency Improvement topics:
Heat, steam, and H
2
supplies
Rotating equipment
Processing units (e.g. CDU, FCCU)
CHP/cogeneration
Asset management and carbon footprint
Computer monitor and control
Media Partners:
Who should attend:
Plant engineers
Utility, reliability and maintenance managers
Operations and technical managers
Procurement managers
Strategic planners and business executives
Environmental, health and safety personnel
EPC firms, consultants, equipment manufac-
turers, service providers
Energy suppliers and traders
Venture capitalists in energy and power tech-
nologies
Power Disruption Mitigation topics:
Prevention and protection
Recovery and restart
Electric supply security and quality
System design and operational improvements
Equipment maintenance and safety
Flaring and emission compliance
Register Online: http://irepconference.com
Contact: Neal Robbins
Email: info@irepconference.com
Call: +1 (610) 408-0116 (USA)
Act Soon for our early bird special. Group rates are available.
Sponsorship and exhibition opportunities are available.
Attendees are eligible for Continuing Learning Education (CLE)
credits.
Houston, TX (USA), May 1-3, 2013
IREP
Reliability, Efficiency, and Profitability
International Refinery Energy-Power
Conference & Exhibition
HYDROCARBON PUBLISHING COMPANY
Translating Knowledge into Profitability
SM
irep.indd 1 11/12/12 14:45:03
Everything including the laser
survey
Avevas new plant design software
suite, Everything 3D (E3D), began
shipping in December following a
third-quarter launch of the product.
The companys aim in developing
E3D is to pull together technologies
that will enable users to adopt
what it calls lean working prac-
tices, spanning engineering, design
and construction, with the aim of
speeding project execution and
reducing wastage in the design
process. The key elements of the
product include capture of laser
scanning data to create intelligent
objects, highly accurate 3D geome-
try, 2D drafting tools that fully
integrate with the 3D elements, and
intuitive user interfaces for
common design tasks.
According to Aveva, it is no
longer enough to achieve error-free
designs. The availability of afford-
able new technologies makes it
practical to broaden the reach of
plant design to vendors, fabricators
and construction teams. The cost
savings of a lean approach come
from the ability of people away
from the design offce to communi-
cate status and construction
information as the build
progresses.
An ageing population of engi-
neers in mature markets and an
abundance of inexperience in
emerging markets is driving plant
design software to be easy to use,
quick to be trained on and to
support a faster time to production,
says the company. In addition, the
emergence of mobile and cloud
computing from the consumer
market into the business market
enables remote workers, specifcally
those on the construction site, to
send as-built information back to
the design team via status updates
and laser scan data to enable rapid
corrective actions.
The Cambridge, England-based
company has been best known in
the refning industry for its PDMS
P
plant design software, launched
almost a decade after it was
founded as CADCentre, a spin-off
of University of Cambridge
engineering departments with
government cash support.
E3D supports PDMS data, but its
most signifcant design function is
to support accurate representation
of plant both at the design stage
and as-built. This is where laser
scanning and modelling capability
come to the fore. It is rare to
encounter an operating plant that
matches its pre-build design with
complete accuracy; the older the
plant, the greater the range of
discrepancies between the two. As
a result, a signifcant modifcation
or revamp calls for a complete
survey of the area of a plant that
will be affected. Without laser scan-
ning as a viable option for data
capture in an operating plant,
surveying is a challenging and rela-
tively costly operation. The
hardware and software for data
gathering by laser are now mature
enough to become tools for captur-
ing and transferring raw survey
data for use in modelling software,
enabling plant designers to detect
clashes between the new design
and the plants current condition,
and to create a modifed design that
fts accurately.
The software recognises where 3D
data points represent a pipe or a
beam, for instance, and offers a
choice of the most likely corre-
sponding catalogue items. The user
then selects the appropriate choice,
for example by choosing the correct
pipe specifcation, and the system
generates an intelligent 3D entity in
the design system, superimposed
on the as-built data. Thus, it
becomes possible to reverse engi-
neer a laser survey into an accurate,
validated and usable 3D model of
