2012 PTQ q4

special features
Gas processinG developments

mass transfer
refininG
Gas processinG
petrochemicals
PETROLEUM TECHNOLOGY QUARTERLY
ptq
Q4 2012
cover and spine copy 13.indd 1 12/9/12 16:47:43
Hydrogen plant effciency is paramount to optimal refning operations. With Topses steam
reforming expertise it is possible to futher optimize your hydrogen production:
To ensure economical production it is important to consider factors such as
- feedstock fexibility
- capacity control
- value of steam as by product
in the design and process layout.
Topses complete range of hydrogen technologies and catalysts ensure options for increase of
effciency and reduction of the CO
2
footprint.
D e sig n yo u r p la n t la yo u t to d e te rm in e
yo u r e n e rg y e ff c ie n c y a n d c a rb o n e m issio n s
Choose your footprint
WWW. T O P S O E . C O M
haldor.indd 1 12/9/12 10:59:42
2012. The entire content of this publication is protected by copyright full details of which are available from the publishers. All rights
reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic,
mechanical, photocopying, recording or otherwise without the prior permission of the copyright owner.
The opinions and views expressed by the authors in this publication are not necessarily those of the editor or publisher and while every care
has been taken in the preparation of all material included in Petroleum Technology Quarterly the publisher cannot be held responsible for any
statements, opinions or views or for any inaccuracies.
3 Buying the store
ChrisCunningham
5 ptq&a

19 Processing Trends
25 Challenges and opportunities of 10 ppm sulphur gasoline: part 2

DelphineLargeteau,JayRoss,MarcLabordeandLarryWisdomAxens
33 Dynamic simulation: a tool for engineering problems
DipanjanBhattacharya,HaribabuChittibabu,JesseMummandJaleelValappil
Bechtel
47 Exploiting opportunities with challenging crudes
IndiaNagi-hanspal,MaheshSubramaniyam,ParagShah,SergioMoretti
andJamesNolandDorf Ketal Chemicals
57 Valuable liquids from refnery gases
ZaheerMalikandJohnSlack Linde Process Plants
69 Thermal oxidiser waste heat recovery
RyanTate Zeeco, Inc
77 Simulating and monitoring H
2
plant operations
GeneYeh Saudi Aramco
PrabhasMandal,AbdulaliSiddiquiandFahadAlhemidda
Saudi Aramco Riyadh refnery
83 RFCC units set new standard for propylene production
PatrickWalkerandRaymondPeterman UOP LLC, a Honeywell Company
95 Value of monitoring exchanger networks
LauraCopeland Nalco Company
99 FCC catalysts and additives for cost and emissions control
KennethHindleandMariaLuisaSargentiGrace Catalysts Technologies
JuanVargas Coryton Refnery
105 Stripping sour water: the effect of heat stable salts
RalphWeilandandNathanHatcher Optimized Gas Treating, Inc
111 Production of biolubricant base stock
ShaliniGupta,RajeevKumar,SudhaTyagiandPeddyVCRao
Bharat Petroleum Corporation
119 Mitigation of heat exchanger fouling
BruceWright Baker Hughes Incorporated
ToddHochheiserValero Energy Corporation
127 Innovations in wastewater treatment
DarrenGurneyLinde Gases Division
135 Optimising sweep gas fow in a fare header
SatishMathur Bechtel India
141 Industry News
MarathonPetroleumsGaryville,Louisiana,refneryprocesses490000b/dofprimarilyheavy,sourcrude.
Photo: Marathon Petroleum
Q4 (Oct, Nov, Dec) 2012
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 16.indt 1 13/9/12 14:20:46

STRATEGIC
SOLUTIONS
OPERATING
SOLUTIONS
ORGANISATIONAL
SOLUTIONS
ENVIRONMENTAL
SOLUTIONS
CAPITAL
SOLUTIONS
SOFTWARE
SOLUTIONS
What is
NextGen Performance
SM
?
NextGen Performance is optimising each and every opportunity. Its
addressing challenges before they become problems and applying
solutions before others have fully considered them.
Its the result of two parties you and KBC working in a side-by-side
partnership, combining our knowledge and experiences.
Why partner with KBC?
KBC is a leading independent consultancy to the energy sector
worldwide. With our proven methodologies and industry-leading
software, our experienced consultants consistently demonstrate
innovative and creative thinking, fexibility, and a penchant for identi-
fying and implementing improvements that result in tangible value.
We are dedicated to continually improving our consultancy and
software solutions by being responsive to your requests, keeping
abreast of new technologies, and anticipating the needs of an
ever-changing industry.
We measure our success in terms of your success. Let us help
you achieve NextGen Performance.
Americas
+1 281 293 8200
Europe, Middle East, and Africa
+44 (0)1932 242424
Asia
+65 6735 5488
answers@kbcat.com www.kbcat.com
2011 KBC Advanced Technologies plc. All rights reserved.
Given our global experience in
wide-ranging practice areas and
client sectors, KBC is uniquely
positioned to provide a fully
integrated response to your
most 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 Q4 2012 3
Editor
Chris Cunningham
editor@petroleumtechnology.com
Production Editor
Rachel Storry
production@petroleumtechnology.com
Graphics Editor
Rob Fris
graphics@petroleumtechnology.com
Editorial
tel +44 844 5888 773
fax +44 844 5888 667
Business Development Director
Paul Mason
sales@petroleumtechnology.com

Advertising Sales
Bob Aldridge
sales@petroleumtechnology.com
Advertising Sales Offce
tel +44 844 5888 771
fax +44 844 5888 662
Publisher
Nic Allen
publisher@petroleumtechnology.com
Circulation
Jacki Watts
circulation@petroleumtechnology.com
Crambeth Allen Publishing Ltd
Hopesay, Craven Arms SY7 8HD, UK
tel +44 844 5888 776
fax +44 844 5888 667
ISSN 1362-363X

Petroleum Technology Quarterly (USPS 0014-781)
is published quarterly plus annual Catalysis edition
by Crambeth Allen Publishing Ltd and is distributed
in the USA by SPP, 75 Aberdeen Rd, Emigsville, PA
17318. Periodicals postage paid at Emigsville PA.
Postmaster: send address changes to Petroleum
Technology Quarterly c/o PO
Box 437, Emigsville, PA 17318-0437
Back numbers available from the Publisher
at $30 per copy inc postage.
Vol 17 No 5
Q4 (Oct, Nov, Dec) 2012
Buying the store
O
il refning used to be such a predictable business: a Big Oil logo on
the front gate and operations tweaked to convert crude to meet local
domestic demand and to supply contract customers. Not any more: the
name on the gate increasingly refects a shift away from traditional ownership
as selected majors move out of refning. And the emergence of US Gulf Coast
refneries as leading exporters of fuel products a job that was supposed to be
the preserve of the Middle East has certainly altered their image as servicers
of local markets.
The biggest, and by far the least welcome, development in recent times has
of course been refnery closures. Sites in Europe and the Caribbean are being
transformed for the time being at least into terminalling operations.
That is not much of a trend for the Atlantic-bordering industry to look for-
ward to. Another development, perhaps the start of a trend of sorts, concerns
change of ownership and partial change of use. Perhaps it would be more
precise to say change of emphasis. Delta Air Lines took control of its own
refuelling destiny when it purchased the idled Trainer, Pennsylvania refnery
in June. Effectively, the customer took over the store in a bid to assure sup-
plies of jet fuel at stable prices and to give it a potential edge over competi-
tors. A refnery set up to produce exceptional quantities of jet, with an owner
committed to keeping jet prices down through targeted crude sourcing, could
well deliver an advantage in airspace. After all, with jet prices surging
recently, who wants to be frst to ramp up fuel surcharges when the choice of
carriers remains competitive?
The purchase price of the plant was $180 million and Deltas operating sub-
sidiary, Monroe Energy, plans to spend around $100 million on upgrades that
will take jet production past 30% of its overall output, at the expense of gaso-
line production. Aiming to have the plant in operation by the end of
September, following a years shutdown, the new owner is also looking to
resolve some of the general upward volatility in jet prices by, for the frst
time, taking control of refnery raw materials. Like other beleaguered east
coast sites, Trainer was reliant on imported North Sea supplies when
ConocoPhillips was in charge. With Brent crude trading at a hefty premium
to North American domestic supplies, Monroe wants to switch to supplies of
North Dakota crude as a key element of its plan to lop $300 million from a
near-$12 billion bill for jet last year. Monroe has signed a three-year crude
supply deal with BP that includes guarantees of jet supplies from BPs own
refneries and there is a similar supply deal with Trainers previous owner,
Phillips 66.
Jet fuel is also high on the slate at Saudi Aramcos new 400 000 b/d refn-
ery in Jubail, where the frst of the sites crude distillation units is expected to
be operational by the end of 2012. Jet and diesel are expected to account for
54% of output when the refnery reaches its scheduled full output, late in
2013. With two further mega-refneries on the way, Saudi Arabia is aiming
for self-suffciency in fuels, with a portion of its output reserved for the busy
international market for products. Saudi Arabia has managed to grow its
economy over recent years, hence air miles have risen in number. The new
refneries should ensure that its airports can operate without relying on ship-
loads of imported jet.
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 16.indt 2 12/9/12 19:22:01
I can sum up my favorite part about this
job in one word: Eureka!
Meet J ohn Smegal:
Amateur Detective. Senior Chemist.
Even after 27 yearsin the industry, John Smegal still getsexcited at the mere suggestion of discovering
the next big breakthrough in catalyst technology. Hisproclivity towardsdiscovery helped lead to the
development of more than twenty catalysts, including CRITERIONs industry game-changers, CENTINEL
and CENTERA. And while today youll find John managing the hydrocracking pre-treat and distillate
hydrotreating teams, his favorite moments are still created in the lab - testing countless hypotheses for
that exciting next discovery.
www.CRITERIONCatalysts.com
Leading minds. Advanced technologies.
criterion.indd 1 11/9/12 11:23:20
Q
Can I have some advice on selection of catalyst for running
my FCC pretreat in mild hydrocracking (MHC) mode, with
particular reference to HDS and HDN requirements?
A
Eelko Brevoord, Technical Services Manager, Refning
Catalysts, Albemarle, eelko.brevoord@albemarle.com
Historically, MHC units were simply traditional VGO
hydrotreating units that were operated at a high
temperature. More recently, units and catalysts have
been designed and developed specifcally to boost
conversion.
MHC units are characterised by a relatively low pres-
sure (50-100 bar). Consequently, product nitrogen is
relatively high, making zeolitic cracking catalysts much
less effective than in full-pressure hydrotreaters. At high
product nitrogen (>200 ppm), conversions of up to 25%
are usually achieved. The main catalyst for this situation
should allow a refner to achieve their product quality
targets, while meeting their required cycle length. For
example, CoMo catalysts are often selected for maxi-
mum sulphur removal, while NiMo catalysts have the
best hydrogenation activity and can yield an uncon-
verted product, giving the best conversion in FCC.
Trimetallic NiCoMo catalysts have often been used to
provide a balance of high HDS activity without sacrifc-
ing too much HDN activity, especially in low-moderate
pressure applications. Replacing some of that catalyst
with an amorphous or zeolitic cracking catalyst can
boost the conversion in the MHC unit by a few per cent.
Some MHC units are capable of reducing product
nitrogen to much lower levels. By applying zeolitic
cracking catalysts, conversions of 30% to 50% can be
achieved. In these cases, diesel selectivity is very high,
increasing diesel production over gasoline. Upstream
of the zeolitic catalyst, pretreat catalysts should have
suffcient HDS and HDN activity and adequate stabil-
ity to achieve the required cycle length. The HDS and
HDN activity needs to be tuned to achieve product
quality targets, while also suffciently reducing the
nitrogen to make effective use of the zeolitic cracking
catalyst. Demetallisation catalysts may be recom-
mended to cope with feed contaminants.
Albemarle has considerable experience in this area
and can assist in assessing the possibilities of running
the FCC pretreater in conversion mode. We typically
recommend an investigation of the impact of the
pretreater operation on the FCC unit in order to opti-
mise the refnerys overall proftability.
A
Robert Wade, Advanced Refning Technologies
There are a number of considerations when designing
a catalyst system for MHC service. This mode of opera-
tion tends to have higher fouling rates than
conventional high-pressure hydrocracking (HCU)
service and also has two competing constraints
compared to a single constraint for the typical high-
pressure service. In a MHC operation, cracking
conversion is controlling at the start of run (SOR) and,
as the hydrotreating (HDT) catalyst deactivates as the
run progresses, HDS and HDN conversion tends to
become controlling. It is therefore critical to understand
how the H
2
partial pressure varies throughout the run
and the impact this has on the expected fouling rate of
both the HDT and HCU catalysts. In addition, it is
important to have a good understanding of the nitro-
gen slip to the HCU catalyst so that the activity of this
catalyst is properly balanced against desired selectivity.
Achieving the proper balance of HDS/HDN activity
and cracking activity is a challenge and the system is
best customised on a case-by-case basis. The addition
of HCU catalyst to the system decreases the volume of
HDT catalyst, impacting overall HDS and HDN activ-
ity. Typically, current-generation, high-activity HDT
catalysts are recommended in these services to maxi-
mise HDN activity (minimise nitrogen slip to the HCU
catalyst) and provide suffcient HDS activity to meet
product sulphur targets. The volume and type of HCU
catalyst needs careful consideration to provide the
desired conversion and minimise the production of
excess gas and naphtha. This is especially important as
the cycle progresses and the reactor temperatures
approach end of run (EOR). As mentioned above, this
is a complex process requiring a tailored catalyst
system that is best designed by working closely with
the catalysts supplier.
A
Kevin Carlson, Business Development Manager, Criterion
Catalysts & Technologies, CriterionPublicAffairs@cri-criterion.
com and Aristides Macris, Licensing Technology Manager, Shell
Global Solutions, PT-globalsolutions@shell.com
Looking at our existing FCCPT/FCC complexes, the
pretreat unit objectives have tended to change over the
past decade with the move to clean fuels production.
In the past, the FCC PT units overall value proposition
has been a combination of environmental compliance
via sulphur removal and increased FCC yields. During
the period of 500 ppm sulphur diesel and 300 ppm
sulphur gasoline production, the pretreat objectives
were often set to produce either a side-draw blendable
low-sulphur diesel stream or resultant on-spec FCC
naphtha product, or both, with the bonus being
improved FCC yields via aromatic saturation and
www.eptq.com PTQ Q4 2012 5
ptq&a
Additional Q&A can be found at www.eptq.com/QandA
Q&A copy 12.indd 1 11/9/12 13:22:53
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 desalter water for coke cutting or are there
better recycling options for the task?
A
ITW Marcello Ferrara, Chairman, ITW, mferrara@
itwtechnologies.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 PTQ Q4 2012 www.eptq.com
:
CZlhg Lfbma% ;:L? K^gbg`:]]bmbo^l FZgZ`^k% CZlhg'
d'lfbma9[Zl_'\hf
Rapid changes in fuels supply and demand can move
the relative pricing of a refners products quickly. As a
result, a refner may fnd themselves with a non-
optimal catalyst and non-optimal unit performance.
Operational changes can be made, but will only result
in incremental performance changes. For high turnover
units, changes in catalyst formulation might be able to
keep up these rapidly changing product goals.
However, most units would be more likely to beneft
from the use of co-catalysts offered by BASF.
Co-catalysts are a standalone product category
providing the refner with the fexibility to change the
product slate of the FCC without changing the FCC
catalyst. These products are added at rates of 10-40%
of the catalyst feed and fundamentally change the
performance of the FCC unit in a short period of time.
LCO production could be increased through opera-
tional changes: decreased riser outlet temperature,
decreased catalyst activity, increased feed preheat and
increased HCO recycle. However, these changes do not
necessarily improve the proftability of the FCC due to
the resultant increase in bottoms product and loss of
product volume. Only a change in catalytic performance
can defnitely increase FCC proftability for LCO maxi-
misation. Changing the catalytic performance allows for
increased LCO selectivity lower bottoms make and
can maintain proftable product volume and value. A
catalyst reformulation could provide this change in cata-
lytic performance, but this takes too long to capture the
advantage of the rapidly changing economics.
Similarly, gasoline production can be infuenced
quickly and most effectively using a maximum conver-
sion co-catalyst as opposed to operational adjustments
and base catalyst changes. Gasoline yield is maximised
while maintaining high throughput and coke selectivity.
In one commercial example, HD Ultra, BASFs maxi-
mum LCO co-catalyst, replaced 30% of the catalyst
feed, increasing matrix activity and stability. The resul-
tant product slate saw a 2% increase in LCO and a 2%
reduction in slurry. In another commercial example,
Converter, BASFs maximum gasoline co-catalyst,
replaced 20% of the catalyst feed. Slurry oil decreased
by 4.5%, while conversion jumped by 5.2%, gasoline
grew by 3.5% and LPG increased by 2.1%. Co-catalysts
quickly changed the product slate at these refneries to
maximise proftability. Further examples can be found
in the 2010 NPRA paper AM-10-110: Co-catalysts Provide
Refners with FCC Operational Flexibility.
While operational changes are quick, they have
limited ability to change unit performance and may
have undesirable trade-offs. Base catalyst formulation
can provide fundamental unit performance changes,
but they often take too long to be of use in a changing
market. Usage of either a maximum gasoline co-
catalyst or a maximum LCO co-catalyst allows refners
to rapidly change unit performance and capture the
most favourable economics at all times, maximising
FCC proftability.
J
P^ Zk^ mZk`^mbg` fZqbfnf ikhire^g^ ikh]n\mbhg _khf
ma^?<<ngbm_hkh[obhnl^\hghfb\k^Zlhgl%[nmZk^ma^k^Zgr
g^`Zmbo^^__^\mlh_mabliheb\rhgma^jnZebmrh_hma^kikh]n\m
lmk^Zfl8
:
>kb\Khblbg%Lne_nkK^\ho^krM^\agb\ZeFZgZ`^k%:q^gl%
>kb\'Khblbg9:q^gl'\hf
When increasing the propylene yield, the gasoline
production will reduce. The gasoline properties will
change as follows:
Olefn content will stay relatively constant
Aromatic content will increase signifcantly and
could be a constraint in the gasoline pool
RON will increase; with conventional maximum
gasoline operation, FCC gasoline will reach around 92
RON. When the propylene yield is at 10 wt%, 94-95 is
very common. As a consequence, the gasoline pool will
require less reformate or lower severity reformate. It
will have a positive effect on the aromatic content in
the pool and could compensate the additional aroma-
ticity of the FCC gasoline
RVP will increase slightly because the content of C
5
in the gasoline will increase.
In order to address the last point on RVP, Axens
advises considering adding a TAME unit or an oligom-
erisation unit on the C
5
cut downstream Prime-G+ frst
step, which constitutes an ideal feed pretreatment
/IMJJ,+)*+ ppp'îmj'\hf
MI NI VAP ON- LI NE
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 11.indd 2 1/6/12 12:56:11
Q&A copy 12.indd 2 11/9/12 13:23:03
R
e
f
n
e
r
s
w
o
r
ld
w
ide use our
h
y
d
r
o
p
r
o
c
e
s
s
i
n
g
t
e
c
h
n
o
lo
g
ie
s
and catalyst
s
t
o
d
e
l
iv
e
r
c
l
e
a
n
e
r
p
r
o
d
ucts from low
-
q
u
a
lit
y
f
e
e
d
s
.
L
e
t
u
s
s
h
o
w
you how. (00
1)5
10
.2
4
2
.3
17
7
www.clg
-
c
le
a
n
.c
o
m
Learn how CLGs tailored hydroprocessing catalyst systems
maximize coversion to clean products visit clg-catalysts.com
ICR 512
ICR 180
ICR 185
ICR 250
ISOCRACKING
VGO
clean transportation fuels
ultra-low sulfur diesel (<5 ppm)
high smoke point jet (25-30 mm)
Ultra-low sulfur naphtha (<0.5 ppm)
FCC feed
clg.indd 1 23/2/12 11:55:22
We can help to reduce sludge and sediments entrain-
ment with our patented chemicals for online cleaning.
These transform the sludge into oil, with the additional
advantage of releasing the entrained sediments and
avoiding their carryover.
A
Thomas Meek, Principal Technologist Delayed Coking,
Shell Global Solutions, PT-globalsolutions@shell.com
From our operating experience, we tend not to do this
and it is certainly not recommended at any Shell
delayed coking sites due to:
Potential for desalter carry-under, resulting in oil and
sludge being routed to the cutting water tank, resulting
in:
Oil and solids, (containing Ca, Si, Ni, Va and other
inorganics)
Contamination of petroleum coke
Cutting water system fouling and corrosion
Cutting water odour, resulting in health and safety
concerns for operators performing drum opening and
decoking
Release of vapours into the environment during
coke drum opening or decoking from storage piles and
coke barns.
Our considerations may not apply to other sites.
Q
Can you provide advice on operational or equipment
measures to tackle plugging in the resid lines of our delayed
coker?
A
Thomas Meek, Principal Technologist Delayed Coking,
Shell Global Solutions, PT-globalsolutions@shell.com
In our organisation, we had success applying the
following:
Prevention or mitigation of plugging:
Steam and electrical tracing and insulation are
installed on all residuum lines. Newer heated liquid
medium tracing has not been installed on any of our
units. We understand that this has been installed at
several industry locations with good results
Flushing oil supplies and connections to unit heavy
slop or maintenance decontamination recovery systems
are installed on all residuum systems for fushing lines
and equipment such as exchangers and pumps prior to
stagnation setup
Residuum feed lines routed into delayed coking
furnace passes are steam purged cleaned with high-
pressure steam at all Shell locations under specifc SIS
(Safety Instrumented System) controls. Coke drums
inlet lines are swept clean with steam during normal
operations and with an alternate utility source during
utility outages
Pump warm-up lines with enough capacity to keep
standby pumps suffciently hot to prevent stagnation
solidifcation, equipped with a fushing capability
when the equipment is taken out of service for
maintenance
Turnaround or shutdown activities: residuum lines
and equipment are to be fushed as soon as possible to
prevent stagnation setup. Individual exchangers and
pumps have fushing oil and steam connections for
purging residuum out into connections to unit slop or
maintenance decontamination recovery systems. Best
practice is to have these systems permanently installed
along with nitrogen for air freeing prior to recommis-
sioning and startup. Residuum lines should be
decontaminated and then inspected for heavy oil
deposits such as precipitated asphaltenes or coke. If
these foulants are present, the lines are to be hydro-
blasted clean. Failure to remove foulants can result in
partial or complete plugging with shorter unit run
lengths and lost production. If a line is known to have
heavy oil foulants or coke, it must be included in the
turnaround list for hydroblasting. Lines such as
delayed coker heater outlet and coke drum overhead
vapour lines are to be hydroblasted at every turn-
around. IR camera inspection of lines has been used to
determine the extent of fouling and helps in line clean-
ing planning as well as on-the-run monitoring.
Our operations and considerations may not ft with
every approach.
Q
Our FCC regenerator NOx emissions are consistently over-
spec. Does this mean we need to adjust the balance of oxygen
supply and operating temperature? If so, how?
A
Eric F Griesinger, Marketing Manager, Grace Catalysts
Technologies
Operational variables that often have an effect on NOx
emissions have been found to include: excess O
2,
regen-
erator mixing effciency, platinum-formulated
combustion promoter, antimony-based nickel passiv-
ators and feed nitrogen. Generally, a decrease in excess
O
2
will lead to lower NOx emissions (see Figure 1).
While regenerator hydrodynamics are complex, a
change in dense bed level and/or temperatures may
provide conditions favourable to reduce NOx emis-
sions. Use of platinum-formulated combustion
promoters and/or antimony have also been observed
to increase NOx emissions. When dosed at exception-
ally high rates, even non-platinum CO promoters can
worsen NOx emissions. Oddly enough, while feed
nitrogen has been found to be the major source of NOx
emissions, typically other variables have a stronger
infuence over the actual NOx emissions. Graces
research has concluded that generally less than 5% of
the available feed nitrogen is released as regenerator
NOx emissions. The fact that only a small amount of
feed nitrogen is converted to NOx contributes to the
complexity of controlling NOx through management of
an FCC units operating conditions.
Adjusting regenerator excess oxygen can infuence
NOx emissions. A cross plot of NOx as a function of
excess O
2
can be indicative of your units response.
Typically, at low excess O
2
conditions, the NOx
increases rapidly with excess O
2
. The NOx curve then
fattens out as excess O
2
increases further (see Figure
2). Naturally, if the FCC unit is operating in the steep
part of the response curve, relatively small changes in
excess O
2
will likely yield noticeable changes in NOx.
Q&A copy 12.indd 3 12/9/12 13:39:52
cbi.indd 1 11/9/12 11:34:23
Conversely, if the FCC unit is operating in the rela-
tively at portion of the response curve, changes in
excess O
2
will not have much of an inuence on NOx.
Regenerator temperatures (dense bed and/or dilute
phase) often correlate with NOx emissions.
Unfortunately, it is difcult to predict how an increase
or decrease in these temperatures will inuence NOx
emissions. Observations from numerous commercial
catalytic NOx reduction applications demonstrate that
some FCC units observe lower NOx as temperature
increases, while other FCC units experience lower NOx
through a decrease in regenerator temperature. It is for
this reason, that Grace recommends performing a thor-
ough study of NOx emissions through the use of
statistical modelling via multiple regression analysis.
While the initial goal of any regression modelling is to
develop a useful model to predict NOx emissions, NOx
can be difcult to model. However, detailed multiple
regression modelling can point the FCC unit engineer
to potential operating knobs that are inuencing
NOx emissions, and which way to directionally move
the knobs. Further, it must be noted that such multi-
ple regression modelling efforts are specic to each
individual FCC unit, rarely transferable to other FCC
units, even FCC units of a similar design and operating
under similar conditions.
Historically, it has been found that CO promoter,
even non-platinum, low-NOx combustion promoters,
when dosed at rates in excess of the additions neces-
sary to control afterburn and/or regenerator CO
emissions, can lead to elevated regenerator NOx emis-
sions. When regenerator NOx emissions were not an
issue, the practice of dosing promoter at elevated rates
was widely accepted, as the negative side effects were
typically not noticed nor problematic. As regenerator
NOx emissions limits tighten, optimisation of promoter
dosing rates is crucial. Grace has developed an analytic
method to characterise an FCC units equilibrium cata-
lysts propensity to oxidise CO to CO
2
. This analytical
method, known as the CO Index, can assist the renery
and Graces technical staff to determine the optimum,
FCC unit-specic, CO promoter dosing rate.
While the chemistry behind the formation of NOx
from within an FCC unit regenerator is complex,
successful NOx reduction can be achieved by manipu-
lating the contributing operating variables identied by
analysis and modelling of the specic FCC unit.

A
Ray Fletcher, Engineering Fellow, Intercat, ray.etcher@
intercatinc.com
NOx emissions on the FCC unit are strongly a function
of excess oxygen in the ue gas for units operating in
full combustion, maldistribution in the regenerator and
the use of platinum-based combustion promoters.
There exists a weaker correlation with the nitrogen
content in the feed. The use of antimony for equilib-
rium nickel passivation increases NOx emissions in
many but not all FCC units. NOx emissions are directly
related to CO levels in partial-burn operations.
Control of NOx emissions requires careful trouble-
shooting to determine the root cause of the NOx
formation. For example, are the NOx levels constant
and stable or do they uctuate? Has there been a step
change increase in NOx at any point in time? The
following guidelines are offered as a starting point for
the troubleshooting and control of emissions:
Combustion control NOx emissions are directly
related to excess oxygen levels in full combustion and
CO levels in partial-burn operations. It is recom-
mended that FCC operators control either the excess
oxygen or CO concentrations at the lowest stable levels
possible. NOx emissions pass through a minimum at
zero excess oxygen and zero CO concentration.
However, operations at the zero point generally result
in unstable operations for most unit designs
Maldistribution Maldistribution in the regenerator is
the second leading cause of high NOx emissions
following excess oxygen in the ue gas. There are
several established troubleshooting techniques for
identifying maldistribution in regenerators, which
include: radial and vertical temperature proles, cross
plotting of NOx emissions versus delta cyclone outlet
temperatures (ie, set #1-set#2, set #2-set#3) and look-
ing for pattern shifts, the use of an reaction mix
sampling (RMS) device with a portable gas analyser for
direct measurement of gas concentrations, and gas and
catalyst tracer studies (Tracerco).
Intercat is now employing computational particle
1 0 0 0
1 0
1 0 0
1 2 5 0 1 3 4 0 1 4 3 0 1 5 2 0 1 6 1 0 1 7 0 0 1 7 9 0
N
O

c
o
n
c
e
n
t
r
a
t
i
o
n
s
,

p
p
m
Temperat ure, F
1
O
2
= 0 . 1 %
O
2
= 1 0 %
O
2
= 1 %
O
2
= 4 %
F C C U
Figure 1 Thermodynamics of NOx formation under typical FCC
unit conditions
1 2 0
1 0 0
8 0
6 0
4 0
2 0
0 0 . 5 1 . 0 1 . 5 2 . 0 2 . 5 3 . 0 3 . 5 4 . 0
N
O
,

p
p
m
Excess oxygen, vol %
0
S o m e c o m m e rc ia l u n its
o p e ra te in th is re g im e
with lo w e x c e ss O
2
O th e r c o m m e rc ia l u n its
o p e ra te in th is re g im e e ve n
a t ve ry h ig h e x c e ss O
2
Figure 2 Excess oxygen increases NOx: pilot plant data
Q&A copy 12.indd 4 11/9/12 13:23:26
uid dynamics (CPFD) to simulate the uid dynamics
and coke combustion chemistry in FCC regenerators to
identify design parameters leading to maldistribution.
Reners are able to identify specic regions in their
regenerator for modication. Finally, the impact of the
proposed modications can be checked to verify that
they have met the specied criteria.
In some cases, maldistribution can be minimised by
adjusting, where possible, the amount of combustion
air being delivered to specic regions within the dense
bed. In other cases, modications to spent catalyst
distributors or air distributors during the next shut-
down will be necessary
Combustion promotion Reners utilising platinum-
based combustion promoters are strongly
recommended to utilise a non-platinum-based combus-
tion promoter. Many reners observe a 50-70%
reduction in NOx emissions after the removal of plati-
num from the circulating inventory. The rate of NOx
emissions decrease is not instantaneous due to the half-
life of the platinum oxidation activity. The decay rate is
a function of the concentration in the circulating inven-
tory, catalyst inventory in the regenerator, concentration
of poisons (sulphur), and so on
Equilibrium catalyst Ni passivation The use of anti-
mony will in most, but not all, units increase NOx
emissions. Reners are recommended to discontinue
antimony injection if possible or minimise injections to
the lowest tolerable levels permissible by the wet gas
molecular weight. Catalyst suppliers are able to add
crystalline alumina to most formulations, which encap-
sulates nickel. It may be possible in some units to
eliminate antimony injection by use of this technology
Use a NOx additive Several effective additive technol-
ogies exist to reduce NOx emissions. Typical reductions
are up to approximately 50%. Addition rates are
usually very low at 1-3 wt% and so do not pose a risk
to base catalyst activity dilution. Units suffering from
excessive NOx emissions may benet from the use of
NOx additives.
A
Alan Kramer, Global FCC Additives Specialist, Albemarle,
alan.kramer@albemarle.com
NOx formation is rather complex and is affected by
several operational and design factors. Additive use
can also affect NOx emissions. The factors to consider
are:
Feed nitrogen: about 50% of the feed nitrogen typi-
cally winds up going to coke. Of the coke nitrogen,
about 5-15% ends up forming NOx. Lowering feed
nitrogen may lower NOx emissions
Excess oxygen in the ue gas: decreasing excess
oxygen will decrease NOx. There is a ne balance
between operational stability and ue gas oxygen. The
challenge is to maintain oxygen at the lowest possible
safe level
Regenerator bed level: changing the bed level may
affect NOx, but the effect is highly unit dependent. It
may be worthwhile tracking NOx versus bed level to
see if a correspondence exists on your unit
Catalyst/air mixing: similar to spent catalyst
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
subsidiaries and more than
220 engineering and sales
ofces spread across the world,
SAMSON ensures the safety
and environmental compat-
ibility 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 companies
to form the SAMSON GROUP.
Visit us at
Gastech 2012
8 to 11 Oct 2012
ExCel Centre
London
Main Hall, B30
Q&A copy 12.indd 5 11/9/12 13:23:41
distribution, better mixing usually results in lower
NOx emissions because it minimises zones of high
oxidation in the regenerator
Large regenerator temperature gradients: this is
indicative of poor mixing, and actions to mitigate
gradients without resorting to high levels of combus-
tion promoter use may help lower NOx
Spent catalyst distributor design: better distribution
will usually lower NOx. While not an operational vari-
able, it is a design variable. It could be something to
investigate and to consider improving during the next
scheduled turnaround
Regenerator design: units designed with countercur-
rent catalyst/air fow in the regenerator tend to have
lower NOx emissions
Use of a non-platinum-containing combustion
promoter: switching from a platinum to a non-plati-
num combustion promoter, such as Albemarles
Eliminox, will lower NOx emissions to the same level
of afterburn and CO control
Optimal combustion promoter use: overpromoting or
adding too much of any promoter (platinum or non-
platinum) will increase NOx emissions. Add promoter
frequently in small doses and only as much as is neces-
sary to control afterburn or CO emissions
Antimony use: if antimony additions are a regular
part of operations, make sure the minimum amount of
antimony is added to achieve the necessary passivation
NOx reduction additives: reduction can be between
0% and 80%, but is variable and often hard to predict.
A collection of experiences during EPA consent decree
trials indicated an average reduction of 26% across 28
different FCC units. (Source NPRA AM-07-44). Beware
of copper-containing NOx additives because copper
can drastically increase hydrogen yield.
Regenerator temperature does not signifcantly affect
NOx formation. N
2
fxation occurs at temperatures
greater than typical regenerator temperatures of 700-
750C. If the unit has a CO boiler, it is recommended to
keep combustion zone temperatures low, oxygen levels
below 6-7% and to avoid overfring the boiler.
A
Robert Ludolph, Principal Technologist Cat Cracking, Shell
Global Solutions, PT-globalsolutions@shell.com
From our experience, the combustion environment of a
full-burn regenerator is important in determining how
low the NOx emissions can ultimately become. Key
conditions are fue gas excess oxygen, the radial distri-
bution of combustion air (oxygen), the radial
distribution of spent catalyst, and the uniformity of
oxygen/catalyst mixing and contact time.
In general, NOx directionally increases with rising
fue gas excess oxygen. So by reducing air rate and/or
oxygen enrichment (less oxygen entering the regenera-
tor), a reduction in NOx emissions would be expected.
Also, it is generally accepted that NOx reduces to N
2
after contact with carbon monoxide (CO) and/or coke
carbon carried by spent catalyst (C). So operating in a
manner that promotes better oxygen/catalyst mixing
and contact time will also mitigate NOx.
By reducing oxygen entering the regenerator, less CO
will oxidise to carbon dioxide (CO
2
), thus providing
more opportunity for NOx to react with CO and reduce
to N
2
. Some CO, however, will enter the dilute region
of the regenerator, oxidise to CO
2
and create an unde-
sirable heat release referred to as afterburning.
Afterburning can threaten the mechanical integrity of
regenerator internals, so excessive afterburning must
be avoided.
A full-burn regenerator that mixes well and distrib-
utes oxygen and spent catalyst uniformly will allow a
refner to operate with low fue gas excess oxygen.
Such regenerator operations will tend to generate fewer
NOx emissions along with minimal afterburning. A
regenerator that mixes and distributes less effectively
will be susceptible to afterburning, be less capable of
operating at low fue gas excess oxygen and thus less
capable of mitigating NOx.
Some refners have employed combustion promoters
to reduce afterburning. These catalyst additives accel-
erate the oxidation of CO to CO
2
. Unfortunately, by
doing this, the available CO for reducing NOx is
lowered, which results in higher regenerator NOx
emissions. Other refners have also adjusted their heat
balance or raised their catalyst additions to drive the
regenerator higher in an attempt to reduce afterburn-
ing and potentially NOx.
The reaction system contributes NOx precursors via
the spent catalyst entering the regenerator. Refners
lower these precursors by improving their feed atomis-
ation and dispersion, operating more effective riser
outlet separation systems, improving stripping eff-
ciency and/or lowering feedstock nitrogen content.
The impact on NOx for each of these approaches is
quite unit specifc and diffcult to project.
Q
We encounter severe foaming problems using mixed
amines for gas.
A
Heather Sartain, Product Specialist, MPR Services, Inc,
hsartain@tkinet.com
Using mixed amine for gas sweetening is a common
practice. Any liquid will produce froth or foam if gas is
introduced into the liquid faster than the two phases
disengage. Removing the contaminants that stabilise
foam is the only effcient way to control amine system
foaming. This is done through an abatement process. If
that abatement process does not remove the contami-
nants (surfactants, solids) that cause and sustain
foaming, foaming will occur again. You can reduce and
stop the foaming by removal of the contaminants with-
out adding anything to the amine. This is done through
a mechanical means of abatement in which the foam is
separated from the solution. This can be performed on
an as-needed basis or continually and does not inter-
rupt any operations, thus allowing continuous gas
treating with any mixture of amines.
The available process for the mechanical removal of
foam-causing contaminants is SigmaPure, a unit
provided by MPR Services, Inc. The foam-causing
agents/contaminants are incorporated in the foams
Q&A copy 12.indd 6 11/9/12 13:23:52
structure. Thus, if the amine is foaming and the foam
is separated from the solution, the contaminants that
caused or enhanced the foaming are also removed (see
Figure 1).
A slipstream of the recirculating amine stream is
pumped into the SigmaPure processs foaming column
(shown on the left). Non-oxidising gas is injected into
the solution to adsorb soluble contaminating surfac-
tants and carry them to the bulk gas/liquid interface.
The surfactants are incorporated in the foam above
the gas/liquid interface and removed from the bulk
solution as it propagates up into the drying section.
Free liquid drains from the foam as it moves through
the drying section and into the stripping section of the
process. The gas and liquid fractions of the foam are
separated by breaking the foam with a stripping spray
(normally make-up water). The foam-causing contami-
nants contained in the liquid fraction of the foam are
coalesced with the stripping spray and collected as
foam waste, while the gas fraction of the foam is recy-
cled back to the foaming column for reuse.
This process has been successfully used to control
foaming in a variety of gas sweetening processes
worldwide. These processes include LNG, natural gas,
rening and chemical plants. Further, it has been used
to control foaming in a wide variety of sweetening
solvent solutions, which include DEA, MDEA, DGA,
DIPA, Flexsorb, Sulphinol and mixed amines. The
mechanical foam abatement application is not plant,
process or solution specic. Furthermore, the use of
this application will not negatively affect the sweeten-
ing solutions in any way.
Mechanical abatement is also more reliable and
endurable than antifoams. Additionally, activated
carbon and mechanical lters are known to contribute
to foaming.
If the solution can foam, it can be controlled with a
portable or permanent foam abatement process.
A
Berthold Otzisk, Technical Consultant, Kurita Europe
GmbH, otzisk@kurita.de
The mixing of amines can provide advantages for gas
sweetening operations. Improved acid gas absorption,
F e e d
R e tu rn
F o a m a te
Figure 1 SigmaPure process for amine foaming control
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
Q&A copy 12.indd 7 11/9/12 13:24:09
Nasty Stuff
Heavy crudes are here to stay.
As long as oil prices remain high,
Canadian, Venezuelan, Deep Water
Gulf of Mexico, Mexican and
other low API gravity crude oils
will play an ever more important
role in supplying world refineries.
And prices promise to remain high
because gainsayers notwithstanding,
Hubbert was right.
A big question is how to best handle
these nasty crudes? Do you
revamp existing units or invest in
new capacity? With refineries now
running flat out, the balance might
seem to favor grass roots expansion,
but given the substantial cost
multiplier over revamps, this could
be questioned. Whichever the case,
however, an inescapable fact is
that the process design of the project
will prove crucial. Between the
charge pump, the desalter and the
units' distillation columns there are
many places where miscalcula-
tions in the process design could
wreck the entire project.
Can you really be sure of attaining
desired crude rates? Desalting
viscous crude is extremely difficult.
Minimizing coking or asphaltene
precipitation in the heaters demands
extreme care. Can you reasonably
expect high diesel and HVGO
recoveries, acceptable levels of
nickel, vanadium, and microcarbon
residue (MCR)? Refiners who cut
deep should not be surprised when
the HVGO product MCR is over 2
wt % and the vanadium content is
in excess of 10 ppmw. Any one of
such difficulties can result in lower
revenue, unstable operation or
even unit shutdown. It is critical to
understand that the inherent
properties of these low API gravity
crudes dictate that exact process
design is of paramount importance.
The point of this litany of possible
problems is to remind you not to
skimp in the early phases of
engineering. From the start of the
LP work through the completion of
front-end process engineering,
actual product yield and qualities
depend on the process design.
The message is clear. Nasty crudes
will continue to make up an
increasing proportion of refineries'
crude slates. But time is precious.
The sooner we face this fact,
unwelcome as it may be, the more
expeditiously we can adapt.
For a more in depth review of
heavy crude challenges, ask us for
Technical Papers 173, 185 and
197.
3400 Bissonnet
Suite 130
Houston, Texas 77005
USA
Ph: [1] (713) 665-7046
Fx: [1] (713) 665-7246
info@revamps.com
www.revamps.com
This exchanger has seen better crude slates
pcs nasty.indd 1 1/6/12 11:58:35
reduced energy requirements or additional carbon
capture are notable benefts. Applied mixtures are
formulations of primary-tertiary amines, secondary-
tertiary amines, tertiary-tertiary amines or formulated
amines with additional activator molecule or hindered
amines.
There are a number of possible reasons why your
mixed amine system is suffering from foaming. In
general, clean amines do not show a foam formation;
this is why contaminants could be the foam initiator.
Contaminants in the feed gas are small particles or
degradation products. When the surface tension is low,
liquid droplets (aerosols) can surround gas bubbles to
create foam. Fine particles such as coke or iron sulphide
form a layer in the flm around the gas bubble. The
installation of an effective amine fltration system helps
to reduce foaming. The fltration equipment should
include a suitable separator, gas liquid coalescer, acti-
vated carbon flter and cartridge flter.
Primary and secondary amines show a fast reaction
with CO
2
to form carbamates. The higher heat of reac-
tion with CO
2
due to the carbamate reaction requires
more energy for regeneration. The regeneration of
mixed amines is more diffcult to manage because of
the different reboiler duty. The result is often a thermal
degradation of primary or secondary amines. A better
temperature control during regeneration will help to
reduce the degradation of the mixed amine.
An underestimated problem is oxygen contamina-
tion. Oxygen can react at low ppm levels with amines
to form carboxylic acids. These acids react with amines
and degradation products to form fatty acids or long-
chain organic acids. Fatty acids increase the
hydrocarbon solubility in the amine solution. Mixed
amines often show a higher absorption tendency of
heavier hydrocarbons such as aromatics, resulting in
foam formation. Maintaining the temperature of the
absorber above the hydrocarbon condensation temper-
ature will help to reduce the foaming potential.
Antioxidants will scavenge oxygen, while surface
active antifoams will eliminate foam. Antifoams for gas
sweetening processes are designed to prevent the
formation of foam and to break existing foam.
Antifoams and corrosion inhibitors play an impor-
tant role in the gas sweetening process. They are
surface active components and must be used with a
thorough understanding of the chemistry. The selection
of less suitable additives can increase the risk of foam-
ing. Overdosing of silicone-based antifoams can create
foaming. At an elevated concentration, the antifoam is
tied up in micelle structures. They form lenses and are
no longer spread on the interface.
Underdosing of an antifoam also increases the poten-
tial for foam formation if there is a need to inject an
antifoam. Activated carbon adsorbs surface active
contaminants and hydrocarbon condensates. Silicone-
based antifoams are removed there as well, which
reduces the antifoam concentration in the mixed amine.
Misuse of additives can produce severe consequences,
which is why a correct application with a proper
concentration is mandatory.
Q
We are planning to process heavy crude feed with a high
aromatics content. What changes will I need to make to the
FCC catalyst to crack those aromatic rings?
A
Eelko Brevoord, Technical Services Manager, Refning
Catalysts, Albemarle, eelko.brevoord@albemarle.com
Aromatic feeds are more hydrogen defcient and are
more diffcult to crack. As a rule of thumb, every 10 kg/
m
3
density increase results in a drop in conversion of
3%. At the same time, aromatics are absorbed on the
catalyst surface and form coke. In order to crack heavy
aromatic molecules to valuable products, the preferred
cracking reaction must take place before dehydrogena-
tion and condensation reactions occur, creating coke. A
very accessible catalyst is required, where large mole-
cules can easily diffuse into the catalyst pores.
Precracking on selective matrices, followed by zeolite
cracking, is the preferred reaction pathway. Albemarle
has developed the Upgrader catalyst family specifcally
for this purpose. Commercial experience has proven its
bottoms cracking characteristics that maximise the refn-
ers proftability. Its metals resistance allows a refner to
maintain a high conversion at high concentrations of
feed contaminants, which often coincides with process-
ing heavy feed. The exact catalyst formulation needs to
be tuned to meet the product quality and selectivity
targets, while also addressing the units constraints.
A
Robert Ludolph, Principal Technologist Cat Cracking, Shell
Global Solutions and David Hunt, Senior Technologist Fluid
Catalytic Cracking Design/Licensing, Shell Global Solutions,
PT-globalsolutions@shell.com
FCC catalysts are not designed to crack aromatic rings.
Feedstocks with a high aromatic content occur through-
out the gas oil and residuum boiling ranges, which can
be categorised as heavy crude feed. Aromatic feedstocks
lack hydrogen content, are diffcult to crack and tend to
produce more coke per unit of conversion than feed-
stocks containing more hydrogen. As a result, delta coke
will increase, the regenerator temperature will also
increase and the catalyst circulation will decline. Also,
since aromatic feedstocks are more diffcult to crack,
unit conversion will decrease, producing more bottoms,
especially if the feedstock is residuum. Lighter products
will have higher olefn and/or aromatic contents.
In general, to address the delta coke increase, it would
be desirable to regain the loss in catalyst circulation.
Reducing the feed temperature to the riser is an option
if coke burning capacity is available. Catalytically, a
reduction in the circulating activity should be consid-
ered. Reducing zeolite content is a common approach
for accomplishing this. Also review the zeolite rare earth
content; potentially, you could improve coke selectivity
while maintaining good zeolite stability.
To address the loss in conversion and bottoms upgrad-
ing, an increase in matrix content should be considered.
Additional matrix can help in maintaining your bottoms
yield depending on how heavy the aromatic feedstock
becomes. Be careful not to let the matrix acidity get too
high; coke selectivity will suffer and may negate the
upgrading beneft of additional matrix.
Q&A copy 12.indd 8 11/9/12 13:24:26
3400 Bissonnet
Suite 130
Houston, Texas 77005
USA
Ph: [1] (713) 665-7046
Fx: [1] (713) 665-7246
info@revamps.com
www.revamps.com
PROCESS
CONSULTING
SERVICES,INC.
upsets from water slugs and
other unpredictable situations
that have damaged internals,
resulting in diluent losses and
high vacuum unit overhead con-
densable oil. Diluent is neither
cheap nor plentiful, and high
vacuum column operating pres-
sure will reduce overall liquid
volume yields. And if the design
of the delayed coker fractionator
is based on todays experience
with conventional heavy feed-
stocks you will be lucky to run
six months.
What all this means is that
special process and equipment
designs are needed to satisfy
the special demands of pro-
cessing oil sands crudes. Such
processes are not generated by
computer based designers who
have little or no experience and
never leave the office. They are
developed only by engineers
with know-how who have real
experience wearing Nomex
suits
and measuring true unit per-
formance in Northern Alberta.
Shouldnt this be kept in mind
by those considering long term
supply agreements?
Oil Sands Crude
Profits and
Problems?
Canadian bitumen production
currently runs about 1 MMbpd,
with some being sold as Synbit
and Dilbit. Over the next 10-12
years output is expected to
increase to 3.5 MMbpd and more
refiners will begin investing to
process it and come to depend
on the Synbit and Dilbit for a
significant part of their supply.
Few today, however, have ever
processed these feeds at high
blend ratios, and are unaware
that conventional process and
equipment designs are not up
to the job. Canadian oil sands
feedstocks are extremely hard
to desalt, difficult to vaporize,
thermally unstable, corrosive, and
produce high di-olefin product
from the coker. If you intend to
lock into a long-term supply,
therefore, it is imperative that you
consider reliability and run length
from a particular design.
Too low tube velocity in the
vacuum heater tubes will lead to
precipitation of asphaltenes. Too
fast a flow rate will erode the
tube bends. If coil layout, burner
configuration and steam rate are
not correct, run length will be
measured in months, not years.
Diluent recovery unit designs
must take into account possible
For a discussion of factors
involved in designing refinery units
to process difficult oil sands feed-
stocks, ask for Technical Papers
#234 and 238.
10 PTQ 01:10 01 PC PTQ 0107 ADF 10/19/07 4:42 PM Page 1
pcs 2.indd 1 1/6/12 21:05:28
So, the zeolite-to-matrix ratio of
the catalyst system will tend to shift
lower if the aromatic feed contains
residuum; the ratio may be similar
or higher if the aromatic feed is
mostly gas oil. Catalyst addition
rates may need to shift higher in
order to manage feedstock metals.
Metal traps or equilibrium fush
catalyst could also need to be
considered. The formulation may
need further adjustment if the feed-
stock nitrogen is also high; aromatic
nitrogen can have a profound
impact on cracking reactions, driv-
ing unit conversion down further.
A
Rosann Schiller, Senior Marketing
Manager, Grace Catalysts Technologies
The aromatic cores of heavy feed
molecules will remain largely
uncracked. Without the proper FCC
catalyst design, these aromatic
compounds will condense on the
catalyst surface and lead to increased
coke yield, high regenerator temper-
atures and lost conversion. To
effectively crack a heavy aromatic
feed, an FCC catalyst with a
balanced zeolite-to-matrix (Z/M)
ratio, an appropriate hydrogen
transfer activity and an optimum
pore structure is required. Industry-
leading coke selectivity, such as
offered by Grace FCC catalysts, is
crucial for units limited by regenera-
tor temperature and/or air blower
constraints. A properly balanced Z/
M ratio minimises coke formation
and maximises liquid product.
The high-activity matrix incorpo-
rated into Graces resid catalysts
possesses the majority of its poros-
ity in the crucial 100-600 Angstoms
() pore size diameter range. This
mesoporosity allows for the free
diffusion of heavy feed molecules
into the matrix, where pre-cracking
is necessary to achieve complete
vapourisation. The selective-active
matrix is also responsible for the
conversion of naphthenoaromatic
rings, reducing aromatic feed
species into gasoline and LCO
range product rather than coke.
Hydrogen-defcit feeds require
higher hydrogen transfer activity in
the zeolite. The balance of mesopo-
rosity and optimised acid strength
greatly improves the selectivity of
the bottoms cracking reactions that
take place without the gas penalty
often observed when cracking
aromatic feeds.
The specifc FCC catalyst reformu-
lation directives for any FCC unit
shifting to a heavier, more aromatic
feed are strongly dependent upon
the current base catalyst technology.
Graces broad FCC portfolio
provides the fexibility to re-optimise
the catalyst recipe for the specifc
refner, feedstock, and projected unit
constraints and objectives.
Q
How effective is software simulation
for calculating mass transfer in trays
and packings?

