ptq

Q4 2014

PETROLEUM TECHNOLOGY QUARTERLY

REFINING
GAS PROCESSING
PETROCHEMICALS

SPECIAL FEATURES
GAS PROCESSING DEVELOPMENTS
MASS TRANSFER

cover and spine copy 22.indd 1

15/09/2014 10:32

Please visit us at

10-13 NOVEMBER

1238_e

Hall 11, Booth 11350

Does your raw natural gas contain hydrogen sulfide,

carbon dioxide, mercaptans or more?
Whatever the impurity, whatever the composition, Air Liquide Global E&C Solutions
has the right treatment.
The composition of natural gas varies
tremendously: almost every source
contains a different blend of impurities.
The options for treatment are almost
as diverse. Thats why offering a solution
specifically designed for your gas field

is crucial. We as your partner of choice

provide solutions for all types of natural gas,
including associated and unconventional
gas, from a single source. Customised
and efficient.

www.engineering-solutions.airliquide.com

air
liquide.indd 1
AIR_2512_018_AZ-Hydrocarbon-Engineering_08-2014_210x297mm_RZ.indd
1

10/09/2014
11:25
05.06.14 09:52

ptq
PETROLEUM TECHNOLOGY QUARTERLY

Q4 (Oct, Nov, Dec) 2014

www.eptq.com

3 Fencing with rules

Chris Cunningham
5 ptq&a
21 Tail gas catalyst performance: part 2
Michael Huffmaster Consultant
Fernando Maldonado Criterion Catalysts
37 Predicting future FCC operations via analytics
Patrick J Christensen, Touseef Habib, Thomas B Garrett and Thomas W Yeung
Hydrocarbon Publishing Company
47 Predicting reactive heavy oil process operation
Glen A Hay, Herbert Loria and Marco A Satyro Virtual Materials Group, Inc
Hideki Nagata Fuji Oil Company Ltd
55 Integrated hydrogen management
Saa Polovina, Danijela Harmina and Ana Granic arac INA Rijeka refinery
69 Structured packing in a CO2 absorber
Ralph Weiland and Nathan Hatcher Optimized Gas Treating, Inc.
75 Reflux in a gas dehydration plant
Sajad Mirian and Hossein Anisi Nitel Pars Co (Fateh Group)
Xiang Yu Hengye Chemical Co
Sepehr Sadighi Research Institute of Petroleum Industry
81 Override control of fuel gas
Rainer Scheuring Cologne University of Applied Sciences
Albrecht Minges and Simon Griesbaum MiRO Mineraloelraffi nerie Oberrhein
Michael Brodkorb Honeywell Process Solutions
87 Snakes and ladders for maximising propylene
Bart De Graaf, Mehdi Allahverdi, Martin Evans and Paul Diddams
Johnson Matthey Process Technologies
97 Troubleshooting a C3 splitter tower. Part 1: evaluation
Henry Z Kister Fluor
Brian Clancy-Jundt and Randy Miller PetroLogistics
105 Overcoming tight emulsion problems
Hernando Salgado Cartagena Refinery, Ecopetrol
Luis Mario Ramgus S. A. Pall Corp.
Rosngela Pacheco Barrancabermeja Refinery, Ecopetrol
111 Enhancing bottoms cracking and process flexibility
Yee-Young Cher, Rosann Schiller and Jeff Koebel Grace Catalysts Technologies
119 Preventing emissions in coke removal
Artur Krueger, Bernd Lankers and Josef Wadle TriPlan AG
125 Troubleshooting steam ejectors
Norman Lieberman Process Improvement Engineering
131 Asphalt quality prediction and control
Zak Friedman Petrocontrol
137 Technology in Action
Marathon Petroleums Detroit, Michigan refinery.

Photo: Marathon Petroleum Corporation

2014. 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 and its supplements the publisher cannot be held
responsible for any statements, opinions or views or for any inaccuracies.

ed com copy 6.indd 1

13/09/2014 11:38

KBC Adv - PTQ Q4 2014.pdf 1 9/3/2014 8:40:35 AM

DEDICATED TO MAKING OUR CLIENTS IN THE HYDROCARBON INDUSTRY WORLD-CLASS BY DRIVING A NEW
STANDARD IN OPERATIONAL EXCELLENCE THROUGH EXPERT CONSULTING AND ADVANCED TECHNOLOGY.

CM

Optimisation and Profit

Improvement from pore to pump
from a deep understanding of the
chemical engineering of assets.

MY

CY

CMY

PVT, thermal hydraulics and process facility

simulation from well bore through refining and
petrochemicals in an open platform, which is
enabled for workflows across the assets.

Sustainable operational excellence from

strategic and organisational consulting,
focused on people and processes that make
the assets work hard.

KBC uses its 35+ years of history working in hundreds of oil and gas facilities worldwide to bring
practical, sustainable solutions to its clients to improve their bottom line performance by using
proven best practices and unit optimisation tools and techniques.

KBC Advanced Technologies

AMER: +1 281 293 8200 ASIA: +65 6735 5488 EMEA: +44 1932 242424

answers@kbcat.com

kbc.indd 1

www.kbcat.com

blog.kbcat.com

10/09/2014 11:50

p
T tq

he European Union has arguably

been the global leader in biodiesel
production and use, with overall
biodiesel production increasing from 1.9
P
ETROLEtonnes
UM TECin
HN2004
OLOGto
Y Qnearly
UARTE10.3
RLY million
million
tonnes in 2007. Biodiesel production in the
US has also increased dramatically in the
past
fewNo
years
Vol 19
5 from 2 million gallons in
2000 to approximately 450 million gallons
Q4 (Oct, Nov, Dec) 2014
in 2007. According to the National Biodiesel
Board, 171 companies own biodiesel
manufacturing
plants and are actively
Editor
marketing
biodiesel.1. The global biodiesel
Chris Cunningham
editor@petroleumtechnology.com
market
is estimated to reach 37 billion
gallons by 2016, with an average annual
Production
growth
rateEditor
of 42%. Europe will continue to
Rachel Storry
be the major biodiesel market for the next
production@petroleumtechnology.com
decade, followed closely by the US market.

Although
Graphics
Editor high energy prices,
increasing
global demand, drought
Rob Fris
graphics@petroleumtechnology.com
and
other factors are the primary
drivers for higher food prices, food
Editorial
competitive
feedstocks have long
tel +44 844 5888 773
been and will continue to be a major
fax +44 844 5888 667
concern for the development of biofuels.
To Development
compete, Director
the industry has
Business
responded
Paul Mason by developing methods to
sales@petroleumtechnology.com
increase
process efficiency, utilise or
upgrade by-products and operate
Advertising
Sales Office
with
lower
quality lipids as
tel +44 844 5888 771
feedstocks.
fax +44 844 5888 662

Feedstocks
Publisher
Biodiesel
Nic Allen refers to a diesel-equivalent
publisher@petroleumtechnology.com
fuel
consisting of short-chain alkyl