the physical plant. The modelled
items can be intelligent, knowing
to which other items they are
connected, just as if they had been
designed in a 3D modelling
Product News
application from the outset. Once
they are captured in this way, it
becomes possible to associate these
objects with their counterparts in
the P&ID and other information
items such as datasheets, plot plans,
instrument loop diagrams and
single line diagrams, essentially
following the normal engineering
and design process but in the oppo-
site direction.
In the build-up to E3Ds develop-
ment, Aveva developed its own
laser model interface. A raft of
acquisitions includes, for instance,
the software division of Z + F UK,
developer of a vendor-independent
interface for laser scanning
software.
When internal development and
acquisitions do not cover the range
of technologies that Aveva would
like to incorporate into its product,
the company develops partnerships
with technology suppliers. For
instance, it has formed an alliance
with Microsoft to speed up the
development of a tablet application
to support its recent technologies.
The mobile tablet application will
be based on Windows Azure tech-
nology, operate on Windows 8
devices and provide interactive
access to 3D design models.
According to Aveva, the relation-
ship with Microsoft will enable it to
deliver a solution that adapts
Microsofts version of cloud
computing to give users greater
control and visibility of their design
model, conceding that some users
may need time to be convinced
about feeding their design data into
a third-party cloud. Mobile tech-
nology will help improve design
productivity, allow more informed
decisions, and deliver fast and high-
quality projects, says Aveva. The
companys Chief Technology
Offcer, Dave Wheeldon, described
to PTQ the enthusiasm he has
measured among customers for
tablet-based operations, in particu-
lar for a large-screen version under
development.
ind news copy 14.indd 1 11/12/12 18:14:06
ABB Analytical 61
ABB Automation 107
ABB Consulting 93
AFPM Annual Meeting 110
Air Liquide Global E&C Solutions 18
Albemarle Catalysts Company IBC
AMACS Process Tower Internals 123 & 125
Aveva 54
Axens OBC
BASF Corporation, Catalyst Division 23
BETE Fog Nozzle 64
Bryan Research & Engineering 41
Burckhardt Compression 104
CB&I 7
Criterion Catalyst & Technologies 4
Curtiss-Wright Flow Control Corporation 68
DuPont Sustainable Solutions 32
Elliott Group 100
Emerson Process Management 9
Enersul 4
European Fuels Conference 118
ExxonMobil Research and Engineering Company 25
Flexim 120
Foster Wheeler 35
Grabner Instruments 11
Grace Catalysts Technologies 37
Haldor Topse IFC
Hoerbiger Kompressortechnik Holding 83
Honeywell Process Solutions 91
International Refnery, Energy & Power 126
ITW Technologies 88 & 102
Johnson Matthey Catalysts 78
Johnson Screens 48
KBC Advanced Technologies 2
KBR 52
Kurita Europe 51
ME-TECH 2013 121
Metso Automation 94
OHL Gutermuth Industrial Valves 77 & 81
Onis International 30
Optimised Gas Treating 109
Process Consulting Services 14 & 17
Sabin Metal Corporation 27
Samson 13
SOGAT 2013 115
Spectro Analytical Instruments 67
Sulzer Chemtech 72
UOP 42
Yokogawa Europe 29
Zeeco 47
Zwick Armaturen 99
Zyme-Flow 62
Alphabetical list of advertisers
For more information on these advertisers, go to www.ptqenquiry.com
ad index copy 15.indd 1 11/12/12 18:49:38
HYDROPROCESSING
TECHNOLOGY SOLUTIONS
For more information on Albemarles top performance catalysts, technologies and related services, visit albemarle.com
Introducing Albemarles new CoMo catalyst
Another novel product from the people
who created Ketjenfne
757, the bestselling

Ultra Low Sulfur Diesel catalyst of all time. Curious?
Call your Albemarle representative to learn more.
IN N O VA TIO N B EY O N D .
ALB-028a_production.indd 1 8/21/12 7:28 PM
albemarle.indd 1 6/12/12 13:16:52
Stimulate the heart of
your hydroprocessing unit
Single source lSO 9001 lSO 14001 OHSAS 18001
www.axens.net
Impulse
TM
, the catalyst technology that combines
the stability you recognize with the activity you need
Imp_PTQ.indd 1 04/12/2012 14:07:08
axens.indd 1 6/12/12 13:17:48

PTQ Q1 2013

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PTQ Q1 2013

Uploaded by

Copyright:

Available Formats

SPECIAL FEATURES

. H e, along w ith fellow scientists from Shells three R&D

Services and Solutions

), Dow (Selexol), and UOP

process (by Prosernat). The

and is marketed by Lurgi AG.

is a registered trademark of Zeeco, Inc. in the U.S.

(suite of chemistries and predictive

Figure 1 Asphaltene content of selected crudes

PKS O rion provides the only truly integrated safety and

When lhe 0ong

MIST ELIMINATORS RANDOM PACKING

757, the bestselling

You might also like