A
Ralph Weiland, President, Optimized
Gas Treating, Inc., ralph.weiland@ogtrt.
com
Truly reliable, predictive simulation
software does not use equilibrium
(ideal) stages, effciencies, HETPs,
residence times of ideal stages, heat
transfer effciency, or any other
information that cannot be read
from a PFD or internals vendor
drawing to translate ideal stages into
real trays or physical depths of real
packing to approximate the perfor-
mance of real tower internals.
Software that does is using the best
guesses of the user of the software
to fll the huge gap between ideality
and reality. Such software is espe-
cially unsuitable for dealing with
packing. The reality is that mass
transfer rates on trays and in pack-
ing depend on column hydraulics,
physical and transport properties,
and the mass transfer characteristics
of the internals. If the model is not
100% mass transfer rate based, it is
either ignoring the tower internals
altogether or is providing only a
very rough estimate of performance.
A strictly mass transfer rate-based
simulator such as ProTreat is the
only reliable way to predict column
performance. Other software is
useful if you have data from the
plant in question, or a similar plant,
that can be used to tune the model.
But such a model needs to be ftted
to data and is therefore not really
predictive. If you are dealing with
packing, mass transfer rate-based
simulation handles all sorts of
random and structured packing very
reliably and is the only way to go.
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: 281-331-595 Fax: 281-585-1Z80
amacs@amacs.com
24-hr Emergency Service:
1-800-231-0077
Q&A copy 12.indd 9 12/9/12 13:40:52
Your 8ingle 8ource for New or Retroht nternals
Why wait 10, 15 or 20 weeks for your delivery! Ask AMACS
about our Fast Track delivery for trays and a complete range
of tower internals. With complete in-house engineering and
fabrication, we can use your existing drawings or modify them
to improve your process.
Trays (numerous options)
Sieve or perforated
Bubble cap trays
Cartridge trays
DuaI ow
Bafe
Valve
www.amacs.com 24hr EMERGENCY 8ERVCE 1-800-231-0077
Mesh-numerous alloys-plastics
Vane (Chevron)
Double Pocket Vane
Candle or Fiberbed
Insertion Misthx
MIST ELIMINATORS
DISTRIBUTORS & SUPPORTS
Manufactured to customer
specihcations or engineered to
meet performance requirements.
COALESCERS
Oil water separations Haze removal
from fuels Removal of tower wet
rehux Caustic treater applications
STRUCTURED PACKING
Woven, sheet metal, and knitted
structured packing. Built to spec
or performance requirement.
RANDOM PACKING
Catalyst bed supports
Basket strainers
Nozzles Outlet/Inlet baskets
Distributors
Hub and header laterals
WEDGE WIRE SCREEN
Wide variety of random packing
types, sizes and materials in
stock! Ask us how our patented
SuperBlend 2-Pac can increase
your capacity and efhciency.
amacs.indd 1 11/9/12 11:30:16
Survey: refnery approaches to energy effciency
Hydrocarbon Publishing Company
According to the US Energy Information
Administration (EIA), energy usage represented 43% of
the overall refnery operating expense, excluding crude
purchases, in 2010.
1
In European refneries, energy
spending can account for as much as 60% of the oper-
ating budget, according to a CONCAWE report.
2
The
fgure has been doubled for the last two decades due
to rising energy costs, increasingly tight fuel standards
and higher refnery complexity. A small fraction in
energy savings can translate into billions of dollars. As
refning margins are rather slim in the down cycle, cost
reduction via energy improvement may turn a loss into
a proft.
Drivers for energy effciency can loosely be catego-
rised as internal or external. Primary internal drivers
for energy effciency include the desire to improve
productivity and proftability, reduce fuel consumption
costs in the face of volatile prices, improve plant safety
and personnel satisfaction, and optimise plant opera-
tions and maintenance. External drivers for energy
effciency can come in the form of government regula-
tions and mandates, industry competition, availability
of cost-effective technologies, and from the public in
regards to the social responsibilities of an industrial
company.
Hydrocarbon Publishing Company conducted a
survey among refners during Q4 2011 in order to help
the industry better understand refners views on
energy effciency and to highlight which approaches to
energy effciency work best. The survey, in particular,
examines where refners can improve energy effciency
in both utility providers (such as heat, steam and
hydrogen) and energy consumers such as processing
units. Forty-four refneries, each with crude oil process-
ing capacities of more than 50 000 b/d, responded to
the survey, which contained 17 questions. Around 40%
of the responses came from refneries located in North
America (18), followed by Europe (11), Asia/Asia-
Pacifc (seven), Latin America/South America and the
Middle East (three each), and Africa (two).
Importance of energy effciency
A large majority of survey respondents (89%) view
energy effciency as important to very important,
while only 9% reported that their refnery viewed
energy effciency as not very important. (2% of
survey respondents did not respond to this question.)
Around 78% reported that energy effciency at their
refnery has improved in the last fve years, while 13
indicated that there has been no improvement or a
drop in energy effciency over the last fve years. When
asked about the reasons behind energy management
projects the participants refneries have performed,
Reduce gasoline cutpoint
70% of respondents reported that their refneries
undertake energy management projects due to a combi-
nation of environmental (lower CO
2
emissions) and
economic (reduced operating costs) incentives. The
remaining 30% indicated that energy management
projects are performed solely for economic incentives.
Energy management team and site-wide assessments
One of the frst steps for improving energy effciency
throughout the refnery is forming and/or hiring some
sort of energy management team and conducting a
site-wide assessment. Refneries cannot improve their
energy effciency without frst understanding the
plants energy consumption and designating someone
to examine opportunities to improve effciency. Almost
half of the survey respondents reported that their refn-
ery had an in-house team designed to examine and
improve energy effciency at their facility. One-third of
the respondents reported using both an in-house team
and a third-party consulting service to manage refnery
energy use. Meanwhile, every respondent indicated
that at least one site-wide assessment had been
performed at their plant. Around 11% of respondents
reported that such assessments were performed more
frequently than every six months, 34% indicated
annual assessments, and 14% reported assessments
take place every two years.
Prioritising energy effciency
Prioritising energy effciency and developing an
energy effciency mindset is important for a successful
energy management programme. One of the frst
places to start developing an energy effciency culture
is with new employees. Teaching employees about the
importance of energy effciency early on can help
them to continually think about opportunities to
improve energy effciency and reduce operating costs
at the refnery. Of the refneries surveyed, over half
reported that energy effciency is currently included
in their new employee training programme. Another
aspect that is benefcial for successfully implementing
energy effciency projects is budgeting money specif-
cally for such projects. Over half of respondents
appear to have realised the importance of setting
aside money for energy effciency projects and
currently do so.
Survey participants were also asked about the role
that energy effciency considerations play in the selec-
tion of new units. When asked about new process
units, only 7% of respondents indicated that energy
effciency was a primary concern when selecting new
units. A greater percentage of respondents, 18%, indi-
cated that energy effciency was a primary concern
when examining auxiliary equipment. For each case, it
is notable that every respondent acknowledged that
Processing Trends
proc trends copy 10.indd 1 11/9/12 13:42:00
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
aveva.indd 1 11/9/12 11:31:49
three levels of investment discussed. The frst set of
projects considered will be viewed as short-term eff-
ciency improvements. These solutions comprise those
identifed as operational adjustments that can be
achieved with relatively little commitment of time and
money; these types of improvements are often referred
to as good housekeeping practices as they primarily
involve getting the most out of the existing confgura-
tion. The next set of solutions will be oriented toward
longer-term goals and will encompass the maintaining
of the frst-level projects while also including the
additional investment opportunities identifed.
Lastly, innovative solutions that are targeted in a
medium- to long-term frame of time incorporate the
higher-end investment options that are described as
major revamps.
In each level of investment, improvements can be
made to utility providers and systems (such as furnace,
boiler, pump, compressor, fan and heat exchanger) as
well as specifc units (such as CDU, coker, visbreaker,
HT and HC, catalytic reformer and alkylation.)
some level of consideration is given to energy
effciency during new unit/equipment selection.
Major sources of energy ineffciency
The three most commonly cited sources of ineff-
ciency include process heaters, steam distribution
system and waste heat, with 74%, 72% and 67%
of the survey participants indicating these
sources, respectively. Fouling and boilers were
also signifcant concerns, with around 50% of
participants identifying these issues as problem-
atic. Figure 1 shows major sources of refnery
ineffciency identifed by survey respondents.
Refners answering Other cited turbines, heat
integration and unplanned shutdowns as addi-
tional sources of energy ineffciency.
Refners seem to recognise the crude distilla-
tion unit (CDU) as a process with signifcant
room for improvement; 76% of respondents
selected the CDU as one of the process units in
their plant with a lot of room for improvement
with regards to energy effciency. The second
most commonly cited unit was the fuid
catalytic cracker (FCC), with 37% of respondents
identifying this process unit. Nearly a quarter of
participants also indicated that there is major
room for improvement in the catalytic reformer,
hydrocracker (HC) and coker. Figure 2 shows
process units with the most room for effciency
improvements based on survey respondents
replies. Refners answering Other cited high
vacuum units, propane deasphalter and aromat-
ics complex.
The most commonly implemented project in
refneries was adjusting operating conditions on
a process unit to reduce energy consumption,
with 93% of respondents indicating that their
refnery has done so to increase effciency in the
last fve years. Adjusting heat exchanger
networks (for instance, rearranging and/or
adding new heat exchangers to increase heat recovery)
and improving insulation closely followed as popular
choices, with 80% of respondents each indicating their
refnery had performed these projects. Less common
projects included installing a cogeneration facility, with
30% of respondents performing this project in the last
fve years. Figure 3 shows energy effciency projects
completed in the last fve years by survey respondents.
Refners answering Other cited performing steam trap
studies, generating steam from waste heat, optimising
heaters, performing online cleaning for heaters, and
installing a magnetic resonator in the naphtha/gas line
of gas turbine/heat recovery steam generator as addi-
tional energy effciency projects.
Three options to improve energy effciency
The surveys results show commitment yet budget
concerns over energy effciency improvements in a
refnery. Refners are recommended to group the
energy-saving opportunities into three categories for
implementation, which will often correlate with the
6 0
8 0
7 0
5 0
4 0
3 0
2 0
1 0
P
r
o
c
e
s
s

h
e
a
t
e
r
s
W
a
s
t
e

h
e
a
t
F
o
u
l
i
n
g
B
o
i
l
e
r
s
D
i
s
t
i
l
l
a
t
i
o
n
H
y
d
r
o
g
e
n

s
y
s
t
e
m
P
u
m
p
s

a
n
d

m
o
t
o
r
s
C
a
t
a
l
y
s
t

d
e
-
a
c
t
i
v
a
t
i
o
n
L
i
g
h
t
i
n
g
O
t
h
e
r
S
t
e
a
m

d
i
s
t
r
i
b
u
t
i
o
n

s
y
s
t
e
m
R
e
s
p
o
n
d
e
n
t
s
,

%
Maj or sources of i neff i ci ency
0
Figure 1 Major sources of energy ineffciency
6 0
9 0
8 0
7 0
5 0
4 0
3 0
2 0
1 0
C
D
U
F
C
C
H
C
C
o
k
e
r
H
T
A
l
k
y
l
a
t
i
o
n
V
i
s
b
r
e
a
k
e
r
O
t
h
e
r
C
a
t
.

r
e
f
o
r
m
e
r
R
e
s
p
o
n
d
e
n
t
s
,

%
Process uni t
0
Figure 2 Refnery units with the highest potential for effciency
improvements
Conclusion
The survey results also reveal the key synergistic role
played by energy management in overall operations
(see Figure 4). Energy management directly affects
refnery margins, as energy savings go straight to the
bottom line. At the June 2012 Industrial Energy
Technology Conference (IETC) in New Orleans,
Louisiana, LyondellBasell received the Energy Award
for reducing energy consumption at its Houston, Texas
(US), plant by 12% via the implementation of an energy
effciency programme. The programme has helped save
the refnery 42 trillion Btu since 2007, with the fare opti-
misation programme alone saving over 500 billion Btu/y.
At an assumed energy cost of $5.50/MMBtu, the refner
saved an average of $46.75MM a year via the implemen-
tation of the energy effciency programme.
The execution of functional
safety assessment and validation
must take place on every safety
project and must be carried out by
functional safety experts
0 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 1 0 0
Respondent s, %
E
n
e
r
g
y

e
f
f
i
c
i
e
n
c
y

p
r
o
j
e
c
t
O th e r
I n sta ll c o g e n fa c ility
S witc h to a lo w
te m p . c a ta lyst
S witc h to a m o re E E
p ro c e ss d e sig n
O p tim ise p u m p /fa n size
I n sta ll a n a ir
p re h e a te r/e c o n o m ise r
A d ju st lig h tin g
F o u lin g c o n tro l
I m p ro ve in su la tio n
A d ju st h e a t
e x c h a n g e r n e two rk
A d ju st o p e ra tin g
c o n d itio n s o n a u n it
I n sta ll h ig h
e ffic ie n c y m o to rs
I m p ro ve h yd ro g e n
p ro d u c tio n /d istrib u tio n
Figure 3 Energy effciency projects completed in the last fve years
Energy
management
Environmental
regulation
compliance
Asset
management
Profit margin
Figure 4 Synergistic role of energy
management
Determination of Sulfur and other elements
at-line and in the laboratory
El ement al Anal ysi s
of Fuel s
To keep pace with the demanding quality requirements of
modern fuels, advanced, precise and easy to use analytical
technology is required. With a complete range of XRF and
ICP spectrometers, SPECTROs unique solutions for at-line
and laboratory elemental analysis are capable of meeting
the most demanding product specification testing
requirements.
Measure your fuels
- at sub 10-15 ppm levels of sulfur 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
www.spectro.com/fuels
Visi t us at
digitalrefining.com
Energy management is also intricately connected to
asset management, as well-planned maintenance can
result in less energy consumption in units generating
and distributing utilities. Additionally, meeting produc-
tion targets and quality standards will require extensive
data collection that can be used in both asset manage-
ment and energy management systems. A number of
energy improvement strategies will go hand in hand
with improvements in terms of plant maintenance, reli-
ability and safety. Maintenance- and reliability-driven
projects such as fouling control, maintaining/fxing
insulation, repairing leaking pipes and conducting cata-
lyst change-outs will typically lead to either direct or
indirect improvements in energy effciency. Asset
management is known to yield proftability and produc-
tivity improvements, primarily through minimising
downtime. According to consultancy ARC, unscheduled
shutdowns coupled with poor maintenance practices
cost the global process industries 5% of their annual
production, which is equivalent to $20 billion.
3
The frm
estimates that 80% of the losses are avoidable. Ineffective
maintenance practices also lead to unscheduled down-
time, costing operators an additional $60 billion a year.
In terms of environmental regulations, faring is an
area that is directly linked to energy management
programmes. Excess faring, in particular, can result in
negative publicity and fnes from local and national
governments. Flare gas recovery sending the fue gas
that would normally be sent to the fare back through
the unit for use as fuel or feed provides a dual beneft
of saving fuel and reducing emissions. In April 2012, US
refner Marathon Petroleum announced an agreement
with the EPA to reduce volumes of gas sent to fares.
The company has already invested around $45MM to
boost combustion effciency at its plants and intends to
spend another $6.5MM to comply with the agreement.
According to the EPA, the proposed changes will cut
emissions from Marathons six refneries by about 5.4K
t/y. In the longer term, energy management is able to
minimise the risk of violating carbon legislation if/when
it is passed and supporting corporate sustainability
goals. Also, with energy consumption monitoring being
built into the asset management system, refners will be
provided with an easy path to monitoring, accounting
and managing carbon emissions if/when it becomes
mandated at a future date.
The surveys questions can be viewed at www.hydrocarbonpublishing.
com/survey/eesurvey.pdf. For further analyses, see Hydrocarbon
Publishing Companys report, Refnery Energy Management: Latest
Technologies and Strategies to Enhance Operational Economics, July 2012.
Contact: Thomas Yeung at tyeung@hydrocarbonpublishing.com
References
1 US Energy Information Administration, Table 2.1d Industrial
Sector Energy Consumption Estimates, 1949-2010 in Annual Energy
Review 2010, (Washington, DC: US Department of Energy, Oct 2011),
http://205.254.135.24/totalenergy/data/annual/pdf/sec2_9.pdf
(accessed 25 Jan 2012).
2 EU Refnery Energy Systems, CONCAWE report no 2/12, Mar 2012.
3 Hollywood P, Making the business case for reliability, Hydrocarbon
Processing, Jun 2012, 15.
Custom Support Grids
Johnson Screens Shaped Support Grid
The new Johnson Screens
patented Shaped Support Grid
matches the contour of the
vessel head. The innovative design
creates increased volume for catalyst or
molecular sieve
in the vessel,
provides uniform
fow across the
entire vessel
cross section,
allows a low cost
bed support and
eliminates the
need for costly
beams and
support ledges.
Leading
the way in
applying innovative
technologies to vessel
internals in Refning,
Petrochemical and Gas
applications for more
than 40 years.
Contact Us Today!
AUSTRALIA - ASIA PACIFIC
TEL: +61 7 3867 5555
salesasiapacifc@johnsonscreens.com
EUROPE - MIDDLE EAST - AFRICA
TEL: +33 5 4902 1600
saleseurope@johnsonscreens.com
NORTH, SOUTH & CENTRAL AMERICA
TEL: +1 651 636 3900
hpisalesnsa@johnsonscreens.com
JAPAN
TEL: +81 55 997 8511
sales.jpn@johnsonscreens.co.jp
www.johnsonscreens.com
Johnson Screens Flat Support Grid
The Johnson Screens fat support grid is custom
designed and engineered
for each vessel,
accommodating
many design
features and
vessel shapes,
ensuring you get
the most
out of your process every time.
HPI Support Grid Advert.indd 1 31/08/2012 8:31:58 AM
j screens.indd 1 11/9/12 11:48:23
Challenges and opportunities of 10 ppm
sulphur gasoline: part 2
A
n emerging worldwide
standard for ultra-low-
sulphur gasoline (ULSG) as
well as the challenges of increased
heavy crude supplies and the
gasoline/diesel imbalance demand
careful consideration and selection
of processing options. Part 1 of this
article (see PTQ, Q3 2012) discussed
commercially proven confgurations
that are available to meet these
constraints and maintain proftabil-
ity. An economic study, presented
here, was also conducted to deter-
mine the best scenario to meet
ULSG requirements: severe FCC
feed pretreatment alone or less
severe pretreatment coupled with
FCC gasoline post-treatment. The
impact of catalytic feed hydrotreater
(CFHT) cycle length requirements,
with and without post-treatment,
was also examined.
An existing refnery reconfgured
to process heavy Canadian crudes
while maintaining its FCC unit was
assumed. The VGO feedstock
consists of a 55 000 b/d blend of
Economic evaluation of processing options for ultra-low sulphur gasoline compares
severe pretreatment with a combination of pre- and post-treat solutions
DELPHINE LARGETEAU, JAY ROSS, MARC LABORDE and LARRY WISDOM
Axens
straight-run VGO and heavy coker
gas oil with 4.2 wt% sulphur. Due
to the refractory nature of this feed,
it has to be hydrotreated in a high-
pressure unit prior to feeding the
FCC unit, and the resulting gasoline
constitutes about one-third of the
total gasoline pool and all of the
pool sulphur. The following three
cases were considered:
Case 1: A high HDS CFHT unit
and FCC unit capable of producing
10 wppm gasoline pool sulphur
without the need for a FCC post-
treatment unit with a CFHT cycle
length of four years to match the
FCC unit
Case 2: A moderate HDS CFHT
designed for a four-year cycle
length with a FCC post-treatment
9 6
1 0 0
9 8
9 4
9 2
0 C a se 1 C a se 2 C a se 3
C
F
H
T

D
H
S
,

%
9 0
3
5
4
2
1
C
y
c
l
e

l
e
n
g
t
h
,

y
e
a
r
s
0
C F H T H D S
C yc le ye a rs
9 2 . 9 % 9 2 . 9 %
9 9 . 5 %
Figure 1 CFHT HDS and cycle length
VG O
C a se 1
C a se s 2 &3
G a s o il ( 6 5 0 F +
D istilla te
( 4 0 0 -6 5 0 F )
G a so lin e
G a so lin e
L C O
S lu rry
C
1
- C
4
4 . 2 wt% S 2 3 0 wtp p m S
3 0 0 0 wtp p m S
CFHT FCC Prime-G+
C
1
- C
4
C a se s 2 &3
C a se 1
ULSG pool
Figure 2 Case studies block fow diagram
axens.indd 1 14/9/12 12:34:09
price of $4 MMBTU, resulting in a
hydrogen cost of $3.300 MSCF.
The hydrogen cost for Case 1 is
almost 25% higher than for Case 2
or Case 3; however, the yield
improvement is quite signifcant
over the lower severity CFHT cases.
Between the lower severity CFHT
cases, the yields and hydrogen
consumption are rather similar, with
the more severe and longer cycle
Case 2 providing a slight improve-
ment in terms of yields over Case 3
commensurate with the small
increase in hydrogen consumption.
With regards to the operating cost
(opex) of the different cases, the
study took into consideration the
hydrogen, octane and utility costs.
Compared to the other factors, the
hydrogen cost was by far the major
contributor to the opex. In addition
to the operating cost, a detailed
total capital investment (TCI) was
developed to estimate the capex for
each case.
The TCI trend illustrated in
Figure 3 clearly shows that Case 1
has a much higher capital require-
ment than the other two cases due
to the signifcantly higher desul-
phurisation and cycle length
requirements for the CFHT.
Both net present value (NPV) and
internal rate of return (IRR)
comparisons are shown in Figures
4 and 5. High-severity CFHT with-
out post-treatment, Case 1, was
considered as the basis, and the IRR
and NPV of the other cases were
compared to Case 1.
The NPV results favour Case 1
with a high HDS/long cycle length
CFHT and no post-treatment over
more moderate HDS CFHT cases
coupled with a post-treatment unit.
On the other hand, the IRR is most
favourable for Case 3 with the
unit (Prime-G+), also designed for a
four-year cycle length to meet
ULSG pool specifcations
Case 3: Similar to Case 2 but with
a two-year cycle length target for the
CFHT unit combined with a Prime-
G+ unit designed for a four-year
cycle length. During the CFHT cata-
lyst change-out, the Prime-G+ unit
will operate at a higher severity to
meet pool sulphur requirements.
For all cases, a relatively high
pressure was selected for the CFHT
to ensure good hydrogen addition
during the whole run. Reactor resi-
dence time was adjusted to meet
the CFHT HDS and cycle length
requirement (see Figure 1). The
very severe level of HDS and four-
year cycle length in Case 1 naturally
lead to a much larger CFHT than
the other cases. High-purity hydro-
gen is supplied from a steam
methane reforming plant.
A block fow diagram illustrating
the three different cases with the
various confgurations along with
the corresponding products consid-
ered for the economics is shown in
Figure 2.
The economic evaluation was
based on a discounted cash fow
(DCF) analysis assuming a depreci-
ation period and a project duration
of 10 years. In addition, a proftabil-
ity index comparison in terms of
net present value (NPV) and inter-
nal rate of return (IRR) was
conducted. The prices for invest-
ment, catalysts, utilities, feedstock
and fnished products were based
on 2011 averaged values, assuming
the plant to be located in the US
serving a domestic market. Prices
are presented in Table 1.
For all three cases considered,
projections on CFHT and FCC
operations were conducted, leading
to expected product yields and
hydrogen requirements. As one
could have expected, the implemen-
tation of a high-severity CHFT
(Case 1) leads to better product
yields in the FCC unit, but has a
major drawback of driving up
hydrogen consumption. Results in
terms of main product yields and
hydrogen cost for each case are
presented in Table 2. The evalua-
tion was based on a natural gas
8 0
1 0 0
9 0
7 0
6 0
T
C
I
,

%

(
c
a
s
e

1
)
5 0
B a se
Figure 3 TCI impact
Feedstock, $/bbl 96
Natural gas, $/MMBtu 4.0
Hydrogen, $/MSCF 3.300
LPG, $/bbl 69
Propylene, $/bbl 140
Butenes, $/bbl 112
Gasoline premium, $/bbl 127
Diesel/LCO, $/bbl 131
Fuel oil, $/bbl 104
Price considerations
Table 1
Case Case 1 Case 2 Case 3
New units CFHT CFHT + CFHT +
post-treat post-treat
Cycle length 4 yr 4 yr + 4 yr 2 yr + 4 yr
Gasoline yield, vol%/VGO feed 61.9 56.3 55.0
Diesel + LCO yield, vol%/VGO feed 27.2 27.6 28.0
Propylene yield, vol%/VGO feed 7.8 7.5 7.3
Butenes yield, vol%/VGO feed 8.8 8.3 8.1
Hydrogen cost, $/bbl feed 4.71 3.73 3.66
Study results product yields and hydrogen requirement
Table 2
axens.indd 2 11/9/12 13:48:33
sulzer.indd 1 11/9/12 12:03:13
eurecat.indd 1 23/2/12 12:02:36
lowest cost CFHT option (moderate
and two-year cycle) coupled with a
four-year cycle post-treatment
Prime-G+ unit.
A sensitivity case was examined
to determine the impact of natural
gas cost on the NPV results. The
fndings are highlighted in Table 3,
where pricing is contrasted to the
2011 basis above. Assuming a
higher natural gas price ($6
MMBTU vs $4 MMBTU), the cost of
hydrogen increases and the differ-
ence in NPV between the three
cases diminishes somewhat.
From an IRR perspective, the
advantage of Case 3 increases when
hydrogen cost increases and the
gap in NPV between Case 1 and 3
decreases.
Surprisingly, Case 2 with a four-
year CFHT cycle in sync with the
FCC cycle does not show an NPV
or IRR advantage over the shorter
cycle Case 3 for either natural gas
pricing scenario. One could have
assumed that designing a CFHT in
sync with the downstream units,
compared to limiting the CFHT
cycle length to only two years,
would be an advantage. However,
the four-year cycle post-treatment
unit brings the additional fexibility
to continuously operate during a
CFHT catalyst change-out. Despite
higher feed sulphur (that could be
partially limited with a change in
crude diet during the CFHT catalyst
change-out), the design of the post-
treatment unit with the Prime-G+
technology is robust enough to
handle this higher severity require-
ment during the catalyst
change-out.
This fexibility is clearly illustrated
in Figure 6, which shows operating
data on a Prime-G+ unit in a refn-
ery processing heavy crudes and
equipped with a FCC CFHT
pretreater. When the CFHT is in
operation, the normal feed sulphur
to the Prime-G+ unit is typically
below 200 wppm. Despite turna-
rounds or operation upsets on the
CFHT unit, which can lead to feed
sulphur as high as 900 wppm, the
product sulphur from the Prime-G+
unit can be maintained at the target
value of 20 wppm at all times.
When processing full-range cut
naphtha (FRCN), the sulphur
content in the product is main-
tained at the target value (20 ppm,
see Figure 6) despite variations in
FRCN quality thanks to the FCC
pretreatment option.
The fexibility brought by adding
a post-treatment to the compulsory
FCC pretreater when processing
heavy crudes should be underlined
and is a major advantage over the
pretreatment alone solution. In
order to produce a gasoline pool
with less than 10 wppm, the refn-
ery becomes a chemical plant with
no margin for error; relying on the
CFHT alone leaves little fexibility.
In summary, coupling a CFHT
with a FCC naphtha post-treatment
unit brings the following
advantages:
The CFHT severity is lowered,
which offers the possibility to
revamp an existing CFHT
It is possible to design the CFHT
unit for a cycle length of two years
instead of four
The Prime-G+ post-treatment
design is simplifed to typically a
single-stage unit
The refnerys reliability and fex-
ibility are improved:
CFHT upset may be compen-
sated by the Prime-G+
post-treatment unit
CFHT severity may be
decreased if needed/permitted
9 5
1 0 5
1 0 0
9 0
8 5
N
P
V

a
t

1
0
%

(
c
a
s
e

1
)
8 0
B a se
Figure 4 NPV results
3
5
4
7
6
2
1
I
R
R
,

a
d
d
i
t
i
o
n
a
l

I
R
R

p
o
i
n
t
s

(
c
a
s
e

1
)
0
B a se
Figure 5 IRR results
Case study Case 1 Case 2 Case 3
NPV @10%: nat. gas = $4 MMBTU (case 2011) Base Base x 0.93 Base x 0.93
NPV @10%: nat. gas = $6 MMBTU Base Base x 0.94 Base x 0.94
Study results hydrogen cost sensitivity study
Table 3
axens.indd 3 11/9/12 13:48:45
or improve the performance of the
FCC unit. Although every situation
is site specic, the combination of
pre- and post-treat solutions around
the FCC unit will often provide the
best combination in terms of exi-
bility and economic benets to most
North American reneries.
References
1 Bonnardot J, et al, Direct production of Euro-
IV diesel at 10 pm sulphur via the HyC-10
process, ERTC 9th Annual Meeting, Nov 2004.
2 Sarrazin P, et al, New mild hydrocracking
route produces 10-ppm-sulphur diesel,
Hydrocarbon Processing, Feb 2005.
3 Roux R, et al, Resid to petrochemicals
technology, ERTC 13th Annual Meeting, Nov
2008.
4 Debuisschert Q, Prime-G+ commercial
performance of FCC naphtha desulphurization
technology, AM-03-26, NPRA Annual Meeting,
Mar 2006.
5 Largeteau D, et al, Benzene management in a
MSAT 2 environment, AM-08-11, NPRA Annual
Meeting, Mar 2008.
6 Debuisschert Q, et al, Technology solutions
addressing gasoline and diesel imbalances,
Platts European Rening Market 4th Annual
Meeting, Sept 2010.
Delphine Largeteau is Technology Manager for
Olens & Light Oil Hydroprocessing, Axens. She
joined Axens in 1998 as Process Engineer. She
holds a degree in chemical engineering from
Universit Technologique de Compigne (UTC)
in France and a masters in rening, engineering
and gas from the IFP School.
Jay Ross is a Technology and Marketing
Manager covering the eld of transportation
fuels. He has over 30 years of experience in the
rening and petrochemical industry. He holds a
degree in chemical engineering from Princeton
University. He has served on the NPRA and
ERTC expert panels, and authored several
patents and numerous technical papers.
Marc Laborde is Strategic Marketing Engineer
in Axens Marketing Department. He holds a
degree in chemical engineering from Ecole
Nationale Suprieure de Chimie in Caen and a
masters in rening, engineering and gas from
the IFP School.
Larry Wisdom is a Senior Executive at Axens
in charge of marketing the companys heavy
ends technologies in North America. During
his 30-year career, he has co-authored more
than 30 papers on heavy oil upgrading and has
been awarded two patents. He holds a BS in
chemical engineering and a MBA in marketing
and nance from the University of Kansas.
FCC unit operation is more
exible in terms of fractionation
quality
FCC gasoline end-point may be
increased when margins favour
gasoline production while still
controlling FCC naphtha sulphur
through post-treatment.
The issue of SOx and NOx control
in FCC ue gas is not addressed in
the above analysis. The high-
severity CFHT (Case 1) may allow
the typical 50 and 40 ppmv targets
for SOx and NOx to be achieved
directly, while a ue gas scrubber
would be necessary to meet such
constraints with Cases 2 and 3. The
addition of the scrubber for Cases 2
and 3 decreases the IRR differential
to Case 1 by one point, while
conversely the NPV advantage over
Case 1 is increased by approxi-
mately 1%.
It is important to note that in
spite of a trend in favour of Case 3,
the conclusion drawn from this
particular study is case specic and
cannot be generalised to other cases
that may have different congura-
tions and project premise.
Conclusion
Most countries are moving towards
limiting the sulphur level in trans-
portation fuels to 10 wppm.
Meeting new ULSG regulations at
10 wppm while using existing FCC
post-treatment assets can be
achieved using commercially
proven solutions in numerous ways
depending on the constraints and
requirements of each renery. Low
renery margins coupled with capi-
tal constraints will likely make the
revamping of existing FCC post-
treatment units the preferred option
for most reners.
As light crudes production
declines, reners will increasingly
process heavier crudes, resulting in
hydrogen-decient gas oil streams
requiring hydrotreating to maintain
1 2 0 0
1 0 0 0
8 0 0
6 0 0
4 0 0
2 0 0
1
/
7
/
2
0
0
4
1
6
/
6
/
2
0
0
5
1
/
6
/
2
0
0
6
1
7
/
5
/
2
0
0
7
1
/
5
/
2
0
0
8
1
6
/
4
/
2
0
0
9
S
u
l
p
h
u
r
,

p
p
m
0
6 0
5 0
4 0
3 0
2 0
1 0
P
r
o
d
.

S
,

p
p
m
0
P ro d . S
S u lp h u r
6
5
4
3
2
1
1
/
7
/
2
0
0
4
1
6
/
6
/
2
0
0
5
1
/
6
/
2
0
0
6
1
7
/
5
/
2
0
0
7
1
/
5
/
2
0
0
8
1
6
/
4
/
2
0
0
9
O
c
t
a
n
e

l
o
s
s
0
6 0
5 0
4 0
3 0
2 0
1 0
P
r
o
d
.

S
,

p
p
m
0
P ro d . S
O c ta n e lo ss
Figure 6 Prime-G+ operation exibility
axens.indd 4 14/9/12 12:35:01
ExxonMobil Global Leader
in Fuels and Lubes Process
Technologies
ExxonMobil Technologies are applied across our
corporation and by licensees worldwide in a growing
list of process industry applications. Take advantage
of our vast experience and ongoing commitment
to continuously improve our industry leading
technologies.
T ECHNOL OGI ES:
3225 Gallows Road, Fairfax, vA 22037-0001, USA
www.exxonmobil.com/tsl
+1-703-846-2568 fax +1-703-846-3872
tsl.email@exxonmobil.com
Advanced Sol ut i ons
for t he Wor l ds Toughest
Energy Chal l enges
Leadership in Process
Technology & Catalysis
development
deployment
discovery
2010
2000
1990
1980
1970
1960
1950
1940
SELECTIVE
LUBE
DEWAXING
(MSDW)
ExxonMobils
new generation
of MSDW
technology
continues to lead
the industry.
Methanol to
Gasoline (MTG)
MTG converts
methanol to sulfur
free gasoline,
including f rst
commercial CTL
application.
FLUID
CATALYTIC
CRACKING
FCC unit in 1942
would become
the industry
standard for
production
of gasoline.
BenzOUT
Converts benzene
contained in the
ref nerys gasoline
blend-stock
into high octane
alkyl-aromatics by
reacting benzene
with light olef ns.
DISCOVERY OF
ZSM-5 ZEOLITE
The discovery of
ZSM-5 revolutionized
many catalytic
processes in ref ning
and petrochemical
industry.
pOWERFORMING
Hydrof nishing
Catalyst
MLDW
Dewaxing
Discovery of
MCM-22
exxon.indd 1 11/9/12 11:38:44
For more information, visit www.uop.com/unifex.
2012 Honeywell I nternational, I nc. All rights reserved.
refinement 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
Dynamic simulation: a tool for
engineering problems
D
ynamic simulation is an
established tool for evaluat-
ing and understanding
transient physical and chemical
processes. Commonly accepted for
the development of operator train-
ing simulators (OTS), recent
improvements and awareness have
shifted the emphasis towards using
dynamic simulation as a process
engineering tool. Process industry
professionals have always known
that dynamic interactions in heat,
mass and momentum transfer are
captured imperfectly by steady-state
process models. Improved dynamic
models have the capability to accu-
rately represent the real-world
transient behaviour of fuids, cata-
lysts, controls and equipment.
Obvious benefts include predicting
design limitations over a wide range
of operating envelopes and overall
validation of design in terms of
performance, operability and safety.
Dynamic process models are
utilised to conduct multiple what-if
scenarios with only minor modifca-
tions. For a pressure relief scenario,
for example, the dynamic model
can be used to evaluate multiple
relieving cases such as fre, blocked
discharge, utility failure, runaway
reaction or control valve failure.
With the help of an event sched-
uler, multiple process upset
scenarios can be programmed into
the same model and then executed
separately. Event schedulers are a
standard feature of commercial
dynamic simulation programs.
While steady-state simulation
models predict discrete process
conditions, they fail to evaluate the
wide range of potential transient
conditions ignored by steady-state
Dynamic simulation can be used to solve engineering problems that require an
in-depth understanding of transient processes
DIPANJAN BHATTACHARYA, HARIBABU CHITTIBABU, JESSE MUMM and JALEEL VALAPPIL
Bechtel
modelling. Say, for example, there
are two hypothetical cases, Case A
and Case B, and that each case
represents different operating condi-
tions for the same process. A
steady-state model would be able to
predict the conditions at either Case
A or Case B, but would not shed
any light on the intermediate condi-
tions as the process transitions from
A to B. A dynamic model, on the
other hand, will predict the entire
trajectory of the process as it moves
from Case A to Case B. Traditionally,
batch and semi-batch processes have
been viewed as attractive candidates
for dynamic simulation (multiple
transient conditions). We know that
continuous processes undergo a
range of transient responses to daily
perturbations caused by climactic
events, equipment, control systems,
start-up and turndown situations as
well as human factors. Impacts of
continuous process upsets are no
less signifcant than those associated
with non-continuous processes.
Currently, Bechtel uses dynamic
simulation extensively as an
engineering tool for designing
gas processing, pipelines and
petroleum refneries. Dynamic
simulation has evolved to provide
signifcant operability, safety and
cost benefts for our natural gas
liquefaction business. Some of the
studies and analyses carried out in
Bechtel are:
Compressor studies, including
anti-surge system design and settle-
out pressure calculations
Analysis of start-up, turndown
and low fow conditions
Gas turbine or motor start-up
analysis
Fuel gas system analysis
Relief and depressurisation
system design
Inlet facility control behaviour
analysis for gas separation plant or
LNG
Distillation column controllability
studies
Hydraulic analysis for low-
pressure tank network systems
HAZOP and operability studies
Controller tuning and control
valve actuator stroke rate design.
This article discusses some exam-
ples of dynamic simulation
applications in brief.
Case study 1:
Hydrocracker/hydroprocessing unit
depressurisation studies
In hydrotreaters and hydrocrackers,
controlled depressurisation is an
important safety feature.
Depressurisation essentially lowers
the reactor and separator loop
pressure. The objective of depressu-
risation is to propagate the system
towards a safe shutdown and to
prevent any runaway situations or
thermal excursions. Controlled
depressurisation protects equip-
ment and minimises the potential
for temperature excursions caused
by exothermic chemical reactions.
Automated depressurisation can be
actuated by several malfunctions in
the unit, as listed below:
Power failure
Hydrogen failure
Fire in the unit
Reaction runaway
Major leak in the reactor or
furnace.
The objective of conducting
dynamic simulations for the hydro-
processing unit shown in Figure 1
is to:
betchel.indd 1 11/9/12 14:43:44
The reactor effuent cooling is
reduced because the cooling
medium,whichisthefeed,stops
The reactor outlet temperature
will start increasing and the reactor
effuent fowing to the separator
will see an increase in temperature
anddecreaseinpressure.
For this study, power failure was
considered as the worst case.
Duringthisscenario,feedandrecy-
cle gas failure were considered and
the system was depressurised. The
depressurisation valve was sized to
bring down the system pressure to
a certain specifed value over a
periodof15minutes.
Reaction details were a critical
input for the dynamic model, but
were very complex. Therefore, a
conservative estimate of the heat of
reaction was used. It was crucial to
interface with the process licensor
about the reaction details. The
worst possible scenarios, like end-
of-run conditions, should be used
as a base case. In end-of-run condi-
tions, the hydroprocessing reactors
typicallyoperateathighertempera-
tures, which translates into more
residual enthalpy in the system
during depressuring for a power
failurescenario.
Some of the key inputs to the
dynamic model as well as relevant
modelling considerations are as
follows:
Reactor volumes were calculated
rigorously (including void volume)
and were input to the model. The
hydraulics also played a critical
part and had to be modelled accu-
rately. It is important to have close
interaction with the process licen-
sors to incorporate maximum
fdelityintothemodel
Heat exchangers were confgured
realistically so that the heat transfer
coeffcient scaled up and down
depending on the fow. This was a
very important component of the
simulation,aseithertheshellorthe
tube side of the exchangers could
experience reduced or no-fow
duringdepressuringsituations
Lags were input into the model
by using transfer functions. A
common approach in dynamic
simulation to capture delay or lag
response is to use a transfer func-
tion block, which is a standard
Size the depressurising valve for
differentrelievingscenarios
Provide peak temperatures and
pressures in the heat integrated
reactor and stripper circuit over a
one-hour period. This is used for
validating metallurgy selection.
Depressurisation conditions can
often be the controlling case for
temperature and pressure design
limits for the heat exchangers,
drums and piping in the reactor
loop
Determine downstream stripper/
fractionator feed enthalpy and oper-
ational conditions to calculate relief
load.Thedynamicsimulationmodel
may also be extended to determine
stripper/fractionatorreliefload.
To simulate the problem effec-
tively, it is critical to understand
the events during depressurisation.
The following happens during the
depressurisationscenario:
The unit feed stops and the
depressurisationvalveopens
F e e d
Recycle gas
compressor
Reactor
Air fin cooler
Depressurisation
valve
To product
separator
To
fla re
To
strip p e r
Figure 1 Hydroprocessing unit
282
470
376
188
94
0 5
399.5C
399.5C
391.5C
358.6C
358.2C
355.2C
297.7C
10 15 20 25 30 35
C

,
t
e
l
t
u
o

r
o
t
c
a
e
R
Time, minutes
0
E she|| side out|et
G process out|et
B tube side out|et F process out|et
D she|| side out|et
C tube side out|et
A tube side out|et
Reactor out|et
281.0C
418.8C
408.8C
325.6C
249.6C
235.5C
162.7C
137.6C
129.5C
Figure 2 Temperature profle of reactor and downstream effuent exchangers during
power failure
betchel.indd 2 11/9/12 14:43:55
kobelco.indd 1 11/9/12 11:50:37
feature in commercial dynamic
simulation software
Pipe and equipment volumes
were input into the model to deter-
mine hold-up. For higher fdelity,
it is preferable to incorporate metal
mass in the model. This is slightly
less conservative, as the mean
metal temperature is always less
than the bulk fuid temperature
During power failure, the air-
cooled exchangers still provide
partial cooling by natural convec-
tion. HTRI runs were conducted to
estimate the cooling effects due to
natural convection. HTRI data at
varying fows and process condi-
tions were utilised to get a duty
profle for natural convection
during power failure. Polynomials
were developed and incorporated
in the dynamic model to establish
the natural convection effect.
It can be seen in Figure 2 that the
thermal effect gradually travels
down the feed effuent exchangers.
So, for each equipment, there is a
time domain profle of increasing
temperature and falling pressure
during the scenario. It can be seen
that the reactor temperature
decreases gradually while the effu-
ent exchangers outlet temperatures
go up gradually.
The temperatures and coincident
transient operating pressures for
every exchanger and associated
piping were compared with mechan-
ical design pressure/temperature
ratings. It was also evident from the
dynamic model that the tempera-
tures do not increase immediately
because of the system lag. The pres-
sure has dropped considerably by
the time the temperature increases
signifcantly. Without dynamic
simulation, the metallurgy would be
chosen at a very high temperature
and corresponding high pressure.
Therefore, it is overly conservative
and unnecessary to design the
system metallurgy for high tempera-
ture and coincident high-pressure
conditions as it is not realistic. Some
conservatism is incorporated in the
dynamic model to eliminate uncer-
tainties in the modelling approach
and thermodynamics. Figure 3
shows the fow rate through the
depressurising valve, as well as the
decrease in system pressure.
The other main objective of the
study was to estimate the stripper
feed condition at different time
points during the depressurisation
scenario (see Figure 4). This was
done to evaluate the stripper/
fractionator relief load. The feed
bottoms heat exchangers in the
stripper undergo temperature tran-
sients during power failure, thereby
increasing the stripper feed
enthalpy. Also, due to high temper-
atures, the level in the separator
drops, which can eventually result
in gas blow-by to the stripper. The
model was confgured to check the
consequences with respect to the
stripper.
Table 1
6 0 0 0
5 0 0 0
4 0 0 0
3 0 0 0
2 0 0 0
1 0 0 0
0 . 0 1 . 1 2 . 2 3 . 3 4 . 4 5 . 5 6 . 6 7 . 7 8 . 8 9 . 9 1 1 . 0
P
r
e
s
s
u
r
e
Ti me, mi nut es
0
2 . 5 e
4
2 e
4
1 . 5 e
4
e
4
5 e
3
M
a
s
s

f
l
o
w
0 e
0
M a ss flo w th ro u g h va lve
S yste m p re ssu re
Figure 3 Depressurising fow rate and system pressure
L C
Product
separator
Stripper
Feed-bottoms
exchanger
D e p re ssu risa tio n to fla re
P ro d u c t
Figure 4 Stripper/fractionator overview
1 . 2
1 . 0
0 . 8
0 . 6
0 . 4
0 . 2
0 1 0 0 . 2 2 0 1 3 0 2 4 0 3 5 0 4 6 0 4 7 0 4 8 0 4
S
e
p
a
r
a
t
o
r

l
e
v
e
l
,

m
t
s
Ti me, seconds
0 . 0
0 . 6
0 . 5
0 . 4
0 . 8
0 . 7
1 . 0
0 . 9
0 . 3
0 . 2
0 . 1
V
a
p
o
u
r

f
r
a
c
t
i
o
n

a
t

c
o
l
u
m
n

i
n
l
e
t
0 . 0
Va p o u r fra c tio n
S e p a ra to r le ve l
Figure 5 Separator liquid inventory depletion and corresponding increase in vapour
fraction to stripper feed
betchel.indd 3 11/9/12 14:44:08
dynamic simulation are now
common across all oil and gas
projects. Dynamic simulation is
routinely used for:
Sizing of compressor anti-surge
valves
Testing of start-up torque require-
ments and other start-up studies
HAZOP and relief scenario
analysis around the compressor
system.
Dynamic simulation studies for
For a conservative estimate, the
separator was considered to be
operating at the lowest liquid level
during steady state. The model was
also set up with the proper nozzle
elevations. It was found that gas
blow-by happens quickly after the
beginning of the scenario. No credit
was taken for instrumentation and
multiple scenarios were evaluated,
including the following:
Level control valve set to open
100% during the scenario
Level control valve stuck at last
position.
Figure 5 shows the time-
dependent trend for the stripper
feed conditions during depressuri-
sation scenarios. It also shows the
increase in vapour fraction at the
stripper/fractionator inlet during
the depressurisation phenomenon.
This is critical information for
evaluating relief scenarios in the
stripper/fractionator overhead
condensing system.
This case study shows that by
building one model, multiple objec-
tives can be met, which in turn
provides real value in terms of cost
savings, and helps to identify
hidden safety and operational risks.
For example, precipitous drops in
the separator liquid level and early
onset of gas blow-by are some of
the risks that can be quantifed
accurately and mitigated properly.
Case study 2: Compressor studies
and settle-out pressure (gas plants
or LNG)
Detailed compressor studies using
1 8 . 5 1 9 . 0 1 9 . 5 2 0 . 0 2 0 . 5 2 1 . 0
Ti me, mi nut es
C o m p re sso r in le t
C o m p re sso r d isc h a rg e
S e ttle -o u t
p re ssu re
Figure 6 Compressor settle-out
ABB C onsulting.
Safe, reliable, efficient
operations.
O il and G as operators aiming to achieve operational excellence in the face of
ageing assets, harder to reach reserves, cost pressures and sk ills shortages,
often require expert assistance. ABB C onsulting offer the oil and gas industry
expertise in; industrial energy efficiency, inspection, specialist functional
engineering, integrity management, operations improvement, process safety,
projects services, technical software and technical training. We benefit customers
by reducing risk, optimising cost and improving manufacturing efficiency.
w w w . a b b . c o m /c o n s u lti n g
A B B C o n s u lti n g
Tel: +44 (0)1925 741111
E-mail: contact@ gb. abb. com
PTQ Q4 2012 (125x178) Sep 2012.indd 1 07/09/2012 12:06:00
betchel.indd 4 11/9/12 14:44:19
For mission-critical equipment, trust the #1 online monitoring system
Who monitors the most reciprocating compressors?
When you look at the numbers, it all addsup to PROGNOST Systems:
|..gost |rst.||oJ b.so c| cr||ro J|.grcst|c systons cr .oc|p.cc.t|rg ccnp.ossc.s
|o .c.|Js tcp .o|ro.s .rJ ||| p.cJJco.s t.Jst ||OC|OS
systems
Ovo. 20 yo..s c| oxpo.|orco |r ncr|tc.|rg .oc|p.cc.t|rg n.c||ro.y
||OC|OS |.s cro |ccJs .oc|p.cc.t|rg n.c||ro.y
OJ. JoJ|c.t|cr orsJ.os t|.t ycJ. syston .on.|rs bost|rc|.ss
reply@prognost.com www.prognost.com
prognost.indd 1 1/6/12 11:42:08
testing load sharing and surge
control algorithms are also gaining
momentum. Once a compressor is
installed, optimal tuning of the
surge controller in the eld is a
difcult process due to the risk of
the machine entering into surge.
For this reason, dynamic simulation
is used as a tool where the optimal
surge controller tuning can be
tested ofine. These projects are
done in conjunction with the
vendors, providing complete
compressor control solutions.
Compressor settle-out study
If dynamic models are congured
for compressor studies, settle-out
simulation is an automatic outcome
of the trip analysis. The settle-
out pressure for a compressor
system is the pressure at which the
discharge and suction eventually
equalise after shutting down (see
Figure 6).
Settle-out pressure is used to set
design parameters for piping and
equipment on the suction side of
the compressor. It plays an impor-
tant part in start-up because higher
settle-out pressures correspond to
higher torque requirements.
Predicting the settling pressure by
steady-state (including pseudo
steady-state) methodologies is both
tedious and prone to inaccuracy.
Traditionally, settle-out pressure
is calculated manually by breaking
the compressor loop into small
isobaric sections. Manual calcula-
tion is tedious, however, with
increased chances of error because
of the uncertainty of predicting
physical properties as the pressure
changes in the compressor loop.
Dynamic simulations, on the other
hand, will incorporate the effects of
physical property changes.
In practical scenarios like turn-
down or start-up, frequently the
compressor will operate at recycle
conditions very near surge condi-
tions. This operating point is
normally selected to be around 10%
to the right of the surge line of a
compressor map. At this point,
depending on the compressor curve
characteristics, the head can be
much higher than at normal operat-
ing conditions, which translates to
a higher discharge pressure.
Therefore, when conducting
settle-out pressure simulations, it is
imperative to incorporate the high-
est possible discharge pressure in
the model. This means that the
operating point should be at the
highest possible head, which is
typically the point corresponding to
the surge-control line. Using the
highest discharge pressure as a
starting condition will yield the
highest settle-out pressure for a
particular scenario.
Driver conguration study
During compressor driver selection,
two critical factors to consider are
driver operability and reliability.
Ensuring the adequacy of a selected
driver requires a comprehensive
evaluation of the power require-
ments of the compressor.
Dynamic simulation is a valuable
tool that can be utilised to predict
effectively both the transient oper-
ating conditions and the total
torque requirements for a compres-
sor train. A single model can be
used to investigate a wide range of
scenarios, including those that can
impact the design and selection of a
driver. Typically, the following
simulations are conducted for
driver studies:
Compressor start-up from shut-
down conditions
Compressor restart from settle-
out conditions.
Information from these simula-
tions can be used to verify the
design and specications of a gas
turbine or motor by comparing the
total torque required for start-up to
the output capabilities of the driver.
Additionally, they can be used to
assess the adequacy of associated
recycle valves, pipe sizing and
layout, and permissible starting
system pressures.
The simulations can also help
identify any potential design or
operational issues that may arise.
For instance, if the selected driver
does not provide sufcient torque
to restart from settle-out condi-
tions, different strategies can be
evaluated. Solutions can include
venting to reduce pressure or
selecting a different motor or
turbine.
For any compressor transient
study, the following inputs are
critical:
Piping and equipment volumes
in the compressor loop
Compressor train inertia
Opening and closing characteris-
tics of valves
Delays in the system
Compressor performance curves.
For driver studies, an additional
key input is the motor or gas
turbine start-up torque curve as a
function of speed. The time it takes
the driver to accelerate a compres-
sor from zero to full speed is
proportional to the torque available
for acceleration and inversely
proportional to the inertia of the
system. The torque available for
acceleration is essentially the differ-
ence between torque delivered by
the driver and torque required by
the compressor. If the acceleration
torque is too low, the motor or
turbine will stall and this can cause
damage.
For motors, it is important to
consider voltage drop during start-
up. Typically, the voltage is 80% of
normal, so torque curves for 80%
voltage should be used in consulta-
tion with the vendor as an input
into the dynamic model.
Accurate modelling of accelera-
tion or deceleration of the
compressor train is another critical
component of transient studies.
The rate of change of speed is
impacted by system inertia (I), fric-
tional losses (P
f
) and the power
consumed by the compressor (P
c
).
For deceleration, residual power
from the driver (P
r
) can also have
an impact. The total power of the
system is proportional to the prod-
uct of the system torque and
angular velocity (or rotational
speed):
P = P
c
+ P
f
-P
r
=
*
=
*
N
*
2
Compressor Driver
I
C
I
D
(
C
) (
D
Figure 7 Angular velocity referenced to the
compressor (I
C
) and driver (I
D
)
betchel.indd 5 11/9/12 14:44:33
( is torque, is angular velocity,
and N is the rotational speed.)
The total torque of the system is
equal to the product of the inertia
of the train (I) and the rate of
change of speed:
= I
*
d
= I
*
dN
*
2
dt dt
Substituting this calculation for
torque into the frst equation and
rearranging yields the relationship
between the rate of change of speed
and the inertia and power of the
system:
dN
=
P
c
(t) + P
f
(t) - P
r
(t)
dt I
*
(2)
2