(methyl or ethyl) esters, made by the

Circulation
transesterification
of triglycerides,
Jacki Watts
commonly known as vegetable oils or
Louise Shaw
animal
fats. The most common form
circulation@petroleumtechnology.com
uses methanol, the cheapest alcohol
available,
to Publishing
produce Ltd
methyl esters.
Crambeth Allen
Hopesay,
Craven Arms
8HD, UKare priThe
molecules
in SY7
biodiesel
tel +44 844
5888 776
marily
fatty
acid methyl esters
fax
+44
844
5888 667 created by trans(FAME), usually
esterification between fats and methanol. Currently, biodiesel is produced
PTQ (Petroleum
Technology
Quarterly)
(ISSNplant oils.
from
various
vegetable
and
No: 1632-363X, USPS No: 014-781) is published
First-generation
food-based
feedstocks
quarterly plus annual Catalysis edition by Crambeth
Allen Publishing
Ltd and
is distributed in
the USsuch as
are
straight
vegetable
oils
by SP/Asendia, 17B South Middlesex Avenue,
soybean
oil Periodicals
and animal
fatsat New
such as
Monroe NJ 08831.
postage paid
Brunswick,lard,
NJ. Postmaster:
send
address changes
to
tallow,
yellow
grease,
chicken
fat
PTQ (Petroleum Technology Quarterly), 17B South
and
theAvenue,
by-products
of the production
Middlesex
Monroe NJ 08831.
BackOmega-3
numbers available
from acids
the Publisher
of
fatty
from fish oil.
at $30 per copy inc postage.
Soybean oil and rapeseeds oil are the
common source for biodiesel production in the US and Europe in quantities that can produce enough biodiesel to be used in a commercial market
with
currently
applicable

Fencing with rules

nvironmental regulation is the mother of invention in refining technology.

The impact of successive clean fuels measures, originating in Europe and
the US, continues to reverberate at refinery sites around the world as the
technology developers output works to continually upgrade the refining
industrys impact on air quality, directly or via its products. With poorer
quality crudes entering the mainstream of raw material flows into refineries,
the pressure for further and better invention does not ease up.
This all, of course, means additional expenditure for the refiners themselves. It is no surprise when they demand a long, hard look at any proposal
to require that cleaner products or cleaner emissions leave a refinery.
The latest significant step along the road to regulation in the US was due
late in October when the call for evidence was scheduled to close on the
Environmental Protection Agencys (EPA) proposed extension of emissions
limits from refineries. Introduced in May, the proposed rule arises from the
EPAs risk and technology review of two existing emissions standards for
refineries. This extension of the rules calls for additional emission control
requirements for storage tanks, flares and coking units at petroleum refineries, as well as a requirement with refiners neighbours in mind to monitor
air concentrations at the fenceline to check for accidental emissions.
According to the EPA, implementing the rule will result in a reduction in
the US of 5600 t/y of air pollutants, and 52 000 t/y of volatile organic compounds (VOC). More specifically, the EPA wants to amend the operating
requirements for refinery flares to ensure a high level of combustion efficiency for waste gases that go to flare. The agency also wants controls on
delayed coking units with the aim of preventing exemptions to emission limits during start-up and shutdown. It expects that the measures will lead to a
reduction in emissions of BTX of 1800 t/y and VOC emissions would fall by
19 000 t/y. The additional flaring measures would result in significant reductions in particulate emissions and VOCs.
Upgrading and tweaking the technologies to achieve all of this would also,
says the EPA, lead to 700 000 t/y less carbon dioxide entering the atmosphere. The all-in cost of implementing the rule would be $240 million, with
annual running costs of $40 million, with resulting negligible impact on
product costs.
Refiners are not so sure about the EPAs latest certainties. The American
Fuel & Petrochemical Manufacturers (AFPM), the US refiners trade body
and voice in Washington, takes the view that further tightening of the emissions rules is not cost effective. The AFPM said earlier in the year that it had
evaluated the EPAs analysis of risk on the basis of emissions data provided
to the agency. That analysis, according to the AFPM, showed that risk levels
were not appreciably higher than they were when the rule was applied
first time around, pointing out that the EPA determined at that time that further action was unnecessary.
The AFPM went on to say that the level of concern for risks expressed in
the original rule does not justify the further actions that the EPA wants to put
forward. Fenceline monitors, for instance, are not justified by the latest risk
analysis. This critique signed off by saying that the EPAs latest proposed
steps are neither cost effective, nor do they provide substantial health benefits.
CHRIS CUNNINGHAM
PTQ Q4 2014

ed com copy 6.indd 2

15/09/2014 11:14

Advanced R & D means results,

like our newest tail gas catalyst: C-834.

Meet Karl Krueger:

Research Scientist. Tail Gas Catalyst Expert.
Activity, pressure drop, cost... According to Karl Krueger, Criterion research scientist, these among
other factors are critical considerations when selecting a tail gas catalyst. He should know. Karl and
his colleagues have just designed C-834, specifically to provide exceptionally high activity in low
temperature operations while continuing to help refiners realize lower operating costs and extended
cycle lengths. This breakthrough catalyst joins the ranks of Criterions range of advanced tail gas
treating catalysts including C-234, C-534 and C-734, which account for the majority of the worlds
installed capacity.
Leading minds. Advanced technologies.

www.CRITERIONCatalysts.com

criterion.indd 1

10/09/2014 11:40

ptq&a
Q

Suspended catalyst in our FCC fractionator is

accumulating in the bottom fraction. What is the most
effective way of clarifying this slurry oil?

Zhe Cui, FCC Technologist, zhe.cui@shell.com,

Bob Ludolph, Principal Technologist - Catalytic Cracking,
robert.ludolph@shell.com and Kevin Kunz, Team Lead, FCC
Design & Licensing, Kevin.kunz@shell.com, Shell Global Solutions

The suspended catalyst in the FCC fractionator bottom

fraction (slurry oil) is uncaptured catalyst leaving with
the reactor overhead vapour (feed for main fractionator). The slurry yield is determined by the FCC units
riser/reactor design, feed and catalyst properties, reactor conditions, desuperheating section temperatures,
and by the design and operation of the bottoms circulating reflux and bottom quench. With a properly
operating reactor cyclone system, a catalyst in slurry
concentration of 0.15% is usually achieved (typical
bunker fuel oil, 0.03-0.15 wt% catalyst in slurry).
However, if the solid concentration in the slurry oil is
higher, either some slurry clean-up system is needed,
or the operating condition of the FCC unit needs to be
reassessed and optimised. This could include redesign
or repair of the existing reactor cyclones.
Shell has evaluated all major technologies for application in removing catalyst fines from the slurry oil.
These options include hydrocyclone, electrostatic
separator, backwash filtering, and centrifugation.
A hydrocyclone separates FCC particles from slurry
oil based on centripetal force and fluid resistance. Due
to the advantages of low cost, no moving parts, low
pressure drop and low maintenance requirement,
hydrocyclones are the first choice for resolving slurry
ash issues when a cyclone redesign is not possible.
However, the performance of the hydrocyclone is relatively sensitive to the feed parameters (temperature,
flow rate, solid concentration). So hydrocyclones are
specifically designed for individual FCCs.
Electrostatic separators are designed and made
specifically for slurry oil so they can provide the
required separation with fairly low pressure drop. But
the capital cost and footprint requirement can limit the
application. Also, direct experimentation is required
for a good design for a specific FCC system.
Backwash filtration can also provide highly efficient
solid separation. Backwash filtration does require high
energy consumption, higher pressure drop, and higher
capital cost than hydrocyclones.
Centrifugation technology has advantages of small
volume and low pressure drop. Also, it can provide

some of the best solid separation performance.