*
N (t)
In a direct driver system (speed
of the driver is equal to the speed
of the compressor), the total inertia
(I
Total
) of a compressor string is
equal to the sum of the driver iner-
tia (I
D
) and the load inertia of the
compressor (I
C
), as shown in Figure
7:
I
Total
= I
D
+ I
C
Often, the compressor and its
associated driver are linked via a
gearbox and operate at different
speeds. In these cases, it is impor-
tant to designate whether speed
and inertia are referenced with
respect to the compressor or the
driver, as depicted in Figure 7.
When calculating the total inertia, it
is necessary to ensure that the load
or driver inertia is properly
converted to the frame of reference.
This conversion is a function of the
square of the speed ratio. For exam-
ple, the total inertia with reference
to the driver is calculated as
follows:
I
Total
= I
D
+ I
C

*
f
C
p
2

D
For improved accuracy, the iner-
tia of the gearbox and any load
couplings should also be included
when determining the total inertia.
Compressor restart
An important consideration in the
design and selection of a refrigera-
tion compressor driver is the ability
to restart after an unplanned shut-
down. If a compressor driver is
unable to provide enough torque to
restart after settling out, for
instance, the system would need to
be depressurised prior to the restart.
An extended downtime and loss of
refrigerant via depressurisation can
be very costly, so in many cases it
may be desirable to ensure that
the compressor can be restarted
quickly without the need for
depressurisation.
In this case study, dynamic simu-
lation was used to determine
whether a refrigeration compressor
could be restarted after an
emergency shutdown without
depressurising. The system evalu-
ated was a four-stage centrifugal
compressor driven by a gas turbine.
A dynamic model of the compres-
sion system, which had been
developed to encompass both shut-
down and start-up exercises, was
utilised. Key inputs included
compressor curves, rotational iner-
tia information, pipe fttings,
equipment and pipe volumes, check
valve locations and control valve
characteristics.
Two starting conditions for start-
up were considered, the frst with a
combined settle-out pressure across
all stages and the second with indi-
vidual settle-out pressures for each
stage. In the frst case, it was
assumed that the pressure would
settle out across all check valves
and the initial pressure at each
0 2 0 4 0 6 0 8 0 1 0 0
T
o
r
q
u
e
Speed, %
L o a d to rq u e c o m b in e d se ttle -o u t p re ssu re
L o a d to rq u e in d ivid u a l se ttle -o u t p re ssu re s
Ava ila b le d rive r to rq u e
Figure 8 Start-up curves of compressor for both the individual and combined settle-out
pressure cases
H
e
a
d
Vol umet ri c f l owrat e
With re d u c e d re c yc le va lve c a p a c ity
With o rig in a l re c yc le va lve c a p a c ity
S to n e wa ll
S u rg e
S u rg e lim it lin e
M
a
x
im
u
m

s
p
e
e
d
N
o
r
m
a
l
s
p
e
e
d
M
in
im
u
m

s
p
e
e
d
Figure 9 Compressor map showing the operating point trajectory for both the original
and reduced valve capacity cases
betchel.indd 6 12/9/12 13:44:23
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
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
stage would therefore be the same.
In the second case, it was assumed
that the initial pressures would be
different due to negligible leakage
across the check valves and a short
downtime between compressor trip
and restart.
The simulation was carried out
by ramping up the compressor
speed according to a typical accel-
eration schedule provided by the
gas turbine vendor. The load torque
of the train, which is the sum of the
compressor load torque and tran-
sient rotational torque for all stages,
was tracked along with the operat-
ing conditions at each stage.
Figure 8 compares the load torque
demand for both cases against the
net torque available from the gas
turbine. The results indicate that
the driver provided suffcient
torque to restart the compressor
without the need for depressurisa-
tion for both cases.
Additional considerations
compressor surge
Although torque requirements are a
signifcant consideration in
compressor start-up, it is also
important to make sure the
compressor can be started without
operational issues, such as ensuring
that the compressor does not surge
due to insuffcient fow throughout
the entire acceleration process. In
the case outlined previously,
although the torque available from
the driver was suffcient, the recycle
valves were oversized and caused
excessive fow through the compres-
sor, which is ineffcient and can
potentially cause damage over
time.
Figure 9 shows the characteristic
operating map for one stage of the
compressor. The operating point
trajectory for the start-up case from
individual settle-out pressures is
plotted, along with a second case in
which the recycle valve capacity
had been reduced. In the original
case, the compressor operated
largely beyond the typical operat-
ing boundaries in the excessive fow
or stonewall region. With reduced
recycle valve, the operating point
followed a more desirable trajectory
that remained within operating
limits.
Although reducing the recycle
valve capacity improved the oper-
ating conditions during start-up,
the added resistance in the recycle
line resulting from smaller valve
sizes increased the overall load
torque requirement for start-up, as
shown in Figure 10. In this case,
although the load torque require-
ment increased, the driver still
provided enough torque to
restart the compressor without
depressurisation.
This case study demonstrated that
dynamic simulation can be used to
not only confrm the restart capabil-
ities of a compressor driver, but
also to identify and address design
and operational issues that can
impact the system during restart.
Case study 3: Stabiliser column
operation gas plants or LNG
Dynamic simulation was used to
evaluate the turndown operation of
a condensate stabiliser column.
During normal operation, the stabi-
liser column was designed to process
slugs or heavier hydrocarbons from
an inlet slug catcher or separator.
Light hydrocarbons and water are
stripped from heavier liquids
through the use of a reboiler. When
there was no liquid feed to the stabi-
liser, it was desirable to operate the
column in a hot standby mode in
order to ease the transition back into
normal operation in a short period
of time. Figure 11 shows an over-
view of a condensate stabiliser
column for a typical LNG plant.
0 2 0 4 0 6 0 8 0 1 0 0
T
o
r
q
u
e
Speed, %
L o a d to rq u e c o m b in e d se ttle -o u t p re ssu re
L o a d to rq u e in d ivid u a l se ttle -o u t p re ssu re s
Ava ila b le d rive r to rq u e
Figure 10 Start-up curves of compressor for both the original and reduced valve capacity
cases
L iq u id
fe e d
S trip p in g
g a s
L iq u id
re c yc le
Stabiliser
feed drum
Condensate
stabiliser
Condensate
stabiliser
reboiler
Bottoms cooler
Recycle pump
O ve rh e a d
va p o u r
B o tto m s
liq u id
Figure 11 Condensate stabiliser column
betchel.indd 7 11/9/12 14:44:52
The initial design did not allow
for a make-up source of condensate
to maintain traffc in the column.
Therefore, in order to maintain
operation above minimum turn-
down conditions, liquid from the
bottoms recycle cooler was recycled
back to the column. To maintain
0 4 8 1 2 1 6 2 0 2 4
S
t
a
g
e

t
e
m
p
e
r
a
t
u
r
e
Ti me, hours
To p o f c o lu m n
B o tto m o f c o lu m n
Figure 12 Temperature profle of the column as it transitions into hot standby mode
0 4 8 1 2 1 6 2 0 2 4
V
a
p
o
u
r

f
l
o
w
Ti me, hours
To p o f c o lu m n
Figure 13 Vapour fow profle of the column as it transitions into hot standby mode
0 4 8 1 2 1 6 2 0 2 4
L
i
q
u
i
d

f
l
o
w
Ti me, hours
To p o f c o lu m n
vapour traffc, a stripping gas
source, typically used to reduce
condensate vapour pressure during
normal operation, was utilised.
Steady-state simulations indicated
that this confguration ultimately
resulted in a fat temperature profle
across all stages of the column and
a loss of mass transfer as the intro-
duction of stripping gas
continuously removed light hydro-
carbons from the recycled liquid. In
order to maintain continuous opera-
tion above turndown limits, the
column would need to be transi-
tioned into full recycle and the
source of make-up vapour would
need to be switched to recycled
vapour from the overhead.
A dynamic model of the stabiliser
column was used to simulate the
transition from normal operation to
hot standby mode in order to deter-
mine how long it would take for
the column to cool down and for
fows to drop below minimum
turndown limits.
Figure 11 is a basic fow diagram
of the column and relevant equip-
ment. In order to model the system,
the hydraulics of piping and equip-
ment, as well as the internals and
dimensions of the stabiliser were
rigorously modelled. Key inputs
included nozzle locations, tray
sizing and spacing, weir heights,
elevation data, vessel dimensions,
control valve specifcations, piping
dimensions, pump specifcations
and heat exchanger specifcations.
The control schemes for fow, pres-
sure and liquid levels were also
modelled in detail.
The dynamic model was initially
confgured to match steady-state
operation at normal operating
conditions. The transition to hot
standby mode was initiated by
ramping down the liquid feed to
the stabiliser. Stripping gas was
automatically introduced into the
column to maintain the minimum
vapour turndown rates.
The temperatures and fow rates
at each tray were recorded through-
out the transition in order to
evaluate the turndown operation of
the column. Figure 12 shows the
temperature profle of the column
over the course of the transition.
Figures 13 and 14 show the vapour
Figure 14 Liquid fow profle of the column as it transitions into hot standby mode
betchel.indd 8 11/9/12 14:45:02
and liquid ow rates, respectively.
Based on the results, it was deter-
mined that the column could
operate between four and six hours
before reaching the minimum turn-
down ow rates, and that it would
take eight to 12 hours for the
column to cool down.
By using a dynamic model to
simulate the transition into hot
standby, it became feasible to
predict the maximum amount of
time the mode of operation could
be utilised. This example shows
how dynamic simulations can be
utilised to evaluate specic opera-
tional issues involving time in
this case, how long the hot gas
bypass mode could be utilised
that cannot be predicted using
conventional steady-state methods.
Conclusion
The case studies in this article
demonstrate how dynamic simula-
tion can be used to solve key
engineering problems that require
an in-depth understanding of tran-
sient processes. The versatility of
dynamic models and the unique
operational insights they offer can
help engineers design safe, operable
and cost-effective processing
facilities, as well as provide meas-
urable value in terms of predicting
transient conditions and mitigating
all types of project risks. These
advantages have established
dynamic simulation as a valuable
tool that can be employed through-
out the lifecycle of a project.
Further reading
1 Moore J J, Garcia A, Blieske M, Kurz R,
Brun K, Transient surge measurements of
a centrifugal compressor station during
emergency shutdowns, Proceedings of the 38th
Turbomachinery Symposium, Turbomachinery
Laboratory, Texas A&M University, College
Station, Texas, 14-26, 2009.
2 Patel V, Feng J, Dasgupta S, Ramdoss P, Wu
J, Application of dynamic simulation in the
design, operation, and troubleshooting of
compressor systems, Proceedings of the 36th
Turbomachinery Symposium, Turbomachinery
Laboratory, Texas A&M University, College
Station, Texas, 95-104, 2007.
3 American Petroleum Institute, API RP521,
Guide for pressure relieving and depressuring
systems, 5th edition, Jan 2007.
4 Ernest J B, Depew C A, Use of dynamic
simulation to model HPU reactor depressuring,
Hydrocarbon Processing, 72, Jan 1995.
5 Thomas P J, Simulation of industrial processes
for control engineers, Elsevier Science, ISBN:
0750641614.
Dipanjan Bhattacharya is a Senior Engineering
Specialist with Bechtel Oil, Gas and Chemicals
in Houston, Texas. He has over 15 years of
experience in simulation, process control and
design, LNG technology and renery operations.
He holds BS and MS degrees in chemical
engineering from Jadavpur University, India
and University of Oklahoma, respectively.
Haribabu Chittibabu is an Engineering
Specialist with Advanced Simulation Group,
Bechtel Oil, Gas & Chemicals, Inc. He is
A e r z e n - o n e s t e p a h e a d
Oi l -f ree gas c ompressi on!
Chemical and petro-chemical processes handle a wide variety
of gases. Dry compression is preferable in the case of con-
taminated and aggressive gas applications for which Aerzen
offers a complete spectrum of solutions involving positive dis-
placement blowers, boosters, and oil-free screw compressors.
Decades of experience in the f eld of custom engineered gas
compression equipment packages ensure maximum up-time
and minimal life cycle cost.
Challenge us and draw upon our know-how.
Please pick up our leaf ets A1-004, G1-151 and V1-067 for fur-
ther information or visit us on Internet at www.aerzen.com
Aerzener Maschinenfabrik GmbH
Reherweg 28
.
31855 Aerzen /Germany
.
Phone: +49 51 54 /8 10
Fax: +49 51 54 /81 91 91
.
info@ aerzener.de
.
www.aerzen.com
B l o we r s . . . . C o mp r e s s o r s . . . . G a s me t e r s B l o we r s . . . . C o mp r e s s o r s . . . . G a s me t e r s
Hall 8.0
.
Stand D 38
120 x 190_PTQ.indd 2 22.05.2012 19:52:38 Uhr
responsible for the technical execution of
dynamic simulation studies in LNG, rening
and gas plants. He holds a BS in chemical
engineering from the University of Madras,
India, and a MS in petroleum rening and
petrochemicals from Anna University, India.
Jesse Mumm is a Simulation Engineer in
the Advanced Simulation Group at Bechtel,
Houston, Texas. He holds a degree in chemical
engineering from the University of Minnesota.
Jaleel Valappil is a Senior Engineering Specialist
with Bechtel Oil, Gas and Chemicals Advanced
Simulation Group in Houston, Texas. He holds
BS and PhD degrees in chemical engineering
from the Indian Institute of Technology and
Lehigh University in Pennsylvania, respectively.

betchel.indd 9 12/9/12 13:45:29
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
Exploiting opportunities with
challenging crudes
T
he rening industry has
changed signicantly to
accommodate feedstock vari-
ations in recent years. Operations
have become more efcient and
treatments have been developed to
allow routine, protable processing
of crudes that, until recently, were
considered very challenging. Today,
some low-API gravity crudes actu-
ally sell at a premium because their
characteristics are now considered
favourable.
Traditionally, challenging or
opportunity crudes are those with
undesirable properties such as low
gravity, high viscosity, high pour
point, high nitrogen content, high
metals content or high acidity as
Chemical treatment techniques solve many of the processing problems
encountered in rening opportunity crudes
INDIA NAGI-HANSPAL, MAHESH SUBRAMANIYAM, PARAG SHAH, SERGIO MORETTI and JAMES NOLAND
Dorf Ketal Chemicals
indicated by total acid number
(TAN) in mgKOH/g. These charac-
teristics are often interdependent,
confronting the renery with
complex, multifaceted problems.
Reneries in the past were
designed and built to process specic
crudes or crude blends, many of
which have become difcult or costly
to obtain. Global markets have
become more dynamic and trading
volumes (see Figure 1) are increas-
ingly forcing reneries to become
more exible in their choice of feed-
stocks. In reality, this is easier said
than done, and signicant changes
must be made to operate protably
under the feedstock constraints most
reneries face today.
Common feedstock constraints
Low gravity/high viscosity
Crudes with low gravity and high
viscosity are becoming the norm,
and they cause signicant separa-
tion problems in renery desalters.
The desalting process is governed
by Stokes Law (Equation 1):
V
s
=
2 (
p
-
f
) (1)
9
gR
2
where:
V
s
= settling velocity
p
= density of particles
f
= density of uid
= uid viscosity
g = gravitational constant
R = radius of spherical object.
According to Stokes Law, as the
U S A
C a n a d a
M e x ic o
S o u th & C e n tra l A m e ric a
E u ro p e & E u ra sia
M id d le E a st
A fric a
A sia P a c ific
Figure 1 Global oil trade movements Source: BP, 2012
dorf metal.indd 1 11/9/12 13:56:58
VALUE CREATION
United States (316) 828-5110 | Canada (905) 852-3381 | Europe +39-035-2273411 | Asia +82-2-3276-7500
www.koch-glitsch.com
PROVALVE is a trademark of Koch-Glitsch, LP and is registered in the US and various other countries. K KOCH-GLITSCH is a trademark
of Koch-Glitsch, LP and is registered in the US. YOU CAN RELY ON US is a trademark of Koch-Glitsch, LP. PROVALVE
technology
is protected by patents in the US and various other countries.
The unique valve
design of the
PROVALVE
tray
combines fouling resistance
with high capacity.
THE choice for fouling in trayed applications
The patented PROVALVE
valve has no moving parts and a wide operating range. Its

rugged, durable construction reduces the potential for fouling and provides long tray life.
Tests of the PROVALVE
tray by Fractionation Research Inc. confrmed that it delivers the

highest capacity of any device in its category.
YOU CAN RELY ON US.
kochglitsch.indd 1 11/9/12 11:52:24

fuels, and fewer refners are opting
for sour feedstock because of the
increased demand placed on
hydrodesulphurisation units to meet
fnished product specifcations.
High metals content
Although metals such as iron,
nickel, vanadium, magnesium,
sodium and calcium are inherent in
crude oil in various forms, levels
have increased over the years due
to deeper exploration and changes
in extraction methods. Organic
calcium is particularly troublesome
because it increases desalting
conductivity and causes fouling. In
some crude oils, calcium may be
present in the carbonate form,
which can interfere with desalter
operation and cause fouling issues
in the preheat train.
Fuel prices factor into the equa-
tion, too. Specifcations for fnished
fuels impose limits on metals
content, and fuels with higher levels
of metals often trade at a discount.
On the other hand, attractive
price differentials on high metal
crudes offer opportunities to
enhance the gross refnery margin,
justifying chemical treatment costs.
For all of these reasons, metal
movement through the refnery
must be monitored and controlled.
Fortunately, metals removal tech-
nology continues to advance,
offering new opportunities for
modern refneries. One recent
development maintains an acid-free
environment in the desalter while
effectively removing metal salts.
Operational constraints
Installation of new units and/or
unit upgrades are costly and time
consuming, and such changes often
density difference between the oil
and the water narrows, the settling
velocity also declines, indicating a
requirement for extra crude resi-
dence time in the desalter.
Desalters are rarely retroftted to
deal with increasing densities.
Furthermore, as viscosity increases,
higher desalter operating tempera-
tures are required and heat energy
balances are often optimised for the
design feedstock, not the opportu-
nity feedstock. All of these
parameters must be considered to
process high-density crudes.
High solids content is a common
problem in low-gravity crude oil.
The dilbits of Canada and South
America contain high levels of
solids that pose severe processing
problems in the desalter. In addi-
tion, the solids can settle as sludge,
reducing storage tank capacity and
increasing wastewater treatment
plant loading. This can result in
environmental penalties if unit
throughput is not reduced, so solids
must be cost-effectively removed as
early in the process as possible (in
the desalters). Studies have shown
that the type and extent of emul-
sion formed depends heavily on the
solids contained in the crude (see
Table 1).
High pour point
Pour point is the lowest tempera-
ture at which the crude will fow.
Since some paraffns can crystallise
at higher temperatures than aromat-
ics, pour point is often an indicator
of crude paraffnic content. Table 2
contains examples of high pour
point crude oils.
Crudes with higher pour points
often require heating in the crude
tanks prior to processing. As a
result, refneries may need to
hold such crudes in separate
tanks, increasing storage space
requirements.
Other challenges presented by
high pour point crudes include
transportation and pumping prob-
lems, higher vacuum gas oil and
vacuum residue (residuum), and
desalter dehydration constraints.
High nitrogen content
Crudes with a high nitrogen content
tend to increase corrosion, degrade
kerosene stream quality (aviation
turbine fuel in particular) and
increase salt loading in FCC units.
Nitrogen-containing compounds in
crude oil react with chloride ions to
form ammonium chloride salts in
secondary units. These salts often
deposit in the crude unit and FCC
main fractionating columns, where
they cause pressure drop increases
and tray plugging.
High nitrogen streams require
hydrotreating for purifcation to
meet fnal product specifcations,
and a high nitrogen content also
reduces hydrotreater and FCC cata-
lyst life through poisoning.
High TAN
TAN denotes the amount of acidic
species in a crude oil as measured
in mgKOH/g. These acidic species
are a combination of light organic
acids and naphthenic acids.
Naphthenic acids cause the most
severe corrosion problems in atmos-
pheric and vacuum units. Ideally,
high TAN crudes are processed
with sour crudes to provide
sulphidation of the metal surfaces,
which offers some protection from
high temperature corrosion.
However, global standards are
imposing increasingly stringent
limits on sulphur levels in fnished
Solids Bentonite Organo- Kaolinite Calcium Barite Water-wet Oil-wet
philic clay carbonate silica silica
Volume, ml 14.6 15.0 13.4 11.9 11.3 11.0 10.7
Type of emulsion O/W W/O O/W O/W O/W O/W W/O
Major size
range, m 0.1-8 0.1-8 2-18 6-10 6-25 8-15 8-15
Shape Platelets Platelets Platelets Approx. Approx. Approx. Approx.
spherical spherical spherical spherical
Source: Yan, Wang, Jiang, Fan, & Su, 1997
Effect of various solid particles on the type and stability of emulsions
Table 1
Crude Country Pour Point, C
Mandji Gabon +9
Oguendgo Gabon +9
Zeit Bay Egypt +9
Congo Export Blend Congo +21
Nile Blend Sudan +33
Darr Sudan +39
Example high pour point crudes
Table 2
dorf metal.indd 2 11/9/12 13:55:31
z4369_210x297mm_3mmBS_cmyk.indd 1 9/5/2012 2:50:41 PM
linde.indd 1 11/9/12 12:17:06
involve a cascade of related consid-
erations that add more complexity
and increase the cost burden.
Inevitably, the tank farm is the
last upgrade considered when refn-
eries choose to increase their
refnery capacity, if it is considered
at all. Yet tank farm operations are
crucial to the successful processing
of crude oil.
Tank farm capacity provides
settling time, reducing the amount
of water entering the refnery and
enhancing desalter operations. As
attempts are made to increase refn-
ery throughput without adding
tank capacity, settling time
decreases, allowing more water and
salt into the refnery and making
operations more diffcult. With this
in mind, many refners rely on
chemical additives to increase
throughput when additional tank
farm capacity is not an option.
Demulsifcation chemistry with
metals removal
A refnery in Europe was consider-
ing alternative feedstock processing
to increase refnery margins by
processing residuum along with
crude. Its ideal blend to deliver
target returns was 70% crude with
30% residuum.
To accomplish this, the refnery
invested in chemical treatment to
process the heavy feedstock. A
demulsifer helped it maximise the
removal of impurities in the resid-
uum. However, the low gravity of
the feedstock degraded desalting
effciency, causing catalyst contami-
nation downstream as well as
coking issues.
Residuum naturally contains
metals, which, if not removed eff-
ciently in the desalter, can cause
serious problems. Metals are a
prime cause of rag layer build-up
in the desalter. They can form
naphthenate soaps that upset
desalter operation, increasing water
toxicity in effuent water treatment.
When desalted crude contains too
much calcium, catalyst poisoning
can occur, interfering with FCC
operation. A high calcium content
also causes fouling problems in
the hydrodesulphurisation (HDS)
system and increases ash content,
degrading coke quality.
This refnerys increase in resid-
uum processing also increased the
amount of feed routed to the down-
stream cracking units that produce
lighter, more valuable products. In
turn, this increased the amount of
phenolic water routed to the sour
water stripper (SWS).
As with tank farms, SWSs are
rarely upgraded when making other
changes to refnery confgurations.
Loading increases, performance
suffers and this reduces effciency
elsewhere in the refnery.
In this case, the concentration of
impurities to be removed in the
SWS increased and the quality of
the water rerouted to other units
within the refnery declined, caus-
ing desalter performance problems.
The refnery was suffering from a
vicious cycle. The residuum it was
processing increased the load on its
downstream cracking units, which
increased the load on the SWS. As
a result, the stripper sour water pH
increased.
This stripper sour water served as
the refnerys source of desalter
wash water. Since alkaline desalter
environments tend to stabilise
desalter interface emulsions, desalt-
ing effciencies degraded as wash
water pH increased. This, in turn,
made it more diffcult to process
the residuum, and the cycle contin-
ued to repeat.
To solve the problem, the refnery
opted for a demulsifer with a
uniquely formulated metals removal
agent specially engineered for heavy
crude blends (<22 API). The demul-
sifer controlled the emulsion and
the metals removal agent featured a
neutral pH that helped improve
desalter performance in spite of a
high wash water pH.
Conventional metals removal
products range from highly acidic
(pH <1) to moderately acidic (pH 3
to 4). Refners are understandably
concerned that although such prod-
ucts can control desalter wash
water pH, they may also attack the
metallurgy of the lines.
Corrosion is already a problem in
many refnery units, and refners
want to make sure that additive
usage will not make these problems
worse. In this case, the product
selected was a unique, neutral
(approximately pH 6) metals
removal agent developed by Dorf
Ketal that allows refners to use
additives for pH control and metals
removal without corrosive side
effects.
Desalting and dehydration eff-
ciencies improved, allowing the
refner to process the residuum and
crude in the desired proportions.
Figure 2 shows the results.
This refnery exemplifed many of
the noteworthy changes in refnery
operations that have taken place in
the past 10 years (see Figure 3):
Crude (and residuum) compati-
bility because the refnery was
60
90
80
70
50
40
30
20
10
20.0
41.0
0.0 0.0 0.0
k
n
a
|
B

,
y
c
n
e
i
c
i
f
f
e

n
o
i
t
a
r
d
y
h
e
D
%
l
o
v
0
5 minutes after app|ying vo|tage
App|y vo|tage for 8 minutes
After 20 minutes
40.0
50.0
52.0
55.0
80.0
85.0
46.0
Figure 2 Dehydration effciency at a European refnery
dorf metal.indd 3 12/9/12 13:48:45
crude unit and contaminant
ingress.
Many heavy crudes today are
treated with several chemicals in
the oilfeld to improve production
rates. These oilfeld chemicals end
up in the refnery units and, due to
their acidic and/or aminic nature,
they often show up frst in the
crude unit overhead when unex-
pected peaks are seen in pH,
chlorides and iron content, indicat-
ing foreign species in the crude. A
full ionic analysis may be needed to
identify the species causing these
fuctuations.
Heavy oil upgraders in the
Orinoco region have been known to
use emulsifers such as calcium
hydroxide in high TAN crudes like
Merey-16 (Orinoco bitumen and
Mesa crude oil). High TAN crudes
contain naphthenic acids of varying
molecular weights.
Generally speaking, the more
naphthenic acids in the crude, the
higher the TAN of that crude.
Naphthenic acids are not corrosive
in the liquid phase, but they are
harmful at their boiling point,
which in most refneries happens to
be in the heavy vacuum gas oil cut
of the vacuum distillation unit. At
certain temperatures, fuid veloci-
ties are high and liquid impinges
on process surfaces in transfer lines,
return bends and restricted fow
areas.
In addition to the amount of
naphthenic acid present, other
factors that can contribute to the
corrosivity of crudes include
the concentration of sulphur
compounds, the velocity and turbu-
lence of the fow stream in the
units, and the liquid/vapour inter-
face location in the unit.
When calcium hydroxide is used
during the production of these
acidic crudes, an even more aggres-
sive compound is produced:
calcium naphthenate. In reservoirs
where alkalinity is high in produced
water, calcium naphthenate may
also occur naturally.
Calcium naphthenate has the
unusual property of not being solu-
ble in oil or water. As a result,
calcium naphthenate compounds
accumulate at the oil-water inter-
face within the desalter, creating a
not designed for the target blend
Reduced tank preparation time
did not allow the water to settle out
effectively
Increased water recycles increased
water stream pH because sour
water stripping was ineffcient
Increased crude viscosities due to
residuum processing
Increased contaminants in the
residuum from the refning process.
Finished fuel specifcations are a
complicating factor that reduces
refnery fexibility in such feed-
stock changes. Flash point
specifcations for diesel limit the
amount of kerosene in the blend.
Low end point naphtha results in
colder overhead conditions and a
higher probability of corrosion.
Refners must take all of these
factors into account when evaluat-
ing discounted feedstocks.
Impact of oilfeld chemical treatment
on feedstock and unit operations
A large Asian refnery faced related
problems that were even more
complex. This refnery typically
processes more than 12 crudes in a
single blend, and the blend changes
at least every three days and often
more frequently. Crude compatibil-
ity is often compromised, increasing
the likelihood of fouling in the
I n c re a se d
c o n ta m in a n ts
I n c re a se d
c o n d u c tivity
I n c re a se d c ru d e
visc o sitie s
I n c re a se d wa te r
re c yc le s
R e d u c e d ta n k
p re p a ra tio n tim e s
C ru d e
c o m p a tib ility
I n c re a se d a c id ic a n d
a m in ic sp e c ie s
30
35
25
20
15
10
5
}
r
S
(

m
u
i
t
n
o
r
t
S
}
n
Z
(

c
n
i
Z
}
O
a
C
(

m
u
i
c
|
a
C
}
3
0
2
e
F
(

n
o
r
I
}
g
M
(

m
u
i
s
e
n
g
a
M
}
a
N
(

m
u
i
d
o
S
}
3
0
2
|
A
(

m
u
i
n
i
m
u
|
A
}
2
0
i
S

s
a
(

n
o
c
i
|
i
S

,
e
n
i
|
i
r
t

n
i

d
e
t
c
e
t
e
d

t
n
e
m
e
|
E
0.29 0.41
0.92
0.10 0.03
6.55
29.45
28.72
Figure 3 Recent changes in refneries
Figure 4 ICP test results for metals in desalter interface following Merey-16 processing
Crude Location TAN, mgKOH/g
SJV California, USA 4.3
Marlim Brazil 1.1
Dalia Angola 1.2
Lokele Cameroon 2.7
Heidrun North Sea 2.9
Shengli China 1.39
Examples of high TAN crudes
Table 3
dorf metal.indd 4 12/9/12 13:49:15
thick interfacial pad/emulsion that
substantially degrades the desalting
process.
Along with calcium naphthenate,
compounds such as calcium
hydroxide, calcium carbonate and
other solids resembling ankerite
and kaolinite have been identied
in desalter interfaces and triline
samples. A study recently
conducted at a renery processing
high percentages of Merey-16 indi-
cated the presence of such solids
through inductively coupled plasma
(ICP) testing (see Figure 4).
If the crude has a naturally high
TAN but is low in metals content,
treatment for metals removal typi-
cally is not considered necessary.
However, as the crude is processed,
calcium napthenate decreases
desalting and dehydration efcien-
cies while increasing the oil content
in the brine water and posing
signicant environmental threats.
Once this problem has been iden-
tied, a metals removal agent can be
used in conjunction with conven-
tional emulsion-breaking technology.
The product is injected into the
desalter wash water so that it mixes
thoroughly with the crude and
reacts with the calcium naphthenate
to form water-soluble metal salts,
controlling the interfacial emulsion.
This is sound in theory, but as expe-
rience showed at the renery in
Asia, it is considerably more
complex in practice.
As mentioned earlier, high TAN
crudes are frequently processed along
with high sulphur crudes to provide
protection from high temperature
corrosion. High temperature corro-
sion inhibitor functionality often
depends on sufcient inherent
sulphur in the crude to assist with
sulphiding metal surfaces.
When this renery processed high
TAN crude with high sulphur
crudes, a thick interfacial pad
formed in the desalter. A conven-
tional metals removal agent was
then added to control the emulsion,
but, like most such products, it was
ineffective due to the presence of
H
2
S.
The result can be seen in Figure
5. Black insoluble particles were
formed that deposited in the
desalter and contaminated the brine
water. These particles further
degraded desalter performance and
had a dramatic negative impact on
efuent treatment plant loading.
To solve the problem, Dorf Ketal
engineered a new product to bypass
the reaction pathway that produced
the black insolubles, providing opti-
mum solubility for the salts in the
brine water and preventing efuent
treatment problems. The results are
apparent in Figure 5 (right-hand
photo). A patent application has
been approved for this new metals
removal process in H
2
S
environments.
Impact of high TAN crudes on unit
metallurgy
The ability to remain exible when
processing crude blends is a key
consideration in managing renery
operating margins. In South
America, reneries often process
high TAN crudes because they are
readily available locally, along with
ample supplies of sweet crudes.
The difculty there is with the
availability of the high sulphur
crudes they need to help control
corrosion. If these crudes are not
readily available then the cost of
importing them may exceed the
Figure 5 Black particles forming due to the presence of H
2
S (left) and after treatment
with Dorf Ketal additive (right)
R e a c tive
su lp h u r
S c a le
F e
N a p h th e n ic
a c id s
C o rro sio n
p ro d u c ts
R -S H
H
2
S
( R C 0 0 )
2
F E ( F e S )
x
R C 0 0 H
R -S -R
Figure 6 Naphthenic acid corrosion cell
dorf metal.indd 5 11/9/12 13:56:11
return from refning domestic sweet
crude.
A refnery in Brazil faced such a
problem. The refnery wanted to
process high TAN crudes with
sweet crudes in the blend, but high
sulphur crudes were not consist-
ently available at an acceptable
cost.
Naphthenic acid is most danger-
ous at its boiling point (broadly
between 200-350C). The naphthenic
acid reacts with the scale chemistry
on the metal surface in order to
form corrosion products. This then
exposes fresh metal and the cycle
begins again.
The corrosion cell with naph-
thenic acids is shown in Figure 6, in
which iron (Fe) represents the metal
surface of the pipe to be protected.
As can be seen in the fgure, reac-
tive sulphur can be present in a
number of forms that adhere to the
metal surface.
Several options other than manag-
ing crude blend TAN and sulphur
can provide high temperature
corrosion control. Metallurgy
upgrades can help, but would have
been too expensive and time
consuming in this refnery. The full
metallurgy of the entire crude unit
(atmospheric and distillation
columns) would have required
assessment prior to processing high
acid crudes. Such a survey would
have been extremely complex
because refnery metallurgies vary
in susceptibility to acid corrosion,
and suitable metallurgy is costly.
To compensate for the corroding
effects of sulphur and naphthenic
acid in this case, an austenitic
stainless steel with approximately
2.5% molybdenum (minimum)
would have been benefcial. Figure
7 illustrates the benefts.
However, a high temperature
corrosion inhibitor was clearly a
simpler, lower-cost alternative for
this refnery and so a number of
corrosion inhibitors were tested in
feld trials.
Conventional high temperature
corrosion inhibitors are based on
phosphate or thiophosphate esters,
but these compounds form reaction
products that can cause fouling.
One of the products tested was a
patented, new-generation inhibitor
that was very different from the
others.
This new chemistry was effective
under the required low sulphur,
high TAN conditions because it did
not rely on sulphur-containing
crudes to pre-sulphide metal
surfaces. The product is stable at
high temperatures and works in
synergy with metal surface chemis-
try to prevent fouling. As a result,
the refner gained the feedstock
fexibility it needed to operate prof-
itably and manage corrosion
without high sulphur crudes.
Conclusion
Refneries confront many challenges
in todays dynamic markets.
Opportunity crudes are just one of
many, along with increasingly
stringent global environmental
restrictions on sulphur, mercury,
arsenic, lead and other components
in process streams.
As this article has shown, treat-
ment chemistry is evolving to meet
these challenges and, as new
problems arise, innovative
chemistry will provide effective
solutions. These include powerful
chemical tools for processing oppor-
tunity crudes, which enhance
feedstock fexibility and prevent
costly shutdowns and throughput
reductions.
One thing seems certain: many of
the crude oils that today are consid-
ered challenging opportunities for
adventurous refners will become
so desirable that they will trade at a
premium tomorrow.
India Nagi-Hanspal is Lead Refnery Engineer:
Technical Services with Dorf Ketal Chemicals,
Mumbai, India, and is primarily responsible
for providing refnery process improvement
strategies through unit monitoring. She
provides chemical solutions to Dorf Ketal site
teams globally and 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 and
leads the companys chemical developments.
He holds a number of patents for oil treatment
chemicals and engineers products to address
emerging issues in the hydrocarbon industry.
He holds a PhD in chemistry from Indian
Institute of Technology, Mumbai.
Email: drmaheshs@dorfketal.co.in
Parag Shah works in Global Refnery Technical
Services with Dorf Ketal Chemicals where he
is instrumental in software development for
desalter adequacy testing and monitoring
fouling in preheat exchanger trains. His
experience includes more than eight years in
the worlds largest grassroots refnery and he
holds a BEng in chemical engineering from
Mumbai University.
Email: paragshah@dorfketal.com
Sergio Moretti is Latin America Managing
Director with Dorf Ketal Chemicals in Brazil and
has led the implementation of operations in
the region. His focus is on sales, marketing and
technical support of refnery, petrochemical
and fnished fuel additives, and he holds a BEng
in chemical engineering from Universidade
Federal do Rio Grande do Sul, Brazil.
Email: Moretti@dorfketal.com.br
James Noland is Senior Director of the
Process Chemicals Division of Dorf Ketal USA,
LLC. He has been involved in the refning and
petrochemicals industry for over 17 years.
He leads the process chemical division for
North America and holds a BEng in chemical
engineering from Mississippi State University,
USA. Email: james.noland@dorfketal.com
3 0
4 5
4 0
3 5
2 5
2 0
1 5
1 0
5
C a rb o n ste e l 5 % c h ro m iu m 9 % c h ro m iu m 1 2 % c h ro m iu m
C
o
r
r
o
s
i
o
n

r
a
t
e
,

m
p
y
0
Figure 7 Corrosion rates with different metallurgies
dorf metal.indd 6 11/9/12 13:56:25
ZZZDLUOLTXLGHFRP
And |t makes A|r L|qu|de G|oba| E&C So|ut|ons your partner of cho|ce.
Looking back on decades of operatio-
nal expertise within the world leader
in gases for industry, health and the
environment, we develop creative, safe,
reliable and competitive solutions for
our customers proposing a sustainable
worldwide oer of best-in-class plants
in a dynamically changing market
place. Your global success will be built
on our global capabilities organized in
a network spread around virtually all
continents with a strong local footprint
to serve our clients.
Get in touch with us. We'll be happy to
advise you and answer your questions.
1
2
0
2
_
e
''' '
AIR_2185015_AZ_PTQ_Q4-2012_RZ.indd 1 04.09.12 10:05
airliquide.indd 1 11/9/12 11:27:58
Flexible H
2
S Removal
LO CATs exible technologies and worldwide services are backed by over three decades of
reliability. Over 200 licensees in 29 countries are meeting todays compliance requirements by
turning nasty H
2
S into elemental sulfur with the use of LO CAT technologies. Ecient, eective
and environmentally sound, LO CAT is the technology of choice for H
2
S removal / recovery.
LO CAT
www.merichem.com
Sweet Solutions
C
M
Y
CM
MY
CY
CMY
K
PTQ-LO CAT.pdf 1 9/10/2012 12:23:14 PM
merichem1.indd 1 11/9/12 11:59:23
Valuable liquids from refnery gases
R
efnery off-gas comes from a
variety of units within the
refnery. The quality and
quantity of the off-gas are depend-
ent on the type of crude and the
complexity of the refnery. The
majority of the fuel gas stream is
generated in the fuidised catalytic
cracking (FCC) unit, where the
long-chain hydrocarbon molecules
are broken into lower molecular
weight chains or lighter hydrocar-
bon components. Signifcant
amounts of olefns are also
produced during hydrotreating.
These olefns are valuable and can
increase revenue. Often, the off-gas
also contains a large amount of
hydrogen that, if recovered, can
provide economic value. Hydrogen
can be recovered by utilising pres-
sure swing adsorption (PSA) or a
cryogenic process.
Table 1 shows the range of
compositions of off-gas compo-
nents. Table 2 refects the refnery
gass contaminants. Several of the
contaminants have to be removed
from the feed gases, either to meet
product specifcations (see Table 3)
With appropriate treatment schemes, refnery off-gases can be valuable sources
of olefns and hydrogen
ZAHEER MALIK and JOHN SLACK
Linde Process Plants
or to mitigate operational hazards.
After removal of the hydrocarbon
liquids, the residual gases are an
excellent feedstock for a hydrogen
plant utilising steam methane
reforming technology.
In order to prepare the gases for
ethylene liquids recovery, the
Component Mol%
Hydrogen 10-50
Carbonmonoxide 0.1-1
Nitrogen 2-10
Methane 30-55
Ethylene 5-18
Ethane 15-20
Propylene 1-6
Propane 1-6
Butadiene 0-0.15
Butylene 0.1-0.5
Iso-andn-butane 0.5-1
C
5
+ 0.2-2
Typical FCC off-gas composition
Table 1
Component Preferred method Detector tube Expected
of analysis available? concentration
Hydrogensulphide Titrator Yes 25ppmv
Carbondisulphide Titrator Yes 1ppmv
Sulphurdioxide Massspec Yes 1ppmv
Methylmercaptan Titrator Yes 700ppmv
Ethylmercaptan Titrator No 120ppmv
Propylmercaptan Titrator No 50ppmv
Dimethyldisulphide Titrator No 10ppmv
Methylsulphide Titrator No 5ppmv
Carbonylsulphide Chromatograph Yes 50ppmv
Othermercaptans Titrator Yes 50ppmv(totalmercaptantest)
Carbondioxide Chromatograph Yes 400ppmv
Ammonia Chromatograph Yes 500ppmv
Arsine Massspec Yes 1ppmv(normallyinppbrange)
Acetylene Chromatograph Yes Trace<500ppmv
Methyl-acetylene Chromatograph No Trace
Mercury(allforms) Altecabsorptionmethod Yes 10ppmv
Lead Massspec No 200ppmv
Propadiene Chromatograph No <1mol%
Butenes
2
Chromatograph No <1mol%
Butadiene Chromatograph No <1mol%
Pentadiene Chromatograph No <1mol%
Benzene Chromatograph Yes <1mol%
Toluene Chromatograph Yes <1mol%
Xylenes Chromatograph Yes <1mol%
Hydrogencyanide Massspec Yes <100ppmv
Nitricoxide Massspec Yes <500ppmv
Nitrogendioxide Massspec Yes <500ppmv
Oxygen Chromatograph Yes <1mol%
Carbonmonoxide Chromatograph Yes <1mol%
Methylacetate Massspec No 120ppmv
Alcohols Massspec Yes 50ppmv
Aldehydes Massspec No 100ppmv
Ketones Massspec Yes(acetone) 100ppmv
Bitumen Unknown No Unknown
Amines Massspec Yes <500ppmv
Notes:
1.Mercaptansheavierthanpropyl-andcanbegroupedastotalmercaptans.
2.Thebuteneanalysisshouldbespecifcforn-butene,isobutene,trans-2-butene,cis-2-butene,2-methyl-1-butene,
2-methyl-2butene,3-methyl-1-butene.
3.The above compounds are contaminants from a product specifcation viewpoint. Other compounds such as
hydrogensulphide,carbondioxide,mercury,butadienes,pentadienes,nitricoxideandoxygenalsohaveapotentially
negativeeffectfromasafetyorequipmentviewpoint.Therefore,thesecomponentsmustbeidentifedregardless
oftheproductspecifcations.
4.Sodium,chloridesandfuoridesmaybepresentintracequantities.
Refning off-gas contaminants: method of analysis and expected concentration
Table 2
linde.indd 1 11/9/12 14:02:09
Arsine
Acetylene
NO,NO
2
,NOx(nitrogum)
Dienes (1-3 butadiene, acetylene,
propadieneandsoon).
The composition of the off-gas
depends on the type of crude, the
cracking severity and type of cata-
lyst used for cracking. A typical
range of components in the FCC
gas is listed in Table 1; the trace
componentsarelistedinTable2.
Before a treating system can be
designed, several trace components
must be properly analysed (see
Table 2). The concentration of the
components shown in Table 2 is
dependent on refnery complexity
andtypeofcrude.
The prospective products of the
refnery liquid recovery units are:
Ethylene
Ethane
Ethylene/ethanemix
Propylene
Propane
GasolineorC
5
+
Residuegasorfuelgas.
The product specifcation is
dependentonthemarket.
Table3liststypicalproductspeci-
fcations for chemical- and
polymer-gradeethylene.
Ethanesoldasanethylenecracker
feedstock must meet certain purity
requirements. Table 4 lists typical
specifcations for the ethane
feedstock.
Propylene may be sold as chemi-
cal, refning or polymer grade.
Propylene is separated from
propane using C
3
splitter fractiona-
tion. Table 5 lists chemical- and
refning-grade specifcations, while
Table 6 gives a polymer-grade
product specifcation. Table 7
shows a HD-5-grade propane
specifcation.
Contaminant removal
Mercury removal
Mercury is present in natural gas,
natural gas-associated condensates
and in refnery off-gases as organo-
metallic and inorganic compounds,
and in the elemental (metallic)
form, depending on the gas source.
Theelementalformcanbefoundin
either the vapour or liquid phase.
The organometallic (typically dime-
thylmercury,methylethylmercury,
following contaminants must be
removed to meet the product speci-
fcations listed in Tables 3-7:
Hydrogen sulphide (H
2
S) and
carbondioxide(CO
2
)
Ammonia
Chlorides
Water
Mercury
Carbonylsulphide(COS)
No Specifcation Chemical grade Polymer grade
1 Composition
a.Ethylene,minimum,mol% 99.5 99.85
b.Ethane,maximum,ppmv 1000 400
c.Propane,maximum - -
d.Propylene,maximum,ppm-mol 10 10
e.Carbondioxide,ppmv 50 5
f.Methane,maximum 500ppmv 0.15%byvol
g.Acetylene,maximum,ppmv 5 5
h.Moisture,maximum,ppmv 2 1
i.Ammonia,maximum,ppbw 10 0.5to1
j.Carbonmonoxide,maximumppmv 10 0.5
k.Carbonyls,maximum,ppmw - 5
l.Hydrogen,maximum,ppmv - 1
m.Methanol,maximum,ppmv - 5
n.Oxygen,maximum,ppmv 20 5
o.Totalsulphur,maximum,ppmw - 1
p.Totalchlorides,ppmv - 1
2 Corrosiveness
Copperstripat100F No.1 -
3 Volatilesulphur,ppmv 10 -
4 Watercontent,maximum,ppmv 10 0.5to10
Typical ethylene product specifcations
Table 3
No Specifcation Field grade Cracker feed
1 Composition
a.Ethane,maximum,mol% 95 99
b.Carbondioxide,ppmw 500 1
c.Methane,maximum 3.0mol% 150ppmw
d.Ethylene,maximum,ppmw NA 1
e.Propaneplus,maximum 3.5mol% 1ppmw
f.Acetylene,maximum,ppmw NA 1
g.Olefns,maximum,ppmw NA 1
h.Aromatics,maximum,ppmw NA 1
i.Ammonia,maximum,ppbw 1 1
k.Arsine,ppmw 10 10
l.Phosphene,ppmw 10 10
2 Corrosiveness No.1 No.1
Ethane product specifcations
Table 4
No Specifcation Chemical grade Refning grade
1 Composition
a.Propylene,minimum,LV% 99.5 60-65
b.Carbondioxide,ppmw 1
c.Methane,ethylene,ethane,maximum,LV% 0.5 1.5-2.25
d.Propane,maximum,mol% 0.5 Remaining
e.Butanes,butenesLV% 0.5-1.00
2 Corrosiveness
Copperstripat100F No.1
3 Volatilesulphur,ppmw 185
4 Dryness Nowater Nowater
Chemical- and refning-grade propylene product specifcations
Table 5
linde.indd 2 11/9/12 14:02:24
Ultra Low Sulfur
High Value LPG Products
REGENULS uses new FIBER FILMpatented technology to achieve
Ultra Low Sulfur levels in LPG product by reducing disulde oil (DSO)
like no other. See how next-generation technology, with lower capital
costs and a smaller footprint can enhance protability
of downstream LPG processing.
www.merichem.com
Sweet Solutions
REGEN ULS
C
M
Y
CM
MY
CY
CMY
K
PTQ-REGEN ULS.pdf 1 9/10/2012 12:16:43 PM
merichem2.indd 1 11/9/12 12:00:49
safe & sound
Integrating people, process and knowledge for better performance.
Experion
PK S O rion provides the only truly integrated safety and

security solution, supporting complete operational integration through
a single dashboard. Additionally, a new Universal Safety I/O module
supporting control in the eld, the Universal Safety Logic Solver,
safeguards the process even in the event communication with the
Safety M anager is interrupted. This advance reduces cost while
increasing availability and efciency.
2012 Honeywell International, Inc. All right reserved.
www.honeywellprocess.com
honeywell.indd 1 11/9/12 11:44:18
or diethyl mercury) and inorganic
forms (such as HgCl
2
) of mercury
condense into the liquid phase in
any hydrocarbon fractionation
column. Vapour-phase elemental
mercury is a primary source of
corrosion in aluminium heat
exchangers.
Elemental mercury that leaves the
plant with the hydrocarbon liquid
streams is a primary source of
corrosion for the aluminium equip-
ment in olefns liquid recovery
plants. Mercury also poisons the
selective hydrogenation catalysts in
olefn plants and can pose inhala-
tion hazards to workers.
The organometallic and inorganic
forms of mercury usually end up in
the condensate stream from the
natural gas or refnery liquid recov-
ery plant. These compounds are
important environmental toxins
that are easily absorbed and accu-
mulated by biological organisms.
The presence of these compounds
in gas condensate streams leads to
waste disposal problems and safety
hazards to workers.
Although relatively high levels of
elemental mercury were discovered
in the Groningen (Holland) feld as
early as 1969, the frst recorded cold
box failure attributed to mercury
corrosion was in the aluminium
spiral-wound heat exchanger of the
LNG plant at Skikda, Algeria,
1
in
1974. Since this time, mercury in
natural gas has become a major
concern in cryogenic gas processing
industries. These industries,
processing refnery gases including
liquefed natural gas (LNG), lique-
fed petroleum gases (LPG) and
olefns recovery, often use brazed
aluminium heat exchangers in their
cold boxes. Mercury corrosion of
aluminium exchangers has led to
several additional failures since the
problems at Skikda. In addition,
mercury accumulation can lead to
poisoning of catalysts used in olefn
processes, personnel safety hazards
and waste disposal diffculties
Mercury sources
Elemental mercury is a natural
contaminant present in the
produced natural gas in various
concentrations at certain geographic
locations. The concentration of
elemental mercury in the gas stream
is often expressed in g/Nm
3
,
which is a very small unit of meas-
ure. Generally speaking, elemental
mercury levels have been found to
be the highest in Southeast Asian
gases (up to 400 g/Nm
3
in the
vapour phase) and lowest in US
Gulf Coast gases (as low as 0.02
g/Nm
3
), although wide variation
is known to occur even within
regions. However, even at the very
lowest natural concentrations, it is
still desirable to reduce the amount
of mercury before any cryogenic
processing, either in refnery or
natural gas streams. The mercury
can accumulate to higher concentra-
tions over time.
The mechanism for the formation
of organometallic mercury is not
known. Current theory suggests
that elemental mercury may react
with the walls of reboilers and cata-
lytic furnaces to form active species,
which can then react with methane,
ethane, ethylene, the products of a
refnery off-gas, and other organic
compounds.
How to remove mercury
Mercury is removed from LPG
plants to protect aluminium heat
exchangers against corrosion.
Sulphur-impregnated activated
carbon is used to remove mercury
from the gaseous hydrocarbon feed-
stock. Mercury removal by the
activated carbon is much more eff-
cient if moisture is not present or is
removed from the stream before
coming into contact with activated
carbon. The maximum attainable
mercury concentration in the
treated gas stream is not pressure
dependent. A typical mercury
removal system is shown in
Figure 1.
The mercury removal process can
be a single- or dual-bed adsorption
Component Export spec PPBW spec
Propylene, wt% 99.5 min 99.45 min
Propane, wt% 0.4 max 0.55 max
Hydrogen, wppm 1 max 1 max
Methane + ethane, wppm 200 max 350 max
Ethylene, wppm 10 max 10 max
Acetylene, wppm 1 max 1 max
Methyl acetylene, wppm 1 max 1 max
Propadiene, wppm 1 max 1 max
1,3 Butadiene, wppm 1 max 1 max
Butanes + butylenes, wppm 10 max 10 max
Water, wppm 2 max 2 max
Oxygen, wppm 1 max 1 max
Methanol, wppm 1 max 1 max
Carbon monoxide, wppm 1 max 3 max
Carbon dioxide, wppm 3 max 5 max
Arsine, ppbw 20 max 20 max
Carbonyl sulphide (COS), ppbw 20 max 20 max
Elemental sulphur, ppbw 500 max 500 max
Polymer-grade propylene production specifcations
Table 6
No Specifcation HD-5 grade
1 Composition
a. Propane minimum, liquid vol% 90
b. Butane and heavier, liquid vol% 2.5
c. Pentanes and heavier None
d. Temperature at 95% evaporation, F -37
2 Vapour pressure at 100F, max 208 psig
3 Corrosiveness
Copper strip at 100F No. 1
4 Volatile sulphur, ppmw 185
5 Dryness No water
HD 5-grade propane specifcation
Table 7
linde.indd 3 11/9/12 14:02:46
Typical operating conditions for
hydrolysis catalyst are a temperature
of 100-250F (38-120C) and a pres-
sure of 150-440 psig (10-30 barg)
COS removal by adsorption
COS removal by adsorption is
performed on the dry feed gas. The
system is very similar to a molecu-
lar sieve dehydration system. Two
beds with switching valves are
installed; one bed is in adsorption
mode, while the other is regener-
ated and cooled. The system usually
has a cycle time of 168 hours (one
week), while the offine bed can be
heated for fve hours and cooled for
four hours; the rest of the time is
standby time. The system requires
a regeneration temperature of 550F
(287C) and generally the dry fuel
gas is used to regenerate the beds.
The regenerated fuel gas containing
the entire COS can be returned to
the refnery fuel system. A typical
system is shown in Figure 2.