However, the capital cost and operating cost (energy
consumption) need to be considered. Furthermore, the
maintenance of a centrifugal slurry separator is more
extensive compared to other separation technologies.
Each technology has its advantages and disadvantages, and applications are unit specific. For a specific
application, a comprehensive analysis, including the
source of the catalyst fines, temperature, residence
time, pressure drop, fines concentration, flow rates,
unit capacity, particle size distribution, and desired
capital cost budget is necessary. Working closely with
vendors of the various solutions will provide guidance
in resolving the excessive fines content in slurry oil
effectively and economically.
The catalyst concentration in the slurry oil should be
minimised not only from an erosion point of view; it
can also impact the sale value of the slurry oil.
Therefore, it is of great importance to select a proper
solid separation technology especially when higher
catalyst concentration is observed, which can happen
later in the FCC operating cycle. In Shell units, as the
first barrier, a high efficiency cyclone system is
installed in the reactor to minimise the catalyst
fines loss to the main fractionator. Furthermore, with
the help of the technologies mentioned, catalyst
concentration in the slurry oil can be controlled to
below 0.05%.

Joe Nguyen, Research Associate,

Joe.Nguyen@bakerhughes.com

Baker

Hughes,

Slurry oil has historically been clarified that is,

reduced in solids content by one of three general
methods. These are centrifuging, filtration, and longerterm passive settling. Centrifuging is not as effective as
other methods since the particulate solids in a slurry
oil (mostly FCC catalyst fines) are not removed by the
FCC reactors cyclones and are finely divided and
light. Filtration systems can be effective at removing
solids, but these filtration systems can also be costly in
terms of equipment and operation. The most costeffective method is passive settling in storage vessels
aided by temperature and slurry settling chemicals.
Elevated storage temperature reduces viscosity, allowing solids to settle more easily. Chemical surfactants
added to the slurry oil can promote agglomeration of
smaller fines into larger particles, which settle at faster
rates. The one drawback to passive settling is that time
is required for slurry settling, even allowing for the
temperature and chemical aids.

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

www.eptq.com

Q&A copy 15.indd 1

PTQ Q4 2014 5

10/09/2014 12:36

Charles Radcliffe, Technical Service Engineer Europe,

Johnson Matthey, Charles.Radcliffe@Matthey.com

I read this as being a problem of catalyst fines settling

in the bottom of the main fractionator. The important
thing with handling slurry oil is to maintain a high
enough velocity to ensure that catalyst cannot settle.
High velocities also reduce residence times and so
reduce coke formation due to thermal cracking. For
this reason, main fractionator slurry systems are
normally designed with a small diameter, <50% of the
main column diameter heel. A short-term fix for the
questioner is to try and increase slurry pumparound,
and if they have a large heel to install a lining in the
bottom of the fractionator with steel skinned
refractory.
Catalyst fines can be removed using a gravity settler
or filter system. The problem with settlers is that to
work efficiently they need to have residence time at
high temperature to lower viscosity. This leads to coke
formation and potentially flash point problems on
slurry product. There is also the problem of what to do
with the settler underflow. Recycling will increase FCC
delta coke. Filtration systems reduce the amount of
high fine material (from back washing) but are high
maintenance equipment.
Most refiners work to ensure good reactor cyclone
efficiency to minimise fines carry-over, and purge this
with the slurry product.

Eva Andersson, Market Manager Refinery, Alfa Laval,

evae.andersson@alfalaval.com

High speed centrifugation technologies have been used

for decades to remove cat fines onboard marine vessels,
where engines demand less than 10-15 ppm cat fines in
the oil. The same technology can be applied on
unblended FCC slurry in refineries but the high speed
centrifuge used must be designed to handle higher
temperatures in combination with more abrasive sludge.

Is there an effective way of cleaning contaminants

from deasphalted oil so that it can be used as a feed to a
hydrocracker?

Deepak RD, Hydrocracking Technical Service Engineer,

Deepak.Rd@CRI-Criterion.com,
and
Deepak
Agarwal,
Hydrocracking Technical Service Engineer, D.Agarwal@CRICriterion.com, Criterion Catalysts & Technologies

In order to maximise the margin of upgrading heavy

fuel oil into lighter products, one operation variable
that has been leveraged by refiners is the degree of
lift employed at the deasphalting unit. Further,
depending on DAO application and yield requirements, SDA (solvent deasphalting) units are designed
with solvents ranging from propane to hexane. The
type of solvent used is based on the contaminant level
and yield of DAO from the unit. For base oil production, propane deasphaltene is used to produce high
quality and low yield DAO. Normally, C4 and heavier
hydrocarbon is used as a solvent when the DAO is
processed in a hydrocracker and higher DAO yield is

6 PTQ Q4 2014

Q&A copy 15.indd 2

desired. This scenario increases both feed severity and

catalyst deactivation rate in the hydrocracker processing the DAO stream. Typical DAO stream composition
metals (5-100 wppm), Conradson carbon (210 wt%),
asphaltenes (heptane insolubles of 500-1000 wppm or
more) and distillation end-point (1300-1500F or more)
are detrimental to achieving economic catalyst life and
conversion at hydrocrackers. There are several ways to
reduce contaminants in a DAO stream. Once the
process to reduce contaminant carry-over is in place,
effective use of catalysts is the key factor to success.
For new unit design, a separate guard reactor could
be designed for contaminant removal. For an existing
hydrocracker, a proper choice of the Demet catalyst
system will be required to handle the contaminants in
the DAO stream. From a hydrocracking perspective,
the contaminants most detrimental to pretreat and
cracking catalyst performance are metals and
asphaltenes. Fortunately, both these show the sharpest
partitioning in SDA units, with DAO, even from deep
extraction, often having a lower C7-insolubles
(asphaltenes) value than the corresponding HVGO.
With proper understanding of the DAO chemistry, a
demetallisation catalyst system can be individually
designed to protect the downstream hydrocracking
catalysts. This has been fully proven by the successful
operation at a European refinery of the combination of
a KBR Rose solvent deasphalting unit with a Shell
fixed-bed hydrocracker, specifically designed to
process a high percentage of deeply extracted DAO in
combination with VGO. The upgrading complex delivers high conversion coupled with high distillate yields
of high quality and with a long, sustained run length.
After three years continuous operation, the demetallisation catalyst was replaced, while both pretreat and
cracking catalysts were retained for the next cycle.
It is important to know the level of removal of
metals, heteroatoms, Conradson carbon residue (CCR)
and asphaltenes. The feed composition and operating
conditions determine the right catalyst choice. These
variables determine the factors to tailor make catalyst
systems that meet each refinerys needs. In an existing
unit with existing HCU reactors, there would be a
maximum reactor volume available for the demet catalysts which then restricts the maximum amount of
DAO that could be processed in the HCU feed based
on the DAO quality. The performance of Criterions
demetallisation catalyst system has been the key factor
that enables ULSD and high quality jet fuel to be
produced directly from DAO over economic cycle
lengths. In addition, Criterion has developed step-out
HDM/HDN/HDCCR catalysts with exceptionally
high HDN and HDCCR activities, good contaminant
metals tolerance and excellent stability. Using innovative manufacturing technology, Criterion has
additionally developed a new type of active site that
resists metal and coke deactivation. In applications
used, it has demonstrated stable activity at high metal
deposition levels. Metal species compete with each
other and hence the effective use of customised layering of catalysts is vital to sustain effectiveness.

www.eptq.com

10/09/2014 12:36

STRATEGIES THAT DELIVER SOLUTIONS

Profiting from Refinery/Petrochemical Integration

With changing feedstock supplies and flat or declining demand for gasoline, its more important
than ever that refiners derive maximum value from every molecule.
CB&Is integrated hydrocarbon processing solutions enable refiners to handle varied feedstock
supplies, maximize clean distillate fuels production and increase FCC light olefin yield to take
advantage of growing petrochemical feedstock demand.
Our broad portfolio of both refining and petrochemical technologies, combined with our
execution expertise, will help you maximize unit flexibility and achieve margin benefits in the
widest range of scenarios.
PROCESS PLANNING AND DEVELOPMENT
LICENSED TECHNOLOGY AND CATALYSTS
FULL-SCOPE EPFC SERVICES
PROJECT MANAGEMENT AND CONSULTING
AFTERMARKET SERVICES