Hazards of nitrous oxide
accumulation in cryogenic olefns
recovery units
Sources of NOx

compounds
NOx is found in FCC gases. It is
produced during combustion of
coke on the FCC catalyst in the FCC
regenerator. Some fue gas from the
regenerator is entrained into the
reactor with the recirculation cata-
lyst, together with oxygen and
nitrogen, where these light gases
comingle with gases produced in
the FCC reactor and are carried
forward to the FCC light ends
recovery units. When C
2
+ recovery
from FCC off-gases is practised, the
nitrogen, oxygen and NOx may
accumulate in the coldest section of
the recovery facilities. Variations in
FCC plant operating conditions
may have a major effect on the
concentration of NOx in FCC gases.
The design of a fuid coker unit (a
regenerator-coke burner associated
with a reactor) is similar in concept
to an FCC unit. Thus, coker gas, if
fed to a C
2
+ recovery unit, may be
another source of NOx.
While other sources of NOx in
feeds to ethylene recovery units
may exist, the only identifed source
is FCC gas and, by analogy, fuid
coker gas.
system. The system can be designed
to reduce mercury to a concentra-
tion of <0.01 g/Nm
3
. The chemical
reaction converts the mercury to
mercuric sulphide. The sulphide
product is then retained in the
pores of the carbon granules. The
mercury capacity of the activated
carbon can be 15-20% by weight:
Activated carbon specifcation
Density 37lb/ft
3
Sulphurcontentbyweight 10%
Arsine
The presence of an arsine compound
(AsH
3
) in a hydrocarbon stream
may act as a poison for the selective
hydrogenation catalyst. The pres-
ence of arsine is a specifc problem
in C
2
splitter columns, since its boil-
ing point is between ethane (C
2
) and
ethylene (C
2
=) so it tends to build
up in the column without leaving
with either the overhead or bottoms
product.
Arsine is generally present in
refnery gases in concentrations of
less than 1 ppm. This small quantity
is enough to poison the selective
hydrogenation catalyst.
A sacrifcial bed flled with ZnO
or CuO material is used to retain
arsine. Arsine removal material is
available from several catalyst
suppliers.
Carbonyl sulphide
Traditional treating with caustic or
amine solution is not effective in
removing COS from a hydrocarbon
stream. COS can be removed from
refnery off-gases in the following
ways:
Removal by hydrolysis
Removal by adsorption (in our
opinion, this is a better approach).
COS removal by hydrolysis
There are several catalysts available
to hydrolyse COS to H
2
S and CO
2
by the following reaction:
COS+H
2
OCO
2
+H
2
S
Some catalysts perform the
hydrolysis of COS to the reaction
products H
2
S plus CO
2
and the
subsequent retention of both of
these acidic compounds (H
2
S and
CO
2
).
These catalysts may be used for
the purifcation of either liquid or
gas streams. The stream to be puri-
fed must contain equilibrium water
(typically 300 ppmw) but must not
contain free water, which would
destroy the material. These catalysts
do not require activation before use.
F e e d
g a s
Tre a te d
g a s
Figure 1 Typicalmercuryremovalsystem
F e e d
Tre a te d
g a s
C O S +re g e n
g a s to fu e l
Figure 2 Regenerative-typeCOSremovalsystem
linde.indd 4 14/9/12 13:06:57
Integrated solutions for
enhanced pipeline management
Amid increasing demand, higher energy costs, potential security threats, and
regulatory and environmental pressures, your primary concerns remain the same:
ensure safe and reliable operations, 24/ 7, while improving asset integrity and
operational eff ciency.
Together, Schneider Electric and its newly-acquired company, Telvent, combine
the competencies of a global energy management specialist with strong process
automation, an integral security offer, and best-in-class project management
services to help you achieve your business and operational objectives.
EcoStruxure
: an integrated approach
EcoStruxure architecture acts as a solution ecosystem, offering guaranteed
compatibility across key application areas, and leveraging open standards
across both Schneider Electric and third-party offers.
Operations solutions from Telvent
Telvent operations and control solutions are built to support your needs along
the supply chain, for safe and secure operations while improving eff ciency
and situational awareness.
Enterprise solutions from Telvent
Beyond the control room, Telvent solutions ensure that your oil or gas pipeline
measurements can be converted into accurate f nancial data and submitted
into your accounting system.

Schneider Electric and Telvent solutions work together to provide you with a single,
tightly-integrated pipeline management solution.
Energy management
Flow measurement and
pipeline operations logistics
Operator training and
pipeline design
Control room management and
Enterprise SCADA
Leak detection and pipeline
operations applications
System security
Energy management systems
Process automation system
Metering EFM proving
Complete gas pipeline
management solutions
2012 Schneider Electric. All Rights Reserved. Schneider Electric is a trademark owned by Schneider Electric Industries SAS or its affiliated companies.
All other trademarks are propert y of their respective owners. www.schneider- electric.com Tel: +33 (0)1 41 29 70 00
Bring this ad to booth C122 and
stand a chance to win an Apple TV
!
Enroll in Energy University
today!
Classes are FREE, for a limited time only!
Visit www.SEreply.com Key Code 24506P
The
Perfect Fit
>
>
>
AD_PETROTECH_0912.indd 1 9/6/12 8:45 AM
schnieder.indd 1 11/9/12 11:24:27
action loves reaction
Chemical reactions require chemical catalysts. As the
global leader in chemical catalysts, BASF acts through
continuous product and process innovations in collaborative
partnerships with our customers. The result is a broad
chemical catalyst portfolio backed by dedicated customer
and technical service and enabled through the strength of
BASF - The Chemical Company.
At BASF, we create chemistry.
www.cat alyst s.basf.com/process
n
Adsorbents
n
Fine Chemical Catalysts
n
Environmental Catalysts
n
Catalysts for Fuel Cells
n

Catalysts for Oleochemicals & Other Biorenewables
n

Oxidation & Dehydrogenation Catalysts
n

Petrochemical Catalysts
n

Polyolef n Catalysts
n

Ref ning Catalysts
n

Syngas Catalysts
n

Custom Catalysts
basf.indd 1 11/9/12 11:33:02
NOx accumulation in the cold train
in cryogenic conditions
While other NOx compounds exist,
the principal compounds are
produced by the following reaction
mechanism:
NO is oxidised to NO
2
according to
the following reaction:
2NO + O
2
2NO
2

(The above reaction requires
excess oxygen at low temperature
but is a slow reaction.)
Equilibrium favours formation of
NO
2
at low temperatures and high
pressures
Reaction rate actually increases at
low temperatures
Conversion of NO to NO
2
is
expected to be highest at tempera-
tures below -100C (-148F)
NO further reacts with NO
2
to
N
2
O
3
as per the following reaction:
NO + NO
2
N
2
O
3
(known as blue oil
liquid)
(The above reaction requires cryo-
genic temperature and high
pressure.)
N
2
O
3
accumulation is most likely
to happen in areas of the plant
where the temperature is -125C
(-193F) or below
Upon warm-up, the N
2
O
3
rapidly
disassociates according to the
following reaction:
N
2
O
3
NO + NO
2
NO
2
+ NO
2
N
2
O
4
(The above reaction is favoured at
cryogenic temperatures and high
pressures.)
Diolefns such as 1,2-butadiene or
1,3-butadiene are the most problem-
aticdiolefns
Alkynes are more reactive than
paraffns or naphthenes and readily
combine with other elements
such as hydrogen, chlorine and
bromine:

N
2
O
3
+ (unsaturated compounds) NOx/HC
(The above reaction is favoured at
cryogenic temperatures.)
NOx/HC is called nitrous gum:
N
2
O
3
+ NH
3
NH
4
NOx
(The above reaction occurs if ammo-
nia is present; normally, it is washed
out with caustic.)
The ammonia salt is very unstable
and can explode when heated.
Formation of nitrated hydrocar-
bons occurs as follows:
NO
2
reacts exothermally with
olefns and diolefns, forming
nitrated gums (NOx gums)
Gums of different olefns/diole-
fns have different decomposition
temperatures
NOxgumsofconjugateddiolefns
form very rapidly at cryogenic
temperatures
NOxgumsofconjugateddiolefns
are highly unstable at cryogenic
temperatures
NOx gums are soluble in cold
methanol (MeOH).
Nitrous gum at cryogenic temper-
atures is accumulated in the coldest
sections of the plant (aluminium
heat exchangers), and whenever the
plant is warmed up the nitrous gum
decomposes at a rapid rate and is
known to cause explosions that
damage equipment.
Key properties of these materials
and some other compounds of inter-
est are summarised in Table 8.
Figure 3 shows a vapour pressure
curve for N
2
O
3
:
2
Both NO
2
and N
2
O
3
are solid at
low cryogenic temperatures
Solubility of NO
2
and N
2
O
3
in
hydrocarbons is expected to be low
Therefore, NO
2
and N
2
O
3
can
freeze out and deposit in the coldest
section of the plant
A signifcant deposit of both NOx
and NOx salts has been measured
in colder sections of ethylene plants
N
2
O
3
can also react with NH
3
to
form NH
3
NOx salts, which are also
very unstable salts. Generally, caus-
tic treating followed by water wash
can remove NH
3
from the feed gas.
Reducing the risk posed by NOx
gums and salt
Refnery off-gas contains hydrogen
and H
2
S. Use of nickel sulphide
catalyst (NiS) will essentially
convert 100% of the oxygen to H
2
O.
Since the oxygen is an essential
element in the formation of nitro
gums and salts, the risk of explosion
is minimised. NiS catalyst requires
NO NO
2
N
2
O
3
NH
4
NO
2
NH
4
NO
3
Melt point -258F +15.3F -152F +140F +338F
-161C -9.3C -102C +60C (explodes) +170C
Boiling point -240F +69.8F +38.3F +410F
-151C +21C +3.5C (decomposes) +210C
Appearance Colourless gas Red-brown gas Blue oil Solid white Solid colourless
crystals crystals
Physical properties of NO
X
compounds
2
Table 8
1 0
6
7
4
5
8
9
1 1
1 2
1 3
1 4
1 5
3 0 0 2 9 0 2 8 0 2 7 0 2 6 0 2 5 0 2 4 0 2 3 0 2 2 0 2 1 0 2 0 0
L
o
g
(
V
P
)
,

p
s
i
Temperat ure, F
1 6
L a n g e s C h e m istry H a n d b o o k , 1 3 e d .
+A I C H E D I P R R d a ta b a se
Figure 3 Plot of N
2
O
3
vapour pressure vs temperature
linde.indd 5 11/9/12 14:03:12
a small amount of H
2
S to keep the
catalyst active; a small amount of
H
2
S in the feed is required.
One alternative to the above is to
convert all triple-bond components
such as acetylene to their olefnic
form by the use of a catalyst.
This requires an extensive feed gas
treatment since any form of sulphur
is poison to the catalyst and
complete removal of all sulphur
is essential (see Figures 4 and 5).
Another alternate is not to provide
any kind of treatment but to build a
methanol (MeOH) wash system.
Periodically, when a high pressure
drop across the cold equipment
occurs, the unit is shut down for a
short period and fushed with meth-
anol to remove the gums and salts
because their solubility in methanol
is very high.
The authors are grateful to Linde and its
management for providing the resources to
write this article.
References
1 Hall R T, Hazards of NOx accumulation in
Cryogenics Ethylene Recovery Facilities, (part
of AICHE NOx Report).
2 AICHE NOx Subcommittee Report Article
15E.
G a s-to -liq u id
re c o ve ry
Feed gas
compression
Gas
dehydration
Selective
hydrogenation
Polishing
dehydration
Final
cooling
Arsine
removal
Mercury
removal
COS
removal
Amine
treating
Caustic
treating
Water
wash
COS
hydrolysis
F e e d
g a s
Wa te r
Wa te r
H
2
S +C O
2
Tra c e
H
2
S +C O
2
I f h yd ro lysis
c a ta lyst is u se d
I f re g e n e ra tive typ e
N o te 1
Figure 4 Typical treating block fow diagram if non-sulphurised catalyst is used
Note: acetylene and other triple-
bond components are saturated
to olefns
G a s-to -liq u id
re c o ve ry
Feed gas
compression
Gas
dehydration
Polishing
dehydration
Final
cooling
Arsine
removal
Mercury
removal
COS
removal
F e e d
g a s
Wa te r
Wa te r
N o te 1
H
2
S +C O
2
O p tio n a l O p tio n a l
I f re g e n e ra tive typ e
Amine
treating
Sulphurised
catalyst
Caustic
treating
Water
wash
Figure 5 Typical treating block fow diagram if sulphurised catalyst is used
Note: Acetylene and other triple-bond components are
saturated to double bonds and oxygen is converted to water
linde.indd 6 11/9/12 14:03:26
enersul.indd 1 11/9/12 11:40:09
Selas Fluid has supplied innovative heater and furnace technologies to reneries and
petrochemical plants worldwide for more than six decades.
As an approved supplier of renery heaters, petrochemical furnaces and incineration technologies to major oil and
gas companies, Selas Fluid is prepared to surpass your expectations. We have built a history of proven results, with
thousands of installations, by providing continuous innovation, quality products and reliable customer support.
Selas Fluid is a full-service partner to the rening, petrochemical, and chemical industries worldwide.
Equipment supply
Process design and engineering
Modular construction
Erection services
Revamp, retrot and upgrade
Consultation and feasibility studies
Start-up and operator training
Spare parts and technical services
Selas Fluid
Subsidiary of The Linde Group
Headquarters:
Five Sentry Parkway East
Blue Bell, PA 19422 USA
Telephone: 610-834-0300
Texas Ofce:
16225 Park Ten Place
Suite 250
Houston, TX 77084 USA
Telephone: 281-717-9090
Al-Khobar:
Linde Arabian Contracting
Telephone: +966-3-887-0133
www.lindeus-engineering.com
selas.indd 1 11/9/12 12:01:56
Thermal oxidiser waste heat recovery
A
s environmental regulations
continue to grow more strin-
gent, gas processing plants
must continuously adapt to the
requirements of the current regula-
tory environment. In an effort to
keep up with emissions regulations,
thermal oxidisers are being applied
in a wider role. The majority of gas
plants utilise regenerative chemical
solvents in an absorption column to
sweeten the gas stream. Elevated
temperatures are used to regenerate
these solvents. Heat recovery in the
form of hot oil is a logical addition
to many thermal oxidiser systems
and one that can reduce the
overall operating cost of solvent
regeneration without introducing
unnecessary complexity in overall
plant operation.
While thermal oxidisers are not
typically viewed as a proft centre,
they are a necessity for maintaining
environmental compliance. They
can, however, be incorporated into
a plant in a way that maximises
plant effciency while only margin-
ally increasing the complexity of
the overall system. This article will
address the basics of thermal oxidis-
ers in gas processing plants and
provide insight on how best to inte-
grate thermal oil heat recovery into
new or existing plants. Heat recov-
ery can signifcantly reduce the fuel
gas needed to regenerate solvent
solutions, positively impacting the
plants bottom line.
Thermal oxidisers role
The basic theory of operation of a
thermal oxidiser sounds simple
purifcation by fre. The waste
streams to be destroyed by thermal
oxidation are maintained for a set
Capturing the heat energy used for waste gas destruction enables processing
plants to enhance their effciency and minimise operating costs
RYAN TATE
Zeeco, Inc
period of time at an elevated
temperature in an oxygen-rich envi-
ronment in order to achieve the
desired destruction effciency.
Proper design of the burner and
thermal oxidiser chamber ensures
adequate mixing of the combustion
products. This process is more
commonly referred to as the three
Ts of combustion: time, tempera-
ture and turbulence.
In a typical gas processing plant,
the inlet gas is sweetened and
dehydrated with solutions that
depend on the addition of heat for
regeneration. The vent streams
generated by the amine and glycol
reboilers can contain hydrogen
sulphide, BTEX (benzene, toluene,
ethylbenzene and xylene) and
hydrocarbons, which must all be
oxidised in order to maintain envi-
ronmental compliance for the plant.
For the typical gas plant using
amine sweetening and glycol dehy-
dration, the stream with the largest
fow rate is generally the fow from
the amine reboiler. This stream is
rich in carbon dioxide and satu-
rated with water a combination
that can be considered the perfect
fre extinguisher. Burning this
stream in a thermal oxidiser at a
temperature suffcient to destroy
the hydrocarbons, hydrogen
sulphide and BTEX requires a
signifcant amount of heat input.
This heat commonly originates from
burning auxiliary fuel gas (sales
gas). Flash gas can often be used to
offset consumption, but it is gener-
ally not present in suffcient
quantities to eliminate the usage of
auxiliary fuel gas.
A less common process, but one
that is gaining wider acceptance in
the industry, is the use of membrane
technology to separate carbon diox-
ide from the inlet gas. This
technology presents its own chal-
lenges, as the permeate gas from
this process contains a signifcant
amount of hydrocarbons, which
represent a source of heating value
that can be returned to the process,
provided they can be burned effec-
tively. This separation process
employs a special selective
membrane that allows the more
permeable components of the inlet
stream (carbon dioxide) to pass
rapidly to the lower pressure side
of the membrane. The slower
components (methane) remain
primarily in the residue gas. Both
single- and two-stage membrane
systems can be employed in a gas
processing facility.
A single-stage membrane process
may be insuffcient to remove all of
the carbon dioxide and water
vapour from the inlet stream. In
these particular situations, one
option is to employ a hybrid
system, where amine and glycol
units are installed downstream of
the primary membrane system to
meet the sales gas specifcations. In
cases where a single- or two-stage
membrane system is suffcient to
meet sales gas specifcations with-
out any additional downstream
dehydration or sweetening, some
form of permeate off-gas must still
be destroyed in a thermal oxidiser.
Role of heat recovery in thermal
oxidiser operation
The common element in the gas
sweetening processes described
is that in order to maintain
compliance with environmental
zeeco.indd 1 11/9/12 14:05:23
oxidiser design. Since all of the fue
gas passes through the coil, the
maximum amount of heat can be
recovered subject only to the
constraints of the thermal oil outlet
temperature and any sulphur dew
point issues. This makes the direct-
fred design in the vertical
orientation very cost effective from
a capital equipment standpoint. The
direct-fred confguration also offers
advantages when over-fring the
thermal oxidiser burner. For
instance, the thermal oxidiser can
be designed to fre excess fuel gas
and air to supplement the waste
gas duty in the thermal oxidiser coil
to provide additional fexibility.
This is especially helpful for cases
where the fow from the amine
reboiler vent may be temporarily
routed to the fare. In these cases,
an increase in the burner fring rate
can compensate for the loss of mass
fow from the amine reboiler vent
to maintain a reasonable duty in
the coil. The additional cost to
increase the capacity of the combus-
tion air fan and burner is relatively
small when compared with the
overall cost of the equipment and
the additional fexibility it provides.
Application
Direct-fred systems are best
applied in series with the direct-
fred heater. First, the oil is heated
by coil in the thermal oxidiser, with
the oil generally fowing at a
constant rate. The oil outlet temper-
ature is allowed to fuctuate based
on the duty generated in the ther-
mal oxidiser. The oil then continues
to the fred heater, where the
desired outlet temperature is
achieved through modulation of the
heaters fring rate. The fred heater
can be designed to generate enough
duty to operate the entire plant at a
reasonable throughput. The thermal
oxidiser offsets the required duty in
the heater and this saves fuel gas
by reducing the heater fring rate.
While it is tempting to eliminate the
fred heater entirely in this applica-
tion, this action is at the expense of
overall plant fexibility. Through
careful planning of the hot oil
circuit, such confgurations are
possible, but careful attention
should be paid to the possible
regulations, the associated waste
streams must be destroyed. The
energy required to achieve proper
destruction does not have to go to
waste, however. Heat recovery
through the use of a heat transfer
fuid represents a valuable method
to minimise the commercial impact
of thermal oxidiser operation.
Implementation of these technolo-
gies in a gas processing plant makes
good business sense and demon-
strates a commitment to minimising
the environmental impact of the
overall operation.
The use of heat transfer fuids in
gas plants for the purpose of regen-
erating solvents is very common. In
most applications, a fred heater is
used to heat the circulating oil in
the plant. Thermal oil heat exchang-
ers can be added to thermal
oxidisers to offset the duty required
from the fred heater. In certain
applications, thermal oil heat
exchangers may be able to replace
the fred heater entirely. The ther-
mal oil heat exchanger incorporated
in the thermal oxidiser system
shares its basic design with the
convection coil commonly applied
in fred heater applications. Plant
effciency and fexibility are
enhanced while commonality
is maintained to the maximum
extent.
There are several ways to incor-
porate this capability in a thermal
oxidiser and the best solution
depends upon the overall process
and the amount of operational fex-
ibility desired.
Thermal oxidiser heat recovery:
theory of operation
Direct-fred or straight through
design
The simplest application of thermal
oil heat recovery in a thermal
oxidiser is the direct-fred or
straight through design. In this
system (see Figure 1), the entire
volume of fue gas generated by the
thermal oxidiser passes through the
coil. These units can be oriented
vertically or horizontally, with the
deciding factors usually being the
available plot space and the size of
the heat transfer coil. The vertical
orientation is often attractive due to
the smaller plot space required and
the cost savings associated with the
smaller vent stack mounted on the
outlet of the coil. Most plants are
already accustomed to accessing
and maintaining the elevated
convection coils on fred heaters, so
placing the oil coil at the outlet of a
vertical unit does not usually pose
an additional hardship.
For larger gas plants, a horizontal
thermal oxidiser with the heat
transfer coil located at grade is the
preferred orientation due to the
coils size and easy access. While
horizontal systems require a sepa-
rate standalone vent stack, the stack
generally represents a much smaller
percentage of the overall equipment
cost due to the larger size of the
thermal oxidiser and coil.
Advantages
The advantages of the direct-fred
approach are readily apparent. The
system is very simple, with a mini-
mum amount of additional
equipment required when
compared to a basic thermal
F u e l
g a s
A c id g a s
F la sh g a s
O il o u t
O il in
Burner
Thermal
oxidiser
Thermal oil
heat
exchanger
Stack
C o m b u stio n
a ir
Figure 1 Vertical direct-fred thermal
oxidiser
zeeco.indd 2 11/9/12 14:05:33
*ad copy 2.indt 1 1/6/12 14:52:43
What you can do
with a
touch of blue.
Improve your refnery proftability by maximizing the production of clean
transportation fuels with our leading residue upgrading technologies.
Deliver the best in refnery hydrogen production while reducing
your operating costs with our unique Terrace Wall reformer design.
Enhance the effciency 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
limitation of plant fexibility in vari-
ous modes of operation. The
thermal oxidiser is frst and fore-
most a tool for waste destruction.
Upstream process upsets generally
fnd their way to the thermal
oxidiser in one form or another,
and it is helpful to have the fred
heater available as a constant source
of heat input that is not subjected
to the potential fuctuations of the
waste stream fow rates and
compositions.
Indirect or extractive design:
theory of operation
An alternative to the direct-fred
approach is the extractive system.
In this type of system (see Figure
2), the coil is located outside of the
main fue gas path. The fue gases
are pulled across the coil though a
system of refractory lined duct
work using an induced draft fan.
The slipstream of fue gas is pulled
from the main thermal oxidiser
chamber and passed across the coil.
The induced draft fan is located
downstream of the coil, where it is
protected from the high tempera-
ture of the fue gas. Through design
of the coil and by maintaining fow
rates in the proper operational
range, outlet temperatures down-
stream of the coil are maintained in
a range that does not jeopardise the
overall reliability of the induced
draft fan. After the fue gas passes
through the induced draft fan, the
cooled gases are returned to the
thermal oxidiser chamber near the
outlet, where they mix with the
main fue gas stream prior to being
exhausted to the atmosphere.
Advantages
The extractive confguration is an
especially attractive option when
the hot oil duty required from the
coil is much smaller than the avail-
able duty present in the fue gas.
The induced draft fan can be oper-
ated by variable frequency drive
and set to only draw the required
amount of fue gas across the coil.
This allows for precise control of
the total duty recovered in the coil.
The extractive system can be used
in vertical and horizontal confgura-
tions. Vertical units are attractive
due to the reduced plot space
requirements coupled with the
convenience of having the coil
located at grade.
By using an isolation damper in
the inlet ducting of the coil, the coil
can be completely isolated from the
fue gas fow, allowing for easy
maintenance and inspection with-
out the need to take the entire
thermal oxidiser off-line. The ability
to isolate the coil also offers advan-
tages during start-up and
commissioning. Basic commission-
ing and refractory dry-out for the
main thermal oxidiser chamber can
be conducted even if the oil circuit
has yet to be commissioned.
Application
Extractive systems offer exceptional
fexibility, especially when other
sources of thermal oil heating
sources are limited. The additional
control provided by the induced
draft fan allows for very precise
regulation of the oil outlet tempera-
ture, even when faced with
fuctuating waste fow rates.
Extractive systems are popular in
gas plants utilising membrane tech-
nology. Due to the relatively high
hydrocarbon content in the perme-
ate gas from the membrane system,
the available heat duty from the
fue gas is often much greater than
that which is needed to be returned
to the plant. A thermal oil heat
exchanger placed directly in the
fue gas path would generate more
duty than could be effectively
returned to the plant, since the oil
circulation rate required to prevent
degradation of the oil places
constraints on the coil design.
Hybrid plants using a single-stage
membrane with downstream amine
treating generate permeate gas with
a considerable hydrocarbon content,
providing a large amount of availa-
ble energy for solvent regeneration.
While the hydrocarbon content in
the single-stage permeate is signif-
cant, it is still too lean to be used as
a conventional fuel and must be
burned in a specially designed ther-
mal oxidiser. Thermal oil heat
recovery can then be incorporated
downstream of the oxidiser to meet
or supplement the required reboiler
duty.
In cases where the inlet gas may
contain heavier hydrocarbons (C
6
+)
or BTEX, gases that are known to
be detrimental to membranes,
molecular sieve units are commonly
used upstream of the membranes to
remove these components.
Molecular sieves require periodic
F u e l g a s
A c id g a s
F la sh g a s
C o m b u stio n
a ir
O il o u t
O il in
Thermal
oxidiser
Stack
Burner
Thermal oil
heat exchanger
Induced
draught fan
Figure 2 Vertical extractive thermal oxidiser
zeeco.indd 3 14/9/12 12:35:48
the fuel consumed elsewhere in the
process. The payback period
depends on the operators cost for
fuel compared with the additional
cost of adding the heat recovery
unit. Generally, in most applica-
tions, the payback period is less
than one to two years.
Case studies
In the following case studies,
specic examples of both types of
thermal oil heat recovery congura-
tions are examined to provide more
insight on specic applications.
Case 1: 9.37 MMNm
3
/day
cryogenic plant
This Western US cryogenic plant is
designed for maximum ethane
recovery and an inlet gas ow rate
of 9.37 MMNm
3
/day. Molecular
sieves are used for dehydration of
the inlet gas. As Figure 3 shows,
the thermal oxidiser is a direct-red
vertical design. The main waste
ow originates from the amine
reboiler vent gases, along with a
signicant amount of ash gas that
offsets the total burner fuel gas
consumption. The balance of fuel
gas comes from residue gas. The
plant utilises a common heating oil
header. Other heat inputs to the
system include a conventional red
heater, along with waste heat recov-
ery units installed on the exhaust of
two turbine generators. Installing a
control valve in series with each
heat source and the common heat-
ing medium header provides
additional exibility. This allows
the plant to modulate the ow of
thermal oil to the thermal oxidiser
coil to achieve a desired outlet
temperature. This ow rate is estab-
lished within a predened range
calculated to ensure that maximum
lm temperature limitations are not
exceeded for any operating condi-
tion. An air cooler is installed in the
system to allow for heat rejection to
the atmosphere in cases where total
heat generation exceeds plant
capacity.
The capability of the thermal
oxidiser burner to accommodate
both ash gas and residue gas
allows the plant the ability to gener-
ate a signicant portion of the total
heat duty required at a very low
regeneration. Heat recovery incor-
porated in the thermal oxidiser can
be used for this purpose as well as
for any inter-stage heating required
in two-stage membrane systems.
Due to the intermittent nature of
molecular sieve regeneration, the
extractive system provides the ex-
ibility to meet the changing heating
demands during the regeneration
cycles.
Reliability
The added efciency and cost
savings associated with incorporat-
ing thermal oil heating in the
oxidiser also comes with an addi-
tional focus on the reliability of the
oil circulation system. Constant oil
ow is a requirement to avoid high
lm temperatures that can degrade
the heat transfer uid. The thermal
oxidiser safety system incorporates
additional permissives that ensure
that oil ow is maintained at all
times during operation. For direct-
red systems, oil ow must be
established through the coil before
starting the thermal oxidiser and a
minimum ow must be maintained
at all times. The ring rate of the
thermal oxidiser burner is corre-
lated with the waste ow rates
rather than the outlet temperature
of the oil, so this limits the ability
to modulate the burner ring rate
to regulate oil temperature. Oil
outlet temperature is monitored by
the thermal oxidiser control system
to ensure that it remains within
acceptable limits. Overall, the relia-
bility of the hot oil circulation
system is paramount for any gas
processing plant, regardless of
whether the thermal oxidiser is part
of the system. The most critical
aspect of integrating a thermal
oxidiser into the hot oil system is a
thorough understanding of the
basic thermal oxidiser control
system and how it will respond to
insufcient oil ow.
Economics
The economics of the application of
heat recovery in a thermal oxidiser
are very simple. The thermal
oxidiser must be operated to main-
tain environmental compliance. The
amount of heat recovered offsets
Figure 3 3D model of a direct-red thermal oxidiser
zeeco.indd 4 11/9/12 14:05:52
cost. The maximum amount of ash
gas is routed to the thermal oxidiser
to be combusted in the burner until
the thermal oxidiser reaches a
specic temperature set point, at
which point excess ash gases are
ared.
Case 2: 2.28 MMNm
3
/day
membrane plus amine hybrid plant
Based in Southeast Asia, this hybrid
membrane plus amine plant oper-
ates in a gas eld with a high
carbon dioxide content. The single-
stage membrane system performs
the primary sweetening of the high
carbon dioxide inlet gas (approxi-
mately 38% by volume). The
permeate gas from the membrane
skid along with the acid gas from
the amine reboiler are destroyed in
the thermal oxidiser. At maximum
ow rates, the total heat release of
the permeate gas, acid gas and fuel
gas are approximately 50.4
MMkcal/h. Of that total, only
3.9 MMkcal/h of heat energy is
required for the amine reboiler.
This heat energy is recovered by an
extractive system (see Figure 4) to
allow the exibility of operating the
thermal oxidiser at high through-
puts without jeopardising the
thermal transfer oil due to high lm
temperatures. This also provides
the plant the exibility to handle
varying compositions in the perme-
ate gas stream without signicantly
impacting the operation of the ther-
mal oil heat exchanger. An air
cooler is not required by this
system, as the heat recovery coil
can be completely isolated from the
system when heat recovery is not
required. This is especially useful
during start-up operations such as
refractory dry-out where the oil
circuit may not be functional.
Conclusion
Thermal oxidisers play an impor-
tant role in emissions control for
gas processing plants. Direct oxida-
tion of the waste gas streams is a
simple and effective way to ensure
that emissions requirements are
maintained, but it comes at the
expense of fuel consumption.
By capturing a portion of the heat
energy used for waste gas destruc-
tion and using this to meet some or
all solvent regeneration needs,
processing plants can enhance their
efciency and minimise their over-
all operating costs. Through
assessment of the total duty require-
ments of the plant, heat recovery
can be implemented in a thermal
oxidiser system for both new and
existing plants with a positive
impact on plant economics, exibil-
ity and environmental impact. The
important aspect is to understand
the heating and process require-
ments for the plant and to pick the
best conguration and coil size to
suit that need. The economics are
easily calculated. For most modestly
sized plants, the option makes
sound nancial sense. Proper inte-
gration of heat recovery in a
thermal oxidiser system can
enhance the efciency of gas
processing plants while still main-
taining the high level of reliability
and availability that is critical for
protable plant operation.
Ryan Tate is an Applications Engineer for Zeeco,
Inc, a combustion equipment manufacturer
based in Broken Arrow, Oklahoma, USA. His
experience includes the design, construction
and commissioning of thermal oxidiser systems
in locations around the world. He holds a BS
in mechanical engineering as well as an MBA
from Oklahoma State University.
Figure 4 3D model of an extractive system thermal oxidiser
zeeco.indd 5 11/9/12 14:06:03
M 8 8 L l
^K'Z/
uS
Selecting the best amine/solvent for gas treating is not a trivial task. There are a number of amines available to remove con-
taminants such as CO
2
, H
2
S and organic sulfur compounds from sour gas streams. The most commonly used amines are
monoethanolamine (MEA), diethanolamine (DEA), and methyldiethanolamine (MDEA). Other amines include diglycolamine
(DGA), diisopropanolamine (DIPA), and triethanolamine (TEA). Mixtures of amines can also be used to customize or optimize
the acid gas recovery. Temperature, pressure, sour gas composition, and purity requirements for the treated gas must all be
considered when choosing the most appropriate amine for a given application.
Primary Amines
1 MLA CC
S
u
MLA P
S CC

MLA
1

MLA uCA
CC
S CCS 1

MLA uCA
Secondary Amines
1 uLA CC
S
MLA
8 uLA MLA
1

ulA
P
S
ulA CCS S
1
ulA Aul SuLllnCL
Tertiary Amines
A MuLA
P
S CC
S
C P
S C
P
S MuLA P
S
CC

uLA P
CC

1
8 MuLA

C

Mixed Solvents
l
S l
MuLA CC
MuLA M
MuLA
CC

M
SuLllnCL
S C
A
P
S
Choosing the Best Alternative
C

M

A M
1

1

M

l M

C AC8
nM l M

8 8 L
Process
Selecting the Best Solvent
for Gas Treating
Insight:
bre.indd 1 12/9/12 13:50:23
Simulating and monitoring H
2
plant
operations
D
aily monitoring of H
2
plant
operation on-site is essential
so that we can spot potential
operating problems in time.
Therefore, there is a need to
develop a user-friendly Visual Basic
computer program in-house for
refnery process engineers to moni-
tor catalyst activity and important
operating parameters, as well as
simulate H
2
plant operation. This
article describes the development of
such a computer program for the
Riyadh refnerys conventional and
modern H
2
plants, and cites actual
examples to show how this program
has been used to simulate and
monitor H
2
plant operations.
A hydrogen plant includes a
hydrogenator, desulphuriser, steam
reformer, high and low temperature
shift converter, as well as a CO
2

removal unit and methanator for a
traditional H
2
plant (see Figure 1),
or a pressure swing adsorber (PSA)
for a modern H
2
plant (see Figure
2). Monitoring critical operating
parameters including the steam-to-
carbon ratio (S/C), methane and
CO slips, as well as catalyst activity
is essential for smooth hydrogen
plant operation.
Our current practice is that refn-
eries send operating data to the
catalyst vendor periodically for
evaluation of catalyst performance.
The catalyst vendor, after evalua-
tion, sends the report back to the
refneries. This practice takes time
and the feedback from the vendor
could not provide in-time warning.
To facilitate H
2
plant operation, a
Visual Basic program should calcu-
late steam reforming equilibrium
composition, steam reforming cata-
lyst activity in terms of approach to
A computer program was developed to enable a refnery to monitor its hydrogen
production daily to foresee potential operating problems
GENE YEH Saudi Aramco
PRABHAS MANDAL, ABDULALI SIDDIQUI and FAHAD ALHEMIDDA Saudi Aramco Riyadh refnery
equilibrium (ATE), high tempera-
ture shift (HTS) and low
temperature shift (LTS) reaction
equilibrium composition, HTS and
LTS catalyst activities in terms of
rate constants, ATE for HTS and
LTS reactions, and hydrogen prod-
uct fow and purity for conventional
and modern H
2
plants.
The computer program should
provide a scientifc method to
monitor H
2
plant operation and
check catalyst performance quanti-
tatively. The catalyst replacement
can be scheduled based on the
trend of catalyst activity. This
program should allow an inexperi-
enced engineer to monitor and
Hydrogenation/
desulphurisation
Steam reforming HTS
LTS Methanation CO
2
removal
H yd ro c a rb o n fe e d
a n d h yd ro g e n S te a m
9 6 % h yd ro g e n
Figure 1 Traditional H
2
plant confguration
Hydrogenation/
desulphurisation
Steam reforming
PSA Shift
H yd ro c a rb o n fe e d
a n d h yd ro g e n S te a m
9 9 . 9 % h yd ro g e n
Figure 2 Modern H
2
plant confguration
aramco.indd 1 11/9/12 14:13:20
Steam reforming activity in terms
ofATEandS/Cratio
Steam reformer absorbed heat
duty
Equilibrium composition and

equilibrium and actual fow rates at
HTSconverteroutlet
HTS catalyst activity in terms of
rate constant and ATE for HTS
reaction
theLTSconverteroutlet
LTS catalyst activity in terms of
rate constant and ATE for LTS
reaction
Dew point at the LTS inlet
Actual composition and fow
rates at the Benfeld unit and meth-
anator outlets for the conventional
H
2
plant
Actual composition and fow rate
for H
2
product and tail gas stream
foramodernH
2
plant.
Simulation of H
2
plant operation
The developed computer program
could be used to simulate the whole
H
2
plant operation. Some examples
are described below.
CO
2
removal capacity
Feeds to the H
2
plant include refn-
ery off-gas, natural gas, LPG and
light naphtha. The heavier the feed,
thehighertherateofCO
2
produced
for a fxed amount of H
2
produced.
The developed program has been
usedtosimulatetheCO
2
absorption
capacity of a Benfeld unit. Light
feed could be used to increase H
2
production if CO
2
removal capacity
islimited.
For example, the feed to the
Riyadh refnerys conventional H
2
plant is platformer off-gas. The
typical H
2
production rate is 55 350
nm
3
/hr and the CO
2
production
rate at the Benfeld unit is 8400
nm
3
/hr.WhenthePlatformerisout
of service, LPG is used as feed
instead. The simulation shows that
maximumH
2
productionusingLPG
feed is 47 800 nm
3
/hr due to the
limitation of CO
2
removal capacity
of 14 500 nm
3
/hr in the Benfeld
unit.
simulateH
2
plantoperationandcan
be used to identify dubious analyti-
cal data and conduct operational
variablestudies.
Visual Basic program development
The programs are developed using
an Excel spreadsheet for easy input
and output for the Riyadh refnerys
modern H
2
plant and conventional
H
2
plant, respectively. To increase
accuracy,anon-idealgasmixtureis
assumed and the fugacity coeff-
cient is calculated using the
Soave-Redlich-Kwong equation of
state for each component (CH
4
, CO,
CO
2
, H
2
, H
2
O, N
2
and NH
3
). The
fugacity coeffcient is then used in
calculating the equilibrium compo-
sition. The equilibrium composition
for the steam reformer is calculated
by solving material balance equa-
tions for each element (C, H, O and
N) and equilibrium equations for
steammethanereforming(Equation
1), water gas shift (Equation 2) and
ammonia synthesis (Equation 3).
Nevertheless, the equilibrium
composition for a shift converter is
calculated by solving material
balance equations of each element
and the equilibrium equation of the
water gas shift reaction only
(Equation2):
Methane steam reforming
CH
4
+ H
2
O CO + 3 H
2
H
298K
= 49 kcal/
mol (1)
Water gas shift
CO + H
2
O CO
2
+ H
2
H
298K
= -10 kcal/
mol (2)
Ammonia synthesis
1/2 N
2
+ 3/2 H
2
NH
3
H
298K
= -10.9 kcal/
mol (3)
ATE is an indicator of the steam
reforming catalyst activity, which is
equal to T
out
- T
eq
for the endother-
mic steam reforming reaction.
Nevertheless, ATE equals to T
eq
-
T
out
for the exothermic water gas
shiftreaction.(T
out
isoutlettempera-
ture and T
eq
is the temperature at
which the outlet gas is in equilib-
rium with respect to the particular
chemicalreaction.)
To facilitate monitoring of the
HTS and LTS catalyst performance,
catalyst activities in terms of rate
constants for shift reactions are
calculated according to Equation 4.
The plot of catalyst activity can
facilitate the planning of shift cata-
lystreplacement:
where:
k
0
: Forward rate constant for HTS
orLTSreaction
GHSV:Gashourlyspacevelocity
K: Equilibrium constant for HTS or
LTSreaction
CO
0
:COcompositionatshift
converterinletin%
H
2
O
0
:H
2
Ocompositionatshift
converterinletin%
CO
2
0
:CO
2
compositionatshift
converterinletin%
H
2
0
:H
2
compositionatshift
converterinletin%
a=K-1
b=-K
(
1+
H
2
O
0)
-
CO
2
0
-
H
2
0