A World of Solutions
Visit www.CBI.com

cbi_ptq_4q2014_ad_sep_2014.indd
1
cbi.indd 1

8/13/2014
8:48:42 AM
10/09/2014
11:35

Alex Yoon, Senior Technology Advisor for Advanced

Refining Technologies, alex.yoon@grace.com

The solvent deasphalting (SDA) process is itself a way

of cleaning contaminants from a portion of vacuum
resid the deasphalted oil (DAO) to be processed in
a hydrocracker. Depending on the severity of SDA
operation (extraction target of DAO), contaminant
levels can vary signicantly. The most common
contaminants are Ni and V containing resid molecules.
These can vary typically from 1 ppm to 30 ppm. There
are also asphaltenes and other high CCR causing molecules which would be considered contaminants in a
hydrocracker operation. The former can be as high as
1000 wppm while the latter can be as high as 13 wt%.
High sulphur, high nitrogen and low API are also typical characteristics of the DAO.
DAO contaminant removal for the hydrocracker is
accomplished within the hydrocracker using appropriately designed catalysts, typically in the rst two beds
of the reactor. Advanced Rening Technologies utilises
catalysts with high macro- and mesoporosity that have
been successfully used in many vacuum resid processing units worldwide. These catalysts have successfully
processed various DAOs with up to 12 wt% CCR, 30
ppm metals, and up to 75% DAO blended in the feed.
These catalysts also have excellent hydrotreating
functionality. However, high contaminant levels can
rapidly reduce their activity and this results in a tradeoff with downstream hydrocracking catalysts. Higher
contaminant levels with a given operating target

require more of these contaminant removal catalysts

which then reduces the reactor volume available for
other hydrotreating and hydrocracking catalysts. A
rener wishing to extend cycle length can increase
loading of high macro/mesoporosity catalysts, but this
reduces the volume of hydrotreating catalysts.
To alleviate the downside of the trade-off, ART has
also developed and begun deploying next generation
catalysts that exhibit two to three times the relative
volume activity of conventional catalysts for this type
of service. With this next generation technology in the
catalyst load, the reduced hydrotreating catalyst
volume will exhibit longer cycle length. A rener wishing to process more DAO can also utilise this next
generation technology in a similar way to improve
protability. An optimum catalyst loading for maximum performance requires close cooperation of
renery operating staff and catalyst supplier experts.

George Anderson, Global Hydroprocessing Specialist,

Albemarle Corporation, george.anderson@albemarle.com

Deasphalted oil (DAO) tends to contain relatively high

amounts of nickel (Ni) and vanadium (V). In addition,
depending on the crude source, it may contain catalyst
poisons (such as arsenic) and may pick up solids (such
as particulates and iron scale) during its transport to
the hydrocracking unit (HCU). Particulates and iron
scale can be removed effectively by ltering prior to
the HCU or by use of appropriate bed grading and
traps at the top of the hydrocracking pretreat (HC-PT)

Our success formula for innovative water and process management

Permanent research and development form the core of our innovative products for chemical water and process management. As a worldwide market
leader, we continually develop new technologies in our own research laboratories. From applications for various industries to improving plant and
operational safety and efficiency, Kurita always places a high priority on conserving water resources. With our unique technology for membranes and
intelligent treatment concepts for reverse osmosis, we significantly increase the efficiency of raw and wastewater treatment.
Our approach is as precise as our outcomes: Innovative products and environmentally friendly treatment programs are the essential elements of our
success formula for creating solutions to meet your individual needs.
Kurita Europe GmbH . Industriering 43 . 41751 Viersen . Germany
Phone: +49 2162 95800 . www.kurita.de

4032_00KU_LY_1_2seitige_Innovation_engl_MS_04062014.indd 1

8 PTQ Q4 2014

Q&A copy 15.indd 3

04.06.14 12:48

www.eptq.com

10/09/2014 12:36

B
Tr

Helping refiners transform

difficult feeds into the cleanest
products possible.

The global hy roprocessing partner of hoice .

c
d
Recent partnerships include:

Saudi Aramco-Total: Jubail, Saudi Arabia, ISOCRACKING Technology

GS-Caltex: Yeosu, Korea, LC-FINING Technology

ROSNEFT: NovoKuybyshev, Russia, ISOCRACKING, ISODEWAXING and ISOFI

NISHING Technologies

PetroChina: Sichuan, China, UFR and VRDS Technologies

Hydroprocessing Technologies and Catalysts

www.chevronlummus.com

clg.indd 1

10/09/2014 11:38

reactor. Similarly, special contaminant traps such as KF

16 MAC can be used in the guard zone of the HC-PT
catalyst system to remove poisons (in this case maximum arsenic capture) from the feed and protect the
active catalysts. Finally, wide-pore HDM catalysts (for
instance, KFR 22) can be effectively employed to
remove Ni and V from the HCU feed. With this protective guard zone up front in the HC-PT reactor, high
activity NiMo catalysts can then be utilised to remove
nitrogen (N) and sulphur from the DAO/VGO feed
blend to enable proper performance of the cracking
catalyst(s) and hydrogenate aromatics to make the feed
more crackable.
Feeds containing DAO tend to have very high backend boiling points, which implies they will have
relatively high Ni and V content and high PNA
content. In addition to requiring a larger guard zone
than typically needed for most VGO feeds, it is
important to ensure that an active NiMo HDN catalyst
with relatively large median pore diameter (MPD) be
used in a significant portion of the HC-PT catalyst bed
to avoid accelerated deactivation of the catalyst due to
coking and pore mouth plugging. KF 860 has excellent
HC-PT activity stability in applications where feeds
have very heavy back-end boiling points.

We need to replace the hydrocarbon gas detectors and

associated wiring in our tank farm. Is a wireless set-up reliable
enough for this critical safety system and what capital cost
savings would a wireless option offer?

Diederik Mols, Business Manager Industrial Wireless,

Honeywell Process Solutions, diederik.mols@honeywell.com

Wireless technology based on the ISA100 wireless

industry standard is technically able to provide a
better than wired performance. Of course it will be
essential that the wireless network is properly
designed and implemented. Reliability can be further
enhanced by designing in redundant communication
paths, redundant gateways and network management.
As a rule of thumb on a given project, wireless technology realises capital cost savings up to 50% and
reduces execution time by up to 80%.
Wireless technology as an integral part of safety
systems is in operation at oil and gas facilities both offand onshore. Wireless stationary gas detectors are in
operation at multiple offshore platforms in Norway.
Wireless stationary gas detectors combined with
solar power panels, sounders and beacons serve as
perimeter monitoring system at a gas producing plant
in Qatar, meeting a stringent end to end latency
requirement of three seconds. Wireless personal gas
detection is selected and in the process of being
commissioned at a large sour gas producing facility in
the UAE, providing near real time data of gas levels as
well as gas- and man-down alarms of personnel.
Honeywell Process Solutions delivered turnkey solutions to each of these three examples. In 2003,
Honeywell was the first vendor to introduce industrial
wireless enabled technology.

10 PTQ Q4 2014

Q&A copy 15.indd 4

How do we effectively remove aerosols from the

hydrogen stream leaving our catalytic reformer?

Sravan Pappu, Refinery Purification Manager, Johnson

Matthey, Sravan.Pappu@matthey.com

Typically, aerosols found in the hydrogen product

stream of the catalytic reformer are removed through
engineering methods. In design, a flash separator
should be sized to give adequate disengaging time for
vapour-liquid separation. For those anticipating problems or experiencing such problems with the design of
the flash vessel, which is expensive to replace, a demister pad is used to prevent liquid droplets from
propagating with the vapour stream. Long or poorly
insulated runs of piping around the hydrogen product
stream can cause stream cooling and subsequent
condensation. Such liquid can have an adverse effect
on chloride guard performance as well as carrying
over unwanted liquids to downstream users via a
hydrogen header.
Given that most catalytic reformers utilise a chloride
guard on the hydrogen product, many end users have
removed aerosols to both protect downstream users
and maximise performance of the chloride guard bed.
This is done using a three-step system similar to those
employed in recontacting sections. The first is a cooler
which condenses out any heavier components that
would have condensed out later in the line. Next, a
knockout pot is used to drain accumulated liquids,
many times for useful recovery. Finally, the third is a
superheater which heats the gas above dewpoint. The
end result of this system is a super-heated, dry,
hydrogen-rich stream.

Since a switch to heavier crude blends we are losing

around 25C from our HVGO cutpoint. Are there any process
schemes to overcome this?

Charles Radcliffe, Technical Service Engineer Europe,

Johnson Matthey, Charles.Radcliffe@Matthey.com

The lowering of the HVGO cut point could be caused

by a number of problems:
1. Overloaded packing, possibly due to coking in the
bottom of the HVGO and top of the wash oil sections.
A switch to structured packing may help, otherwise
talk to the packing manufacturers.
2. Overloaded draw tray due to inadequate pump
capacity or increased vapour/liquid traffic. If pumps
are fully loaded, try running the spare(s) to see if
the cut point is improved if so debottleneck the
pump(s). If the tray is flooding, this will require a
revamp to fix.
3. Flooding will increase the tower DP lower cut point
at a given flash zone temperature. Raising this will
result in increased cracking which will increase the
vacuum package loading and further reduce the cut
point.
4. As well as reviewing the design and operation,
Tracerco could be invited to scan the tower for
evidence of flooding or maldistribution problems.

www.eptq.com

10/09/2014 12:36

We create
chemistry
that makes
tight oil
feedstocks
love flexible
catalyst
technology.

BASFs flexible catalyst technology and advanced

technical services provide solutions to improve product
yields, solve heat balance problems, and improve
metals tolerance for refineries processing tight oils.
BASF, the leading supplier of FCC catalysts to the tight
oil market, provides a broad foundation of knowledge
and operating strategy to deliver value to your refinery.
www.catalysts.basf.com/refining

basf.indd 1

10/09/2014 13:54

pcs myth.indd 1

8/3/12 09:55:39

Heat stable salts forming in our amine plant circulation

loop are leading to frequent unplanned shutdowns. Is there a
means of continuous removal of these contaminants without
shutdown and amine clean-up?

Nate Hatcher, Vice President, Technology Development,

Optimized Gas Treating, nate.hatcher@ogtrt.com

There are at least two non-distillation technologies for

continuous removal of HSSs from amine systems:
one is based on ion exchange (EcoTec in Canada
and MPR Services in the USA), the other uses electrodialysis (Dow and ElectroSep) although it may
not be continuous. HSSs are generated in the amine
unit itself from components coming in with the gas.
HCN is a typical and common source for HSS
formation and once in the amine there is not much
you can do to prevent its hydrolysis to formate.
However, another source is oxygen incursion and yet
another is SO2 breakthrough in TGTUs. These generate
cyanate and thiocyanate which can be prevented by
keeping out oxygen and SO2. Care must be taken not
to over or underclean the system of HSSs. Some residual HSS can be beneficial for assisting the amine
regeneration and result in improved treating in leanend pinched columns. The optimal HSS level can be
predicted with the ProTreat rate-based gas treating
simulator.

calibrated on a monthly basis as a preventative maintenance function.

Proper design is a critical element for a successful,
online pH system. Maintaining adequate water flow is
necessary. Managing the potential for contaminants to
interfere is also important. Oil is the most serious
threat to reliable operation. To manage that, devices
can be included to maintain the probes in a clean
condition, although at some sites, these are not
required.
The first benefit of an online pH measurement is to
provide an alarm when the system goes out of the
control target. For online pH measurement, an alarm
for low or high pH should trigger an investigation.
The closed-loop control of amine injection is automatic

The first benefit of an online pH

measurement is to provide an
alarm when the system goes
out of the control target

Is there a means of online pH testing of process water

from our crude unit?

when low pH events occur; however, occasional events

of acid-bearing crude oils can lead to maximum pump
flow, leaving the system below the minimum pH
target. Operator intervention is then required to determine the best course of action. For systems with tramp
amine in the crude feed, the upper pH target can be
exceeded. For refiners where this is a common occurrence, the upper alarm limit may need to be adjusted
to accommodate the nature of the operations with
amine contamination.
Note that the pH that is analysed is only a look at
the bulk overhead water. The distribution of both acid
and amine through the overhead is a function of many
different variables. A thorough analysis should be
conducted to ensure that the proper amines are used to
minimise overhead corrosion, featuring the Baker
Hughes Topguard corrosion risk monitor. Assuming
the proper amines are used and other parameters, such
as wash water, are properly applied, an online pH
system is an excellent method of controlling the feed of
amine to the overhead system.

Baker Hughes operates a closed-loop pH control

system to regulate the injection of overhead amine on a
real-time basis. Overhead water is directed through a
cooler and past a pH probe. The amine feed pump is
connected to the pH controller and varies the feed of
amine as required. As a safety precaution, the overhead pH is tested by each shift to ensure that the
on-line overhead pH analyser is providing accurate
results. On a weekly basis, the overhead pH is rigorously tested to ensure that the calibration of the
overhead unit is still acceptable. Deviations of more
than 0.5 pH units trigger immediate recalibration of
the pH control unit. The overhead unit is also

There is a wide range of corrosive components which

can damage the equipment of a crude unit. Hydrogen
chloride (HCl) is an acid gas, which is very aggressive,
when it is dissolved in condensed water to form
hydrochloric acid. The corrosivity of hydrochloric acid
depends on its concentration, measured by pH.
Regular pH online monitoring is a fundamental
component of controlling corrosion potential by keeping pH in the optimum corrosion control range. The
optimum pH control range of the overhead accumulator water is between 5.5 and 6.5. It is common practice
to inject a neutralising amine such as Kurita CI-6110,
which is a very powerful blend of selected aminebased corrosion inhibitors. Such corrosion inhibitors

Charles Radcliffe, Technical Service Engineer Europe,

Johnson Matthey, Charles Radcliffe@matthey.com

Most amine systems include a slip stream filter, which

normally treats about 10% of the circulation. A well
designed filter should be able to maintain the solids at
a level which minimises fouling. Sponge oil absorbers
are prone to foaming, resulting in LCO carry-over into
the downstream amine scrubber. LCO contaminated
amine will tend to increase fouling and foaming in
other scrubbers. Antifoam additives are quite effective
in minimising this.
Where packed beds foul, refiners will often bypass
them for 12-24 hours and steam clean the packing.