CO
0
CO
0
CO
0
c =K
H
2
O
0
-
CO
2
0
*
H
2
0
CO
0
CO
0
CO
0
x: CO conversion in the high

temperature or low shift converter
E:Activationenergy,HTS,29Kcal/
gmol;LTS,15Kcal/gmol
R: Gas constant, 1.987 cal/g
mol-K
T: HTS or LTS converter outlet
temperatureinK
Input
Inputdatainclude:
Hydrocarbon feed composition
andrate
RecycleH
2
andsteamrates
Reformer outlet pressure and
inletandoutlettemperatures
Actual dry gas composition at
reformeroutlet
Inlet temperature for the HTS
andLTSconverters
Pressure and temperature at the
HTSandLTSconverteroutlet
Actualdrygascompositionatthe
HTSandLTSconverteroutlet
H
2
recoveryatPSA.
Output
Outputdatainclude:
thesteamreformeroutlet
(4)
aramco.indd 2 11/9/12 14:13:33
Steam reformer absorbed heat duty
The steam reforming reaction is
highly endothermic. The reaction
heat is supplied by burners in the
reformer furnace. The heat duty of
the steam reformer is typically
designed to match the required H
2
production rate with certain types
of feedstocks.
For example, with a limitation of
CO
2
removal capacity of 14 500 nm
3
/
h in the Benfeld unit, the refnery
plans to use sales gas (mainly meth-
ane) feed instead of LPG to increase
H
2
production during platformer
catalyst regeneration. However, the
steam reformer absorbed heat duty
in the radiant section is limited to
1.49 E+08 btu/h. For the fxed steam
rate of 58 400 kg/h, the H
2
produc-
tion rate for sales gas feed (46 500
nm
3
/hr) is actually lower than that
for LPG feed (47 800 nm
3
/h) since
the sales gas feed required a higher
heat duty due to less heat generated
from the water gas shift reaction.
Subsequently, within the furnace
heat duty limitation, the H
2
produc-
tion rate for sales gas feed could be
increased to some extent with a
lower H
2
purity by reducing the S/C
ratio.
Table 1 summarises the effect of
various feed types, feed rates and
S/C ratios on the absorbed heat
duty, CO
2
loading and H
2

production rate and purity for a
conventional H
2
plant. The feed-
stocks to this conventional H
2
plant
include platformer off-gas, sales
gas and LPG. The plant is
constrained with an absorbed heat
duty of 1.49E+08 btu/hr in the
radiant section and a CO
2
loading
of 14 500 nm
3
/hr. With the LPG
feed, the maximum H
2
product
rate is 47 800 nm
3
/hr with a purity
of 96.3% due to a CO
2
loading
Table 1
Feed Feed rate, Steam rate, S/C ratio Absorbed heat CO
2
loading
b
, H
2
product stream H
2
purity, % Pure H
2
nm
3
/hr kg/hr duty
a
, btu/hr nm
3
/hr rate, nm
3
/hr production, nm
3
/hr
Plat off-gas 30 020 54 970 6.8 1.17E+08 8400 55 350 97.1 53 745
Sale gas 12 700 58 400 5.8 1.49E+08 11 300 46 500 95.7 44 500
Sale gas 15 300 44 000 3.7 1.49E+08 11 900 51 000 92 46 920
LPG 4365 58 400 4.5 1.46E+08 14 500 47 800 96.3 46 030
a: Absorbed heat duty is limited to 1.49E+08 btu/hr.
b: CO
2
absorption capacity in the Benfeld unit is limited to 14 500 nm
3
/hr.
Effect of feed type, S/C ratio on furnace heat duty, CO
2
loading and H
2
production for a conventional H
2
plant
Table 1
*)+IMJJ,+)*+ ppp'îmj'\hf
be carried out (hazard and
operability). Some phrasing, such
as SIL ready or Equates to SIL,
can be puzzling. However, there
are two ways to clarify the confu-
sion. First, look for a monitoring
system whose vendors provide
genuine SIL certifcates issued by
recognised certifcation institutions.
Keep in mind that the SIL rating
must cover not only the safety
system itself, but also the inherent
components in the safety loop from
sensor to the emergency shutdown
device (ESD). Second, be aware that
SIL ratings should not only be high,
but relevant to your application.
For instance, SIL certifcation for
monitoring over-speed protection is
of no signifcance to a reciprocating
compressor user, but a rating for a
safety system that performs
segmented RMS vibration analyses
may be signifcant for your
machine.
Hi^kZmbg`\hg]bmbhg\aZg`^l
The enduring popularity of recipro-
cating compressors is due in part to
consumption and increased produc-
tivity. Tracking performance can
also provide other benefts, such as
early warning of impending gas
leakages. Unfortunately, losses in
effciency often go undetected by
many monitoring systems that
focus on vibration, piston position
and temperature only. Machine eff-
ciency, like other key parameters,
should be monitored continuously
and the monitoring system should
provide comprehensive analyses
that identify how you can restore
optimum operation. They can begin
by detecting any changes in
dynamic pressure during operation
and performing specialised p-V
analyses to identify components
causing reductions in effciency.
They should also incorporate infu-
ences resulting from todays
compressor regulations. Finally,
they must analyse other key values,
such as compression cycle, piston
rod load and piston rod reversal.
LBE\^kmb\Zmbhg
When it comes to machine
protection, the abbreviation SIL is
very prominent in sales literature.
It is important to understand what
SIL certifcation means and what
it does not mean. Safety integrity
level (SIL) ratings were established
to defne a metric for evaluating a
systems level of operational relia-
bility with regard to safety, as
defned by IEC 61508. As it applies
to machine monitoring systems, a
SIL rating refers to the probability
of failure on demand (PFD value)
of the protection system. A key
point to remember is that SIL
ratings have nothing to do with
monitoring precision, which is
represented by false trips and
missed detects. Keep in mind that,
before a SIL-rated safety system
comes into play, operators have to
determine the appropriate SIL
rating for the machinery that has to
be safety protected. In other words,
IEC 61508 is a risk-based standard
and, in order to apply it, criteria for
the tolerability of risks must be
established for the machine; for
example, a Hazop study must
prognost.indd 2 1/6/12 17:36:56
aramco.indd 3 12/9/12 13:51:52
outlet temperature can be calcu-
lated easily and translated into
operating costs. Lower reforming
catalyst activity also results in
higher tube skin temperatures and
consequently a decreased life for
the tubes. By monitoring key
performance indicators including
ATE, methane slip, pressure drop
and TWT, the expected beneft of a
steam reforming catalyst replace-
ment can be calculated with
reasonable accuracy.
Dew point in LTS converter
The LTS converter should operate
20C above dew point to avoid
water condensation and below 250-
260C to avoid accelerated
deactivation. The program calcu-
lates the dew point at the LTS
converter inlet according to
Equation 5:
(5)
where:
P
H
2
0
= Water partial pressure in
atmosphere absolute [atm. abs.]
For example, the LTS pressure
indicator of a hydrogen plant
wrongly shows a lower pressure
than that of the Benfeld and meth-
anator downstream. The LTS
pressure indicator could be
malfunctioning. It was found that
the margin between the LTS inlet
temperature and water dew point
is 10-15C for the same operating
period. The recommended margin
is 20C to avoid water condensa-
tion. Water can weaken and break
the LTS catalyst, resulting in
powder formation, which could
plug the pressure sensor line and
cause a lower pressure reading.
Analytical data
The program can be used to check
analytical data including methane
slip and CO slip at the reactor
outlet. For example, laboratory
analysis indicates that the methane
slip is 1.2%, but the program shows
that the equilibrium methane slip is
5.0% at the same operating condi-
tions. The lab data is inaccurate
since the actual methane slip should
be the same or higher than the
equilibrium methane slip.
limitation. With the sales gas feed,
the maximum H
2
product rate is 46
500 nm
3
/hr with a purity of 95.7%,
which is constrained by absorbed
heat duty. Consequently, the H
2

product rate can be increased to 51
000 nm
3
/hr with a purity of 92%
by reducing the S/C ratio from 5.8
to 3.7.
Ammonia content in export steam
Sales gas contains 6.6% N
2
; there-
fore, ammonia is produced in the
steam reformer. The developed
program could be used to simulate
the equilibrium ammonia content at
the steam reformer outlet. The
actual ammonia content at the
reformer outlet is close to that at
equilibrium since the condensate is
recycled. The ammonia formed will
be concentrated in the condensate,
which is used to produce export
steam. High ammonia content in
the export steam can cause equip-
ment failure downstream.
For example, the H
2
plant uses
sales gas as feed. According to the
simulation, the ammonia content at
the reformer outlet is 300 ppm to
400 ppm. Export steam would be
contaminated with a few hundred
ppm of ammonia and cannot be
used as stripping steam.
H
2
production rate
The computer program can be used
to simulate the H
2
production rate
by changing feed type, feed compo-
sition, feed fow rate, S/C ratio,
steam reformer temperature and
shift converter temperature. Using
this program, the H
2
production
rate can be estimated accurately
before any change in actual
operation.
H
2
plant performance monitoring
Steam reforming and shift
converter catalyst activity
ATE is calculated for steam reform-
ing, HTS and LTS catalysts. ATE is
an indicator of steam reforming
catalyst activity, which should be
monitored to know the perform-
ance of the catalyst. The higher the
ATE, the lower the steam reforming
catalyst activity. The optimal ATE
for shift converters should be
between 15F and 20F; therefore,
the HTS and LTS converter inlet
temperature can be adjusted
accordingly.
As the steam reforming catalyst
deactivates over the life of the
charge, the ATE will increase. To
maintain the desired H
2
purity or
capacity, it will be necessary to
increase the reformer outlet temper-
ature, increase the S/C ratio or
replace the steam reforming
catalyst.
The extra fuel consumption neces-
sary to maintain a higher reformer
Feed rate
a
, Reformer P
out
, Reformer T
out
, S/C ratio Absorbed Methane H
2
production,
nm
3
/hr kg/cm
2
g C heat, btu/hr slip, % nm
3
/hr
17750 22 810 4 1.78E+08 4.7 46400
17750 25 810 4 1.74E+08 5.4 45000
17750 22 830 4 1.89E+08 3.5 48500
17750 22 810 5 1.95E+08 3.0 49900
19000 22 810 4 1.91E+08 4.7 49600
a:Feedcomposition:C
1
,86.6%;C
2
,4%;C
3
,0.6%;C
4
,0.2%;C
5
,0.1%;H
2
,1.7%;N
2
,6.6%;CO
2
,0.2%.
b:HTST
in
,364C;T
out
,406C;PSAH
2
recovery,85%.
Effect of operation parameters
b
on absorbed heat, methane slip and H
2
production
rate for a modern H
2
plant
Table 2
2 100 P C] [ T
4
O H dew
2
+ =
The extra fuel
consumption
necessary to
maintain a higher
reformer outlet
temperature can be
calculated easily
and translated into
operating costs
aramco.indd 4 12/9/12 13:52:10
H
2
recovery in PSA
The program can also be applied to
monitoring the percentage of H
2

recovery in the PSA system by
comparing actual and simulated H
2

product fow rate, tail gas fow rate
and the composition of the tail gas
stream.
Operating variables study
The computer program has been
used to study the effect of operat-
ing variables such as reformer
pressure and temperature, S/C
ratio, feed type and feed rate on the
H
2
production rate and methane
slip at the reformer outlet for a
modern H
2
plant. It can be utilised
for unit design and revamp.
Table 2 summarises the H
2
production rate, absorbed heat and
methane slip as a function of
reformer outlet pressure and
temperature, and the S/C ratio for
a modern H
2
plant. With a fxed
feed composition, the HTS converter
inlet and outlet temperatures as
well as H
2
recovery, H
2
production
and absorbed heat increase, while
the feed rate and the S/C ratio
increase, the reformer pressure
decreases or the reformer outlet
temperature increases. Methane slip
at the reformer outlet decreases,
while the S/C ratio increases, the
reformer pressure decreases or the
reformer outlet temperature
increases. Methane slip should not
vary with a change of feed rate and
other operating variables fxed.
Conclusions
Visual Basic computer programs
have been developed for the Riyadh
refnerys conventional H
2
plant and
modern H
2
plant, respectively. The
feedback for these programs is very
positive. The refnerys engineers
are currently using them to monitor
and simulate their H
2
plant
operations.
Gene Yeh is an Engineering Specialist in the
Process & Control Systems Department of
Saudi Aramco in Dhahran, Saudi Arabia. With
more than 26 years experience in oil refning,
catalyst manufacturing and R&D, his support
areas include catalyst and adsorbent selection,
hydrogen plant, hydroprocessing and naphtha
reforming. He is a registered professional
engineer in the state of Louisiana, USA, and
holds a BS in chemical engineering and a ME
and a PhD in chemical and fuels engineering.
Email: gene.yeh@aramco.com
Prabhas K Mandal is a Senior Operations
Engineer at Riyadh refnery, Saudi Aramco,
Saudi Arabia. With more than 20 years
experience in technical services, operations
and commissioning, his support areas include
distillation, hydroprocessing and sulphur
recovery. He holds a BTech degree in chemical
engineering and a MTech in petroleum
engineering.
Abdulali Siddiqui is a Group Leader for the
hydrogen, hydrocracker and reforming units
of Saudi Aramcos Riyadh refnery, supporting
the selection of catalysts for these units. With
more than 30 years experience in refning
and petrochemicals, he is a graduate chemical
engineer.
Email: abdulali.siddiqui@hotmail. com
Fahad Alhemidda is a Process Engineer for the
hydrogen plant at Riyadh refnery. With more
than 10 years experience in petroleum refning
and gas processing, he is a chemical engineering
graduate from King Saud University, Saudi
Arabia.
steam methane reforming process is
the cost of purchasing natural gas,
which is consumed as feedstock
and fuel.
Depending upon the price differ-
ence between natural gas (or other
hydrocarbon feedstock) and coke,
the unit production cost of hydro-
gen can reportedly be higher for
steam methane reforming compared
to coke gasication, especially when
the natural gas price is on the high
side of the wide volatility range.
For reneries with captive coke
production and also a need to
import expensive natural gas,
producing hydrogen from coke
gasication should be considered
and would minimise the impact of
natural gas price volatility on the
renerys overall protability.
M^d Lnmbdgh ls lrocess Lnglneerlng Munuger
wlLh lluor. He holds Sc, MSc und uLngr
degrees ln chemlcul englneerlng und u MA
degree, ull Irom Lhe UnlverslLy oI Kunsus.
>fZbe3m^d'lnmbdgh9nhk'\hf
Dôbg Mnkbgb ls u lrocess Munuger uL lluor
wlLh over 12 yeurs' experlence ln Lhe process
lndusLrles. He holds u buchelor's degree Irom
Lhe UnlverslLy oI 1exus uL AusLln.
ppp'îmj'\hf IMJJ,+)*+**,
with the costs associated with the
higher purity make-up hydrogen.
Additionally, alternate process
schemes are available to signicantly
increase the hydrogen partial pres-
sure at the inlet of the
hydroprocessing reactors without
relying on make-up hydrogen puri-
ties higher than 97%.
The hydroprocessing units of a
typical renery operate at different
reaction pressure levels, which
generally increase as the feeds
become heavier, and some licensors
offer a range of gasier operating
pressures. The different operating
pressures of the hydroprocessing
units and the required hydrogen
make-up rates will need to be
reviewed to determine the optimal
supply pressure of hydrogen from
the new coke gasication plant to
minimise the overall compression
horsepower and the number of
compression stages.
KêbZ[bebmrZg]\hlml
Compared to other renery process
units, such as the FCC unit, which
can be operated continuously up to
ve years between major turna-
rounds, the coke gasier typically
requires a scheduled shutdown or
turnaround about once every two to
six months, and this requirement
may vary from one licensed unit to
another. To improve the reliability
and availability of a coke gasica-
tion unit, spare gasiers are typically
installed; for example, 2 x 100% or 3
x 50% gasiers. For the overall gasi-
cation unit, the gasication and
quench/cooling sections have been
reported to cause higher percent-
ages of unplanned shutdowns
relative to the remaining sections.
<hg\enlbhg
Compared to the conventional and
common method of producing
hydrogen from the steam reforming
of light hydrocarbons, gasifying
coke to produce hydrogen is gener-
ally more capital intensive. This
results in a high percentage of capi-
tal recovery factor in the unit cost
of hydrogen production from coke
gasication. On the other hand,
about 70% or more of the unit
hydrogen production cost from a
ProTreat Taking the Guesswork Out of Sour Water Stripping
0oI RoaI
Don'I 0uoss!
Sour Water Stripping iS Challenging
1utn yout concopIuaI dosIgn InIo a vItIuaI pIanI
wIIh Iho IndusIty's mosI advancod gas-ItoaIIng
sImuIaIot.
- ProTreat uses a real, mass transfer rate-based tower model.
- No ldeal stages, absolutely no translatlon from theory
to reallty.
- Completely predlctlve, wlth outstandlng
rellablllty to slgnlcantly reduce rlsk.
- l0 years of proven, hlghly successful
customer performance.
AccutaIoIy sImuIaIo sout waIot sItIppots,
rtodIcI ammonIa dIsItIbuIIon In amIno sysIoms
- ProTreat" 5.0 mass transfer rate-model calculates ammonla
absorptlon and strlpplng, wlth heat stable salts and caustlc
ln[ectlon, lncludlng pH predlctlon.
- Accurate, rellable slmulatlon of renery & syngas appllcatlons.
- Next-generatlon sclence for deslgn, optlmlzatlon,
troubleshootlng of all plant conguratlons.
rto1toaI y.o Is Iho onIy sImuIaIot IhaI can accutaIoIy
ptodIcI sout waIot sItIppot potIotmanco and Iho
opotaIIon oI amIno unIIs conIaInIng ammonIa.
www.ogt rt .com +1 281 970 2700 info@ogt rt .com
Loatn how sIaIo-oI-Iho-atI sImuIaIIon scIonco can bonohI you.
fluor.indd 4 2/6/12 17:58:50
aramco.indd 5 12/9/12 13:52:39
TRANSFORMING TODAYS PRODUCTS
FOR TOMORROWS INNOVATIONS
Market Leader in Catalyst Services, Products and Tolling
Catalyst Services:
Regeneration Reactivation Presulding Density Grading Length Grading
Catalyst Acquisition Metal Reclaim Drying Resale Quanta
Products:
Claus Catalysts Adsorbents Purication Dehydration FCC-additives
Inert Support Media
Tolling:
Extrusion Calcination Impregnation Forming Drying Milling Blending
Classifying
GLOBAL PRESENCE, WORLDWIDE SATISFACTION
Porocel_adv_FC_210x297mm.indd 1 20-02-12 09:26
porocel.indd 1 23/2/12 12:16:42
RFCC units set new standard for
propylene production
W
orldwide demand for
propylene continues to
rise at a rapid pace.
Consequently, propylene continues
to command a substantial price
premium over most other refned
products. Refners have long
exploited this price structure by
designing and operating their
conventional fuid catalytic cracking
(FCC) units to produce higher
levels of propylene. In recent years,
refners have extended this practice
to converting residual feedstocks to
propylene. This article explores the
technical challenges and opportuni-
ties presented when targeting high
propylene production from a resid-
ual fuid catalytic cracking (RFCC)
unit. The discussion will focus on
the operation of propylene-targeted
grassroots RFCC units designed by
UOP and commissioned in the past
few years. The discussion will also
include a summary of advances in
RFCC operations and reliability.

Unique features of resid FCC
operations
Resid FCC operations differ from
conventional VGO FCC operations
in several important ways: feed
characteristics, operating condi-
tions, catalyst and equipment
design.
RFCC feed characteristics
Resid is the heaviest fraction in the
barrel of crude oil boiling at the
highest temperatures. Relative to
VGO, resid contains larger quanti-
ties of polycyclic aromatics and
smaller quantities of paraffns.
Relative to VGO, resid has a lower
API gravity and lower hydrogen
content. Conradson carbon and
RFCC units are being commissioned and operated to convert hydrotreated and
straight-run resid to high levels of propylene
PATRICK WALKER and RAYMOND PETERMAN
UOP LLC, a Honeywell Company
heterocyclics of sulphur, nitrogen
and oxygen, as well as organome-
tallics of nickel and vanadium tend
to concentrate in the heaviest frac-
tion. See Figure 1 for a description
of these phenomena.
1

Aromatic rings cannot be opened
in an FCC unit due to the low
hydrogen partial pressure.
Consequently, the potential conver-
sion of resid feedstocks to smaller,
more valuable molecules is limited.
Often, about 75% of the Conradson
carbon in the feedstock will wind
up as additional coke,
2
which is
subsequently burned in the regen-
erator, releasing large quantities of
heat.
RFCC operating conditions
When processing heavily contami-
nated feedstock in an RFCC unit,
contaminant coke will deposit on
the catalyst as it passes through the
reactor. Contaminant coke is unique
to RFCC unit operations and is
formed in two ways: Conradson
carbon in the feedstock will prefer-
entially wind up as coke; metals in
the feedstock, which accumulate on
the catalyst particles, catalyse the
formation of coke. Collectively, these
two types of contaminant coke
contribute very little to conversion.
Additional coke (catalytic coke)
must be produced in suffcient quan-
tity to ensure an acceptable level of
conversion. A third type of coke
(catalyst circulation coke) can also
be carried into the regenerator due
to imperfect stripping.
3
Catalyst
circulation coke is not tightly
adhered to the catalyst particle
but rather constitutes heavy
6 0
1 0 0
8 0
4 0
2 0
1 0 0
M
o n o n u c le a r a ro m
a tic s
R e sid u e
R
e
s
i
n
s
A
s
p
h
a
l
t
e
n
e
s
D
in
u
c
le
a
r a
ro
m
a
tic
s
T
r
ic
y
c
lo
p
a
r
a
ffin
s D
ic
y
c
lo
p
a
r
a
f
f
in
s
M
o
n
o
c
y
c
lo
p
a
ra
ffin
s
is
o
-p
a
ra
ffin
s
n-p a ra ffin s
T
r
in
u
c
le
a
r
a
r
o
m
a
t
ic
s
G a s C 1 -C 4 N a p h th a /g a so lin e L ig h t g a s o il
K e ro , je t, d ie se l, N o .2 F O Va c u u m g a s o il Va c u u m re sid .
H e a vy g a s o il L u b e fra c tio n
2 0 0 3 0 0 4 0 0 5 0 0 6 0 0 7 0 0 8 0 0 9 0 0 1 0 0 0
C
o
m
p
o
s
i
t
i
o
n
,

v
o
l
%
Approxi mat e carbon number
0
Boi l i ng poi nt s, F
5 6 7 8 9 1 0 1 1 1 2 1 3 1 4 1 5 1 6 1 7 1 8 2 0 2 2 2 4 2 6 2 8 3 2 3 6 6 0 +
Wa x e s
H e te ro c yc lic
Figure 1 Relative quantities and boiling range of the major hydrocarbon classes in crude
from the Ponca City Field, Ponca City, Oklahoma Source: Marcel Dekker
1
uop.indd 1 11/9/12 14:26:24
activity that is lower than that asso-
ciated with VGO catalyst.
Cracking catalyst also deactivates
via the sodium and vanadium that
accumulates on the catalyst. Sodium
neutralises some of the acid sites
within the zeolite structure, render-
ing those acid sites permanently
inactive. Vanadium deposited on
the catalyst particle surface oxidises
in the regenerator to V
2
O
5
, which in
the presence of steam becomes
mobile vanadic acid (H
3
V
04
) that
migrates throughout the catalyst
particle and then destroys the
zeolite silica-alumina framework.
Sodium poisoning is usually
caused by sodium chloride that
enters the crude unit during a
desalter upset. This is well control-
led in most refneries by using
double desalters in the crude unit.
Fresh catalyst typically has 2000 to
3000 wppm sodium left over from
the manufacturing process. Additive
sodium beyond 5000 wppm on
equilibrium catalyst can cause a
rapid decline in catalyst activity.
Vanadium is organically bound
within organometallic molecules
and tends to concentrate in the
heaviest fraction of the feedstock.
With modern vanadium trapping
technology, concentrations of up to
7000 wppm vanadium on equilib-
rium catalyst can be tolerated while
maintaining suitable catalyst
activity.
Nickel is also organically bound
within organometallic molecules
and tends to concentrate in the
heaviest fraction of the feedstock.
Nickel is a dehydrogenation agent
and catalyses the formation of H
2
and coke. However, nickel deposits
on the surface of the catalyst parti-
cle and, once deposited, tends to be
stationary. The signature symptom
of nickel poisoning is high H
2
formation, which results in high
dry gas production. This can be
countered by modern nickel trap-
ping technology, lift gas and
antimony to passivate the dehydro-
genation impact of the nickel. New
nickel will deposit on top of old
nickel, rendering old nickel inac-
tive. As a consequence, H
2
production tends to rise rapidly
with the initial nickel deposits, but
then tapers off with further
hydrocarbon remaining in the emul-
sion phase.
4
Thus, effcient stripping
is critical for resid operations. The
total coke yield (lb coke/lb feed) is
controlled by the operating condi-
tions of the plant. For a clean
feedstock, reasonable conversion
might be achieved with a coke yield
of around 5 wt% on feed. For a resid
feedstock, the coke yield might be
greater than 10 wt% on feed in order
to reject the contaminant coke while
maintaining catalytic coke at a
quantity suffcient for reasonable
conversion.
The weight fraction of coke (of all
types combined) that deposits on
the catalyst during a single pass
through the reactor is known as
coke (lb coke/lb catalyst). coke is
a function of the feed properties,
catalyst properties, operating condi-
tions and reactor technology. For
heavily contaminated feedstock,
delta coke can be greater than 1.0
wt%.
The regenerator temperature is a
strong function of coke. At
constant operating conditions,
higher coke resulting from
elevated contaminants in the feed-
stock and higher metals on the
catalyst leads to a higher regenera-
tor temperature. Operations at
higher regenerator temperatures
can reduce the useful working life
of the regenerator internal equip-
ment while increasing operating
costs due to accelerated catalyst
deactivation.
Therefore, when presented with
operating conditions must be
adjusted to cool down the regenera-
tor. One way to do this is to operate
the regenerator in partial combus-
tion, with a fue gas CO
2
/CO mole
ratio of about 2 to 6. This is effec-
tive because burning coke partially
to CO releases about one-third as
much heat as burning coke all the
way to CO
2
. However, this tech-
nique alone is insuffcient for
heavily contaminated feedstock and
would likely result in a severe oper-
ating cost penalty due to high
catalyst consumption rates. For
direct removal of heat from the
catalyst in the regenerator via a
catalyst cooler is required in addi-
tion to operating in partial
combustion.
Decreasing the CO
2
/CO ratio or
increasing the catalyst cooler duty
have a similar impact on the opera-
tion of the unit. Both measures
result in a reduction in the regener-
ator temperature, increasing the
catalyst-to-oil ratio, and thereby
increasing the conversion of feed to
gasoline and lighter products. The
coke is ultimately burned in the
regenerator and recovered as super-
heated high-pressure steam via the
catalyst cooler and the heat recov-
ery section downstream of the CO
incinerator. This way, incremental
contaminant coke is rejected from
the feedstock and burned to
produce high-quality steam suitable
for driving large turbines such as
the main air blower or the wet gas
compressor.
RFCC catalyst
Cracking catalyst for resid feedstock
should have several features
beyond those characteristics of
cracking catalyst for VGO feed-
stock. More specifcally, the catalyst
structure must be designed to
accommodate the large polycyclic
aromatic molecules associated with
resid feedstocks. This is usually
accomplished by the incorporation
of matrix components that contain
a signifcant pore volume. The cata-
lyst also needs a low coke
character to counter the high coke
tendencies of the contaminated
feedstock. Finally, the catalyst needs
to be tolerant of the metals in the
feed, notably nickel and vanadium.
Modern cracking catalyst deacti-
vates rapidly at temperatures above
around 1350-1375F (730-745C).
This is due to hydrothermal deacti-
vation, caused by high temperatures
and a steam environment, whereby
the zeolite structure literally
collapses. Most refners have
discovered that the operating costs
associated with high catalyst
consumption become prohibitive at
regenerator temperatures above
1375-1400F (745-760C). Thus, there
is a need for a catalyst exhibiting
low coke characteristics, which is
accomplished through specifc cata-
lyst design features as well as
targeting an equilibrium catalyst
uop.indd 2 11/9/12 14:26:41
dupont.indd 1 23/2/12 15:04:26
increases in nickel. Incremental
nickel levels much above 8000
wppm produce very little incre-
mental H
2
.
5
Resid FCC equipment
Heavily contaminated resid is often
processed in an RFCC unit employ-
ing a two-stage regenerator with
catalyst coolers. This technology
was rst commercialised in 1983 at
the Ashland Oil Renery located in
Catlettsburg, Kentucky.
6,7
Processing resid with a two-stage
regenerator
Referring to Figure 2, fully regener-
ated catalyst exiting the
second-stage regenerator is trans-
ferred through the regenerated
catalyst standpipe to the wye. From
the wye, the catalyst is lifted with a
mixture of steam and sponge gas
part way up the riser to the UOP
Optimix feed distributors. Cracking
now commences and continues as
the catalyst is lifted further to the
UOP VSS riser termination device
followed by a single stage of
cyclones. Spent catalyst then
proceeds down the stripper contain-
ing UOP AF stripping technology.
The spent catalyst is transferred
through the spent catalyst stand-
pipe into the rst stage of the
regenerator, where the catalyst is
partially regenerated. The partially
regenerated catalyst is then trans-
ferred into the second stage of the
regenerator via ow-through cata-
lyst cooler(s), with the excess
catalyst transferred through the
circulation catalyst standpipe. In
the second stage, regeneration of
the catalyst is completed and cata-
lyst is returned to the wye.
Combustion air is delivered to
both stages. Flue gas exiting the
second stage is transferred up into
the rst stage, where it co-mingles
with incoming fresh air. The
combined ue gas exits the rst-
stage regenerator to a CO
incinerator followed by a waste
heat recovery section, which
produces high-pressure super-
heated steam.
The rst-stage regenerator oper-
ates in partial combustion with a
CO
2
/CO mol ratio of about 3.0.
Since the combined net ue gas
departs from the rst stage, the
regenerator overall operates in
partial combustion, thereby cooling
off the regenerator. However, the
catalyst exiting the rst stage is
only partially regenerated and the
activity is suppressed by the
remaining coke. The second stage
operates in complete combustion
with excess oxygen. This ensures
the catalyst returned to the reactor
is fully regenerated and the active
catalyst sites are restored to the
maximum extent possible.
Processing resid with a combustor
regenerator
Several reners process resid in
FCC units employing a UOP
Combustor Regenerator. This type
of regenerator operates in full
combustion, so the catalyst is fully
regenerated in a single stage.
Operating in full combustion elimi-
nates the need for a CO incinerator
because the ue gas exiting the
regenerator has minimal carbon
monoxide.
Spent catalyst exiting the reactor
stripper is transferred through the
Figure 2 UOP RFCC unit two-stage
regenerator with catalyst cooler
Figure 3 UOP Combustor Regenerator
uop.indd 3 11/9/12 14:27:01
arrange a larger amount of cooling
capacity around the regenerator.
This expands the working range of
combustor regenerators into terri-
tory once reserved for two-stage
regenerators. An 8ft-diameter cata-
lyst cooler is capable of removing
the heat equivalent of about 1.5
wt% Conradson carbon from an
80 000 b/d RFCC unit.
In general terms, RFCC refers to
FCC units processing feedstock
containing 2.0 wt% Conradson
carbon or greater. Based on this
defnition, Figure 4 shows UOPs
RFCC experience, grouped by type
of regenerator. Figure 5 describes in
spent catalyst standpipe into the
combustor (lower vessel) where the
catalyst is fully regenerated. During
regeneration, the catalyst is lifted by
the main air up into the regenerator
(upper vessel) where it is collected
and distributed (see Figure 3).
Some of the collected catalyst is
returned to the combustor through
the catalyst cooler and cooled cata-
lyst standpipe. Additional catalyst
is returned to the combustor
through the recirculation catalyst
standpipe, which increases the
combustor temperature and acceler-
ates the combustion kinetics. The
remaining catalyst is returned to
the reactor via the regenerated cata-
lyst standpipe.
Reactor side equipment is similar
to that previously described for a
two-stage regenerator.
Feedstocks containing as much as
2-3 wt% Conradson carbon can be
reasonably processed in a combus-
tor regenerator without a catalyst
cooler. For Conradson carbon levels
exceeding 2-3 wt%, the regenerator
temperature becomes excessive, and
a catalyst cooler is required to
manage catalyst deactivation and
maintain reasonable conversion.
Processing feedstock containing
greater than 4 wt% Conradson
carbon was once considered the
exclusive domain of two-stage
regenerators. However, in recent
years, several refners have proc-
essed increasingly contaminated
residual feedstock (up to 6 wt%
Conradson carbon) in combustor
regenerators with catalyst coolers.
A major advantage of a combustor
regenerator is the ability to produce
the lowest combined levels of CO
and NO
x
compared to all other
available regenerator technologies. It
is not unusual for emissions from a
combustor regenerator to be less
than 50 wppm for both pollutants.
Catalyst coolers drive conversion
Catalyst coolers provide the fexibil-
ity to maintain conversion when
feed quality deteriorates. This might
happen when taking down a feed
hydrotreater for catalyst replace-
ment or when processing more
contaminated crudes.
Structural and mechanical design
of the plant typically limits the
number of catalyst coolers that can
be reasonably arranged around the
regenerator. Early-generation cata-
lyst coolers were limited in size to
6ft or 7ft diameter. In recent years,
several 8ft-diameter catalyst cool-
ers as well as two 9ft-diameter
catalyst coolers have been commis-
sioned. In addition to larger
diameters, tube lengths have
increased. Both measures increase
the heat transfer surface area and,
hence, the capacity of a catalyst
cooler.
Table 1 shows the relative duty of
each size catalyst cooler. With
larger catalyst coolers it is easier to
Shell diameter, ft. 6 7 8 9
Relative duty, % 100 150 200 250
Catalyst cooler duty ranked by shell diameter
Table 1
S in g le -sta g e
S in g le -sta g e with
c a ta lyst c o o le r
Two -sta g e with
c a ta lyst c o o le r
0 2 4 6 8 1 0
Feed Conradson Carbon, wt %
O p e ra tin g u n it
Figure 4 UOP RFCC experience, grouped by type of regenerator; single stage = combustor
regenerator or bubbling bed regenerator
3 0
4 0
2 0
1 0
3 4 5 6 7 8 9
F
e
e
d

N
i
+
V
,

w
t

p
p
m
Feed ConCarbon, wt %
0
Two sta g e re g e n e ra to r with c a ta lyst c o o le r
C o m b u sto r re g e n e ra to r with c a ta lyst c o o le r
Figure 5 UOP RFCC units processing heavily contaminated feedstock
uop.indd 4 11/9/12 14:27:11
greater detail the subset of the most
contaminated feedstocks processed
in units having one or more catalyst
coolers.
How propylene is produced in a
catalytic cracking unit
Where propylene is produced
Propylene is produced primarily in
the riser section of the FCC unit
through a complex combination of
thermal and catalytic reactions (see
Figure 6).
Operating variables that impact
propylene production
The principal operating variables
that impact propylene production
are reactor temperature, ZSM-5
concentration, hydrocarbon partial
pressure and feed properties. Table
2 describes how each of these prop-
erties impacts propylene production.
Equipment design can also have a
profound impact on propylene
production.
8

High hydrocarbon partial pres-
sure tends to shift the equilibrium
of cracked products from olens to
parafns. Hence, higher propylene
production is achieved at lower
riser absolute pressure and higher
riser steam concentration.
Vanadium has a negative impact
on ZSM-5 activity. UOP has
observed this impact both in
commercial plants and through
laboratory testing. ZSM-5 appears
to be less efcient in units process-
ing feedstocks with high levels of
vanadium.
Feed properties
Propylene production is highly
dependent on feed properties.
Typical RFCC feedstocks are highly
aromatic (low UOP K) and conse-
quently resistant to conversion at
typical riser residence times. Typical
RFCC feedstocks also have a low
hydrogen content, frequently 12.0
to 12.5 wt% hydrogen, while
propylene requires a disproportion-
ate share of the hydrogen at 14.3
wt%. Co-products including
propane and dry gas require an
even greater share of hydrogen.
Hence, the amount of hydrogen
available from the feedstock can
limit the potential to produce
propylene.
Conradson carbon is another
important factor. Much of the
Conradson carbon winds up as
coke, thereby reducing the potential
pool of liquid and gas products,
including propylene.
Impact of hydrotreating
Table 3 shows a comparison of the
feed properties and corresponding
catalyst properties operating in two
RFCC units. One of the units proc-
esses hydrotreated resid feedstock,
while the other processes straight-
run resid feedstock. Note the high
50% boiling point temperature and
corresponding high concentration
of heterocyclics and organometal-
lics consistent with Figure 1. Note
also that the heterocyclics and
organometallics are substantially
lower for the hydrotreated
feedstocks.
As mentioned previously, nickel is
a dehydrogenation agent that
increases the molecular H
2
produc-
tion and delta coke (which manifests
as a higher regenerator tempera-
ture). The impact of nickel at levels
listed in Table 3, and even higher,
can generally be accommodated.
As a catalyst poison, the maxi-
mum achievable vanadium
concentration is self-limiting at
constant catalyst activity. If the cata-
lyst addition rate is decreased, the
vanadium concentration increases
and catalyst activity declines, forcing
a higher addition rate to maintain
the activity, which then reduces the
concentration of vanadium.
Resid to propylene
For the purposes of this discussion,
primary conversion means tradi-
tional conversion of feedstock to
Operating variable Propylene yield
Increased reactor temperature
Increased ZSM-5 concentration
Increased hydrocarbon partial pressure
Improved feed properties
Relative impact of operating conditions on propylene production
Table 2
Figure 6 Propylene production in a
catalytic cracking unit
Most of the
propylene is
produced by
catalytic and thermal
cracking in the riser
Like dry gas, some
of the propylene is
produced by thermal
cracking in the
hottest part of the
riser just above the
feed distributors
uop.indd 5 11/9/12 14:27:23
gasoline and lighter, while
secondary conversion means
conversion of gasoline boiling-
range olefns (produced during
primary conversion) to LPG olefns.
Propylene is produced by both
types of conversion. ZSM-5
promotes only secondary conver-
sion. With multiple handles
available for increasing propylene
production, and a unique set of
trade-offs associated with each
handle, optimisation testing is
required to determine the most
proftable operating point.
Case study
UOP has closely monitored optimi-
sation testing at several RFCC units.
In one such unit processing heavily
contaminated resid, comprehensive
testing was done at two levels of
primary conversion at high and low
ZSM-5 addition rates. Primary
conversion was adjusted by chang-
ing the reactor temperature.
Propylene production increased
with both primary conversion and
the ZSM-5 addition rate. At low
primary conversion, the incremen-
tal propylene attributable to ZSM-5
was more than three times the
increment at high primary
conversion.
ZSM-5 appeared to be less effec-
tive at higher primary conversion.
This phenomenon could be the
result of the lower supply of gaso-
line olefns available for secondary
conversion via ZSM-5. Light olefn
equilibrium may also contribute to
the diminishing response.
8

ZSM-5 zeolite deactivates in the
unit and is fushed out by equilib-
rium catalyst withdrawals and
therefore must be added continu-
ously. The higher the ZSM-5
addition rate, the higher the second-
ary conversion, but with
diminishing returns. When the
ZSM-5 addition rate was subse-
quently decreased for both primary
conversion levels, the propylene
yield decreased substantially while
the butylenes yield changed very
little, suggesting that propylene
selectivity is more sensitive to ZSM-
5 deactivation than butylenes
selectivity.
The addition rate for the high
ZSM-5 cases was 5 to 6 t/d, which
is a signifcant operating expense.
The major driver for this high ZSM-
5 addition rate was the high
cracking catalyst addition (and
withdrawal) rate required to main-
tain catalyst activity given the high
concentration of vanadium in the
feedstock, whereby ZSM-5 is
fushed out of the unit.
As with most FCC units, various
unit constraints were lifted at lower
reactor temperature and conversion.
This allowed for higher throughput,
which resulted in a higher gross
margin compared with lower
throughput at higher conversion.
The overall optimum operating
condition appeared to be a higher
feed rate with a lower reactor
temperature and with higher ZSM-
5 addition. However, given a
changing market price differential
between gasoline and propylene,
the optimum operating point will
shift, so re-optimisation is required
from time to time.
Figure 7 provides a broader
summary of the typical propylene
yield from various RFCC units
when operating in propylene mode.
Note the consistent correlation
between propylene production and
feedstock quality.
Note that several of these units
comfortably produce 10-11 wt%
propylene. Even higher levels are
readily achievable, depending on
market economics.
Advances in unit operations and
reliability
Main column bottoms system
fouling
Circulating slurry and net slurry
exchangers are more likely to suffer
from fouling on the tube side of the
3 0
4 0
2 0
1 0
3 4 5 6 7 8 9
F
e
e
d

N
i
+
V
,

w
t

p
p
m
Feed ConCarbon, wt %
0
Two sta g e re g e n e ra to r with c a ta lyst c o o le r
C o m b u sto r re g e n e ra to r with c a ta lyst c o o le r
H yd ro tre a te d re sid .
3 3 lb C
3
H
6
p e r b a rre l fe e d
S tra ig h t ru n re sid .
3 0 lb C
3
H
6
H yd ro tre a te d re sid .
3 5 lb C
3
H
6
S tra ig h t ru n re sid .
2 7 lb C
3
H
6
Figure 7 propylene production is a strong function of feedstock quality
Feed type Hydrotreated resid Straight-run resid
Feed quality
Base gravity, API 21.3 17.5
ASTM 50% boiling point, F 961 990
Conradson carbon, wt% 4.2 8.5
Sulphur, wt% 0.41 2.20
Total nitrogen, wppm 1200 2000
Nickel, wppm 4.2 20.0
Vanadium, wppm 6.9 17.0
Catalyst properties
Nickel, wppm 2500 7600
Vanadium, wppm 4500 6500
Conversion*, LV% 84.2 83.0
Propylene yield, wt% 10.7 8.2
*Gasoline ASTM 90% @ 380 F
Impact of hydrotreating on feed quality and catalyst properties
Table 3
uop.indd 6 11/9/12 14:27:36
High pressure. Extreme temperatures. volatile products. lt's all part of the job in
hydrocarbon processing. But so is the goal of maximizing safety integrity. We make
the process more secure with our innovative valves and controls, which is why the
industry relies on us to keep their workers safe and their plants running smoothly.
/ea|n mo|e aoo0t o0| o|ant oe|fo|mance so|0t|ons at CWFC.com
Eng|nee||ng t|ansfo|mat|on.
9PZROHZ
HS^H`ZILLU
WHY[VM
[OPZQVI
A part we can do without.
c wright.indd 1 1/6/12 21:16:30
exchangers when processing resid.
Exchanger fouling can limit heat
removal from the main column,
thereby increasing tray traffc and
overhead condenser duty beyond
capacity. As the cold side of the
circulating slurry is used to heat
feed, exchanger fouling can also
limit the feed temperature to a sub-
optimal range. Fouling can also
force a frequent reduction in feed
rate while cleaning the exchangers.
The composition of slurry
exchanger foulant has been inspected
on a number of plants and frequently
described as a thin homogeneous
layer of solid hydrocarbon.
Laboratory analysis has confrmed
the composition to be hydrocarbon,
free of catalyst fnes. This type of
fouling is due to the condensation of
polynuclear aromatics, and is a func-
tion of the residence time,
temperature and composition of the
circulating inventory. A useful
observation regarding this type of
fouling is that the slurry viscosity
will increase rapidly at incipient
fouling, so fouling can be predicted
and mitigated by monitoring the
slurry viscosity.
Fouling is also a function of the
tube side velocity. Since slurry oil
contains abrasive catalyst fnes, a
higher velocity will scrub away
foulant as it is produced. Indeed,
high velocity can cause tube inter-
nal erosion, so UOP specifes both
lower and upper limits for tube
side velocity.
The circulating slurry system is
similar to a continuous stirred tank
reactor. In this case, the reactants
are polynuclear aromatics, which
condense to form the foulant.
Production increases with residence
time. Decreasing the bottoms level
or increasing the net slurry produc-
tion reduces the residence time and
therefore condensation reactions.
At bubble point, the composition
and temperature in the circulating
pool track together and are directly
controlled by the LCO draw rate.
Fouling can be substantially control-
led by reducing the LCO draw rate,
but this presents a penalty as LCO
is downgraded to slurry. Quenching
(sub-cooling) the slurry pool with
cooled slurry is an effective way to
break the composition-temperature
relationship. Therefore, additional
LCO can be drawn and the slurry
composition will become heavier,
but without increasing the pool
temperature, partially mitigating
fouling.
Early generations of quench
systems consisted of a single nozzle
with no internals, which cooled off
that portion of the pool nearest the
nozzle but less so far away from
the nozzle, so localised high
temperatures persisted, contribut-
ing to fouling.
Exchanger fouling can be effec-
tively mitigated by using proper
equipment design and procedures,
and by implementing a rigorous
monitoring program. A key equip-
ment design element is a quench
distributor, which evenly distrib-
utes cool slurry throughout the
fractionator bottoms pool, thereby
minimising localised hot spots. The
quench rate is adjusted to maintain
the target pool temperature. The
normal operating liquid level is
maintained on the low side to mini-
mise the residence time in the pool.
Also, the circulating slurry fow
rate is set to maximum at all times.
This results in a high tube side
velocity and maximises the amount
of cooled slurry available for
quench. Also, a rigorous monitor-
ing program is required whereby
the slurry viscosity, tube side veloc-
ity and exchanger overall heat
transfer coeffcient are tracked as a
function of the LCO draw rate.
With these design and procedure
features, the time between cleaning
can be extended substantially.
Reactor dipleg coke fouling
Coke deposits inside the reactor
and associated equipment are more
common in resid FCC units due to
the coke precursors in the feed-
stock. Coke deposits are most
commonly found in stagnant or
cooled regions where precursors
might plate out on a surface.
Perhaps the most problematic coke
deposits attach to the interior of the
cyclone near the stagnant region on
the back side of the gas tubes. Coke
deposits at this location can be on
the same scale as the dipleg diame-
ter. These deposits can be dislodged
as the gas tube contracts and
expands during a thermal cycle
caused by an emergency shutdown
and restart. If the deposits then fall
into the dipleg, blocking catalyst
fow, high reactor side catalyst
losses can occur during restart.
This scenario has occurred through-
out the industry since refners
frst began processing resid.
Consequently, UOP now recom-
mends installing coke anchors
(similar to refractory anchors)
around the gas tube for RFCC units.
The coke anchors serve to capture
coke deposits that might otherwise
dislodge during a thermal cycle.
Coke anchors have been practised
occasionally since 1999
9
and are
gaining wider acceptance as a way
to mitigate this problem.
In addition to coke anchors, UOP
recommends 45-degree fapper
valves for the reactor cyclone
diplegs in RFCC units. In the event
a piece of coke dislodges from a gas
tube, a 45-degree fapper valve has
6 0 0 0
8 0 0 0
7 0 0 0
5 0 0 0
4 0 0 0
3 0 0 0
2 0 0 0
1 0 0 0
6
/
3
/
0
8
7
/
3
/
0
8
8
/
3
/
0
8
9
/
3
/
0
8
1
0
/
3
/
0
8
1
1
/
3
/
0
8
1
2
/
3
/
0
8
1
3
/
3
/
0
8
1
4
/
3
/
0
8
1
5
/
3
/
0
8
1
6
/
3
/
0
8
N
i
c
k
e
l

/

1
0
*
a
n
t
i
m
o
n
y
,

w
t

p
p
m
0
0 . 3 0
0 . 4 0
0 . 3 5
0 . 2 5
0 . 2 0
0 . 1 5
0 . 1 0
0 . 0 5
H
y
d
r
o
g
e
n

y
i
e
l
d
,

w
t
%

f
r
e
s
h

f
e
e
d
0
H yd ro g e n yie ld
N ic k e l
1 0 * A n tim o n y
Figure 8 Impact of antimony on H
2
production
uop.indd 7 11/9/12 14:27:48
an advantage because the dipleg is
completely free draining when the
fapper valve is wide open.
Dry gas production
Given the high nickel levels associ-
ated with processing resid, dry gas
production, and specifcally molec-
ular H
2
production, can be quite
substantial. This is problematic
because H
2
cannot be economically
recovered from dry gas and is thus
valued as fuel gas, thereby
downgrading its value. Since H
2

has a low molecular weight, small
quantities (on a wt% basis) can
substantially increase the fow rate
of dry gas, which might hydrauli-
cally bottleneck the wet gas
compressor and downstream gas
plant. Furthermore, since resid
feedstocks are generally hydrogen-
defcient (less than 12.5 wt%
hydrogen), diverting just 0.1 wt%
to molecular H
2
diminishes the abil-
ity to produce propylene. Previously
discussed well-proven methods for
controlling nickel-catalysed molecu-
lar H
2
production include using an
antimony additive. Figure 9 shows
a dramatic drop in H
2
yield when
antimony is frst introduced into a
unit with high nickel concentration
on the circulating catalyst.
Catalyst handling
A widely practised technique for
purging metals out of the unit is to
fush with low-metals purchased
equilibrium catalyst. This reduces
the harmful effects of the metals
and reduces fresh catalyst costs. For
this purpose, UOP specifes
purchased E-Cat hoppers for RFCC
units (see Figure 9). In some units,
the hoppers were originally charged
from one-ton supersacks. This
became a bottleneck because it took
a long time to transfer catalyst from
the supersacks to the hoppers.
Tilting truck charging facilities were
subsequently added for charging
the hoppers from shipping
containers. Load cell-based dosing
systems were also added to accu-
rately record the amount of catalyst
and additive transferred into the
unit.
Summary
In the past few years, several RFCC
units have been commissioned and
operated to convert resid to high
levels of propylene utilising UOPs
resid FCC process technology.
Included among this class are units
that process hydrotreated resid as
well as straight-run resid.
Processing heavily contaminated
resid using UOP Combustor
Regenerators with catalyst coolers
is gaining wider acceptance along-
side two-stage regenerators with
catalyst coolers. Catalyst coolers
expand the fexibility of these units
to process heavily contaminated
resid feedstocks, enabling high
conversion levels and high propyl-
ene production rates.
The optimum operating conditions
for maximising an RFCC units prof-
itability is a function of many
variables. In one case study, the
optimum was found to be at maxi-
mum throughput with a slightly
reduced conversion and a high
concentration of ZSM-5 additive.
Numerous other advances in
conventional RFCC processing tech-
nology have occurred in the past
few years. These include measures
to control main column bottoms
system fouling, reactor cyclone
dipleg fouling and dry gas produc-
tion, as well as advanced catalyst
handling facilities.
Optimix, VSS and AF are trademarks of UOP.
References
1 Venuto P B, Habib E T, Jr, Fluid Catalytic
Cracking with Zeolite Catalysts, Marcel Dekker
Inc, New York, 1979.
2 Stokes G M, Mott R W, FCC resid processing
an overview, AIChE 1988 Annual Meeting, 2
Dec 1998.
3 Upson L L, The basic heat balance, 4th Annual
Katalistiks FCC Symposium, 18-19 May 1983,
Amsterdam.
4 Rosser F S, Schnaith M W, Walker P D,
Integrated view to understanding the FCC NO
x

puzzle, 2004 AIChE Spring Meeting.
5 Schnaith, et al, Resid cracker handles record
nickel levels, Proceedings Second Int. Pet. Conf.
& Exbn. Petrotech-97, New Dehli, 503-512.
6 Hemler C L, Lomas D A, Tajbl D G, Residual
catalytic cracking: theory and practice, API
49th Midyear Meeting, 14-17 May 1984.
7 Cabrera C A, Hemler C L, FCC heat balance
control effect on resid processing effciency and
fexibility, AM-86-60, NPRA Annual Meeting,
23-25 March 1986, Los Angeles, CA.
8 Knight J, Mehlberg R, Creating opportunities
from challenges: maximizing propylene yields
from your FCC, AM-11-06, NPRA Annual
Meeting, 20-22 March 2011, San Antonio, TX.
9 NPRA Q&A October 2000 Question 4:
What single thing have you done to improve
the reliability of your FCC unit during the past
year?.
Pat Walker is a Technical Services Fellow
located at UOPs headquarters in Des Plaines,
Illinois. He is responsible for home offce
support of FCC units and resid FCC units
worldwide. He holds a degree in chemical
rngineering from the University of Illinois and
is a registered professional engineer.
Raymond Peterman is Regional Service
Manager for Honeywells UOP. He holds an
MBA from Pepperdine University and a degree
in chemical engineering from Northwestern
University. He is Six Sigma Green Belt certifed.
To re g e n e ra to r
Fresh cat.
hopper
Purchased
e-cat.
hopper
ZSM-5
hopper
Load cell
No. 1
Load cell
No. 2
1-ton bag loading
(old system)
New tilting truck
loaders (2)
Figure 9 Improved catalyst handling facilities
uop.indd 8 11/9/12 14:27:59
The sci ence of recovering and rening precious metal
catalysts is straightforward: state of the art technology.
The ar t of this process, however, is what makes Sabin
different from all others: thats the knowledge, experience,
and expertise gained from seven decades of successfully
serving thousands of organizations around the world.
Wed be pleased to count you among them.
Learn more at
sabinmetal.com
Sabin ExpTheDif-TableOfElements.indd 1 2/27/12 1:41 PM
sabin.indd 1 12/9/12 11:01:19
www.uhde.eu
ThyssenKrupp Uhde
Get more out of your coal.
Only too often do we fail to see the treasures that are right in front of us.
With our solids gasification technology you can make more out of any
feedstock. Why not contact us: you might just be surprised!
As a leading EPC contractor, we also have a proprietary portfolio of
technologies. And we network intelligently within the ThyssenKrupp Uhde
group based on our business philosophy Engineering with ideas.
th uhde.indd 1 11/9/12 11:16:18
Value of monitoring exchanger networks
H
eat exchanger fouling has a
direct impact on proftabil-
ity. Over time, fouling leads
to higher energy consumption,
higher maintenance costs, reduced
feed rates and shorter intervals
between turnarounds. The relation-
ship between fouling and energy
becomes more signifcant when you
consider the link between addi-
tional fuel gas consumption, higher
CO
2
emissions and the detrimental
impact on a refnerys energy inten-
sity index (EII). The environment
and the total cost of operation
(TCO) are negatively impacted.
Proven energy savings can be
realised when the fouling of a crude
unit preheat network exchanger can
be effectively monitored. Monitoring
will determine how fouling in a
network changes with time. Crude
units see the highest charge rates
and the largest temperature increase
of any refnery unit,
1
so the benefts
of a successful monitoring and foul-
ing control programme can be
signifcant.
This article will include a brief
review of crude unit heat exchanger
fouling mechanisms, how fouling
affects energy management costs,
and potential solutions.