Ralph Kajdasz Senior Area Manager, Baker Hughes,

Ralph.Kaidasz@bakerhughes.com

www.eptq.com

Q&A copy 15.indd 5

Berthold Otzisk, Kurita Europe GmbH, otzisk@kurita.de

PTQ Q4 2014 13

10/09/2014 12:36

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.

Nasty Stuff

This exchanger has seen better crude slates

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.

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 miscalculations in the process design could
wreck the entire project.

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

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
3400 Bissonnet
Suite 130
Houston, Texas 77005
USA

pcs nasty.indd 1

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.

Ph: [1] (713) 665-7046

Fx: [1] (713) 665-7246
info@revamps.com
www.revamps.com

1/6/12 11:58:35

significantly reduce the corrosion potential by neutralising the acids and elevating the pH of water
condensate.
A side-stream of the process water can be routed
from the overhead accumulator drum to an online pH
meter (ATEX Eex). If the pH meter is connected to the
neutralising amine pump it is possible to avoid
under-dosage or over-dosage. It is an additional benefit to optimise the neutralising amine consumption.
The neutralised process water can be used as wash
water for the desalter or crude unit overhead heat
exchangers.

Charles Radcliffe, Technical Service Engineer Europe,

Johnson Matthey, Charles.Radcliffe@matthey.com

Online pH probes are commonly used in monitoring

refinery sour water streams. They are quite reliable but
are vulnerable to probe breakage and fouling, and so
need regular checking (weekly is common). Buffering
can also be a problem if they are used to control acid/
alkali dosing using standard PID control. A more
sophisticated control algorithm may be required if
off-line titration indicates there is significant buffering.

Our heavy crude feed is delivering very high chloride levels

downstream of the desalter and caustic injection cannot cope.
What further treatment do you recommend?

Andrea Fina,
afina@chimec.it

Process

Technogical

Unit,

CHIMEC,

Certainly the best way to reduce the chloride content

in a main fractionator overhead system is to maximise
the performance of the desalting system by means of a
proper tuning of the process parameters and the chemical treatment:
Revise the mix valve pressure drop to find the
lowest achievable value of salt downstream the
desalter
For heavy and high viscous crudes, a higher wash
water flow rate is advisable (up to 7 wt% vs crude
flow rate). A proper wash water flow rate assures
higher mixing between the wash water and the salts in
the crude oil
Split the wash water injection into two different
points: before the cold train (20-30% of wash water)
and before the mix valve (80-70% of wash water). This
configuration is aimed to improve the mixing between
the wash water and the salts, thus better salt removal
in the desalting vessel
Maximise the oil residence time by decreasing the
interface level, on condition of maintaining the effluent
water free of oil
Revise the demulsifier program for a higher dewatering and desalting efficiency.
A further strategy would be to implement a dewatering program on the crude oil directly in the storage
tank. Being chloride salts mainly dissolved in water, by
applying a demulsifying programme directly in the
storage tank, an important amount of chloride will be
discharged during the tank water draining operation,

www.eptq.com

Q&A copy 15.indd 6

thus a lower load of chloride will be processed in the

desalting system. Chimec has developed specific
demulsifiers to be applied in the crude storage tank.
A high salt content downstream of the desalter leads
to increased chloride content in the main fractionator
overhead system. If the partial pressures of HCl, NH3
and amines in the overhead system are high, depending on the temperature, ammonium and/or amine
chloride salts are formed. Once precipitated, these salts
promote fouling and under-deposit corrosion.
Chimec has developed an innovative and referenced
technology, Salt Dispersant Technology. Applied
together with a corrosion inhibitor program
(neutraliser and film forming corrosion inhibitor), this
technology assures complete protection against fouling
and corrosion, promoted by salt precipitation above
the water dew point and wherever the condensed
water is not enough for removing precipitate salts.

Gerald Hoffman Technical Manager, Baker Hughes,

Gerald.Hoffman@bakerhughes.com

As crude slates change to more attractively priced

crudes, such as heavy Canadian crudes and shale oils,
desalting can present increased challenges. Salt
removal may suffer due to a variety of factors, such as
incoming complex emulsions, asphaltene instability,
increased solids loadings (and smaller particle sizes),
and the presence of high melting waxes. Solids, waxes,
and asphaltenes can concentrate at the interface, causing the interface to grow. As the interface grows
downward, water residence time decreases, which can
lead to oily solids in the effluent brine. If the level is
increased to clean up the brine, electrical grids can
short out. Of greater consequence, emulsion carryover
with the desalted crude leads to a variety of downstream impacts, including higher overhead chlorides,
preheat and furnace fouling, and metals contamination
of downstream catalyst and products.
When processing heavy crudes, the first reaction to
these negative impacts is often to reduce mix valve
pressure drop. While this can help reduce emulsion
band size, it also reduces salt removal and dehydration
efficiency. If oily under-carry is occurring, the desalter
level may be raised, which only serves to reduce crude
oil residence time and move the problem downstream;
however, these are the wrong steps to take if good salt
removal is the goal.
A more appropriate approach is to first understand
the root cause of the problems. The Baker Hughes
Crude Oil Management programme is a suite of tools
and capabilities designed to better manage opportunity
crude desalting. The toolkit includes both operational
strategies (crude blending and desalting best practices)
and treatment methods (primary emulsion breakers,
tank pretreatment, asphaltene stabilisers, wax dispersants, acidification, and solids release additives). The
philosophy behind Crude Oil Management is to treat
the problem, not the symptom. A few of the tools are
discussed below.
Crude pre-treatment technologies can address
several of the challenges posed by heavy crude

PTQ Q4 2014 15

10/09/2014 12:37

processing. Pre-treatment chemicals are specifically

designed to function at tank conditions, and can be
used to address asphaltene instability, wax stabilsed
emulsions, and solids impacts.
Crude compatibility: the Baker Hughes Asphaltene
Stability Index Test (ASIT) provides guidance on crude
oil blending to minimise the risk of forming asphaltene
solids. If a blend of poor compatibility is processed,
asphaltene stabilisers can be injected to improve
asphaltene stability.
Crystallised waxes: high melting point waxes in
some crudes can stabilise emulsions. Wax dispersants
are an effective means of managing these concerns.
Solids: oil coated inorganic solids are a major contributor to emulsion formation in the transport and storage
of crude oils. These solids can be the primary cause of
oily under-carry with the effluent brine. Pre-treatment
chemicals are designed to break emulsions in the storage tank, de-oiling solids and improving the overall
quality of the crude oil. This leads to improved tank
dewatering, more consistent crude quality, and fewer
desalter and crude unit upsets.

Crude pre-treatment
technologies can address several
of the challenges posed by
heavy crude processing
Treatment of heavy crudes at the desalter requires
specialty emulsion breaker technology. Xeric heavy oil
demulsifiers are designed to manage the challenges
posed by heavy crudes, providing exceptional salt
removal and dehydration efficiency.
Last but not least are the adjunct technologies: acidification and solids release:
Desalter
acidification:
Excalibur
contaminant
removal technology. Excalibur technology is designed
to remove contaminants such as metals (calcium, iron)
and amines from the crude oil. The acidification also
improves emulsion resolution, yielding higher quality
effluent brine
Solids release Jettison solids release additives
(SRA). Solids are a major cause of desalter emulsions,
downstream fouling, and catalyst poisoning. They also
tend to carry oil out with the effluent brine. Patented
Jettison SRA products are designed to aggressively
de-oil solids and enable clean separation from the oil.
The result is effluent brine high in oil-free solids,
meaning these contaminants are no longer fouling up
the refining processes.
The use of these technologies, in conjunction with
best practice operation of desalters, allows the safe
processing of heavy crudes, free of the negative downstream impacts.
ASIT, Crude Oil Management, JETTISON, EXCALIBUR, and TOPGUARD
are marks of Baker Hughes Incorporated.