Fouling mechanisms
What is fouling? It is the formation
of deposits in process equipment
that impedes the transfer of heat
and increases the resistance to fuid
fow. Several physical, operational
and chemical factors can combine
to form these deposits. Most crude
preheat deposits have low thermal
conductivity and reduce heat trans-
fer. Fouling can have a substantial
economic impact upon a refners
A rigorous exchanger simulation model can be used to calculate the true cost of
fouling in crude preheat networks
LAURA COPELAND
Nalco Company
proftability when it causes through-
put reductions due to hydraulic
limits or furnace tube temperature
limits. Fouling always leads to
energy losses when fuel has to be
increased to the furnace to make up
for lower crude temperatures
coming from the fouled crude
preheat network.
What causes fouling?
There are three main operational
factors that lead to fouling: blend-
ing crudes, the velocity through the
process and crude quality. Changes
to any of these factors can lead to a
change in fouling throughout the
process.
When crudes are blended
together, there is the potential for
instability that can lead to fouling.
If the crudes are processed individ-
ually, the fouling potential can be
different than if two or more crudes
are blended together. For example,
a refner could be processing a
heavy, low API crude and have
very little fouling, but when they
blend a light, high API crude with
it they see increased fouling in their
crude preheat train. The introduc-
tion of another type of crude has
caused instability in what is
normally a stable crude. Preheat
train monitoring can be used to
support the refners decision-
making process when implementing
a strategy to prevent fouling depo-
sition due to the incompatibility of
crudes.
2
Figure 1 shows the results
of testing done on the Nalco
Fouling Potential Analyzer (FPA),
where each crude individually has
a lower fouling potential than when
they are blended together. The FPA
value is the infection point of each
trend, and a lower FPA value
equals less stability. In this exam-
ple, when crude A is blended with
crude B, the stability decreases and,
therefore, there is a higher potential
for fouling.
Another operational factor that
can cause fouling is a change in the
0 . 0 6
0 . 1 0
0 . 0 9
0 . 0 8
0 . 0 7
0 . 0 5
0 . 0 4
0 . 0 3
0 . 0 2
0 . 0 1
4 5 4 7 4 9 5 1 5 3
R
e
l
a
t
i
v
e

a
b
s
o
r
b
a
n
c
e
FPA val ue
0
C ru d e C
C ru d e B
C ru d e A
F P A va lu e d e c re a se d
= le ss sta b ility
Figure 1 FPA trend that shows crude blending impact on stability
nalco.indd 1 11/9/12 14:47:02
ited when stressed by heat. The
deposition of inorganic salts can
result in fouling if the renery has
no desalting capability or if the
desalters are not working properly.
Finally, one more potential type of
fouling material is polymeric gums
that can form if a reactive stream
is added to the crude oil. Figure 2
shows an example plot of normal-
ised furnace inlet temperature
(NFIT). This shows how a change
in incoming crude to a renery
could have a signicant impact on
fouling. More discussion of NFIT
can be found in the next section of
this article.
Cost of fouling + cost of fouling
control = total cost of operation
It is possible to operate a crude
preheat to achieve the lowest TCO
by calculating the total cost of foul-
ing and the total cost of fouling
control. The costs of fouling are all
related to the extra fuel burned in
the furnace due to the fouling layer
inhibiting heat to be transferred to
the crude. As the fouling increases
in the different exchangers, the
crude exiting each exchanger leaves
at colder and colder temperatures.
That lower temperature crude must
be heated up to the xed furnace
exit temperature in order for the
rener to meet their target cut
points. This additional fuel due to
fouling is difcult to calculate with-
out a proper heat exchanger
simulator being run on a regular
basis.
Reners will also change the
pumparound rates to manipulate
the cuts in the atmospheric tower
for maximum protability. This will
also add or delete heat from the
preheat, but this is not due to foul-
ing. A proper monitoring
programme will be able to distin-
guish the difference between
temperature losses due to opera-
tional changes from temperature
losses due to fouling. This can be
achieved by calculating a NFIT
using a base set of operating condi-
tions. The NFIT will be equal to the
actual furnace inlet temperature
(FIT) as long as the operating condi-
tions remain the same. When
pumparound ow rates or tempera-
tures change, the heat load to the
velocity through the process.
Initially, most heat exchangers are
designed and sized to achieve a
maximum heat transfer at design
throughput conditions. Often these
design conditions achieve very low
fouling rates due to the high veloc-
ity and proper bafe design and
spacing. As throughput changes or
heat exchangers are added or rede-
signed, velocity changes and the
rate of fouling can also change.
Particulates travelling along with
the crude have a greater potential
to fall out and cause fouling at
lower velocities. A proper monitor-
ing programme should take into
account the velocities of the various
streams, how they are changing
with time, and the impact on foul-
ing and temperatures throughout
the system. Just as many exchanger
networks may encounter a clean-
ing effect from a sudden increase
in throughput, they can also experi-
ence a fouling event with a
sudden lowering of throughput. A
well-monitored system can locate
which exchangers tend to foul the
most as a result of decreased
throughput (velocity).
The third operational factor that
can lead to fouling issues through-
out the renery is the quality of
the incoming crude. Crude oil can
contain asphaltenes and inorganic
materials that can contribute to
fouling in the system. Asphaltenes
are the most common fouling
material found in the hot preheat
train. They are naturally stabilised
by resins that prevent them from
agglomerating, but asphaltenes can
easily agglomerate when destabi-
lised and cause fouling. Another
type of foulant is existent debris
such as sand or sediment carried
in the crude oil that may be depos-
Figure 2 Example trend of NFIT; circle indicates where a crude change took place that led
to increased fouling
4 8 5
5 0 0
4 9 5
4 9 0
4 8 0
4 7 5
4 7 0
4 6 5
4 6 0
1 4 N o ve m b e r 3 J a n u a ry 2 2 F e b ru a ry 1 2 A p ril
N
F
I
T
,

F
4 5 5
4 8 0
4 4 0
4 0 0
3 6 0
3 2 0
2 8 0
1
9

M
a
r
c
h
1
6

A
u
g
u
s
t
1
3

J
a
n
u
a
r
y
1
2

J
u
n
e
9

N
o
v
e
m
b
e
r
8

A
p
r
i
l
5

S
e
p
t
e
m
b
e
r
2

F
e
b
r
u
a
r
y
1

J
u
l
y
2
8

N
o
v
e
m
b
e
r
N
F
I
T
,

F
2 4 0
F I T
N F I T
N F I T a n d F I T
d iffe re n c e d u e
to o p e ra tio n a l
c h a n g e s
Figure 3 Example showing the difference between NFIT and FIT that can be seen due to
operational changes
nalco.indd 2 11/9/12 14:47:12
preheat will change and affect the
FIT, causing the FIT and the calcu-
lated NFIT to be different. The
difference will be the result due
solely to operating changes between
the base case and the current case.
Figure 3 is an example of the
differences that can be seen between
FIT and NFIT. The NFIT will show
the temperature decline due to foul-
ing, while the FIT will show the
temperature decline due to both
fouling and operational changes.
The NFIT trend is useful to show
the impact of changing any variable
that has an effect on the fouling
rate.
The decline in temperature due to
fouling can be converted into lost
BTUs (energy) that must be made
up in the furnace by burning extra
fuel. Incorporating furnace ef-
ciency and cost of fuel, the NFIT
reects a cost of fouling.
An antifoulant programme could
be added to improve (reduce) the
cost of fouling. The cost of the anti-
foulant would not be a cost of
fouling, but it should be considered
as cost of fouling control. The only
costs that should be considered as
fouling costs are those costs that
occur due to fouling, such as
increased furnace fuel spend, lost
throughput margin or even envi-
ronmental penalties from ring the
furnace harder. The costs of fouling
control are the total spend the
rener makes to clean or keep an
exchanger network clean. This
would include the maintenance
cleaning costs, extra fuel to the
furnace (if an exchanger is taken
off-line to clean), lost throughput
margin (if rates are reduced to
clean), antifoulant chemical (if
used), cleaning chemical (if used)
and any other cost the rener
absorbs when taking action to
reverse or control the existing foul-
ing in the exchangers.
The most common method of foul-
ing control is to take exchangers
off-line and mechanically clean them
by hydroblasting or lancing the
inside and outside of the exchanger
tube bundle. Whatever the cleaning
process, the rener should add all
the costs associated with the clean-
ing to determine the optimum time
to clean. Calculating the cleaning
cost is relatively easy, but knowing
when to clean is the hard part.
Solutions
In order to calculate the true cost of
fouling, a proper monitoring
programme is critical. The total
spend on fouling and fouling
control discussed in this article is
the total cost of operation for the
crude preheat. The optimum TCO
is the lowest combined cost of foul-
ing and spend on fouling control.
3

Each exchangers contribution to
the furnace inlet temperature is
maximised by cleaning exchangers
at the optimum cleaning cycle
frequency, and good exchanger
network simulators will calculate
the optimum time between clean-
ings. The optimum TCO is totally
dependent on the fouling rate. If
the unit starts to process a crude
that fouls at a faster rate, there will
be exchangers in the network that
will need to be cleaned more often
and, therefore, the optimum TCO
will increase. The same holds true
if the unit starts to process a crude
that is less fouling in nature the
optimum TCO will decrease. By
trending the optimum TCO for each
data set the rener can see how
much is being spent and so manage
cleaning to achieve the lowest
possible spend.
This includes all costs: cleaning,
fouling, chemical, lost production,
and so on. What it does not include
is a discount the rener receives for
processing opportunity crudes. The
rener will now be able to see the
added cost of processing opportu-
nity crude(s) and include these
economics into future buying
decisions.
Antifoulants can also decrease the
fouling rate, but they add to the
cost of fouling control. The addition
of antifoulants has to reduce the
fouling rate enough to lower the
overall TCO of the preheat to justify
the added cost. By using a proper
simulation model on a regular
basis, the rener can evaluate the
benets of using an antifoulant (or
not) and will always be able to
schedule the right exchanger for
cleaning in time to keep the preheat
operating at the lowest possible
total cost.
Conclusion
Crude preheat networks can be
managed to achieve the lowest
possible total cost of operation. It
requires a rigorous exchanger simu-
lation model that can normalise
input data to calculate the true cost
of fouling. The same model should
be used to calculate the optimum
cleaning cycle frequency for each
exchanger in order to determine the
true cost of fouling control. In this
way, antifoulant chemistries can be
evaluated based on the impact on
the total cost of operation.
Do:
Use a rigorous exchanger simula-
tion model
Normalise the furnace inlet
temperature
Base cleaning decisions on the
maximum impact on the furnace
inlet temperature when the TCO
calculation shows it is time to clean
Evaluate discounted crude
purchases based on impact to TCO
Evaluate antifoulant chemistry
based on impact to TCO.
Do not:
Make cleaning decisions based on
individual exchanger data (no way
to achieve lowest TCO)
Normalise the data based on
crude ow only can cause you to
miss >75% of the operating changes
that affect the furnace inlet
temperature.
References
1 Worrel E, Galitsky C, Energy Efciency
Improvement and Cost Saving Opportunities
for Petroleum Reneries, an ENERGY STAR
Guide for Energy and Plant Managers, 2005.
2 Wiehe I A, Kennedy R J, The Oil Compatibility
Model and Crude Oil Incompatibility, Energy &
Fuels, 14, 56-59, 2000.
3 Mason B, McAteer G, Nalco Company, Energy
Services Division (USA), Crude Preheat Energy
Management Leads to Sustainable Energy
Savings, Hydrocarbon Processing, 105-110,
Sept 2008.
Laura C Copeland is Global Industry
Development Manager with Nalco in Sugar
Land, Texas. She holds a BS in chemical
engineering from The University of Iowa and a
MBA from Northwestern University.
Email: lccopeland@nalco.com
nalco.indd 3 11/9/12 14:47:23
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
FCC catalysts and additives for cost and
emissions control
I
n 2011, Grace Catalysts
Technologies commercialised the
REpLaCeR series of low and zero
rare earth catalysts for both hydrot-
reated and resid feed applications.
REpLaCeR catalysts utilise proprie-
tary zeolites (Z-21 and Z-22) and
state-of-the-art stabilisation methods
to deliver similar, or even improved,
performance compared to more
traditional rare earth-based FCC
technologies. They received market
acceptance and offered signicantly
reduced catalyst costs for reners
when rare earth prices were at their
highest levels. These catalysts
continue to be used in more than 50
FCC units globally, even in an envi-
ronment where rare earth prices
have dropped from their peak in
mid-2011.
Based on the Z-21 and Z-22 tech-
nologies, there have been a total of
69 applications to date, which
includes 27 applications of zero rare
earth catalysts for low-metal hydrot-
reated feed and VGO applications.
For resid applications, the develop-
ment of rare earth-free catalysts is
much more challenging due to the
additional demands placed on
zeolite stability. However, signicant
advances in combining metals resist-
ance functionality with the rare
earth-free Z-21 and Z-22 zeolites
resulted in the development of the
REduceR catalyst for resid applica-
tions. This rare earth-free catalyst
can be blended with rare earth-based
resid FCC catalysts, thus reducing
the overall rare earth requirement
and the costs associated without
sacricing performance.
One example of an application of
REduceR catalyst is at an FCC unit
in a UK renery. The FCC unit is a
A rare earth-free catalyst enabled a renery to reduce its FCC catalyst costs and
a non-platinum combustion promoter helped control NOx

emissions
KENNETH HINDLE and MARIA LUISA SARGENTI Grace Catalysts Technologies
JUAN VARGAS Coryton Renery
UOP high-efciency design with a
catalyst inventory of 290 tonnes and
a fresh catalyst addition rate of 13-
15 t/d. The FCC unit operates in
full burn; the regenerator tempera-
ture is typically in the vicinity of
740C with an excess oxygen
content of between 0.7 and 1.3
vol%. This is a resid unit that proc-
esses 60-100% atmospheric resid,
with a feed concarbon of 3.5 wt%.
The e-cat metal levels are approxi-
mately 3500 ppm V and 2500 ppm
Ni. The renery began the catalyst
trial in August 2011 by blending
30% of the REduceR catalyst with
its previous Nektor catalyst. The
objective of the trial was simply to
maintain the performance of the
Nektor catalyst while reducing
overall rare earth content from 3.1
to 2.2 wt%.
Table 1 shows the feed properties
and specic catalyst addition rate
(SCAR) for the 100% Nektor cata-
lyst as well as the subsequent
period when 30% of the REduceR
catalyst was blended. It can be seen

Nektor (Jan 11 to Jul 11) REduceR 30% (Jul 11 to Oct 11)
Average feedstock properties
Specic gravity, kg/L Base Base + 0.01
Sulphur content Base Base + 17%
Basic nitrogen content Base Base
Concarbon, wt% Base Base + 0.1
Nickel content Base Base - 18%
Vanadium content Base Base + 14%
SCAR Base Base + 5%
Feed properties and SCAR
Table 1
80
90
85
75
70
65
60
55
0
1
0
2
/
2
1
/
7
2
1
1
0
2
/
1
/
7
2
1
1
0
2
/
2
/
7
2
1
1
0
2
/
3
/
7
2
1
1
0
2
/
4
/
7
2
1
1
0
2
/
5
/
7
2
1
1
0
2
/
6
/
7
2
1
1
0
2
/
7
/
7
2
1
1
0
2
/
8
/
7
2
1
1
0
2
/
9
/
7
2
1
1
0
2
/
0
1
/
7
2
1
1
0
2
/
1
1
/
7
2
1
1
0
2
/
2
1
/
7
2

,
T
A
M
50
100 Nektor 70 Nektor
+ 30 REduceR
Figure 1 E-cat microactivity
grace.indd 1 11/9/12 14:55:50
remained similar with the use of
30% of this catalyst, while bottoms
yield was slightly reduced. To
summarise the performance of the
REduceR catalyst, rare earth content
of the catalyst was reduced from
3.1 to 2.2 wt% while maintaining,
and even improving, catalyst
performance. The renery estimated
that the new catalyst provided cost
savings in the region of $500 000
per month.
Case study: a non-platinum
low-NOx

combustion promoter
Increasing environmental legisla-
tion is forcing reners to reduce
their emissions of nitrogen oxides
(NOx). The FCC unit regenerator
is typically the single largest source
of NOx within a renery and can
account for as much as 40-50% of
total renery NOx emissions. These
emissions are the result of nitrogen
impurities in the feed depositing
on the catalyst during the cracking
reaction. When the coke is burned
off in the regenerator, a portion of
the nitrogen is converted to NOx.
Since NOx emissions contribute to
acid rain, to the formation of
ground level ozone, as well as to
respiratory health impacts, various
environmental agencies have been
tightening NOx emission standards
over the last decade. There are a
variety of NOx reduction options,
both catalytic and hardware
oriented, available to help reners
comply with limits on NOx content
emitted from the FCC unit regen-
erator ue gas stream. However,
NOx-reducing additives are
becoming an increasingly popular
option among reners, as they
require no additional infrastructure
or chemical reactants. Indeed,
NOx-reducing additives are now
considered to be the best available
technique from the Integrated
Pollution Prevention and Control
(IPPC) Directive. Grace Catalysts
Technologies has developed a
series of NOx-reducing additives,
including XNOX W, which is a
non-platinum low-NOx

combustion
promoter.
Traditional platinum-based
combustion promoters are effective
at reducing carbon monoxide (CO),
but unfortunately also dramatically
that feed properties and SCAR were
very similar for both periods. The
very slight increase in SCAR (by
5%) can be attributed to an increase
in feed vanadium content by 14%.
Vanadium deactivates FCC cata-
lysts, causing product yields to
deteriorate.
Microactivity was not affected by
the use of 30% REduceR catalyst
(see Figure 1). Furthermore, the
FCC test run data shown in Figure
2 demonstrate the performance
obtained using the REduceR cata-
lyst. It can clearly be seen that dry
gas, LPG, gasoline and LCO yields
1 2
1 4
8
1 0
6
4
2
0 6 1 2 1 8 2 4 3 0 3 6 4 2 4 8 5 4 6 0 6 6 7 2 7 8 8 4
N
o
r
m
a
l
i
s
e
d

d
r
y

g
a
s

y
i
e
l
d
,

V
o
l
%
Test run
0
Te st ru n R E d u c e R 3 0 %
Te st ru n N E K T O R
2 0
2 5
1 5
1 0
5
0 6 1 2 1 8 2 4 3 0 3 6 4 2 4 8 5 4 6 0 6 6 7 2 7 8 8 4
N
o
r
m
a
l
i
s
e
d

L
P
G

y
i
e
l
d
,

V
o
l
%
Test run
0
1 2
1 4
8
1 0
6
4
2
0 6 1 2 1 8 2 4 3 0 3 6 4 2 4 8 5 4 6 0 6 6 7 2 7 8 8 4
N
o
r
m
a
l
i
s
e
d

g
a
s
o
l
i
n
e

y
i
e
l
d
,

V
o
l
%
Test run
0
1 5
1 0
5
0 6 1 2 1 8 2 4 3 0 3 6 4 2 4 8 5 4 6 0 6 6 7 2 7 8 8 4
N
o
r
m
a
l
i
s
e
d

L
C
O

y
i
e
l
d
,

V
o
l
%
Test run
0
1 2
1 4
8
1 0
6
4
2
0 6 1 2 1 8 2 4 3 0 3 6 4 2 4 8 5 4 6 0 6 6 7 2 7 8 8 4
N
o
r
m
a
l
i
s
e
d

s
l
u
r
r
y

y
i
e
l
d
,

V
o
l
%
Test run
0
Figure 2 FCC unit test run data
grace.indd 2 11/9/12 14:55:08
zymeflow.indd 1 11/9/12 11:56:36
increase NOx emissions due to the
fact that platinum catalyses the
oxidation of intermediate nitrogen
species, such as ammonia and
cyanide gases. Low-NOx combus-
tion promoters were introduced to
alleviate this problem by retaining
the CO oxidation function while
controlling the sharp increase in
NOx. XNOX W is a non-platinum-
based, low-NOx combustion
promoter, which with a dual func-
tionality actually reduces NOx
levels at the same time as retaining
the CO oxidation function.
The UK renery discussed above
has also recently trialled this low-
NOx combustion promoter. Before
the use of the promoter, the ren-
ery previously used the high-activity
Grace CP 3 platinum-based combus-
tion promoter on a regular basis.
Platinum is known to catalyse NOx

formation from intermediate nitro-
gen species. The basic nitrogen in
the FCC feedstock was typically 500
to 600 ppm, which led to uncon-
trolled NOx emissions of around 20
mg/Nm
3
. In late 2010, the renery
started antimony injections to help
passivate nickel from the feedstock.
The use of antimony is also known
to increase NOx formation from the
regenerator, and the combined use
of the platinum-based combustion
promoter with antimony resulted in
a sharp increase in NOx formation.
Due to this sharp increase in NOx

formation, the injection of antimony
had to be stopped. Grace
recommended switching from
the platinum-based combustion
promoter to the non-platinum low-
NOx combustion promoter to
reduce NOx emissions. Generally, a
replacement ratio of between 1.5-
2.0: 1 is required for XNOX W to
provide the same combustion
promotion as CP 3. As Figure 3
shows, the use of XNOX W allowed
the renery to reinitiate its addi-
tions of antimony while still
keeping NOx emissions under
control.
An excess of oxygen in the regen-
erator is also known to increase
NOx emissions. As Figure 4 shows,
the use of the non-platinum low-
NOx combustion promoter
provided NOx reduction even at a
high oxygen excess. In addition, its
use had no effect on afterburning
performance (see Figure 5). The
overall performance comparison of
CP 3 versus XNOX W is shown in
Table 2, where it can be seen that
the latter resulted in a 65% reduc-
tion in NOx emissions.
The rare earth-free REduceR
catalyst for resid applications has
provided signicant catalyst cost
savings for many reners and, as
the case study here shows, it is
maintaining and even improving
product performance compared to
traditional rare earth-based cata-
lysts. The non-platinum low-NOx

combustion promoter has allowed
this UK renery to control NOx
6 0
8 0
7 0
5 0
4 0
3 0
2 0
1 0
2
7
/
5
/
2
0
1
0
2
1
/
6
/
2
0
1
0
1
6
/
7
/
2
0
1
0
1
0
/
8
/
2
0
1
0
4
/
9
/
2
0
1
0
2
9
/
9
/
2
0
1
0
2
4
/
1
0
/
2
0
1
0
1
8
/
1
1
/
2
0
1
0
1
3
/
1
2
/
2
0
1
0
7
/
1
/
2
0
1
1
1
/
2
/
2
0
1
1
2
6
/
2
/
2
0
1
1
2
3
/
3
/
2
0
1
1
1
7
/
4
/
2
0
1
1
1
2
/
5
/
2
0
1
1
6
/
6
/
2
0
1
1
1
/
7
/
2
0
1
1
2
6
/
7
/
2
0
1
1
N
O
x
,

m
g
/
N
m
3
0
3 0 0
2 0 0
1 0 0
0
S
b

o
n

e
c
a
t
,

p
p
m
C P 3 XN O X W
S b a d d itio n s
S B o n e c a t
N O in flu e g a s
N O x b a se lin e
Figure 3 Controlling NOx emissions while injecting antimony using XNOX W combustion
promoter
6 0
1 0 0
9 0
8 0
7 0
5 0
4 0
3 0
2 0
1 0
0 0 . 5 1 . 0 1 . 5 2 . 0 2 . 5 3 . 0 3 . 5
N
O
X
,

m
g
/
N
m
3
Excess O
2
0
XN O X W
C P 3
Figure 4 Effect of excess oxygen on NOx emissions
CP 3 (Pt-based) XNOX W (low NOx)
Sb additions Sb additions
NOx in ue gas, mg/Nm
3
18 68 13 24
Basic N in feed, ppm 565 509 473 550
Feed rate, m
3
/hr 377 376 358 323
Concarbon, wt% 3.43 3.47 3.50 3.09
Regen temperature, C 741 745 744 731
Dilute phase pressure, BarG 2.63 2.79 2.88 2.88
O
2
excess 1.16 0.85 1.02 1.14
NOx reduction - - - 65%
Performance comparison of CP 3 vs XNOX W
Table 2
grace.indd 3 12/9/12 13:54:12
joined Petroplus in 2007 and has eight
years experience working as a process and
advanced process control engineer for units
including FCC, HF alkylation, isomerisation
and hydrocracking. Before joining Petroplus
he worked in the Hamaca Project for
the commissioning of the Ameriven
Heavy Crude Upgrader in Venezuela, and holds
a masters degree in chemical engineering from
Simon Bolivar University, Caracas, Venezuela.
emissions while still using anti-
mony injection.
GRACE, REPLACER, REDUCER, XNOX, and CP
are trademarks, registered in the United States
and/or other countries, of W R Grace & Co.-
Conn. NEKTOR is a trademark of W R Grace &
Co.-Conn. This trademark list has been compiled
using available published information as of the
publication date of this brochure and may not
accurately refect current trademark ownership
or status. Grace Catalysts Technologies is a
business unit of W R Grace & Co.-Conn.
Copyright 2012 W R Grace & Co.-Conn. All
rights reserved.

Kenneth Hindle is Regional Technical Service
Manager with Grace Catalysts Technologies.
After working as a Research Fellow at the
Centre for Theory and Application of Catalysis,
Queens University, Belfast, UK, he joined
Grace to work in R&D before moving into
sales covering UK and Sweden. He holds a PhD
in chemistry from the University of Glasgow,
UK.
Maria Luisa Sargenti is Technology
Coordination Manager with Grace Catalysts
Technologies, working within the marketing
team. Before joining Grace she held positions
in technical sales and marketing with Nalco,
and has more than 10 years experience in
chemical water treatments for different
industrial applications, including refning
processes and oil feld chemicals. She holds
bachelors and masters degrees in chemical
engineering from Simon Bolivar University,
Caracas, Venezuela.
Juan Vargas is Lead Engineer for the
cracking area at Coryton Refnery, UK. He
3 0
4 0
3 5
2 5
2 0
1 5
1 0
5
7 1 0 7 2 0 7 3 0 7 4 0 7 5 0 7 6 0 7 7 0
A
f
t
e
r
b
u
r
n
i
n
g
,

C
Regen. T, C
0
C P 3 ( P t-b a se d )
XN O X-W ( L o w-N O x )
XN O X-W +S b
Figure 5 Effect of XNOX W combustion promoter on afterburning
HydroCOM
www.hoerbiger.com
For more information:
compressor-mechatronics@hoerbiger.com
Experience real capacity
control and energy savings
HydroCOM is an effcient, stepless,
dynamic and fully-automated control
system. Several different methods
are available for capacity control
for reciprocating compressors.
Only HydroCOM, however, achieves best
results in terms of energy savings, speed
and accuracy. It improves process control,
optimizes performance and pays for itself
within a short period of time.
HydroCOM_ptq.indd 1 06.09.2012 11:50:15
grace.indd 4 11/9/12 14:55:34
Swagelok
Pressure Regulators are now an even bet ter choice for all
your pressure regulator needs. Why? Well, alongside our proven experience
and expertise, our range now covers sizes from
1
/8 to 4 in. and all your
regulator needs high-ow capabilit y, two-stage, back-pressure and
vaporizing models. With our regulators you get accuracy, sensitivit y and
pressure stability. In short total predictability. Exactly what you would expect.
Visit swagelok.com/pressure.
Make the unpredictable
totally predictable.

2
0
1
2

S
w
a
g
e
l
o
k

C
o
m
p
a
n
y
swa5573-10_RegAd_PTQ_210x297_ENG_v01AR_20120821.indd 1 8/21/12 11:48 AM
swagelok.indd 1 11/9/12 12:06:05
Stripping sour water: the effect of heat
stable salts
I
n a recent article dealing with a
live steam-injected stripper (PTQ,
Q3 2012), we discussed how
steam usage affected the stripped
water residual ammonia and hydro-
gen sulphide levels, how ammonia
distributed itself within the stripper
in an unexpected way, and how
Murphree (1925) vapour effciencies
varied with location within the
tower at various stripping steam
rates.
4
The discussion was limited to
non-phenolic sour water. In the
present article, we examine the effect
of heat stable salts (HSS) on sour
water stripper (SWS) performance,
how the injection of caustic soda can
spring ammonia from the sour
water, and how caustic injection can
worsen H
2
S stripping if it is injected
at the wrong place, or too much
caustic is injected. The analysis uses
a mass transfer rate-based simula-
tion model for sour water stripping
and for assessing the distribution of
ammonia in conventional amine
treating systems.
Sour water sources
The sour water generated in refn-
eries comes from numerous sources.
Most refnery sour water systems
contain very little CO
2
, but H
2
S
levels can become very high. The
capacity of ammonia solutions for
H
2
S is a direct result of the weak
acid-weak base reactivity between
H
2
S and ammonia. The potentially
high H
2
S content can make sour
water extremely foul, and H
2
S
removal from the sour water to
quite low levels is mandatory to
avoid unacceptable pollution levels.
Many sour water sources have been
noted in the excellent review article
by Asquith and Moore.
1
Mass transfer rate-based simulation models sour water stripping and assesses the
distribution of ammonia in conventional amine treating systems
RALPH WEILAND and NATHAN HATCHER
Optimized Gas Treating, Inc
Sour water is generally classifed
as phenolic or non-phenolic. Non-
phenolic water contains almost
exclusively NH
3
, H
2
S and possibly
a trace of CO
2
. It is generated by
refnery hydrotreating (hydrodesul-
phurisation, or HDS) units. When
stripped of contaminants, non-
phenolic water can typically be
recycled for reuse in the HDS unit
as wash water, or it can be used as
make-up water to the crude desalt-
ing process. Phenolic (or more
broadly, non-HDS) water typically
contains HSS, phenols and caustic.
Finally, it may be useful to point
out that ammonia and hydrogen
sulphide have almost unlimited
solubility in water when they are
present together. This is an interest-
ing consequence of the fact that the
reactive component of the solvent,
ammonia, is volatile and, if present
in the gas phase, it will continue to
absorb as long as it becomes proto-
nated as a result of H
2
S
co-absorption. Thus, it is conceiva-
ble that a particular sour water
stream may be a lot more concen-
trated than the solubility of either
ammonia or H
2
S by itself might
suggest.
SWS uses either steam generated
by a reboiler, directly injected
steam, or even a hot hydrocarbon
stripping vapour to shift chemical
reaction equilibria by applying heat.
Stripping vapour is the gaseous
solvent used to remove and carry
the ammonia and H
2
S out of the
system. It functions by:
Heating the sour water feed to
the boiling point
Reversing chemical reactions
Diluting the partial pressure of
the gases stripped by furnishing
excess vapour.
Figure 1 shows a typical SWS
column with heating by the injec-
tion of live steam and with the
possibility of injecting caustic soda
to one of the trays in the column.
Typical energy usage in the strip-
ping process is in the range 1.0-1.5
Cooler
Pump-1
SWS
4
1 1
7
1
3
2
8 1 0
9
6
5
F e e d /
b o tto m s
S WS A G
N a O H
in je c tio n
L ive
ste a m
S trip p e d
w a te r
S o u r
w a te r
Figure 1 SWS using live steam and caustic injection
ogrt.indd 1 11/9/12 16:56:14
taken to ensure that disposal of the
spent caustic is not completely reli-
ant on this destination, or the tail
will begin to wag the dog.
When adjusting the pH of the
water to spring ammonia chemi-
cally, the adjustment is usually
made by metered injection of caus-
tic onto a tray far enough down the
column that most of the H
2
S has
already been stripped out and
ammonia is the main remaining
component. The metering rate is
normally controlled to a set point
on the pH measurement in the
stripped water after it has been
cooled. Caustic injection on a lower
tray generally works better than
injection directly into the SWS feed
itself because the H
2
S concentration
is already small on lower trays.
However, pH is extremely respon-
sive to caustic addition, so the
measuring and control elements
should be as close together in time
as possible if rather large fuctua-
tions in pH are to be avoided. As
we shall see, no more caustic than
is absolutely necessary should be
injected because excess caustic can
permanently bind H
2
S into the solu-
tion and eventually this will fnd its
way into biological treatment
ponds, either reducing the effcacy
of the microbial population or
unnecessarily increasing the biolog-
ical oxygen demand.
Traditionally, SWSs have been
modelled as a series of equilibrium
stages, with stage effciencies being
quoted anywhere in the range from
15% to 45%, that is ranging over a
factor of three. However, since the
mid to late 1980s, the mass transfer
rate-based approach to simulating
amine contactors and extractive,
azeotropic and reactive distillation
has been in successful commercial
use. The extension to sour water
stripping is a natural progression
and, in December 2011, the ProTreat
simulation package saw the addi-
tion of a commercial mass transfer
rate-based SWS model. The remain-
der of this article uses this model to
explore how HSS affect the perform-
ance of SWSs and how caustic
addition can be tailored to provide
the optimum amount of stripping
of H
2
S and ammonia from HSS-
laden sour water.
lb of 50 psig equivalent saturated
steam per gallon of sour water.
To minimise heat exchange
surface, an external reboiler often
uses higher pressure (temperature)
steam than is typical in an amine
regenerator because amine thermal
degradation is not a limiting factor.
However, there is a practical limit
of 400-450F, where coking heavy
hydrocarbons can lead to fouling
and solids deposition in the reboiler
and, of course, corrosion is always
a concern.
Stripped sour water specifcations
for NH
3
and H
2
S can be highly
dependent on local requirements.
Typical targets for NH
3
are 30-80
ppmw in the stripped water and
undetectable to less than 0.1 ppmw
for H
2
S. Typical recent installations
involve 35-45 actual trays.
1,3
It is common in refnery cracking
units (FCCs and cokers) for the sour
water generated to contain organic
and inorganic acid impurities from
HSS precursors and, just as for
amine units, ammonia partially in
the protonated form. It cannot be
thermally regenerated because the
HSS responsible for the protonation
is completely non-volatile and
cannot be removed by boiling it
into the stripping steam. In such
cases, it is quite common to inject a
small amount of strong base
(NaOH) to shift the pH into a range
where ammonium ion (NH
4
+) shifts
back to NH
3
. Spent caustic from
Merox-type units is commonly used
for this purpose, but care must be
Total fow, lb/h 150 000
Temperature, F 135
Pressure, psia 70
Water, mol% 96.4
Hydrogen sulphide, mol% 1.5
Carbon dioxide, mol% 0.1
Ammonia, mol% 2.0
Formate, ppmw 200
Thiocyanate, ppmw 100
Sour water feed conditions
Table 1
1 0
1
1 0
3
1 0
2
1 0
1
1 0
2
1 0
3
1 0
4
1 0
5
1 0
6
0 . 7 5 1 . 0 0 1 . 2 5 1 . 5 0
H
2
S

i
n

s
t
r
i
p
p
e
d

w
a
t
e
r
,

p
p
m
w
t
1 0
7
St eam, l b / Sour wat er, gal l ons
3 0 0 p p m w t
0 H S S
1 0 0 0 0
1 0 0 0
1 0 0
1 0
0 . 7 5 1 . 0 0 1 . 2 5 1 . 5 0
N
H
3

i
n

s
t
r
i
p
p
e
d

w
a
t
e
r
,

p
p
m
w
t
St eam, l b / Sour wat er, gal l ons
1
3 0 0 p p m w t
0 H S S
Figure 2 Effect of HSS on stripped water quality
ogrt.indd 2 11/9/12 16:56:25
Case study
Figure 1 shows the simplest possi-
ble confguration of a SWS with
caustic injection. For this case study,
the same stripper as outlined previ-
ously was used.
4
It contained 40
one-pass valve trays on 2ft spacing
with 2-inch weirs. Sour water was
fed at 235F to tray 6 (from the top),
live steam saturated at 50 psig
entered below the bottom tray and,
in all cases, the column was sized
for 70% of jet and downcomer
food. Caustic could be injected on
any tray in the column. Table 1
gives the conditions of the sour
water used for this case study.
Effect of HSSs on stripped water
quality
Figure 2 shows the effect of the
stripping steam rate and the
absence versus presence of 300
ppmw HSS (composition indicated
in Table 1 on the simulated stripped
water quality with respect to resid-
ual ammonia and H
2
S content). It is
readily apparent from the upper
plot that the presence of HSS forces
the retention of a corresponding
concentration of ammonia, and no
matter how much steam is injected
into the stripper there is a residual
ammonia level that simply cannot
be removed when there are HSS in
the sour water. HSS have a negative
effect on ammonia stripping.
However, they have a benefcial
effect on H
2
S removal because they
are generally stronger acids. A two
order-of-magnitude reduction in
residual H
2
S is possible with only a
modest amount of HSS present.
One way to improve ammonia
removal is to inject caustic soda, a
stronger base than ammonia, onto a
tray somewhere in the column. At
the risk of oversimplifcation, if the
right amount is added, the caustic
will then bind all the HSS that
would otherwise be neutralised
with ammonia, without negatively
affecting H
2
S stripping. If too much
is used, the excess would be
expected to permanently bind H
2
S;
if too little, then less than the full
potential for ammonia release
would be realised. In a broad sense,
this is indeed what happens, but
for accurate answers one must
account for the pKa of the weak
acids and bases involved, as well as
considering the fnite mass transfer
rates of ammonia and H
2
S by using
a genuine mass transfer rate model
of the tower and its trays. There are
two questions to be answered:
where to inject ammonia, and how
much optimally to inject.
Injecting neutralising caustic on
various trays
As a frst approximation, one might
expect that the optimal caustic
injection rate will be just suffcient
to completely neutralise the HSSs
in the feed. This is a neutralisation
ratio of NaOH to HSS of unity. Less
than unity is under-neutralised,
greater than unity is over-neutral-
ised. Figure 3 shows the effect on
the stripped water residual ammo-
nia and H
2
S levels of injection of an
exactly neutralising amount of
caustic soda onto various trays.
Two live stripping steam rates are
considered.
As far as ammonia removal is
concerned, caustic injection at or
above tray 30 results in the optimal
degree of improvement. However,
H
2
S removal is improved only if
caustic is injected below tray 30.
Indeed, if caustic is not injected at
all (below tray 40 is the equivalent
of the tower sump and no stripping
is assumed to take place there) the
best possible H
2
S removal is
obtained, but no improvement to
ammonia stripping results.
Interestingly, and in contradiction to
conventional wisdom, caustic might
as well be injected right into the
sour water feed itself, as there is no
real beneft from injecting it onto a
mid-tower tray, at least not in this
case study. The beneft to H
2
S from
injecting below tray 30 is probably
because by that tray most of the H
2
S
has been removed anyway, and
there is too little left in the water to
be held by the caustic. On the other
hand, the further below tray 30 one
injects caustic, the fewer stripping
trays that are left to remove the
0 2 0 3 0 4 0
Caust i c f eed t ray
1 0
1
1 0
2
1 0
3
1 0
4
1 0
5
1 0
6
H
2
S
,

p
p
m
w
t
1 0
8
1 0
7
1 . 2 lb /g a l
1 . 4 8 lb /g a l
1 2 0
1 4 0
1 0 0
8 0
6 0
4 0
2 0
1 0 2 0 3 0 4 0
N
H
3
,

p
p
m
w
t
Caust i c f eed t ray
0
1 . 2 lb /g a l
1 . 4 8 lb /g a l
Figure 3 Effect of caustic injection on stripped water quality
ogrt.indd 3 11/9/12 16:56:35
sour water as well. For the case
being examined here, and injection
at tray 30, the stripped water pH at
120F is shown as a function of
neutralisation ratio in Figure 5 for
two stripping steam rates. If the
target caustic injection rate is 100%
neutralisation of the HSS, the
treated water pH at 120F is 9.0-9.1.
If the set point in actual use were
8.0, say, no caustic at all would be
injected at a steam rate of 1.2 lb/gal
and only 70% neutralisation would
be achieved at 1.48 lb/gal, leaving
35 ppmw ammonia in the stripped
water rather than the 5 or so ppmw
actually achievable. The value of
the pH set point to achieve the best
ammonia removal without need-
lessly poor H
2
S stripping is rather
an important parameter. It does not
seem to depend to any signifcant
extent on the stripping steam rate,
but it is a parameter whose value
can be predicted and verifed accu-
rately only by using a genuine mass
transfer rate model for SWS.
An additional observation from
the simulations was that when 500
ppmw of MDEA was added to the
sour water (steam rate of 1.48 lb/
gal) and the caustic was added to
tray 30 with a neutralising ratio of
1.0, essentially all the MDEA
reported to the stripped water.
However, the higher consequent
pH (9.48 versus 9.08) caused less
ammonia (5.5 versus 7.8 ppmw) but
more H
2
S (0.0000548 versus
0.0000235 ppmw) to remain in the
stripped water.
Conclusion
Perhaps the most important message
here is that genuine mass transfer
rate-based modelling allows the
construction of a virtual plant on a
computer. There is no reliance on
what might be termed fudge
factors to achieve agreement
between calculations and reality.
The power of the ProTreat SWS
model is the ability to analyse a
given tower under specifed condi-
tions in such detail that the optimal
caustic injection rate and even the
pH control point can be very accu-
rately assessed beforehand.
A mass transfer rate-based SWS
model provides a virtual SWS on a
computer. Engineers can design,
ammonia released; thus, in this case,
spiking the water with caustic is
increasingly pointless the further
below tray 30 one injects. We note in
passing that some of this discussion
becomes increasingly more academic
as the H
2
S levels being considered
fall below detectable limits (<<1
ppmw).
Effect of caustic injection rate
The effect of the extent of HSS
neutralisation on ammonia stripping
is fairly clear cut. As the upper plot
of Figure 4 shows, caustic injection
at a rate just suffcient to neutralise
the HSS springs the maximum
possible amount of ammonia,
although at low stripping steam
rates, a slightly larger amount helps.
With HSS, however, the effect of
under- versus over-neutralisation is
spread over a wide range of neutral-
isation ratios. When the
neutralisation ratio reached 0.5, the
H
2
S residual had already started to
rise and, by the point where a ratio
of about 1.25 was reached, the resid-
ual H
2
S had attained its highest
level. However, if exact neutralisa-
tion is not exceeded by much, quite
satisfactory residual H
2
S levels in
the stripped water can be achieved.
Note, though, that these results
pertain only to the conditions
presented in this case study they
should not be generalised; rather,
the ProTreat SWS model should be
run for the conditions pertinent to
the existent installation.
Set point for pH control and optimal
stripping
When using caustic injection with
phenolic sour waters, it is common
practice to control the caustic fow
to a pH set point. The question is,
what should the set point value be?
The answer almost certainly
depends on the amounts and kinds
of HSS present and probably on the
levels of ammonia and H
2
S in the
0 0 . 5 1 . 0 1 . 5 2 . 0
1 0
1
1
1 0
2
1 0
3
1 0
4
1 0
5
1 0
6
H
2
S
,

p
p
m
w
t
1 0
8
1 0
7
Neut ral i sat i on rat i o, NaOH/ HSS
1 . 2 lb /g a l
1 . 4 8 lb /g a l
1 2 0
1 4 0
1 0 0
8 0
6 0
4 0
2 0
0 . 0 0 . 5 1 . 0 1 . 5 2 . 0
0
N
H
3
,

p
p
m
w
t
1 . 2 lb /g a l
1 . 4 8 lb /g a l
Figure 4 Effect of extent of neutralisation on stripped water quality. Caustic injection was
onto tray 30 from the top of the column
ogrt.indd 4 11/9/12 16:56:46
analyse and optimise SWS units
with unprecedented accuracy and
reliability. Plant operations person-
nel can answer a wide range of
what-if questions to troubleshoot
operations and to optimise existing
units with considerably more con-
dence in the results. Some possibly
valuable lessons include:
The optimal caustic injection rate
is primarily dictated by the kind
and concentration of HSS in the
sour water
No matter how much caustic is
injected to spring ammonia, it will
have a deleterious effect on strip-
ping H
2
S
The tray selected for caustic injec-
tion is unimportant as long as it is
far enough up the stripper for there
to be sufcient trays below the
injection point to strip the addi-
tional ammonia
As long as caustic is added to the
sour water a reasonable number of
trays up from the bottom, the precise
tray for injection is immaterial to the
outcome. In other words, even
adding caustic directly to the feed
water itself produces stripped water
of the same quality as obtained by
injection into the column. Some
practitioners advise to add caustic to
the feed water; others say always to
add it to a tray. In fact, it makes no
difference, so one might as well save
a feed nozzle
The presence of amine in the sour
water results in higher pH and
better ammonia stripping, but
slightly poorer H
2
S removal. All
amine remains in the stripped
water.
In conclusion, this article has
revealed several aspects of strip-
ping phenolic sour water that do
not seem to have been recognised
heretofore. It also highlights the
tremendous benets that can be
obtained from simulating SWSs the
right way using a genuine mass
transfer rate model.
ProTreat is a mark of Optimized Gas Treating,
Inc.
References
1 Asquith J, Moore A, Sour water processing
balancing needs, Proceedings of the 2000
Brimstone Sulphur Recovery Symposium, Vail,
CO.
2 Murphree E V, Graphical rectifying column
calculations, Ind. Eng. Chem., 17, 747-750 and
960-964, 1925.
3 Quinlan M P, Hati A A, Processing NH
3
acid
gas in a sulphur recovery unit, Proceedings
of the 2010 Laurance Reid Gas Conditioning
Conference, Norman, OK.
4 Weiland R H, Hatcher N A, Reliable design of
sour water strippers, PTQ, Q3 2012.
Ralph Weiland began working in gas treating
in 1965. He taught chemical engineering for
30 years at universities in Canada, Australia
and the US, and spent 10 years doing mass
transfer tray development with Koch-Glitsch
LP. He established Optimized Gas Treating in
1992. He holds BASc, MASc and PhD degrees
in chemical engineering from the University of
Toronto. Email: ralph.weiland@ogtrt.com
Nathan Hatcher joined Optimized Gas
Treating, Buda, Texas, as Vice-President,
Technology Development, in 2009. He has
spent most of his career in amine
treating and sulphur recovery. He holds a BS
in chemical engineering from the University
of Kansas and is currently a member of the
Amine Best Practices Group.
Email: nate.hatcher@ogtrt.com
1 2
1 1
1 0
9
8
7
0 . 0 0 . 5 1 . 0 1 . 5 2 . 0
T
r
e
a
t
e
d

w
a
t
e
r

p
H

a
t

1
2
0
F
6
1 . 2 lb /g a l
1 . 4 8 lb /g a l
Figure 5 Setting the pH control point for optimal stripping
1 DURANIT
inert ceramic balls

2 special reformed packings
6 liquid distributors / collectors
7 random packings made of plastic
8 random packings made of metal
9 random packings made of ceramic
10 software and consulting
4 support plates / grids
5 feed devices: gas / liquids
3 droplet separators / demisters
P. O. Box 552, D - 56225 Ransbach-Baumbach
Phone +49 26 23 /895 - 0, info@vff.com
Tower packings,
catalyst support
material and
column equipment.
For further information
please visit:
www.vff.com
1
1
2
2
3
6
6
4
4
4
4
5
7
7
8
10
10
8
9
9
ogrt.indd 5 12/9/12 13:55:00
www.metso.com/oilandgas
Focusing on your safety
We o er peace of mind when it comesto your processsafety and productivity.
Metso isa single source for all your processcontrol needs, from automation systems
to intelligent ow control solutionsthat are proven in the toughest conditions.
Our 30,000 professionalsbased in over 50 countriesdeliver sustainability and
pro tability to customersworldwide.
metso.indd 1 11/9/12 11:42:58
Production of biolubricant base stock
B
iofuels have already been
accepted around the world for
their advantages over conven-
tional petroleum fuels, including
the opportunity for energy inde-
pendency. Now, similar growth is
expected for biolubricants, which
are derived from renewable vegeta-
ble oils for different niche
applications. Recently, the idea of
producing vegetable oil-based biol-
ubricants has led researchers to
develop process technologies for
their commercialisation.
Biolubricants are esters of heavy
alcohols derived from vegetable oil-
based feedstock and have
lubricating properties similar to
those of mineral oil-based lubri-
cants. Even though biolubricants
are priced twice as high as conven-
tional petroleum lubricants,
1

industries are investing in R&D
towards increasing oil recovery
from seeds, reducing the costs of
processes and exploring niche
application areas. The high cost of
biolubricant today is mainly due to
the non-availability of low-cost
vegetable oil feedstock and
lower-value realisation of the
by-products.
Vegetable oils are being investi-
gated as a potential source of
environmentally favourable lubri-
cants because of their combination
of biodegradability, renewability
and excellent performance as lubri-
cants. However, their relatively low
resistance to oxidation and thermal
stability, poor low-temperature
properties and narrow range of
available viscosities limit their
application as industrial lubricants.
Traditionally, biolubricants are
produced from vegetable oils in a
Using microwave technology in biolubricants production enhances product yields
and reduces reaction time
Shalini GUpta, Rajeev KUmaR, SUdha tyaGi and peddy v C Rao
Bharat Petroleum Corporation
reaction vessel, where the mixture
is heated by means of coil jacketing
using steam/hot water under lami-
nar conditions for an extended
period. Some of the key parameters
that affect the transesterifcation
reaction are: reaction temperature,
reaction time, amount of catalyst
and alcohol-to-oil ratio. In view of
this, the present work is mainly
focused on developing a cost-
effective, homogeneous catalytic
route as an alternative to the
conventional synthesis of biolubri-
cant base stock. The technique
under development involves the
infuence of microwave radiation
3

(3000-30 000 MHz) to reduce chemi-
cal reaction times from hours to
minutes and to reduce side reac-
tions to help in improving the
overall yield.
2,3,4

In the present study, various
vegetable oils have been explored
to synthesise higher esters using
heavy alcohols and catalysts.
Process variables
1,5
such as the
molar ratio of alcohols to oils,
higher homologues of alcohols,
catalyst-to-oil ratio, reaction temper-
ature, reaction time and feedstock
quality are investigated. The effect
of variables was monitored with the
help of HPLC and FT-IR analysis
techniques.
6
Furthermore, the
synthesised lubricant base stocks
were characterised for essential
properties such as density, viscosity
index (VI), pour point, ester content,
and so on. These properties of
vegetable oil-based lubricants can
be tailored by modifying the oper-
ating parameters and using suitable
additive formulations and some
chemical modifcations for end-use
applications.
Challenges in production and
commercialisation
For biolubricants to be economically
viable, both the lubricating charac-
teristics and the method of synthesis
are important. Biodegradable esters
are derived from the transesterifca-
tion of vegetable oils with higher
homologues of alcohols. These
products have been developed for
various biodegradable lubricant
applications, including metal work-
ing fuids and food-grade lubricants.
Vegetable oils have an excellent
lubricity, but poor oxidation and
low temperature stabilities. In order
to use plant-based oils as lubricants,
special additives and chemical
modifcations are necessary for end-
use applications.
Keeping these factors in view, the
reaction parameters for the synthe-
sis of biolubricant base stock by
microwave and conventional routes
were optimised. A study of the
effect of residence time for both the
conventional route and the micro-
wave route has been carried out
and the results compared.
Microwave processing provides a
distinct and substantial advantage
over conventional processes
primarily due to the manner in
which microwave radiation inter-
acts with the matter and transfer of
energy.
The data gathered could be used
for optimisation of the processing
parameters for the synthesis of biol-
ubricant base stocks using a
microwave system at shorter dura-
tion. The conceptualisation of the
process scheme has been developed
for scale-up and pilot studies
towards the commercial production
of biolubricant base stocks.
bharat.indd 1 12/9/12 10:16:13
prepared ester samples on a Thermo
Nicolet Nexus instrument.
Typical properties measured for
the vegetable oils used in the
present study are shown in Table 1.
The feedstocks were characterised
in terms of their viscosity and pour
point according to ASTM D 445 and
D97 respectively. Total acid number
was estimated according to ASTM
D664.