16 PTQ Q4 2014

Q&A copy 15.indd 7

Berthold Otzisk, Kurita Europe GmbH, otzisk@kurita.de

As a first step, I would recommend an optimisation of

the desalter system. Salt removal from heavy crude oils
has always been problematical due to difficulties in
obtaining good mixing between brine and dilution
water. Heavy crude oils have a high viscosity and low
differential density between oil and dispersed matter.
An improved mechanical mixing (delta pressure
change) in combination with a higher desalter wash
water rate can help to achieve a better contact performance for salt removal without producing stable
emulsions. The utilisation of an emulsion breaker
programme for heavy crude oils can offer additional
benefits like improved dehydration and desalting
efficiency.
Kuritas patented ACF technology is an appropriate
means, controlling the corrosion potential of the
overhead system, when caustic cannot cope. The additives used for this technology are chemical liquid
formulations, known in the market as ACF. The very
strong organic base ACF contains no metals which
may harm downstream processes. Magnesium chloride
and calcium chloride can be easily converted to hydrochlorides of ACF. In the furnace, these chlorides
thermally decompose into a non-corrosive, volatile
organochloride gas and a neutralising amine
with almost no hydrogen chloride produced. Both
decomposition products are routed to the overhead
system, where the neutralising amine leaves the
system with the accumulator water and the volatile
gas with the gas phase. This is a technical innovation,
which significantly reduces the chloride concentration
and corrosion potential in the overhead system,
even when high chloride concentrations leave the
desalter.

Charles Radcliffe, Technical Service Engineer Europe,

Johnson Matthey, Charles.Radcliffe@matthey.com

Ideally, the problem should be addressed at the

desalter. Initial moves would be to increase wash
water rate, increase mix valve DP, and ensure all the
electrodes are working at full power. The source of the
problem may also be due to the formation of a stable
emulsion, and in this case reducing mix valve DP may
help, and using de-emulsifier additives. This problem
can be detected by taking a sample of desalter product
and keeping it in a temperature controlled bath at
desalter temperature for 24 hours then checking the
relative volumes of hydrocarbon and aqueous
fractions.
If the problem cannot be controlled at the desalter
then increase the strength of the caustic and increase
neutralising and filming amine injection rates to the
crude and FCC overheads, along with increased
frequency of crude and FCC fractionator overhead
water washes.

Nate Hatcher, Vice President, Technology Development,

Optimized Gas Treating, nate.hatcher@ogtrt.com

Poor separation in the desalter may be to blame due to

emulsion carry-over. If stripped sour water is used as

www.eptq.com

10/09/2014 12:37

MIDAS GOLD

The Gold Standard for Resid Upgrading

MIDAS FCC catalysts are Graces high
matrix catalysts with proven success for
deep conversion of resid. MIDAS Gold,
the latest member of the family, is a high
activity formulation that delivers
p

Superior coke-selective bottoms

upgrading

Stable activity against metals

contamination

Formulation flexibility for a customized

yield profile

We offer several distinct MIDAS formulas,

suitable for a broad range of feed and
operating objectives. Let Grace technical
experts show you the best fit for your operation.
grace.com

MIDAS GOLD
The Gold Standard for Resid Upgrading
Grace FCC Catalyst

grace.indd 1

10/09/2014 11:46

make-up to the desalter, verify that the amine content

is not excessive. Amines combine with fatty acids in
heavy crudes to generate soap.

India Nagi-Hanspal, Asst. Manager Technical Services,

Dorf Ketal, indianagi@dorfketal.com

Achieving good desalting performance is a challenge

during heavy crude processing. Processing heavy
crudes requires a good desalter optimisation plan and
use of superior demulsifier chemistry. A good
approach during heavy crude processing can be as
follows:
Optimise desalter wash water rate
Optimise desalter temperature
Optimise mix valve pressure drop
Dorf Ketals proprietary desalter treatment
programmes, comprising a demulsifier and co-additives, have helped refiners to process heavy crudes
(<25 API) with optimum performance
Optimisation of wash water upstream of the overhead condensers can also help reducing chlorides and
corrosion rates in overhead system.

We are not getting the cycle life we were expecting from

our hydrotreating catalyst after presulphiding. How should we
be adjusting process conditions after presulphiding?

George Anderson, Global Hydroprocessing Specialist,

Albemarle Corporation, george.anderson@albemarle.com

There are several potential causes for not achieving

expected hydrotreater cycle length after start-up, and
each of these has different potential solutions. For
instance, the problem could be due to higher than
expected start-of-run (SOR) WABT, and/or due to
higher than expected catalyst deactivation. Both of

It is strongly recommended that

a rigorous root cause analysis be
conducted to find the real causes
of the short cycle so that the
problem does not occur again
these causes are really responses that may be due to
multiple potential root causes. Insufficient information
is provided to do a root cause analysis, so the answers
provided here are very generic.
At a minimum, extending cycle length means that
you should operate your unit at minimum WABT to
meet product specifications for as long as possible.
Thus, you should avoid over-treating so that product
properties are significantly better than the product
specifications.
Maximising activity stability generally implies that
you should also maximise treat gas rate and purity to
maximise hydrogen partial pressure (ppH2). If possible,
you should also reduce the most difficult components

18 PTQ Q4 2014

Q&A copy 15.indd 8

in the feed (heavy ends, cracked stocks, and so on) to

lower hydrogen consumption, exotherms and coking
tendency, and you should remove very light feed
components (such as coker naphtha, FCC naphtha and
light kerosene) to avoid excessive feed vaporisation
that lowers ppH2.
If these recommendations do not achieve the desired
cycle length and/or it is not possible to implement
them or to route an easier feed to your unit, you may
be forced to make a choice between accepting a shorter
cycle length or reducing throughput (lower LHSV and
feed/product rates) for your unit. Obviously, neither of
these options is desirable economically. Therefore, it is
strongly recommended that a rigorous root cause analysis be conducted to find the real causes of the short
cycle so that the problem does not occur again in
subsequent cycles.

Is there a catalyst option for upgrading LCO to ULSD

without an additional process step?

George Anderson, Global Hydroprocessing Specialist,

Albemarle Corporation, george.anderson@albemarle.com

LCO can be upgraded to ULSD in either an existing

diesel hydrotreating (DHT) unit or in a hydrocracking
pretreat (HC-PT) unit without adding an additional
process step. HC-PT is a high pressure process
with very high treat gas rates, and can upgrade
LCO with relatively high throughputs and LCO
contents in the feed. Dedicated LCO hydrotreaters
operating at 85 bar ppH2 inlet can successfully
upgrade 100% LCO feed with ~three-year cycles
(depending on LHSV).
Most DHT units tend to operate in the 15-55 ppH2
inlet range. At the lowest pressures, very little (10%)
LCO can be co-processed to produce ULSD and
achieve a one-year cycle length. As ppH2 and hydrogen treat gas rate increase, higher percentages of LCO
can be accommodated and longer cycle lengths can be
achieved. However, as LCO content in the feed
increases, higher exotherms are generated due to the
greater hydrogenation of olefins and aromatics. The
exotherms and required cycle length may determine
the maximum LCO content in feed to avoid exceeding
maximum temperature limitations for the reactor
materials.
Selection of the preferred catalyst system to maximise ULSD throughput and upgrade to ULSD
specifications is highly specific to the unit capabilities
and limitations as well as the feed properties and
ULSD specifications (especially cetane number,
density and cold flow properties). Albemarles
Stax technology can be used to optimise unit performance with the right combination of CoMo
and/or NiMo catalysts to meet multiple refiner objectives in producing ULSD. For higher pressure units
with inlet ppH2 40 bar, using 20-40 vol% Nebula as
part of the Stax catalyst system has been proven to
be especially effective in increasing LCO upgrade
to ULSD.