Conventional synthesis
5
The vegetable oils were mixed with
an alcohol-catalyst mixture in a
glass reactor with a refux condenser
and magnetic bar and subjected to
conventional heating for two to six
hours.
Transesterifcation reactions were
performed at different reaction
temperatures (150-185C) in the
presence of various catalyst concen-
trations ranging from 0.5-1.5 wt%.
The prepared biolubricant was
subjected to separation using
fltration/decantation or distillation
to remove the by-product. This
mainly neutralises the catalyst used
and also removes the impurities
retained in the upper phase.
Recovery of higher alcohols is done
by distillation under vacuum.
Samples were analysed at different
time intervals to check for conver-
sion in terms of ester content by the
HPLC method. The fnal product
was analysed for its physico-chemi-
cal properties.
Microwave-assisted synthesis
2, 7,8,9
A microwave-accelerated reaction
system equipped with an open vessel
design was employed for the purpose.
The experiments were performed in a
round-bottom fask with an alcohol-
to-oil ratio of 6:1 to 9:1 using C
6
-C
8

alcohols at 150-185C and catalyst
concentrations of 0.5-1.5 wt% for 10-
35 minutes employing the microwave
parameters listed in Table 2.
Results and discussions
Transesterifcation reactions were
carried out at similar reaction
conditions using conventional heat-
ing and microwave synthesis routes
for different times and prepared
samples were analysed at different
time intervals to monitor the
progress of the reactions in terms of
Selection of feedstock, synthesis
and characterisation of base stock
Vegetable oils (karanja, jatropha,
soya bean and linseed) were
obtained from local commercial
suppliers and were used without
further purifcation. A CEM make
MARS 5 microwave system was
used for the transesterifcation reac-
tions. The system was equipped
with a round-bottom fask and
refux condenser with magnetic stir-
rer bar and non-contact infrared
continuous feedback temperature
system, which allows for the
required constant temperature with
continuous stirring.
Solvents such as methyl tertiary
butyl ether (MTBE) and n-hexane,
used as the mobile phase for high-
performance liquid chromatography
(HPLC), were of the appropriate
grade and were degassed prior to
use. HPLC analysis was carried out
on a Shimadzu advanced Vp HPLC
system at a column temperature of
30C. Separations were made on a
Princeton silica column with a parti-
cle size of 5 microns and dimensions
of 250 mm x 4.6 mm. The mobile
phase consisted of a mixture of n-
hexane and MTBE in the ratio 90:10
by volume at a fow rate of 0.8 ml/
min. A RID 10A detector was used
for the detection of eluting compo-
nents. The optimum detector cell
temperature was kept at 30C and
the injection volume was 10 l. Data
acquisition and analysis was carried
out on Class Vp software. FT-IR
spectra were recorded using an ATR
crystal for characterisation of the
6 0
1 0 0
9 0
8 0
7 0
5 0
4 0
3 0
2 0
1 0
L in se e d o il S o yb e a n o il J a tro p h a o il K a ra n ja o il
E
s
t
e
r

c
o
n
t
e
n
t
,

w
t
%
0
6 2
6 1
6 h rs
R e sid e n c e
tim e
4 h rs
2 h rs
7 2
9 6
9 8
7 5 . 5
5 7
7 6
7 8
9 9 9 9
5 9
Figure 1 Conventional transesterifcation route
Feedstock properties Jatropha oil Karanjia oil Linseed oil Soybean oil
Properties
Density, gm/cm
3
@ 15C 0.9125 0.9340 0.9290 0.9234
Kinematic viscosity, cSt @ 40C 35.54 41.24 34.35 38.71
Kinematic viscosity, cSt @ 100C 7.93 8.23 8.08 8.80
Viscosity Index (VI) 136.2 128.31 141.8 140.33
Pour point, C 5.0 18.0 3.0 6.0
Acidity, mg KOH/sample 5.8 10.6 2.4 1.8
Typical properties of vegetable oil feedstock
Table 1
Stage Power RAMP, Control Hold
mins temperature Stirring time,
Max % C mins
I 1200W 100 30 185 Yes 30
Microwave parameters for biolubricant synthesis
Table 2
bharat.indd 2 12/9/12 10:16:26
Main Products
Regeneration of spent catalysts
XpresS ex-situ activation (sulphiding) of fresh and regenerated catalysts
Rejuvenation of catalysts
Resale of once regenerated catalysts
Sale of GUARDIAN material (guard bed catalysts)
Additional Services
Laboratory expertise of spent catalysts
Pilot plant ex-situ activation
Technical support
Storage of catalysts
Tricat GmbH Catalyst Service Bitterfeld

OT Greppin
Tricat-Str. (ChemiePark Areal B-Ost)
06803 Bitterfeld-Wolfen
Germany
Phone: +49 3493 75910
Fax: +49 3493 75999
Internet: www.tricatgroup.de
E-mail: info@tricatgroup.de
Tricat, Inc.
260 Schilling Circle
Hunt Valley, MD 21031
USA
Phone: (410) 785 7900
Fax: (410) 785 7901
Internet: www.tricatgroup.com
E-mail: sales@tricatgroup.com
- TRICAT
tricat.indd 1 23/2/12 12:15:06
Were Not Exaggerating. Our Boilers
are something to brag about.
We custom design and custom build boilers to perform efciently, safely and cleanly.
Your RENTECH boiler will lower operating costs, reduce emissions, and provide faster start-up
and cool-down. You ll nd satised customers on six continents with specialty boilers, HRSGs,
wasteheat boilers and red packaged watertube boilers from RENTECH. We ve been designing
and building boilers for people who know and care since 1996.
WWW.RENTECHBOILERS.COM
rentech.indd 1 1/6/12 13:22:29
ester content by HPLC (see Figures
1 and 2).
Vegetable oils, both edible and
non-edible, were selected for the
synthesis of biolubricants. These
included soybean, linseed, karanja
and jatropha oils. Characterisation of
the vegetable oils by their physical
properties was carried out. Among
these oils, karanja oil was found to
have a high acid value. With a high
acid feedstock, it is essential to ester-
ify the oil prior to subjecting it to a
transesterifcation reaction. Synthesis
of biolubricants was carried out at
temperatures of 150-185C, a molar
ratio alcohol-to-oil of 6:1 to 9:1, and
at catalyst concentrations of 0.5 to
1.5 wt%. The conversion of esters
was found to be more than 98% on
a weight basis in about 35 minutes
by the microwave route. Such a
conversion level could only be
achieved by means of a conventional
route in up to six hours. The prod-
ucts purity was determined by
HPLC. Typical HPLC chromato-
grams of biolubricants prepared in
conventional and microwave condi-
tions are shown in Figure 3. Peak 1
corresponds to biolubricant prepared
by the conventional method and
peak 2 corresponds to biolubricant
prepared under microwave irradia-
tion. The fnal products are
comparable.
The observed trend clearly
demonstrates the infuence of
microwave during synthesis due to
dipolar interaction between micro-
waves and alcohol solution. This in
turn is anticipated to lead to volu-
metric heating,
2,7
thereby resulting
in a rapid rise in temperature
within a few minutes and fnally to
complete conversion within a short
time span. On the other hand, the
conventional route is found to take
about six hours to attain the
required temperature because of
the non-volumetric mode of heating
employed. Furthermore, ionic inter-
actions are expected to be stronger
under microwave conditions due to
a smaller hydration shell compared
to the conventional route. The
increased reaction rate by micro-
wave irradiation is probably due to
elevated temperatures at local reac-
tion sites such as catalyst surfaces.
Microwave treatment brings about
a greater accessibility of the
susceptible bonds, thus making the
chemical reactions more effcient.
2,7
The IR spectrum of synthesised
biolubricant shows a characteristic
ester peak at 1738 cm
-1
(see Figure
4). A peak at 1170 cm
-1
is observed,
corresponding to C-O-C stretching,
and another peak at 1463 cm
-1
is
due to CH
2
asymmetric deforma-
tion bending. Strong bands at lower
frequencies, between 1170 and 1000
cm
-1
, are due to aliphatic esters. An
overtone of ester in the region 3450
cm
-1
is also visible. A peak at 723
cm
-1
is due to the presence of a long
alkyl chain in biolubricant.
Biolubricant base stock obtained
by transesterifcation using the
microwave synthesis route was
characterised for its physico-chemi-
cal properties
1,2
(see Table 3). The
optimised reaction parameters were:
temperature at 185C, C
8
alcohol,
6 0
1 0 0
9 0
8 0
7 0
5 0
4 0
3 0
2 0
1 0
L in se e d o il S o yb e a n o il J a tro p h a o il K a ra n ja o il
E
s
t
e
r

c
o
n
t
e
n
t
,

w
t
%
0
6 3
3 5 m in s
R e sid e n c e
tim e
2 0 m in s
1 0 m in s
7 3
9 6
9 7
7 5
6 1
7 7
7 9
9 9 . 2 9 8 . 9
6 2
6 5
Figure 2 Microwave transesterifcation route
0 . 2 5
0 . 4 5
0 . 4 0
0 . 3 5
0 . 3 0
0 . 2 0
0 . 1 5
0 . 1 0
0 . 0 5
0 . 0 0
0 . 0 0 . 5 1 . 0 1 . 5 2 . 0 2 . 5 3 . 0 3 . 5 4 . 0 4 . 5 5 . 0 5 . 5
V
o
l
t
s
Ti me, mi ns
0 . 0 5
5
3
3
9
0
7
8

5
.
0
0
8
Figure 3 Comparison of HPLC chromatograms of biolubricant
Properties of biolubricant Jatropha oil Karanja oil Linseed oil Soybean oil
biolubricant biolubricant biolubricant biolubricant
Density @15C, gm/cm
3
0.9032 0.9103 0.8961 0.8999
Viscosity @40C, cSt 11.58 12.71 10.3 11.08
TAN (mg KOH/sample) 0.8 1.2 0.4 0.8
Viscosity Index (VI) 125 121 128 131
Yield (%) of ester 96.9 96.2 97.5 98
Pour point,C -15 -5 -20 -18
Physico-chemical characterisation of prepared biolubricant base stock
Table 3
bharat.indd 3 12/9/12 10:16:39
FFA with aliphatic alcohol in the
presence of heterogeneous catalyst
included prior to the biolubricant
reaction. The overall process scheme
(see Figure 5) is continuous in oper-
ation and can be used in scale-up
studies.
Transesterifcation of esterifed oil
and/or vegetable oil with a low
FFA content reacted with aliphatic
alcohols in the presence of homoge-
neous base catalyst for biolubricant
production at near ambient condi-
tions. Removal of impurities and/or
by-product from the reaction
mixture for purifcation has been
included. Flash distillation of excess
alcohols has also been included in
the overall process scheme for
improving the product quality and
meeting specifcations IS: 1115.
Conclusion
A green technology is desirable for
the production of biolubricant to
reduce the dependency on mineral
oil imports and for the cultivation of
energy crops to boost the national
economy. In the present study, an
attempt has been made to synthesise
biolubricant base stock from vegeta-
ble oils by means of a novel and
environment-friendly process. A
microwave synthesis route is advan-
tageous compared with conventional
processes by accelerating the
alcohol-to-oil molar ratio 6:1; catalyst
concentration 1.0 wt%, and residence
time was 35 minutes. The properties
meet the product specifcations for
the application of metal working
lubricants as per IS: 1115. The
obtained data can be used for
conceptualisation of the fnal process
scheme for scale-up studies.
Conceptualisationoftheprocess
The conceptualised process typi-
cally comprised continuous
esterifcation and/or transestifca-
tion of vegetable oils with n-C
6
-C
10
alcohols to produce biolubricant in
the presence of catalysts, with no
yield loss compared to conventional
batch processes. Microwave radia-
tion is used for the reaction systems.
The result is a continuous process
for biolubricant production wherein
simultaneous use of catalysts for
esterifcation and/or transesterifca-
tion of vegetable oil having high
and/or low FFA content with n-C
6
-
C
10
alcohols can be employed at
near ambient conditions.
The process entails esterifcation
of vegetable oil containing high
R-OH
High FFA
Low FFA RX-MW-1 RX-MW-2 Purify
Flash
R -O H -1
O il-1
O il-2 C a t-1 C a t-2
O il
R 1 -O H -2
I m p u rity
P u re -B L
R 2 -R -O H
Figure4FT-IR spectrum of synthesised biolubricant sample (conventional and microwave
route)
Figure5Continuous process of MW synthesis for biolubricant basestock
6 0
1 0 0
3
8
4
8
.
1
8
2
9
5
6
.
7
2
3
0
0
8
.
8
8
3
4
6
1
.
6
0
3
4
6
7
.
7
8
9 0
8 0
7 0
5 0
4 0
3 0
2 0
1 0
0 3 5 0 0 3 0 0 0 2 5 0 0 2 0 0 0 1 5 0 0 1 0 0 0
T
r
a
n
s
m
i
t
t
a
n
c
e
,

%
Wavenumber, cm
1
0
3
0
0
8
.
9
9
2
9
5
6
.
7
7
2
9
2
3
.
7
0
1
7
3
6
.
6
1
1
4
6
2
.
9
9
1
3
7
8
.
9
8
1
2
4
1
.
3
9
1
1
1
8
.
5
3
1
1
1
8
.
7
6
1
0
3
6
.
5
4
7
2
3
.
1
0
7
2
3
.
6
1
1
0
3
9
.
1
4
2
8
5
4
.
6
2
bharat.indd 4 12/9/12 10:16:49
kinetics, saving energy and ftting
into a continuous production scheme
for biolubricant production. The
overall process scheme can be scaled
up under industrial-scale microwave
reactors for optimisation of process
conditions to meet product specifca-
tions for commercialisation. Product
obtained from the microwave
synthesis route has been character-
ised and found to meet the
specifcations for metal working
fuid applications. Conceptualisation
of the process scheme has been
carried out for microwave synthesis
for scale-up studies and commercial
production.
Acknowledgements
The authors express their sincere thanks to
Mr K V Seshadri, ED (R&D) of BPCL for his
continuous mentoring on research activities
and encouragement.
References
1 Akhade P B, et al, Proceedings of Petrotech
2009, P09-708.
2 A Rapid Microwave Assisted Procedure
for Recovery and Assay of Platinum
from Spent Reforming Catalysts Using
Advanced Analytical Techniques, ISFL-2006-
28-0064.
3 Sharma B K, et al, Chemical modifcation
of vegetable oils for lubricant applications,
JAOCS, vol 83, No 2, 2006.
4 Microwave technology a cost-effective
aid for producing biodiesel from high FFA,
Paper 031, ISFL-2008.
5 Nuchter M, Ondruschka B, Jung nickel
A, Muler U, Review commentary: organic
processes initiated by non-classical energy
sources, Phy Org Chem, 2000, 13:579-86.
6 Saifuddin N, Chua K H, Production of ethyl
ester from used frying oil: optimization of
transesterifcation process using microwave
irradiation, Malaysian J Chem, 2004, 6(1):077-
82.
7 Gupta S, Rao P V C, Proceedings of Petrotech
2003, L079.
8 Hillion G, Proriol G, et al, Synthesis of a high-
grade lubricant from sunfower oil methyl
esters, C R Chimie 00 (2004) 000-0009 article
in press.
9 Biswas A, Adhvaryu A, et al, Microwave
irradiation effects on the structure, viscosity
thermal properties and lubricity of soybean
oil, Industrial Crops and Products, vol 25, issue
1, Jan 2007, 1-7.
ShaliniGupta is a Deputy Manager (R&D) with
Bharat Petroleum Corporation, specialising
in the process development of biolubricants,
metal recovery from spent catalyst, resid
upgrading, new product development and
analytical sciences. She holds a masters in
chemistry.
Email: guptashalini@bharatpetroleum.in
Rajeev Kumar is a Deputy Manager (R&D)
with Bharat Petroleum Corporation. His main
areas of interest are crude oil processing,
refning processes, modelling, simulation and
optimisation as well as process development
for biodiesel and biolubricants production. He
holds a masters in chemical engineering from
Indian Institute of Technology, Kanpur, India.
Email: rajeevkumar@bharatpetroleum.in
Sudha Tyagi is a Manager (R&D) at the
Corporate Research & Development Centre of
Bharat Petroleum Corporation, Greater Noida,
Uttar Pradesh, India. Her areas of research
are analytical sciences, biofuels, fuel additives
development and new product development.
She holds a PhD in chemistry from the
University of Roorkee.
Email: sudhat@bharatpetroleum.in

MOBILE COOLING FOR INCREASED
PETROCHEMI CAL PRODUCTI ON
Chillers
Water Pumps
Heat Exchangers
Power Generators
Cooling Towers
Aggreko employs over 4,500 people operating from 165 locations in about 100 countries.
Aggreko has a successful track record in commissioning engineered mobile
solutions that eliminate process bottlenecks and improve production as well as
unit bottom lines.
We can help you meet your objectives:
Provide temporary cooling to increase the distillation capacity in the summer
Improve your Vacuum unit cut quality in summer by improving steam ejectors
capacities
Clean in operation your heat exchangers without unit stoppage
Continue to run your units during cooling tower maintenance
Cool your rundowns and hydrocarbons streams to meet product quality requirement
Set-up contingency plans to secure your utilities
We Rent :
France : 0825 008 007
Germany : 0800 2447356
Spain : 902 22 11 01
Italy : 02 48 40 491
Benelux : (0)3 825 02 71
Netherlands : (0)88 644 1441
Poland : 00800 3311395
www.aggreko.com
bharat.indd 5 12/9/12 10:17:05
C O M P R E S S O R S Q T U R B I N E S Q G L O B A L S E R V I C E
EBARA CORPORATION
www.elliott-turbo.com
Q Cust omer:
Ethylene processor, Texas, USA
Q Chal l enge:
Compressor load instability caused
by variations in suction conditions.
Q Resul t :
Elliott installed Tri-Sens TSx compressor controls
to keep the compressor load and quench fow in phase.
They turned to Elliott
for better control of their process.
The customer turned to Elliott for controls that prevent process instabilities.
Elliotts machinery expertise allied with Tri-Sens controls leadership allows operators
to focus on managing the process, not the machine. Who will you turn to?
The world turns to Elliott.
elliott.indd 1 11/9/12 11:23:54
Mitigation of heat exchanger fouling
T
he naphtha hydrotreater
(NHT) feed-efuent exchang-
ers at a US renery were
experiencing severe fouling. The
heat exchanger fouling was limiting
run length. As the preheat exchang-
ers fouled, the heater inlet
temperature declined, resulting in
an increased potential for two-
phase ow in the heater. Unit
throughput was reduced to manage
the minimum required heater inlet
temperature.
A root cause analysis investiga-
tion was conducted to develop a
clear understanding of the fouling
source. This analysis resulted in the
development of an antifoulant addi-
tive treatment programme that has
signicantly reduced the rate of
fouling. The antifoulant programme
has extended cycle length and
reduced maintenance costs, result-
ing in a yearly economic return of
over 500%. This article will review
the root cause investigation steps,
results of the treatment programme
and benets to the renery.
Fouling in hydrodesulphurisation
(HDS) units can impact throughput,
energy consumption, and shorten
catalyst life. Deposits form in the
feed-efuent heat exchangers and
on the top of the reactor beds. The
economic impact can be severe
from the problems caused by foul-
ing. Solutions include operational
changes, mechanical upgrades
and antifoulant additive treatment
to control specic fouling
mechanisms.
1

Description of unit
This NHT processes straight-run
and coker naphthas from a
combined crude/coker gas plant.
Detailed analysis of potential contributors identies the root cause of fouling in
naphtha hydrotreater feed-efuent exchangers
BRUCE WRIGHT Baker Hughes Incorporated
TODD HOCHHEISER Valero Energy Corporation
The feed from the gas plant consists
of butane through jet boiling range
material. The NHT feed is supple-
mented with purchased naphtha
from an intermediate storage tank.
All feed streams are mixed in a
surge drum and then pumped to
the shell side of the feed-efuent
exchangers. There are four exchang-
ers in series. Prior to entering the
rst exchanger, the naphtha feed is
mixed with hydrogen. The feed-
efuent exchangers are designed to
fully vapourise the naphtha to
prevent two-phase ow in the red
heater. The vapourised naphtha
and hydrogen mixture is heated in
the red heater to the required
reactor inlet temperature. Sulphur
and nitrogen impurities are
converted to hydrogen sulphide
and ammonia, respectively, in the
xed-bed catalyst reactor. The reac-
tor efuent vapour is cooled and
partially condensed in the tube side
of the feed-efuent exchangers and
reactor efuent air n cooler. The
liquid and vapour are separated in
a product separator. The hydrogen
gas from the separator is
compressed and recycled to the
shell-side inlet of the feed-efuent
exchangers. The separator liquid is
fractionated in the NHT gas plant
into butane, light naphtha, reformer
feed and jet fuel. Figure 1 is a sche-
matic diagram of the unit.
Description of problems
A renery conguration change
altered the boiling range of the
NHT feed from C
5
through jet to C
4
through jet. After the conguration
change, the fouling rate of the NHT
feed-efuent exchangers increased
signicantly. Figure 2 shows the
increased fouling rate after modify-
ing the NHT feed to include
butanes and butylenes. The loss of
heat transfer resulted in lower
C ru d e a n d
c o k e r n a p h th a
P u rc h a se d
n a p h th a
Surge
drum
Hydrogen recycle
compressor
Reactor
Furnace
Feed
effluent
heat
exchangers
Separator
Air cooler
M in im u m
te m p e ra tu re
re q u ire m e n t
Figure 1 NHT unit diagram
b hughes.indd 1 14/9/12 12:36:26
grace.indd 1 11/9/12 13:28:45
effective than washing the shell
side.
Deposit analyses
Visual inspections of the heat
exchangers prior to cleaning
revealed that the tube side was
clean, while the shell side was
severely fouled. Since the renery
regularly water washes the tube side
of the exchangers to dissolve and
remove ammonium chloride salts,
the cleanliness of the tube side was
expected. The shell-side deposits
consisted primarily of hydrocarbon-
based materials coupled with lesser
quantities of iron sulphide.
Table 1 shows the results for the
deposits obtained during the root
cause investigation. The tube side
was quite clean and the small
amount of material obtained was
found to be primarily iron sulphide
and iron oxide corrosion by-prod-
ucts that can form due to the reaction
of ammonium chloride salts with the
heat exchanger tubes. The shell-side
deposits were mostly organic but
were coupled with some iron
compounds and chloride salts. The
organic portion contained substantial
amounts of carbon; the hydrogen-to-
carbon atomic ratio revealed that the
deposit was composed of degraded,
partially cyclised polymeric material.
There was also a signicant amount
of nitrogen found in the shell-side
deposits. Nitrogen is commonly
found in deposits from coker naph-
tha due to the presence of amines
and pyrroles, which can take part in
polymer formation.
furnace inlet temperature. A low
furnace inlet temperature is not
sustainable due to heater fouling
caused by two-phase ow. The
reactor outlet temperature was
increased to offset the heat transfer
coefcient reduction by raising the
log mean temperature difference
across the feed-efuent exchangers.
The reactor temperature increase
was an effective method of manag-
ing the required minimum furnace
inlet temperature, although an
energy penalty was incurred for
heat lost through the reactor efu-
ent air cooler. As the feed-efuent
exchangers continued to foul, the
reactor temperature could not be
further increased due to sulphur
recombination at a higher reactor
temperature. Unit throughput was
reduced and eventually a shutdown
was required to mechanically clean
the feed-efuent exchangers.
Root cause analysis steps
and results
In order to understand the causes
of fouling in the NHT, a root cause
analysis approach was employed
that consisted of system and opera-
tions reviews, deposit analyses,
feedstock analyses and laboratory
fouling studies. These pieces of
information were coupled together
to establish the mechanisms respon-
sible for fouling and to develop
mitigation options.
System and operations
The NHT is congured so that a
combination of straight-run and
coker naphthas are fed hot to the
unit surge drum. Purchased naph-
tha supplements the renery feeds
to keep the NHT operating at
capacity. The majority of the
purchased naphtha is delivered to
the plant via barges and is contami-
nated with oxygen. The purchased
naphtha is not oxygen stripped.
Coker uids commonly contain
reactive compounds, including
olens, amines and carbonyls, that
can lead to polymer formation.
Some of these reactions are auto-
catalytic in the presence of oxygen.
The best practice for processing
coker naphtha through an HDS unit
is to ensure that intermediate tank-
age is not utilised, which would
provide time for polymer reactions
to commence. As such, the congu-
ration and operation of this unit
should help to minimise the forma-
tion of polymeric deposits.
Straight-run naphtha typically has
little impact on HDS fouling unless
there is a signicant inux of corro-
sion by-products from the crude
unit overhead system.
Feed to this NHT ows through
the shell side of the preheat
exchangers, while the reactor efu-
ent ows through the tube side.
This conguration is commonly
employed in hydrotreating units
because of the tendency for ammo-
nium chloride salts to form in the
reactor efuent at sublimation
temperature and pressure. These
salts must be removed through
online water washing to maintain
the heat transfer performance of
the exchangers. Online water
washing of the tube side of heat
exchangers is easier and more
6 0
8 0
7 5
7 0
6 5
5 5
5 0
4 5
4 0
3 5
0 5 0 1 0 0 1 5 0 2 0 0 2 5 0 3 0 0 4 5 0 5 0 0 3 5 0 4 0 0
U
,

B
T
U
/
h
r

/

S
F

/

F
Days
3 0
P o st-re va m p
P re -re va m p
Figure 2 Heat transfer coefcient pre- and post-unit revamp

All results in wt%
Tube deposit Shell deposit
Carbon <1 57
Hydrogen Nil 5
Nitrogen Nil 4
Oxygen 10 13
Sodium Nil 3
Sulphur 22 11
Chlorine Nil 1
Iron 63 4
Silicon 1 Trace
Composition Primarily Primarily
inorganic organic
NHT deposit analyses
Table 1
b hughes.indd 2 12/9/12 13:56:08
va|ue acc|ng cata|ysts, abscrbents, acc|t|ves anc
prccess technc|cgy fcr c|| refn|ng prccesses.
[[[NQGEXEP]WXWGSQVIRIVMIW
9/
Te| +44 (0)1642 553601
97%
Te| +1 630 268 6300
Te| +1 732 223 4644
Naphtha HDS
D|ese| HDS
,]HVSXVIEXMRKGEXEP]WXW
vacuum tcwer
,]HVSKIRGEXEP]WXW
FCC un|t
*''EHHMXMZIW
SOx NOx remcva|
l|ght c|efn prccuct|cn
Bcttcms ccnvers|cn
Veta|s traps
/ct|v|ty bccsters
CO cx|cat|cn
+EWPMUYMH
TYVMGEXMSR
S, Hg, C| remcva|
Cruce un|t
vCO HDS
Hycrcgen p|ant
4VSGIWW
HMEKRSWXMGW
Untitled-1 1 31/07/2012 17:18
j matthey.indd 1 11/9/12 11:47:00
Feedstock analyses
Analyses of HDS unit feed streams
provide insight into the fouling
mechanism root cause. Tests that
are utilised to identify potential
contributors to fouling are shown
in Table 2. Several of the analyses
look for components that can take
part in various polymerisation reac-
tions, while other tests identify
inorganic constituents such as iron
sulphide that contribute to deposit
formation.
2
Samples of the NHT feed compo-
nents were analysed to identify
possible fouling precursors. The
analyses summarised in Table 3
revealed that the purchased naph-
tha was relatively free of fouling
precursors. Additionally, the
handling practices of the purchased
naphtha had not changed from
prior cycles when minimal fouling
was observed. After this review of
the purchased naphtha, it was elim-
inated as a root cause.
The combined crude and coker
naphthas contained components
that lead to polymer formation.
Olens, as measured by the
bromine number, and mercaptans
are both signicant fouling contrib-
utors. When these components are
present in the feed stream, they will
produce free radical polymers.
Elevated basic nitrogen was also
detected in the feed. Basic nitrogen
in combination with organic acids
participates in condensation polym-
erisation.
2
The organic acid content
was below the detection limit, so it
is unlikely that these nitrogen-based
reactions were the primary cause of
the high fouling rate.
Laboratory fouling tests
Fouling simulation studies are used
to generate deposits and study the
ability of chemical additives to
control their formation. For these
feed streams, existent gums
3
were
used to measure the as-received
polymer content, and thermally
stressed gums were used to deter-
mine the tendency to produce
additional polymer. The thermal
stress test is also used to select the
best-performing polymer inhibitor.
Table 4 shows a summary of the
gum tests run on the feeds to this
NHT unit.
Dispersion tests are used to meas-
ure the ability of dispersant
additives to hold deposits in solu-
tion. Samples of the gum deposits
from the stress tests or deposits
from the heat exchangers are mixed
with a clear organic solvent along
with various dispersant additives.
The mixtures are shaken and then
allowed to settle. An effective
dispersant will hold the deposit in
solution longer than an untreated
sample. Dispersion tests were run
on the polymeric material formed
from the gum tests in order to iden-
tify an effective product for
controlling deposition of the foulant
material. A dispersant specically
formulated for control of organic
deposits was found to be highly
effective for this application.
Treatment programme
implementation
Based on the root cause analysis, a
Baker Hughes Lifespan treatment
programme was implemented to
control polymerisation of the reac-
tive feeds and to disperse the
organic and inorganic particulates.
Polymer Inhibitor A was used to
treat the coker naphtha stream
before it was mixed with the
straight-run naphtha. A dispersant
was injected at the NHT charge
pump to provide good mixing for
Analytical test Fouling concerns
Bromine number, g Br/100 g Olens
Mercaptan sulphur, ppmw as S Free radical polymerisation
Hydrogen sulphide, ppmw as S Iron sulphide formation
Total acid number, mg KOH/g Condensation polymerisation
Basic nitrogen, ppmw as N Condensation polymerisation
Pyrrole-indole nitrogen, ppmw N Polymerisation
Metals, ppm Inorganic deposits
Filterable solids, ppm Inorganic deposits
Typical HDS feed analyses
Table 2
Analytical test Purchased naphtha Combined crude and coker naphtha
Filterable solids, ppmw <10 <10
API Gravity 58.3 60.9
Bromine number, g Br/100 g 1.4 14
% Saturated H, normalised 94.5 96.2
% Olenic H, normalised 0.2 1.4
% Aromatic H, normalised 5.3 2.4
Mercaptan sulphur, ppmw as S 13 308
H
2
S, ppmw as S <1 10
Total acid number, mg KOH/g <0.05 <0.05
Basic nitrogen, ppmw as N <10 23.6
Pyrrole-Indole nitrogen, ppmw as N 0.2 3.7
NHT feed analyses
Table 3
Sample Test Stress Stress Results, Reduction
medium temperature, F (C) mg/100 ml %
Combined naphtha Existent gums N/A N/A 11 N/A
Combined naphtha Stressed Nitrogen 212 (100) 35
Polymer Inhibitor A Stressed Nitrogen 212 (100) 11 69
Polymer Inhibitor B Stressed Nitrogen 212 (100) 16 54
Combined naphtha Stressed Air 212 (100) 48
Polymer Inhibitor A Stressed Air 212 (100) 15 69
Polymer Inhibitor B Stressed Air 212 (100) 23 52
Gum test results with combined naphtha
Table 4
b hughes.indd 3 12/9/12 10:12:40
Conclusion
The root cause of the increased heat
exchanger fouling rate was the shift
in the NHT feed quality. The
renery conguration change imple-
mented just prior to the increased
heat exchanger fouling altered the
NHT feed to include C
4
. The coker
butanes and butylenes contain a
high concentration of mercaptans
and olens, which lead to free radi-
cal polymerisation.
Identication of the primary
cause of fouling enabled the devel-
opment of an antifoulant additive
treatment programme that was able
to control the rate of heat exchanger
fouling. This programme provided
the renery an economic return of
over 500% by permitting the unit to
run at full throughput rates,
preventing unit shutdowns prior to
scheduled catalyst replacements
and reducing maintenance costs.
LIFESPAN is a trademark of Baker Hughes.
Acknowledgements
The authors wish to extend their gratitude to
Tomasa Ledesma, Baker Hughes Antifoulant
Chemist, for her work in conducting the
fouling studies; the analytical group at the
Baker Hughes Sugar Land laboratory; and Ralph
Kajdasz, Baker Hughes Account Manager, for
his efforts in keeping the programme running
at the renery, and for development and
calculation of the monitoring data.
References
1 Wright B E, The causes and control of fouling in
hydrodesulphurization units a tutorial, AIChE
2002 Spring National Meeting, 3rd International
Symposium on Mechanisms and Mitigation of
Fouling in Rening and Upgrading, Mar 2002.
2 Medine G, Wright B E, Distillate hydrotreater
fouling, AIChE 2008 Spring National Meeting,
Apr 2008.
3 ASTM D 381, Standard Test Method for Gum
Content in Fuels by Jet Evaporation.
Bruce Wright is a Senior Technical Support
Engineer in Baker Hughess Industrial
Technology department in Sugar Land,
Texas, specialising in the hydrocarbon
process industries. He holds a BS in chemical
engineering from Rensselaer Polytechnic
Institute, Troy, New York, a MBA from the
University of Houston, and is a registered
professional engineer in the State of Texas.
Todd Hochheiser is a Renery Optimization
Manager with Valero Energy Corporation. He
holds a BS degree in chemical engineering
from the University of Delaware, an MBA from
the University of California, and is a member of
the American Institute of Chemical Engineers.
treatment of the combined feed
stream, including purchased naph-
tha. This combination programme
has provided excellent fouling inhi-
bition capability in similar units
and offers the advantage of being
able to adjust the treat rates of the
two components as needed.
Unit monitoring tools and trends
In order to verify the performance
of the chemical treatment
programme, a heat exchanger moni-
toring programme was utilised to
compare current operation to prior
cycle performance. The heat trans-
fer coefcient for the feed-efuent
heat exchanger bank was trended
versus run time. Figure 3 shows the
rate of decline of the exchanger heat
transfer coefcient for the last three
cycles. The rst two cycles shown
are prior to chemical treatment,
while the last cycle shown is after
implementation of the chemical
treatment programme. During the
previous two untreated cycles, the
heat transfer capabilities for the
preheat exchangers declined at a
rate of 0.11 U-coefcient units per
day, resulting in a 202-day run and
0.06 U-coefcient units per day for
a 404-day run. For the recent cycle
with the treatment programme in
place, the heat transfer decline was
0.01 U-coefcient units per day,
resulting in a 662-day run. The
treatment programme allowed the
renery to operate the NHT unin-
terrupted until the normally
scheduled turnaround for catalyst
replacement.
Figure 4 compares one of the heat
exchanger tube bundles with and
without the antifoulant treatment
programme. Both photographs are
taken at the end of run prior to
cleaning. The bundle appearance
and deposits from cycle 2 were
consistent with degraded polymers
coupled with iron sulphide. Clearly,
there was signicantly less deposit
with the fouling control treatment
programme in place.
After cycle 2 After cycle 3
1 1 m o n th s o n lin e ( u n tre a te d ) 2 1 m o n th s o n lin e ( tre a te d )
Figure 4 Heat exchanger tube bundle prior to cleaning
6 0
8 0
7 5
7 0
6 5
5 5
5 0
4 5
4 0
3 5
0 1 0 0 2 0 0 3 0 0 4 0 0 5 0 0 6 0 0 7 0 0
U
,

B
T
U
/
h
r

/

S
F

/

F
Days
3 0
C yc le 3 tre a te d
C yc le 2 u n tre a te d
C yc le 1 u n tre a te d
Figure 3 Heat transfer coefcient trends
b hughes.indd 4 12/9/12 10:12:59
global combustion technology
Harsh operating conditions. Challenging waste streams.
Changing emissions regulations. At Zeeco, we understand your
everyday realities. Our thermal oxidisers and combustion systems are
designed to efficiently dispose of sulfur, halogen or nitrogen-bound wastes,
BTEX vapors or amine vent streams. We know combustion: our burners,
flares and thermal oxidisers operate successfully on nearly every
continent and in every kind of environment - and our global
offices and experts stand behind each one.
Global focus. Local expertise.
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, Inc. 2012
Zeeco_PTQ_full_August.indd 1 8/29/12 4:43 PM
zeeco.indd 1 11/9/12 13:26:35
Weir Minerals Lewis Pumps is the world
standard for manufacturing pumps and valves
in the sulphur chemicals industry. O fferin g a
fam ily o f Lewis steam -jacketed su lp h u r p u m p s,
o u tstan d in g reliab ility in h ig h -tem p eratu re
su lp h u ric acid ap p licatio n s, an d a co m p lete fam ily
o f valves, Weir M in erals Lewis P u m p s co n tin u es a
lo n g trad itio n o f o fferin g su p erio r p ro d u cts an d
services.
With its stro n g co m m itm en t to service, Weir
M in erals Lewis P u m p s can sh ip stan d ard
rep lacem en t wear p arts in 7 2 h o u rs to m o st
in tern atio n al airp o rts in em erg en cy situ atio n s.
With a co m p lete lin e o f p u m p s an d valves, Lewis
P u m p s p ro d u cts are rig h t o n targ et. C u sto m ers in
m o re th an 1 0 0 co u n tries can t b e wro n g .
Lewis
Pumps
products
are right
o
n

t
a
r
g
e
t
Excellen t
M in erals
S o lu tio n s
L E W I S
P U M P S
Vertical C h em ical P u m p s
8 6 2 5 G ran t R d .
S t. Lo u is, M O 6 3 1 2 3
T: +1 3 1 4 8 4 3 -4 4 3 7
F: +1 3 1 4 8 4 3 -7 9 6 4
Em ail: sales@ lewisp u m p s. co m
www. weirm in erals. co m
C o p yrig h t 2 0 1 1 , En viro Tech P u m p system s, I n c. All rig h ts reserved . LEWI S an d LEWI S P U M P S are reg istered trad em arks o f En viro tech P u m p system s, I n c. ; WEI R is a reg istered trad em ark o f Weir En g in eerin g S ervices Ltd .
lewis.indd 1 7/3/12 10:48:06
Innovations in wastewater treatment
T
he demand for effcient and
cost-effective wastewater
treatment technology in the
refning and petrochemical sector is
being driven by not only ever-tight-
ening environmental legislation, but
also by the sectors own desire to
follow a meaningful sustainability
agenda and to take its responsibili-
ties around product stewardship
seriously. However, refning and
petrochemical companies are
continually confronted with the
challenge of striking a balance
between making their activities
proftable while ensuring the indus-
trial processes involved in the
production and application of a
chemical product, across its lifecy-
cle, have minimal impact on the
environment.
Treatment of wastewater from
petrochemical plants can be a chal-
lenging and costly matter,
particularly when needing to
comply with the requirements of
operational permits and national
environmental legislation govern-
ing the discharge of treated
wastewater into community treat-
ment plants or natural water bodies
such as rivers, lakes and oceans.
The segregation, collection and
treatment of wastewater play a vital
part in the protection of public
health, water resources and wild-
life. Refning and petrochemical
facilities, as part of their permit to
operate, must demonstrate that they
are successfully able to treat all
their pollution streams to the
appropriate regulatory standards.
One of the most widely used
strategies to meet the ever-rising
demand for water and increasingly
strict regulations governing water
Industrial gases provide a wide range of options for treating wastewater from
petrochemical processes
DARREN GURNEY
Linde Gases Division
protection is through improved
water management and strength-
ened investment in the technologies
that preserve and recycle process
wastewater.
The refning industry converts
crude oil and associated petroleum
gas (APG) into hundreds of refned
products, including petroleum,
diesel fuel, kerosene, aviation fuel,
fuel oils, lubricating oils and
primary feedstock for the petro-
chemical industry, and in doing so
it employs a wide variety of physi-
cal and chemical treatment
processes in which large volumes
of water are utilised, especially for
cooling systems, distillation, flter
backwashing and deionisation tech-
niques. Vessel cleaning, equipment
fushing and surface water run-off
also generate additional volumes of
wastewater that need to be treated.
In turn, the petrochemical industry
produces a multitude of essential
products to modern-day living
including intermediates for the
pharmaceutical industry, aromatic
organics, solvents, alcohols, ketones,
polymers and aldehydes, all of
which are synthesised through vari-
ous process operations, which can
produce large quantities of waste-
water that must be treated.
Given the complex and diverse
nature of refnery wastewater
pollutants, a combination of physi-
cal, chemical and biological process
trains and treatment methods are
usually required before it is fnally
discharged into the aquatic
environment.
Wastewater treatment can be
improved signifcantly by harness-
ing industrial gases such as pure
oxygen, for example, to enhance the
biological assimilation and oxida-
tion processes of wastewater
treatment plants or prevent unde-
sired odours in refnery mains or
storage lagoons and tanks. Carbon
dioxide is a versatile and safe
substitute for corrosive mineral
acids to effectively neutralise alkali
wastewater.
The technology used for refnery
wastewater systems is site specifc
and depends on the nature of
infuent (incoming wastewater)
conditions and the level of treat-
ment required by local regulatory
authorities. However, a typical
refnery wastewater treatment plant
usually consists of physico-chemical
pre- and primary treatment,
followed by secondary biological
treatment and tertiary treatment, if
necessary.
In a refnery wastewater treat-
ment system, two steps of oil
removal are typically required to
achieve the necessary removal of
free oil from the collected wastewa-
ter prior to feeding it to a biological
system. This oil removal is achieved
by using an American Petroleum
Institute (API) or equivalent oil
water separator followed by a
dissolved air fotation (DAF) or
induced air fotation (IAF) unit.
The wastewater is then routed to
the primary treatment clarifer and
to the aeration tank and secondary
clarifer, which constitutes the
biological system. The effuent from
the clarifer is then sent to tertiary
treatment, if required, prior to
discharge. The activated sludge
process is the most widely used
wastewater treatment technology
for the removal of soluble organic
contaminants in the oil refning and
linde 2.indd 1 12/9/12 10:52:30
chemicals facility manufactures
over 2 million tonnes of chemicals
products per annum and the refn-
ery has an annual capacity of 10
million tonnes.
At the petrochemicals plant, an
environmentally friendly CO
2
-based
technology, Solvocarb, is being used
to control alkali aqueous wastewa-
ter pH prior to discharge. The
petrochemical industry. Often the
pH of the raw wastewater requires
reducing before it can be accepted
by the bio-treatment stage, as the
high pH could potentially kill off
the bacteria doing the treatment.
CO
2
: versatile acid alternative for
pH control
In the UK, the industrial gases
technology company BOC Ltd, part
of The Linde Group, has seen
wastewater treatment successfully
implemented at a major plant oper-
ated by one of the worlds leading
petrochemical manufacturers.
The 1700-acre site is highly inte-
grated, exploiting synergies
between the petrochemicals plant
and adjacent refnery. The petro-
Figure 1 Solvocarb-R process
Oxygen tank
Control
unit
pH
electrode
Clear water
Evaporator Solvocarb
reactor
Centrifugal
pump
C O
2
m e te rin g
Wa ste wa te r in le t D isc h a rg e
ppp'îmj'\hf IMJJ,+)*+*)0
Where:
A = 20.95/79.05 x (100-CO
2
-CO-
O
2
) - (O
2
+CO
2
+CO)
CO
2
, CO and O
2
are fue gas
components in mol% carbon diox-
ide, carbon monoxide and oxygen,
respectively.
From these equations, one can
deduce that the hydrogen in coke
may vary between 2 and 15 wt% if
the CO
2
is changed from 14 to 18
mol%. In other words, small inac-
curacies in fue gas analysis can
lead to signifcant variations in the
calculated hydrogen in coke.
Therefore, hydrogen in coke is not
a very reliable indicator of stripper
performance. The following varia-
bles are recommended measured
before and after the stripper revamp
to assess the revamp benefts.
(However, hydrogen in coke can
still be used as a secondary param-
eter to check for consistency
of delta coke/heat balance
calculations.):
Regenerator dense bed tempera-
tures for a given set of riser outlet
temperature and feed preheat
temperature combinations
Product yields corrected for a
specifed boiling range
The combustion air required at a
given oxygen concentration in fue
gas
Feedstock analysis data (Conradson
carbon, nickel, vanadium, and so on)
Coke on the spent catalyst and
regenerated catalyst (to determine
delta coke)
Measured delta coke value
compared with estimated delta coke
value based on fue gas analysis
data.
In other words, the overall
performance of the unit before and
after the stripper revamp should be
used to quantify the benefts of the
stripper revamp.
<hg\enlbhgl
Use of the ModGrid stripper inter-
nals design signifcantly improves
performance and hence proftability
of an FCC unit. It is important to
understand the coke yield, delta
coke, their relationship to the cata-
lyst-to-oil ratio, and the infuence of
these variables on FCC unit
performance. It is also necessary to
utilise the correct measured varia-
bles to assess the stripper
performance after the revamp.
K^_^k^g\^l
* lCC heuL bulunce Ior heuvy olls, Hbe @Zl
ChnkgZe, 21 OcL 198S, 64.
+ lCC HeuL ulunce lundumenLuls wlLh u
Mechunlcul lrospecLlve, NlkA 2008 (CA1-08-
101).
, @kZ\^ =Zoblhg <ZmZeZ`kZf, lull 2009, 106,
22.
=Zebi Lhgb ls ulrecLor, lCC 1echnology, wlLh
Lummus 1echnology, u C&l Compuny. He hus
20 yeurs' experlence ln process deslgn, reseurch,
developmenL und evuluuLlon oI peLroleum
reInlng und synLheLlc Iuel processes, Lhe
mu|orlLy oI hls experlence hus been reluLed
Lo lCC Lechnology. He holds u buchelor's
degree ln chemlcul englneerlng Irom lun|ub
UnlverslLy, lndlu, where he wus uwurded gold
medul Ior Lhe besL ucudemlc perIormunce, und
u musLer's degree ln chemlcul englneerlng Irom
Okluhomu SLuLe UnlverslLy, USA.
KZfZ KZh FZkkb ls u Senlor lrlnclpul lrocess
1echnology Lnglneerlng SpeclullsL wlLh
Lummus 1echnology, HousLon, 1exus. He hus
over 19 yeurs' experlence ln lCC process deslgn,
developmenL und Lechnlcul servlces und wus
Lhe co-lnvenLor oI Lhe lndmux lCC process
developed by lOC k&u cenLre Ior converLlng
heuvy Ieeds, lncludlng resldue, Lo llghL oleIns.
He hus 12 puLenLs Lo hls credlL und holds u
musLer's degree ln chemlcul englneerlng Irom
lndlun lnsLlLuLe oI 1echnology, Kunpur, lndlu.
Ihlm&kôZfik^lneml
leed ruLe + 6%
leed CCk + 0.4%
MeLuls ln Ieed +2.7 ppmw
klser ouLleL LemperuLure +SC
kegeneruLor bed LemperuLure No chunge
;^g^mlh_Fh]@kb]lmkbii^kbgm^kgZel
!>nkhi^Zgk^g^kr"
MZ[e^+
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:
x refining/petrochemicals process specialty chemicals sales
x 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. C.da S.Cusumano 96011 Augusta - Italy E-mail: info@itw.it www.itw.it
HIRING EXPERIENCED PROFESSIONALS WORLDWIDE
lummus.indd 3 2/6/12 17:57:48 linde 2.indd 2 12/9/12 13:58:17
system uses gaseous CO
2
to neutral-
ise alkaline waters through the
production of carbonic acid (see
Figure 1).
The refnery, in compliance with
legislation at the time, had been
discharging wastewater from the
plant into the local river estuary
after adjusting its pH using mineral
acids, such as sulphuric and hydro-
chloric. Variability in discharge pH
and the corrosive nature of strong
mineral acids led to concerns over
potential harm the discharge may
cause to aquatic wildlife resident in
estuaries.
In addition to the plant needing
to fnd a more environmentally
friendly wastewater pH control
solution, it needed to fnd one that
would give it more robust control
over the whole process. In order to
achieve the target pH range through
the use of mineral acids, the
company observed periods of pH
oscillation from too much acidity
dosing, requiring adjustment with
additional alkalinity. This inevitably
leads to extra cost and operating
complexity arising from operating
two pH adjustment processes. The
company ultimately opted for a
single process route involving CO
2
,
which preserves the natural alkalin-
ity of the wastewater and the
process pH control is more stable
over the desired pH control range.
BOC was appointed to design the
pH control system for the newly
designed wastewater treatment
plant.
Owing to strict environmental
permits, wastewater may only be
discharged into the outlet channels
if it is within a narrow pH range
usually between 9 and 6. The
Solvocarb method employs gaseous
CO
2
to neutralise alkaline waters.
When dissolved in water, CO
2

forms carbonic acid, which reacts
with the alkalinity to form a salt.
The neutralisation reaction controls
the pH value to the appropriate
discharge level.
Awarded the contract to design
two Solvocarb systems to neutralise
all plant wastewater, anywhere
between 10 000 and 20 000 cu m/d,
BOC engineers designed each
system to simultaneously mix and
dissolve CO
2
into each 10 000 cu m
tank, controlling the pH to the
appropriate set point. Consent
permits dictated the site could only
discharge pH-corrected wastewater
into the estuary when the tide was
in, allowing the treated wastewater
to dilute effectively into the larger
body of available water in the estu-
ary. However, this meant that the
whole process had to be completed
within a restricted time period of a
six-hour window between the two
tides.
It was critical for the wastewater
to be neutralised in the two tanks
within the time available, which
called for challenging process
hydrodynamics. Large and variable
volumes of wastewater needed to
be brought within the correct pH
range within a fxed timeframe
the wrong pH value could result in
the refnery being unable to
discharge the wastewater, causing
potential bottlenecks and resulting
in backups further up the process
chain. A signifcant amount of test-
ing was conducted before the team
was satisfed that the proposed
system would operate to the
required parameters.
The new wastewater treatment
plant was commissioned in
February 2000 and was completed
on time and on budget. BOC has
continued to supply the plant for
nearly 12 years. The continued
successful operation of the plant
will help safeguard the environ-
mental status of aquatic areas in
which the wastewater is
discharged.
Today, the main driver for treat-
ing effuent high in alkalinity prior
to discharging to the outfall is the
strict regulation to protect the sensi-
tive, biodiverse ecosystem within
the estuary. Using CO
2
to neutralise
an alkali effuent avoids large
swings in the discharge pH, a vital
component in creating a sustainable
and suitable environment for
marine life.
Compared with mineral acids
commonly used in previous years,
CO
2
offers many advantages,
Oxygen tank
Control
unit
Evaporator
Oxygen di ff usi on
Aerat i on
t ank Vent uri pi pes Mi xi ng pi pes
Wa ste w a te r in le t D isc h a rg e
Sewage i mmersi on pump
Figure 2 Solvox-V process
linde 2.indd 3 12/9/12 10:52:49
air or pure oxygen supplied equip-
ment. However, traditional aeration
systems, designed as fxed mass
transfer, often cannot match the
variability or increased oxygen
transfer intensity requirements of
the activated sludge process, espe-
cially when operating temperatures
increase above aeration perform-
ance design values, normally
around 20C. Pure oxygen-based
systems, on the other hand, have
the potential to signifcantly
enhance the aeration process by
augmenting, or completely replac-
ing, portions of the installed
existing aeration systems, even at
elevated temperatures of >30C.
Pure oxygen is able to maintain a
higher aeration intensity because
air only contains one-ffth oxygen
by volume, and the limitation on
oxygen transfer is controlled by the
partial pressure of oxygen in air
and the oxygen solubility at a given
water temperature. The activated
sludge process requires a positive
dissolved oxygen level of at least
>2 mg/l to be effective. As the
temperature increases and the back-
ground dissolved oxygen remains
constant, the driving force of air
effectively reduces. It is the driving
force potential of pure oxygen,
compared with air at the same
temperature, that proportionately
increases the rate of oxygen transfer
into the activated sludge treatment
process. Combined with effective
mixing, a constant supply of
amounting to the best economical
and ecological alternative. CO
2
is
not categorised as a substance that
is harmful to water and does not
lead to the addition of unwanted
anions in the water environment,
such as chlorides and sulphates.
There is also no over-acidifcation
of the wastewater, due to the self-
buffering nature of CO
2
in water,
which produces a fat neutralisation
curve and no corrosion of system
and equipment components. CO
2
is
also far safer than the acids previ-
ously used. Simple to handle, it is
delivered as a liquid cryogen that is
stored in tanks onsite and dosed
automatically into the process.
The fact that CO
2
is sequestered
and effectively removed from the
global carbon cycle means that it is
not available to take part in global
warming; therefore, less CO
2
is
being released into the atmosphere.
This gives the Solvocarb process its
green credentials.
Pure oxygen wastewater treatment
The application of pure oxygen has
been successfully applied in the
activated sludge process to treat a
variety of industrial and refning
operations wastewaters for over 30
years. The activated sludge process
is the most widely used form of
biological treatment for organic
contaminants in aqueous wastewa-
ter globally. There have been
numerous industrial wastewater
installations built around the world
where pure oxygen is routinely
used. The application technology
can be designed into the process
during a greenfeld build or later as
a retroft application to increase the
capacity of an existing asset.
Lindes Solvox wastewater appli-
cations are an example of a pure
oxygen wastewater technology that
can be employed in the original
design of an activated sludge plant
or retroftted as an upgrading tech-
nology for existing wastewater
treatment plants (see Figure 2). The
primary application in most cases
is to increase treatment capacity in
existing plants that are overloaded
or experience wide swings in
dissolved oxygen demand. The
philosophy of Linde has been to
provide, where practical, perform-
ance improvement in existing
assets rather than building new
plant capacity to increase waste-
water treatment throughput. This
approach allows the operator to
harness the benefts of pure oxygen
within the existing plant footprint
and improve operational perform-
ance; for example, lower surplus
sludge production, reduced vola-
tile emissions, lower operational
power consumption, better settle-
ment of biomass and simple
installation.
Effcient oxygen transfer and
adequate process mixing are essen-
tial components of all aerobic
wastewater treatment processes and
these can be readily achieved using
Untreated
wasted water
Process
control
Bioreactor
Bio-treatment and oxygenation
Oxygen and air injection
controlled by DO and pH
feedback loops
Membrane
modules
RO process and
water re-use
Water re-use /
final disposal
Treated
water
S e p a ra tio n o f b io m a ss
/ tre a te d wa te r
A ir
O x yg e n
Figure 3 Axenis wastewater treatment system
linde 2.indd 4 12/9/12 10:52:59
dissolved oxygen ensures that the
micro-organisms providing the
biological wastewater treatment
perform to their maximum poten-
tial, in an environment where
oxygen transfer matches their
demand.
The Solvox concept is to intro-
duce pure oxygen into the
wastewater treatment process via a
family of specially designed and
developed pure oxygen and air
oxygen application equipment.
Equipment is congured to work
alongside existing aeration equip-
ment and meet the oxygen demand
of the biomass generated during
synthesis and oxidation of biode-
gradable contaminants. Within
industrial wastewater treatment
plants, pure oxygen can be used in
conjunction with air to increase
operational efciency or overall
treatment capacity. Often the exist-
ing aeration equipment has reached
its maximum transfer intensity due
to limitations brought about by the
nature of the wastewater, plant
operating temperature or reduced
efciency due to equipment wear
and tear.
Solvox oxygen solutions can also
be used to prevent odour nuisance
from wastewater that is transferred
in large pressurised pipelines and
mains, especially if the retention
times are long. The addition of
oxygen prevents the naturally
occurring bacteria on the pipe walls
from consuming the chemically
bonded oxygen and producing
highly odorous compounds such as
hydrogen sulphide and mercaptans.
Such compounds can quickly result
in the corrosion of pipework
networks and ancillary equipment.
The addition of free oxygen creates
ideal conditions for successful
preliminary treatment and aerobic
sealing of sludges in tanks and
vessels.
Equipment and process knowl-
edge can be applied in a variety of
ways to improve the operational
performance efciency of many
plants and processes. Adding
oxygen using Solvox equipment
requires an initial low capital
investment and, once installed, is
designed to be operated exibly so
it can be adjusted to seasonal needs
or used to track peak loading
during ramp-ups in production
campaigns.
Water reuse and recycling
Recycled wastewater makes it
possible for companies that use
water in their processes to treat the
efuent on-site and reuse or recycle
as much of it as possible back into
the industrial operation. Many
companies are turning to this
alternative against a background of
increasing pressure in the form of
restrictions on allowable discharge
volumes, limitations on abstraction
quotas and mains supply, legisla-
tive pressure and the upward
spiralling increase in the turna-
round costs of water. The latter
considers both incoming water
supply and trade efuent discharge
costs. Lindes wastewater recycling
technology, Axenis, employs a
high-rate membrane bioreactor
We Process Your Energy We Process Your Energy
Because Oil & Gas projects require highly reliable and cost effective solutions we
timely deliver since over 35 years fully guaranteed modular process-equipment
packages tailored to meet specifc customers requirements; for all kind of
Oil & Gas processing applications.
Because each project is unique and highest quality design is paramount to
success we offer a unique portfolio of proprietary gas treatment technologies
licensed by Prosernat. Backed by our strong investment and partnerships in R&D,
with IFP Energies nouvelles and Total, we are proud to bring to the market the
most advanced technologies in the following areas: Natural Gas Sweetening,
Natural Gas Dehydration, Sulfur Recovery, Hydrocarbon Dewpointing, Separation
and CO
2
Capture from Flue Gases.
TECHNOlOGIES aND EquIPmENT FOR OIl & GaS TREaTmENT
YOuR EFFICIENCY, OuR FlExIBIlTY
Polinter/ Venezuela, Maracabo, Modular units for production of butene-1
linde 2.indd 5 12/9/12 13:58:59
treatment stage and uses a combination of pure
oxygen and air to ensure optimal process perform-
ance (see Figure 3). The membrane replaces the
secondary clarifer used in conventional activated
sludge plants.
Against a conventional wastewater treatment solu-
tion, the innovative, patented process offers very
substantial cost savings across the industrial water
lifecycle. This includes much reduced capital invest-
ment, lower energy costs and superb fnal effuent
quality. When combined with reverse osmosis as a
tertiary treatment step to provide high-quality
softened water, Axenis makes wastewater reuse a
possibility for a large number of organisations. By
unlocking the resource they already have available,
operators have the potential to increase production
capacity in a more sustainable manner without
further depleting a fnite natural resource. All this
can often be achieved at much lower cost than
paying for the supply and disposal of process
water.
Advanced oxidation, ozone solutions
Oxygen can also be applied in other wastewater
treatment applications such as ozone water treat-
ment, where in combination with either hydrogen
peroxide (H
2
O
2
) or ultraviolet radiation (UV) it is
effective in partial or total oxidation of many non-
biologically treatable wastewater compounds, colour
removal and as a chlorine replacement for primary
disinfectant. Ozone (O
3
) is a molecule consisting of
three oxygen atoms and is created by passing
oxygen through ultraviolet light or a cold electri-
cal discharge. Under ambient conditions, ozone is
very unstable, readily giving up one atom of oxygen
within its short lifespan of usually less than 10
minutes. However, this process results in a power-
ful oxidising agent called a free radical, which is
toxic to most waterborne organisms. This property
makes ozone a very strong, broad spectrum disin-
fectant and biocide that will oxidise many organic
and inorganic substances. For this reason, it is used
widely throughout the world.
Its strong oxidising properties make ozone an
effective chemical for water treatment, but to use
ozone it must be created on-site and added to the
water by bubble contact. A major advantage of
ozone includes the production of fewer dangerous
by-products in comparison to chlorination, for
example.
Due to the growing concerns over chlorinated by-
product formation, the use of ozone in water-based
evaporative cooling towers is becoming an increas-
ingly attractive option, particularly within process
industries. It can also provide wider environmental
benefts in comparison to the more traditional chem-
ical treatment programmes.
Typically, chemicals such as chlorine and chelat-
ing agents are added to cooling tower water to
control microbiological growth and inhibit mineral
build-up. However, as the volume of water in a
DAMPF TRIX
5000
and 6000 a hot couple!
The new steam hose generation DAMPF TRIX
5000 and 6000

has been designed to meet the highest requirements on hose
material for the safe transport of steam and hot water in the
chemical and petrochemical industry, the building industry, in
gardening, for the heating of tank wagons, as well as for vari-
ous other industrial applications. The hoses have a black, non-
porous and smooth EPDM lining and an EPDM- cover, which is
resistant to ozone, weather UV and abrasion. Both hoses are
electrically conductive (R < 106) and meet all requirements as
per DIN EN ISO 6134.
DAMPF TRIX
5000:
reinforcements: high-strength, hydrolysis and temperature
resistant aramid
working pressure up to 6 bar
temperature range up to +120C (hot water), or
up to +164C (steam)
DAMPF TRIX
6000:
optimized EPDM inner layer material much longer
service life
reinforcements: 2 steel wire braids, excellent resistance
to corrosion
improved resistance against pop corning
working pressure up to 18 bar; burst pressure: min.180 bar
temperature range for saturated steam up to max. +210C,
short term up to +220C at 23 bar (saturated steam)
ContiTech Schlauch GmbH
Postfach 1120
D-34481 Korbach
industrial.hoses@uid.contitech.de
w w w. c o n tite c h . d e /a im
Conti Anzeige Dampf Trix 86x270 03-2012 90x270.indd 1 07.03.12 12:47
linde 2.indd 6 12/9/12 10:53:19
.)IMJJ,+)*+ ppp'îmj'\hf
state gain and some defnition of the
time-dependent path to steady state.
Generally, accuracy for the former is
prioritised over accuracy for the
latter, as (relative) steady-state gains
are used for both control and
economic optimisation, and can also
be used in validating the viability of
the selected control solution. In
addition, the tacit knowledge trans-
fer between process operations and
the control project team discussed
above is often more successful for
some general aspects of the dynamic
relationship (such as dead time and
frst-order time constant) than for
steady-state gain.
Initially, these tools were not inte-
grated with their input data
generation, and control practitioner
organisations would emphasise
their work practices that enable eff-
cient collection of this input data
with minimum explicit effort and
disruption to the process unit.
Nevertheless, in some circles, the
mindset was that identifcation was
a very costly process with long
elapsed times. And this is not to
mention the impact of changes on
controller design as a consequence
of a revised requirements defnition
or increased understanding of the
best solution to the control
problem.
Control product vendors
responded by integrating the data
generation step with the identifca-
tion step in real time. Indeed, these
tools have received signifcant write-
up in the technical press, including
many documented successes. The
frst generation of such tools
required the process that is being
identifed to be operated in open
loop while identifcation is
performed by stepping the process.
In practice, this leads to the selection
of small steps and the identifcation
on small sub-groups of independent
and dependent variables at a time,
in order to minimally interfere with
the process operational goals. As the
tools have evolved, the current state
of the art now allows what is termed
automated closed-loop stepping and
identifcation. This means that the
controller simultaneously controls
the plant during the testing and
identifcation process. This leads to
the largest reduction in effort for
principles), delayed coking and plat-
forming. The degree of technology
openness and toolkit productisation
varies. Commercialisation of such
toolkits represents a challenge, as
often a toolkit is focused on a partic-
ular process unit and there are only
so many installations of that process
unit worldwide. Where available
and maintained, these toolkits
remain a key component of the
control solution for certain proc-
esses. However, the focus and
narrative of control product software
vendors moved relatively quickly
towards data-centric modelling and
effciency tools that have broad
applicability to all control projects.
The earliest advanced control soft-
ware products included control
model identifcation tools. The tools
analyse plant data, and a key output
is statistically validated time-domain
relationships between independent
and dependent variables.
Relationships of independent to
dependent variables are generally
sought in terms of a linear steady-
In today's challenging economic climate, creative thinking, accurate
cost estimates and a commitment to technical excellence are crucial to
the success of capital projects. Mustang has a superior reputation for
providing owners with total project delivery, from feasibility analysis and
front-end engineering to facility startup.
Whether a project is grassroots, revamp, expansion or modernization,
Mustang has the experience and capability to successfully execute
your project to achieve the desired results.
Contact Mustang today!
Total Project Delivery
With a Reputation for Excellence
processplants@ mustangeng.com
www.mustangeng.com
honeywell.indd 2 1/6/12 14:27:49
cooling tower is reduced through
evaporation, the concentration of
water treatment chemicals and their
by-products contained within the
tower increases. To maintain chemi-
cal and contaminant concentrations
at prescribed levels, water is peri-
odically removed from the system
through a process called blowdown.
The blowdown wastewater needs
to be subsequently discharged to a
local wastewater treatment facility
or treated on-site to permit condi-
tions before it can be discharged. A
key beneft of ozone is that it dissi-
pates quickly and reduces the
overall chemical load found in the
discharged water, making it easier
to comply with regulations.
A further wastewater treatment
process is supercritical water oxida-
tion (SCWO). This process takes
advantage of the unique tempera-
ture and pressure properties
displayed by water when it is used
above its thermodynamic critical
point; that is, 374C (705F) and 220
bar (3210 psi). Under these condi-
tions, water develops unique
properties that can be applied to
completely oxidise a multitude of
complex biologically inert organic
compounds, inorganic complexes
and organic sludges.
Virtually all the organic content
of the wastewater is effectively
converted to CO
2
, water and salts
with almost no production of
carbon monoxide, NOx or SOx and
has signifcant environmental
compliance credentials.
SCWO is often categorised as
green chemistry or as a clean
technology, and the elevated pres-
sures and temperatures required for
SCWO are routinely and conven-
iently produced in industrial
applications, including petroleum
refning and process industries. It is
also possible to incorporate the
excess energy into power genera-
tion schemes providing a source of
waste to green energy on-site.
At the other end of the scale, low-
pressure oxidation (LoProx) is a
wet air oxidation application histor-
ically used to pretreat recalcitrant
high-strength industrial wastewater
prior to conventional aerobic bio-
treatment. The treated wastewater
is then blended with biodegradable
wastewater and treated through an
plant.
Conclusion
There are a myriad approaches to
dealing with industrial wastewaters
that have their own place on the
water technology map and there is
no single solution that fts all. The
desired level of treatment required,
available budget and nature of the
wastewater water, among other
factors, determine the elements that
make up the fnal wastewater treat-
ment process train.
Darren Gurney is a Senior Process Engineer
with Linde Gases Divisions global aquaculture
and water treatment business based in
Munich, Germany, and the UK. He holds a
masters in chemical engineering from the
University of Manchester, UK, and is a member
of the British Chartered Institute of Water and
Environmental Engineers.
linde 2.indd 7 12/9/12 10:53:29
VISIT OUR WEBSITE:
www.feminggulf.com
27 28 November 2012, Rio de Janeiro - Brazil
Mastering Basics & Maximizing Effciency
LatAm Plant Shutdown
& Turnaround Summit
Maximize profts by implementing latest
procedures in your company
Do not miss the opportunity to meet industry
experts and market leaders under one roof
Develop business leads
Operational Health & safety in Shutdown
activities
Understanding the Shutdown and Turnaround
lifecycle
Learn how to design an effcient Turnaround
team
How to eliminate and/or cope with catastrophic
failures
2 Exclusive Workshops & 4 Interactive Panel
Discussions
Find you reason to attend Key Event Highlights
LatAm Plant Shutdown & Turnaround Summit will help you learn from international & regional experts of
Oil and Gas refneries, Chemical plants, Bio energy production plants, Power plants, Automotive production
plants and Manufacturing units on the latest updates and developments in turnaround projects and procedures,
understand best practice strategies and learn about the latest technologies, solutions and equipment used to
enhance effciency.
Ajay Nimbalkar | Tel : +971 4609 1570 | E : ajay.nimbalkar@feminggulf.com
Register Today
Simultaneous on-site translations in
PORTUGUESE, SPANISH & ENGLISH
Register
NOW
ajay.indd 1 11/9/12 11:22:09
Optimising sweep gas fow in a fare
header
T
he purging of fare systems
prevents the ingress of air
through the open stack, which
can create an infammable mixture
of hydrocarbons and oxygen inside
the system. Stack purge, together
with purge reduction seals, is
generally used to prevent this, in
addition to a make-up gas for
sweeping the fare headers. Some
estimates of header sweep gas
quantities are suggested by Duggan,
Simpson and others.
1
With rising energy costs, there is
a need to minimise the amount of
gas used for sweeping without
compromising the safety aspect,
which is that there should be no
air ingress in the system. This arti-
cle develops the correlations for
estimating the header sweeping
gas requirement needed to
compensate for the shrinkage effect
due to ambient cooling or rapid
cooling of the gases after a hot gas
release. It also provides guidelines
to facilitate a decision on investing
in instrumented intermittent purge
systems, which can provide further
savings in the quantity of sweep
gas. The practice of using rule-of-
thumb fare header sweeping gas
rates could lead to substantial
annual utility costs in large fare
networks and is therefore not a
desirable practice.
A typical fare network consists
of several sub-headers that collect
the discharge gases from relief
valves, process faring control
valves, emergency depressurising
systems and equipment depressu-
rising for shutdown purposes. In
large complexes, the sub-headers
may terminate in unit knockout
drums for liquid removal before
How to calculate the minimum fow of fare header sweep gas needed to
maintain safe conditions
SATISH MATHUR
Bechtel India
joining the main header leading to
the common fare stack.
The fare stack can have a water
seal to prevent any fashback from
the stack to headers (except in cryo-
genic applications), a seal on the
stack to prevent infltration of air
and a dedicated continuous fare
stack purge. The quantity of stack
purge gas required is dependent on
the size of the fare tip, the compo-
sition of the purge gas, the
composition of the waste gases and
the design of the seal.
2
In most
cases, the seal purge gas quantity is
specifed by the seal supplier.
In addition, the unit area fare
headers and sub-headers are
provided with a sweeping gas
supply in most installations,
although some fare systems with a
water seal and stack designed for
internal explosion may not require
continuous header purge.
3
This is to
compensate for a shrinkage effect
due to ambient cooling or rapid
cooling of the gases after a hot gas
release in the fare header network.
This article deals with header
sweeping gas quantity, which is
independent of the stack purge fow
due to their different objectives, as
explained above. Natural gas avail-
able from the plant fuel gas system
is often used for purging, both at
the stack as well as in the sub-
headers, with a backup gas supply
source for ensuring uninterrupted
gas availability.
Basis
Under normal operation, when the
plant is in steady condition and if
the valves are not passing, there
should be no fow of waste gas into
the fare headers. In this condition,
the stagnant gases in the fare head-
ers can be subjected to volume
shrinkage due to cooling. The
factors infuencing shrinkage are:
Temperature of gas in the header
after stoppage of faring
Ambient temperature
Wind speed
Flare pipe surface area
Heat capacity of fare pipe metal
Thermal conductivity, viscosity
and density of gas.
Method
Heat transfer from the hot fare
gases to ambient is by the following
means:
Inside the pipe: by natural
convection (gases inside the pipe
are considered stagnant when not
faring)
Across the pipe wall: by
conduction
Outside the pipe: by natural
convection, forced convection and
radiation.
Theoretical analysis of natural
convection heat transfer inside
enclosed surfaces is provided in the
Nusselt equation:
4
N
Nu
= a (N
Gr
N
Pr
)
m
where
N
Nu
=h
i
L/k = Nusselt number
N
Gr
= L
3
2
Gt/
2
= Grashof
number
N
Pr
= c/k = Prandtl number.
The factors a and m for hori-
zontal cylinders are given in the
reference. Using these, h
i
, the heat
transfer coeffcient for natural
convection inside the pipe, is
computed.
When a heated fare pipe surface
is exposed to fowing air, the
betchel 2.indd 1 12/9/12 19:05:13
the temperature dropping sharply
to ambient at the air flm outside
the pipe (the thin profle shown in
Figure 1). The analysis provided
here assumes faring for the dura-
tion so that entire pipe length is
heated evenly. On stoppage of far-
ing, as the inside heat transfer
coeffcient reduces to even lower
than the outside (and becomes the
controlling heat transfer resistance),
the average pipe temperature
reaches closer to ambient tempera-
ture. The temperature gradients,
starting from the faring condition
to when the stagnant gas inside the
fare pipe fnally reaches ambient
temperature, are depicted in Figure
1. The bold profles are the time-
dependent temperature gradients
after the stoppage of faring.
The pipe temperature can be esti-
mated, based on the inside and
outside heat transfer coeffcients
and the respective bulk tempera-
tures, as follows:
6
T
pipe
= (T
gas
-T
ambient
) x (h
o
/(h
o
+ h
i
))
Differentiating above and consider-
ing the ratio h
o
/(h
o
+ h
i
) and T
ambient

as constant, we get:
T
pipe
/ T
gas
= h
o
/(h
o
+ h
i
)
The drop in gas enthalpy due to
heat loss can be computed as:
Q
gas
= m
gas
x c
pgas
x T
gas

The drop in pipe enthalpy due to
heat loss can be calculated as:
Q
pipe
= m
pipe
x c
ppipe
x T
pipe
Substituting for T
pipe
we get:
Q
pipe
= m
pipe
x c
ppipe
x T
gas
x h
o
/(h
o
+ h
i
)
Summing up the equations for
pipe and gas enthalpy changes and
equating to the total heat loss to
ambient, we get:
Q = Q
gas
+ Q
pipe
= m
gas
x c
pgas
x T
gas
+ m
pipe
x
c
ppipe
x T
gas
x h
o
/(h
o
+ h
i
)
Rearranging we get:
T
gas
= Q / (m
gas
x c
pgas
+ m
pipe
x c
ppipe
x h
o
/(h
o

+ h
i
))
convective heat transfer outside the
pipe is a combination of forced and
free convection. For this mixed
convection condition, Churchill
recommends the following equation
for computing the heat transfer
coeffcient h:
5
(N
u
- )
j
= (N
uf
- )
j
+ (N
un
- )
j
The forced convection Nusselt
number for the horizontal pipe of
diameter D is given by Incropera
and Dewitt
5
as follows:
Where R
eD
= VD/ = Reynolds
number.
The natural convection Nusselt
number for the horizontal pipe of
diameter D is given by Churchill
and Chu
5
as follows:
Where R
aD
= g...c
p
(t)D
3
/(.k
f
) =
Rayleigh number.
The overall Nusselt number N
u
is
computed using j = 4 and = 0.3.
Once the Nusselt number is
computed, the heat transfer coeff-
cient external to the pipe is
computed as:
h = N
uD
. k
f
/ D
The radiation heat transfer coeff-
cient outside the pipe is computed
using the Stefan Boltzmann Law:

with = 0.95
The overall outside heat transfer
coeffcient will be:

h
o
= h + h
r
Neglecting the heat transfer resist-
ance of the pipe and difference of
pipe inside and outside diameters,
the overall heat transfer coeffcient
U from gas to air is given by:
1/U = 1/h
i
+ 1/h
o
The heat transferred out of fare
gas and the pipe metal to ambient
can be computed as:
Q = U x A x (T
gas
- T
ambient
)
The heat transferred to the ambi-
ent state results in a lowering of the
temperature of the pipe and the gas
inside. Based on computations of
heat transfer coeffcients, it can be
seen that the external heat transfer
coeffcient (due to convection and
radiation from pipe to the ambient)
is higher as compared to the heat
transfer coeffcient inside the pipe
in the stagnant gas, with the differ-
ence being greater at higher ambient
air velocities and in rainstorm
conditions. However, during faring
(prior to the stagnant condition),
the gas velocity inside the pipe is
usually high and it can be assumed
that the inside pipe heat transfer
coeffcient is also high. The pipe
temperature in this period will be
closer to the gas temperature, with
(BT 1JQF "JS
F
N
J
5
Figure 1 Temperature gradients from faring condition to ambient temperature
]
|
|
|
|
|
|
|
|
|
|
|
\
[
j
)
+
|
|
|
+
+ =
5 / 4
8 / 5
4 / 1
3 / 2
3 / 1 2 / 1
,
282000
Re
1
Pr) / 4 . 0 ( 1
Pr Re 62 . 0
3 . 0
D D
D f
Nu
]
2
27 / 8
16 / 9
6 / 1
,
Pr) / 559 . 0 ( 1
387 . 0
06 . 0
|
|
|
|
|
|
|
|
|
|
|
+
+ =
D
D n
Ra
Nu
( )
0
4
0
4
T T
T T
r
h
s
s
]
8/27
]
1/4
betchel 2.indd 2 12/9/12 19:12:56
uhs.indd 1 12/9/12 11:05:47
At constant pressure, the volume
shrinkage is directly related to the
temperature fall computed above.
The sweeping gas compensates for
the fare gas shrinkage, thereby
ensuring the constant pressure
condition in the headers:
V
shrinkage
= V
fare header
x T
gas
/ T
gas-abs
Results
Calculations to estimate sweeping
gas were performed for a 500m
fare header for different pipe diam-
eters and gas temperatures during
faring. Wind velocity was consid-
ered as 2 m/s, the fuel gas price
was $3.0/thousand ft
3
and 10 mm-
thick steel pipe was used.
Calculations were performed for
1ft, 1.5ft, 2ft and 3ft-diameter fare
pipes based on ethane gas in the
fare pipe as well as ethane gas as
the sweeping gas. The results are
plotted in Figures 2 and 3.
Figure 2 shows continuous header
purging with a fuel gas quantity
needed to compensate shrinkage,
thereby avoiding any negative pres-
sure in the header. The annual cost
of sweeping gas is shown. As
Figure 2 illustrates, a higher sweep-
ing gas fow is required when the
initial fare gas temperature and
fare pipe diameter are high. If the
pipe length is more than 500m, the
sweeping gas requirement also
increases proportionately. The
required sweeping gas velocity
(volume of sweeping gas at the
sweeping gas temperature and fare
pressure, divided by fare pipe area)
for different pipe diameters for the
example case is shown in Figure 3.
Instrumented purge systems
The scheme indicated in Figure 4
is often used to provide sweep gas
to the fare header intermittently,
depending on pressure and
temperature measurements in the
header. Since faring is not
expected to be frequent, the use of
this scheme avoids continuous
header purge. However, the
scheme is more expensive than
installing a simple orifce plate
required for a continuous sweep-
ing gas fow system. If, for
example, the cost of a control valve
and other instrumentation shown
in Figure 4 is $20 000, then, refer-
ring to Figure 2, the instrumented
purge system becomes attractive
only above a 200C fare gas
temperature for a 2ft-diameter pipe
and above 110C for a 3ft-diameter
pipe in the above case, with a
payback period of one year. For
cases with a lower sweeping gas
requirement, the orifce plate
system for continuous sweeping
gas can be more economical.
The cost of sweep gas in intermit-
tent purge systems is expected to
be small, considering it is used for
a short duration, which is only after
a faring incident. For rainstorm
conditions, the sweeping gas
requirement can be marginally
higher (about 1% more is estimated)
Table 1
6 0 0 0 0
5 0 0 0 0
4 0 0 0 0
3 0 0 0 0
2 0 0 0 0
1 0 0 0 0
5 0 1 0 0 1 5 0 2 0 0 2 5 0 3 0 0
U
S

D
o
l
l
a
r
s

p
e
r

y
e
a
r
Fl are t emperat ure, C
0
1 ft
3 ft
1 . 5 ft
2 ft
Figure 2 Continuous header purging with a fuel gas quantity needed to compensate
shrinkage
0 . 0 3
0 . 0 4
0 . 0 2
0 . 0 1
5 0 1 0 0 1 5 0 2 0 0 2 5 0 3 0 0
S
w
e
e
p
i
n
g

g
a
s

v
e
l
o
c
i
t
y
,

m
/
s
Fl are t emperat ure, C
0
1 ft
3 ft
1 . 5 ft
2 ft
Figure 3 Sweeping gas velocity required for different pipe diameters
P I T I
F la re h e a d e r
Figure 4 Scheme to provide sweep gas to the fare header intermittently
betchel 2.indd 3 14/9/12 12:37:21
since the stagnant gases in the pipe
remain heat transfer controlling.
The estimation of fare gas
shrinkage shown above is based on
a horizontal layout of the fare pipe,
which is true for designs where the
fare header terminates into a water
seal drum, since the stack is purged
separately to meet the purge require-
ment of the seal near the fare tip.
Where there is no water seal drum,
the header extends to the vertical
section of the fare header, which is
the stack. The length of the vertical
pipe is to be added in this case;
however, there is not a major differ-
ence in the overall sweep gas
estimate if that height is treated as
equivalent horizontal run of pipe,
together with the rest of the fare
horizontal pipework.
Conclusion
Flare header sweeping gas fow is
dependent on gas shrinkage due to
the ambient cooling effect. The
minimum quantity to maintain safe
conditions in the headers can be
estimated according to the method
shown in this article. The equiva-
lent cooling rates (see Figure 5) for
different temperature fare gases for
the case described have been found
to be slightly lower than Simpsons
rates.
1
Sweeping gas fow systems can
be either continuous or intermittent.
Intermittent systems can provide
economy in gas consumption over
the continuous systems; their higher
capital cost becomes justifable in
large networks and/or high-
temperature fare systems.
Acknowledgements
The author wishes to thank Bechtel (India) Pvt
Ltd for their support in completing this study.
References
1 CCPS/AIChE Guideline for Pressure Relief and
Effuent Handling Systems, Ch 5.
2 Bader A, Baukal C E, Jr, Bussman W, Selecting
the Proper Flare Systems, CEP Jul 2011.
3 Pressure-relieving and Depressuring Systems,
API Standard 521 (ISO 23251), 5th ed.
4 Perrys Handbook, 7th ed, Ch 5, Heat and
Mass Transfer.
5 ASTM C680-10, Standard Practice for
Estimate of the Heat Gain or Loss and the
Surface Temperatures of Insulated Flat,
Cylindrical, and Spherical Systems by Use of
Computer Programs.
6 Kern, D Q, Process Heat Transfer.
Key
T, t Temperature
Pr Prandtl number
U Overall heat transfer coeffcient
D Diameter
V Volume
A Area
k Thermal conductivity
g Acceleration due to gravity
Volumetric coeffcient of thermal
expansion
Density
Effective surface emittance between
outside surface and the ambient surroundings
Stefan-Boltzmann constant
Kinematic viscosity

Satish Mathur is a Supervising Engineer in
the Process Department of Bechtel India Pvt
Ltd. He has over 36 years experience in the
oil, gas, petrochemicals and synthetic fbre
industry, from conceptual design through to
commissioning. He holds a BTech degree in
chemical engineering from Indian Institute of
Technology, New Delhi, India.
Email: sxmathur@bechtel.com
1 0 0
1 5 0
5 0
0 1 0 0 2 0 0 3 0 0
R
a
t
e

o
f

t
e
m
p
e
r
a
t
u
r
e

c
h
a
n
g
e
,

C
/
h
r
Temperat ure, C
0
1 ft
3 ft
1 . 5 ft
2 ft
Figure 5 Cooling rates for different temperature fare gases
ppp'îmj'\hf LhnkA^Zor+)*++*
Canadian bitumen blends in the 18-
20 API gravity range.
Heavy Venezuelan crudes have
been processed since the mid 1980s
in high blend percentages by only a
few US refners. In nearly all cases,
the CDU/VDUs did not meet
design capacity, product yields or
reliability. In some instances, none
of these goals were achieved, with
the units requiring major revamps
just to meet their original design
basis. Today, Venezuelan crudes
are increasingly going to India and
the Far East because they are seen
as low-cost alternatives to Middle
East and West African crudes. In
most instances, the lessons learned
in the US have not been incorpo-
rated in the CDU/VDU designs.
Few major engineering and
construction companies have
designed units for Orinoco bitumen
blends, and the ones that have often
repeat the same mistakes on future
designs because they rarely perform
rigorous post start-up audits and
almost never look at unit perform-
ance one to four years after start-up
when reliability problems show up.
Conc|usions
Lessons learned by US refners
processing heavy Venezuelan
crudes and Canadian upgraders
processing bitumen should be
applied to new projects that will
process increasing amounts of
opportunity crudes. These units
should not be designed with
conventional practices used for
light and medium crudes; other-
wise, run length will be short,
product yields and quality poor,
and maintenance costs high.
Opportunity crude properties are
very different and design require-
ments will raise initial costs. These
higher initial costs will pay out over
the life of the project through an
increased VGO product yield, lower
maintenance costs, fewer VGO
contaminants and a longer run
length between decokings.
Scott Go|den bl Z <a^fb\Ze >g`bg^^k pbma
Ikh\^ll <hglnembg` L^kob\^l% Bg\% bg Ahnlmhg%
M^qZl' A^ li^\bZebl^l bg _khgm&^g] ikh\^ll
^g`bg^^kbg` _hk k^g^kr ngbm kôZfil Zg] aZl
Znmahk^] fhk^ maZg *)) m^\agb\Ze iZi^kl
hg kôZfibg` Zg] mkhn[e^lahhmbg` k^g^kr
ikh\^llngbml'>fZbe3l`he]^g9kôZfil'\hf
1ony 8ar|etta bl Z <a^fb\Ze >g`bg^^k pbma
Ikh\^ll <hglnembg` L^kob\^l% Bg\'% bg Ahnlmhg%
M^qZl' Abl ikbfZkr k^lihglb[bebmb^l Zk^
\hg\îmnZe ikh\^ll ]^lb`g Zg] ikh\^ll ]^lb`g
iZ\dZ`^l_hkeZk`^\ZibmZekôZfil'
>fZbe3m[Zke^mmZ9kôZfil'\hf
Steve white bl Z <a^fb\Ze >g`bg^^k pbma
Ikh\^ll <hglnembg` L^kob\^l% Bg\% bg Ahnlmhg%
M^qZl' A^ aZl fhk^ maZg ,) r^Zkl h_ ikh\^ll
]^lb`g ^qi^kb^g\^ _hk k^g^kr kôZfil Zg]
`kZllkhhmlngbml'>fZbe3lpabm^9kôZfil'\hf
Iigure 3 Lô^k^pZlashg^\hdbg`
eptq.com
1echno|ogy Q&A
1echnica| writing
Sty|e Guide
Digita| Lditions
Lvents Listings
Iind them a|| at
ppp'îmj'\hf
betchel 2.indd 4 12/9/12 19:05:50
PLUS WORKSHOP:
How to design and sulphur forming and handling
facility*
*Pl ease not e t hi s i s a separat el y bookabl e event
PLUS ROUNDTABLE:
Sulphuric Acid Coolers - design parameters,
maintenance procedures, NDT, water treatment
and tube failures
8audi Aramco Razi Petrochemical Co. Amine Experts
Axens Cansolv Technologies lnc. (8hell} Enersul
EURECAT U8 lnc. FBN Hudson ltaliana 8pA
Foster wheeler linde Engineering
NEC8, lnc./DuPont 8ustainaole 8olutions
N8 8ilveroack Consultanc] Perma-Pipe, lnc
Rameshni & Associates Technolog] & Engineering llC
8andvik Process 8]stems 8hell 0looal 8olutions ov
8iirtec higi 8pA 8ulphur Experts
T]co Thermal Controls worle]Parsons
Atlantic Copper 8lU Ba]er Technolog] 8ervices 0moH
FlEXlN 0moH 0E Power & water
lndustria 8ulfrica 8A - lsusa
langeloth Netallurgical Co. llC Nol]met 0roup
havdeep Environmental and Technical 8ervices Pvt. ltd
h0RAN Engineering & Constructors ltd 0utotec
0uadrant EPP A0 Razi Petrochemical Co., lran
8iemens Turoomachiner] Equipment
8hC-lavalin lnc. Fenco Division
8outhern Peru Copper Corporation 8d-Chemie A0
8ulzer Chemtech ltd Tenova Ninerals (Pt]} ltd.
28- 31 October 2012, Estrel, Berlin, Germany
S u lp h u r | S u lp h u ric Acid | S u lp h u r D ioxid e
SULPHUR
International onference E xhibition
Gold Sponsors: Silver Sponsors: Bronze Sponsor:
The worl ds l eadi ng sul phur and sul phuri c aci d annual event
Regi st er t oday onl i ne at www.sul phur conf er ence.com
More informat ion on technical programme and abst ract s www.sulphurconference.com
50% DISCOUNT avai l abl e t o Techni cal st af f f rom Operat i ng/ Product i on compani es
Media Partners: Offcial Publications:
Neet and learn of the current challenges, proolems and
targets operating companies are encountering
hetwork with the technical experts from engineering,
technolog] and equipment companies
listen to papers offering the latest innovations and
operations solutions
Hear of the market trends affecting future projects and
highlighting new regions of development
Visit largest exhioition dedicated to sulphur and sulphuric
acid industries
Sul phur and Sul phuri c Aci d Expert s Gat her i n Berl i n!
5
7
0
+

d
e
le
g
a
te
s

a
tte
n
d
e
d
in

2
0
1
1
6
0
+

e
x
h
io
ito
rs
SULPHUR FOCUS: sulphur olock, sulphur plant operation,
sulphur recover], degassing and sulphur handing, and sulphur
tail gas from.
SULPHURIC ACID TECHINICAL FOCUS: 8ulphuric acid control,
equipment, catal]sts, operational experiences from metallurgical
plants, sulphuric acid tail gas, circuit concerns , sulphuric acid tail
gas and sulphur dioxide from.
ptq
PETROLEUM TECHNOLOGY QUARTERLY
www.energyglobal.com
EnergyGlobal
sulphur.indd 1 12/9/12 19:17:46
Steam reformer for Tyumen
refnery
Haldor Topse has selected UOP
technology to purify hydrogen from
a steam reforming unit to be
installed at the Antipinsky refnery
in Tyumen, Russia. The Polybed
PSA system will recover and purify
hydrogen to help the refnery
produce cleaner transportation
fuels.
The new hydrogen unit is sched-
uled to start up in 2013 and is part
of the refnerys plan to increase its
capacity of crude oil processing by
up to 7 million t/y. It will also
enable the production of fuel prod-
ucts that meet the European
Unions Euro-5 emissions standards
aimed at reducing emissions from
light-duty vehicles.
The Polybed PSA system is a
skid-mounted, modular unit
complete with vessels, valve skid,
adsorbents, control systems and
embedded process technology. The
process uses UOP adsorbents to
adsorb impurities at high pressure
from hydrogen-containing waste
streams and subsequently reject
them at low pressure. The system
allows hydrogen to be recovered
and upgraded to more than 99.9%
purity to meet downstream process-
ing requirements.
In addition to recovering and
purifying hydrogen from steam
reformers, Polybed PSA systems
can be used to produce hydrogen
from other sources, including refn-
ery off-gases, ethylene off-gas,
methanol off-gas and partial
oxidation/syngas.
Biomass process for
automotive fuel
France-based researcher IFP
Energies nouvelles (IFPEN) and
Dynamotive of Canada are joining
forces with refnery technology
provider Axens in the development
and commercialisation of a process
to produce transportation fuels
from pyrolysis oil derived from
agricultural waste.
P
According to the developers, analy-
sis and verifcation of economic and
technical parameters have concluded
that the process has competitive
advantages when compared to exist-
ing processes and competing
technologies. Laboratory-scale units
were developed and operated in
Canada and France, where
Dynamotives bio-oil was upgraded
to synthetic hydrocarbons.
The agreements mark the
completion of Dynamotives invest-
ment programme in the process as
IFPEN commits to develop the full
bio-oil upgrading technology.
Dynamotive will provide pyrolysis
oil to IFPEN for the development
programme. Axens will fnalise the
development of the upgrading
technology, and will be in charge
of its industrialisation and
commercialisation.
Feedstocks from which Dyna-
motive has produced its bio-oils
include sugar cane bagasse, bark-
rich forest trimmings, softwoods
and palm fronds. The upgraded
bio-oils were of widely varying
quality and in some instances had
been in storage for several years.
Dynamotives pyrolysis process
converts roughly 85% of the total
biomass feed into solid and liquid
fuels. The balance is utilised to
provide energy to the process.
Yields of diesel/gasoline from
bio-oil have been developed in the
laboratory and, according to
Dynamotive, have been confrmed
as the highest available.
JV takes on Venezuelan
revamp
Venezuelan parastatal PDVSA
Petrleo has selected a consortium
of Korean Hyundai Engineering
and Wison Engineering, Chinas
largest private sector EPC company,
for a $3 billion contract at its Puerto
La Cruz refnery. The Hyundai-
Wison consortium will carry out
engineering, procurement, construc-
tion and start-up assistance of the
environmental and auxiliary units,
Industry News
and a revamp of the atmospheric
distillation units of the Deep
Conversion Project at Puerto La
Cruz, which is located about 250
km east of Caracas. The project will
upgrade refnery facilities to process
heavy crude oil, with a capacity of
210 000 b/d. In addition, the
contract calls for the expansion of
gasoline, diesel and aviation kero-
sene projects and other facilities.
The project is expected to be
completed within 40 months.
Stover to ethanol in Iowa
KBR has been awarded a contract
for engineering and procurement
services for DuPonts frst cellulosic
ethanol plant in Nevada, Iowa. It
will provide front-end engineering,
detailed engineering and procure-
ment services to DuPonts Industrial
Biosciences Group for the plant to
be constructed in the Midwest US.
The ethanol plant is designed to
process 1300 t/d of corn stover and
produce 27.5 million gal/y of etha-
nol. The cellulosic ethanol product
will be used as a blending compo-
nent in gasoline by fuel
manufacturers and will enable them
to meet EPA requirements for etha-
nol derived from cellulosic sources.
The project is due for completion
by the end of 2013.
Petronas expands petrochems
Petronas has awarded CB&I a
contract for the licence and engi-
neering design work of fve
petrochemicals units. The project is
part of the new refnery and petro-
chemicals integrated development
(RAPID) project to be located in
Johor, Malaysia.
Lummus Technology will be
providing technology for a steam
cracker complex comprising ethyl-
ene, butadiene, benzene, isobutylene
and MTBE units.
Hydrogen production for
Gazprom
A subsidiary of Foster Wheeler has
been awarded a contract by JSC
ind news copy 13.indd 1 12/9/12 10:55:17
Gazprom Neftekhim Salavat for the
engineering and material supply of
a new double-cell terrace wall
steam reformer heater and air
preheating system to be built at the
Salavat refnery in the Republic of
Bashkortostan, Russia.
The steam reformer will be part
of a hydrogen production unit
being built by Gazprom Neftekhim
Salavat. The plant is designed to
produce up to 27 500 Nm
3
/h of
hydrogen and is based on Foster
Wheelers hydrogen technology
and process design package. The
work is expected to be completed
in November 2013.
The terrace wall reformer has a
specifc fring arrangement and a
sloped-wall radiant section design
that enable long catalyst tube life. It
can operate with ultra-low-NOx
burners to meet all current environ-
mental emissions standards.
BTX boost for Reliance
GTC Technology is licensing its
GT-BTX Extractive Distillation
process for the production of high-
purity aromatics at Reliance
Industries Jamnagar refning and
petrochemical complex in Gujarat,
India. The licence is part of a
consortium with CB&I Lummus
Technology to supply technology
for a multi-unit complex for
benzene and paraxylene produc-
tion. The new facility will be the
worlds largest for the production
of paraxylene.
Fujairah terminal begins
phase 2
Socar Aurora Fujairah Terminal
FZC (SAFT), the joint-venture
company between Azerbaijans
Socar, Aurora Progress, the Swiss-
based commodity trading house,
and the government of Fujairah,
has started the construction of the
second phase of the 645 000 m
3
oil
products terminal in Fujairah, UAE.
This phase adds 235 000 m
3
to the
tanks already in operation.
The second phase will involve 10
new tanks. Fujairah National
Construction & Transportation is
doing all the civil works on the site,
Belleli Energy is constructing the
tanks and pipelines, and Consilium
Middle East is providing the
electrical, instrumentation and fre
fghting equipment.
According to the joint venture,
the frst phase of the terminal was
completed as planned and received
the frst oil products into the tanks
in March. The next phase will be
fully operational by Q4 2013.
The terminal at present consists
of 645 000 m
3
of storage across 22
tanks and can handle fuel oils,
gasolines, naphthas and middle
distillates, and will be equipped
with both tank-to-tank and in-line
blending facilities. The Port of
Fujairah is providing a new multi-
berth facility for the receipt and
loading of oil tankers. The terminal
is being constructed in three
phases.
Yokogawa, SICK widen
agreement
Yokogawa Electric Corporation and
SICK have signed an agreement
whereby Yokogawa will distribute
and service SICK Process
Analytical instrumentation world-
wide. Three years ago, both
companies announced a distribu-
tion agreement for the European
process analytical markets, as the
product portfolios of both compa-
nies are, to a large extent,
complementary. With this new
agreement, both companies say
that they confrm this strategy and
widen it to a global scale.
Refning industries thrive in
developing regions
While refneries across the US and
Europe are forced to upgrade to
meet increasingly stringent environ-
mental norms, the future refning
industries in the Middle East and
Asia are set to expand sharply,
according to a report by GBI
Research.
The report (Refning Industry to
2017 - Asia-Pacifc, Middle East and
Africa to Emerge as Key Regions for
Infrastructure Investments for
Capacity Additions) says that while
the crude oil refning industry in
developed countries is struggling
to cope with slack demand for
refned products, stringent envi-
ronmental regulations and higher
investment needs, expansions
in developing countries are
being driven by a huge demand
for oil.
The refning capacity growth in
the Asia-Pacifc region will be
driven by China, India, Indonesia
and Malaysia. China and India, in
particular, are expected to expand
their refning capabilities to cater
to their domestic demand, and
intend to turn themselves into
major regional refning hubs.
Indonesia is instead rapidly
expanding its refning capacity to
target the export market. A total
of around 198.1 MMtpa of addi-
tional capacity will be added
through new refneries during
2012-2017.
The Middle East also represents a
major producer and exporter of
crude oil, and is now targeting the
petroleum products export market.
Countries in the region are working
on modernising their existing refn-
eries as well as implementing
capacity expansions. Signifcant
capacity additions are expected to
come online during the next fve
years, with industry growth seen
most prominently in Iran, Iraq and
Saudi Arabia.
Newly planned refneries in the
Middle East are targeting Western
countries as their major export
markets, and they hold the capabil-
ity to process heavy crude and
produce quality products in accor-
dance with the environmental
regulations upheld in developed
countries. Subsequently, the refning
industry, in the US particularly, is
due to face a new wave of competi-
tion. The recession-hit refning
industry in the US has already
witnessed declining margins due to
idling and surplus capacity, and the
refnery infrastructure improve-
ments planned within the Middle
East will lead to surplus capacity
and a decline in the refning
margins of North American
refneries.
Global refning capacity is
expected to grow at a steady rate,
from 4661.7 million metric tons per
annum (MMtpa) in 2011 to 5516.9
MMtpa in 2017, at an average
annual growth rate (AAGR) of
3.1%. Middle Eastern countries are
expected to provide a capacity of
183.7 MMtpa by 2017.
ind news copy 13.indd 2 12/9/12 10:55:29
Grand Ameri ca Hot el
Sal t Lake Ci t y, UT
Oct ober 1-3, 2012
AFPMs Q&A and Technology
Forum is the industr ys most
comprehensive meeting on
refner y operations and process
technology. This event assembles
representatives from throughout
the industr y including refners,
petrochemical producers, catalyst
and chemicals suppliers, plant
automation specialists and
engineering frms and gives
them the oppor tunit y to address
real problems refners face today.
PROVIDING
SOLUTIONS
THe 2012 AFPM (NPRA)
Q&A AND TeCHNOLOGy FORUM
At tendees will gain from the
chance to discuss their challenges
and fnd solutions from the exper t
Q&A panelists and other at tendees.
Todays competitive refning
environment requires at tention to
plant safet y, superior technology,
innovation and excellence in
operations. Beneft from the shared
knowledge of exper ts throughout
the industr y the Q&A and
Technology Forum will keep you
moving in the right direction as you
optimize your plant s operation.
Register today at www.afpm.org.
af pm.or g
afpm.indd 1 11/9/12 11:26:49
4th Unconventional Hydrocarbons Summit 137
ABB Consulting 37
Advanced Refning Technologies 120
Aerzener 45
AFPM Q&A 143
Aggreko 117
Air Liquide Global E&C Solutions 55
AMACS Process Tower Internals 17 &18
Aveva 20
Axens OBC
BASF Corporation, Catalysts Division 64
Bryan Research & Engineering Inc 76
Burckhardt Compression 41
CB&I 9
Chevron Lummus Global 7
ContiTech 133
Criterion Catalyst & Technologies 4
Curtiss-Wright Flow Control Corporation 90
DuPont Sustainable Solutions 85
Elliott Group 118
Emerson Process Management 42
Enersul 67
eptq.com 139
Eurecat France SAS 28
Executive Summit 2013 IBC
ExxonMobil Research and Engineering Company 31
Foster Wheeler 72
Grabner Instruments 6
Haldor Topse IFC
Hoerbiger Kompressortechnik Holding 103
Honeywell Process Solutions 60
Hydrocarbon Publishing Company 23
ITW 46 & 128
Johnson Matthey Catalysts 122
Johnson Screens 24
KBC Advanced Technologies 2
Kobelco Machinery Europe 35
Koch-Glitsch 48
LatAm Plant Shutdown & Turnaround Summit 134
Linde 50
Merichem 56 & 59
Metso Automation 110
Mustang Engineering 132
Nalco Downstream Energy Services 98
OHL Gutermuth Industrial Valves 79
Onis International 71
Optimised Gas Treating 81
Porocel 82
Process Consulting Services 14 & 16
Prognost Systems 38
Prosernat 131
Rentech Boiler Systems 114
Sabin Metal Corporation 93
Samson 11
Schneider Electric 63
Selas Fluid Processing Corporation 68
Spectro Analytical Instruments 22
Sulphur 2012 140
Sulzer Chemtech 27
Swagelok Company 104
ThyssenKrupp Uhde 94
Tricat Catalyst Service 113
UOP 32
Vereinigte Fllkrper-Fabriken 109
Weir Minerals - Lewis Pumps 126
Yokogawa Europe 13
Zeeco 125
Zyme-Flow 101
Alphabetical list of advertisers
ad index copy 14.indd 1 13/9/12 11:37:49
Join the Top Executives
Shaping the Industry
of This Key Region
Russia & CIS
Executive Summit
Downstream Oil & Gas
21 - 23 FEBRUARY 2013
MADINAT JUMEIRAH DUBAI
WWW.EUROPETRO.COM I MOSCOW@EUROPETRO.COM I + 359 2 427 22 05
epc.indd 1 11/9/12 12:04:50
The winning catalyst combination
for your hydrocracker
Single source lSC 9001 lSC 14001 CHSAS 18001
www.axens.net
HRK, HDK, HK series: the expert trio to maximize
cycle length, activity and middle distillates selectivity
Hydro_cats_ad_fin.indd 1 26/7/11 12:35:06
axens.indd 1 8/3/12 10:17:04

2012 PTQ q4

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

2012 PTQ q4

Hochgeladen von

Copyright:

Verfügbare Formate

special features

Gas processinG developments

25 Challenges and opportunities of 10 ppm sulphur gasoline: part 2

contents/ed com copy 16.indt 1 13/9/12 14:20:46

valve has no moving parts and a wide operating range. Its

tray by Fractionation Research Inc. confrmed that it delivers the

kochglitsch.indd 1 11/9/12 11:52:24

PK S O rion provides the only truly integrated safety and

Equilibrium composition and

x: CO conversion in the high

(suite of chemistries and predictive

inert ceramic balls

Tricat GmbH Catalyst Service Bitterfeld

5000 and 6000

Das könnte Ihnen auch gefallen