www.eptq.com

10/09/2014 12:37

UOP_Platinum_Standard-Ad-A4_UOP Ad A4 size UOP6460 9/9/14 3:30 PM Page 1

platinum standard

UOPs propylene production technologies outshine the rest

Low cost feedstocks, high yield products. Theres no better combination for
generating petrochemical profits. As an industry leader in petrochemical process
technology for more than 70 years, UOP continues to deliver proven, flexible
solutions with high-yield returns. UOP advanced Methanol-to-Olefins (MTO) and
OleflexTM processes provide a higher return on investment, smaller environmental
footprint and innovation that is second to none. For advanced MTO, you can use
alternative feedstocks such as coal, natural gas and more, and you can produce
the high-value olefin of your choice, including propylene and ethylene. Recyclable,
platinum-based Oleflex catalysts offer the best performance for environmentally
friendly on-purpose propylene production. From low-energy solutions to eco-friendly
innovations, UOP sets a standard that shines.

1914 - 2014

A Century of Innovation
in the Oil and Gas Industry

Looking for exclusive information on olefins technology?

Scan the code to register for the UOP Portal.
2014 Honeywell International, Inc. All rights reserved.

uop standard.indd 1

10/09/2014 12:16

sandvik.indd 1

10/09/2014 11:57

Tail gas catalyst performance: part 2

The second part of a two-part account of time and temperature effects on tail gas
catalyst performance gives a background to reaction modelling and pilot plant studies
Michael Huffmaster Consultant
Fernando Maldonado Criterion Catalysts

he tail gas unit (TGU) process

has been developed to
remove sulphur compounds
from Claus tail gas in order to
comply with stringent emission
regulations. From the early 1970s to
today, TGUs have been improved
to meet higher levels of performance
for
ever
tighter
environmental requirements and to
reduce capital or operating cost.
Reactor performance is a critical
parameter in achieving TGU environmental performance. Conversion
of sulphur species to H2S is a function of catalyst activity, reactor
space velocity and temperature.
Assessment of the impact of these
principal variables on both catalyst
bed design and performance is the
subject of this article, presented in
two parts. The first part of the article (PTQ, Q3 2014) provided an
introduction to reactor modelling,
process and catalyst development
history, and chemical equilibrium.
The second part develops reactor
modelling with a kinetic reaction
model, the effects of temperature
and space velocity, catalyst activation, catalyst deactivation, and
determining TGU catalyst health
from a commercial unit temperature profile.

Reactor modelling

In order to predict the performance

of TGU reactor systems, a basic
framework of chemical equilibrium,
reaction chemistry and catalyst
activity is used. This, in turn,
provides a tool to evaluate the
effects of space velocity and temperature on reactor performance.
Requisites for good reactor design
and operation must be met to

www.eptq.com

criterion.indd 1

achieve good performance. The

assumption in modelling and
predicting performance is that all
the other things are done correctly.
With good gas distribution across
the catalyst bed and acceptable
pressure drop for process line-up
and inlet concentrations within the
design boundary, one can move
forward to the impact of gas rate
and time and temperature on
performance.
The influence of temperature on
performance has competing effects

To predict the
performance of TGU
reactor systems, a
basic framework of
chemical equilibrium,
reaction chemistry
and catalyst activity
is used
in kinetics and equilibrium, impacting conversion. The kinetics for
reactions of importance are favourably
influenced
by
higher
temperature, proceeding to higher
conversion at a given space velocity. Equilibrium effects from higher
temperatures usually result in
higher equilibrium concentrations
for the species, which the system is
designed to destroy, limiting lower
values for outlet concentration.
Equilibrium considerations were
addressed in the first part of the
article and this part deals with
kinetics.

Conversion is a term of frequent

reference in this article. Conversion
for CO (or COS) is expressed as
disappearance across the reactor
and is adjusted for the equilibrium
back pressure of the reacting
component:

COout COequilibrium
Conversion = 1 - -------------------------

COin COequilibrium

Tail gas reactor temperature has

historically ranged between 200C
and 325C. This fits within the
region of active catalyst functions
and meets the required minimum
temperature for catalytic activity
function, about 200C for low
temperature
catalysts
and
240-300C for conventional tail gas
catalysts. Maximum temperature is
generally limited to 345C (650F).

Kinetics: first order reaction model

Reaction kinetics describes how fast

a reaction proceeds and allows
prediction of how far a reaction
proceeds toward equilibrium in a
reactor system. The reactor system
is defined by gas composition, gas
flow rate, and catalyst type and
volume. A first order reaction
kinetic model with equilibrium is
useful to describe the conversion of
reactants in the tail gas reactor
system and is helpful to understand
performance effects of operating
variables.
Tail gas reaction kinetics can be
represented reasonably with first
order reaction models as a useful
representation of tail gas reaction
kinetics. Experimental data for a
Criterion C-534 tail gas catalyst is
fitted into the reaction modelling

PTQ Q4 2014 21

10/09/2014 12:48

CO conversion, %

100

90

converted, adjusting for equilibrium residual:

Conversioneq = 1 (Caout Caeq)/(Cain Caeq)

and
80

Conversioneq = 1 (Caout Caeq)/( Cain Caeq) =

1- e-k * 3600 / GHSV

where: k = first order activity catalyst, mol/sec

SOR
Cain = inlet concentration of compoEOR
nent A
60
1500
2000
2500
3000
3500
Caout = outlet concentration of
Space velocity, GHSV 1/h
component A
Caeq = equilibrium concentration of
Figure 1 CO conversion vs space velocity
component A, set by WABT and
composition of outlet stream
Reactor loading is represented in GHSV = actual gas hourly space
equations; those models are used to
evaluate design parameters and test the kinetic expression as gas hourly velocity.
operating considerations or trouble- space velocity (GHSV), a ratio of
Reactor outlet concentration will
shooting to improve performance the gas rate per unit of catalyst. The
basis used for these expressions is be the remainder of the inlet not
and assess catalyst activity.
actual volumetric gas flow rate per converted plus equilibrium residual:
Sulphur species conversion:
hour divided by catalyst volumetric
Caout = (Cain Caeq) * (1-Conversioneq) + Caeq
catalyst activity, temperature and
inventory.
space velocity
The relationship between converModelling the reactor as a continu- sion, actual gas hourly space
The reactor space velocity expresous plug flow reactor, first order velocity, GHSV and catalyst activ- sion is:
reactions kinetics provides a means ity, k, is defined in first order
actual gas volumetric flow (vol/hr)
for quantifying the gas loading reaction model as:
GHSV = ------------------------------------ = 1/hr
effect. First order reaction kinetics
catalyst inventory (vol)
models are reasonable models for rate of disappearance of component a = dCa
/
dt
=
k
*
Ca
Reactor residence time is someboth COS hydrolysis, COS sour
times used as another expression of
shift and CO water gas shift
conversion in the tail gas system. and overall for the reactor, the reactor loading, and it is related to
Although the complete expression equilibrium conversion is given by: GHSV:
is complex when considering mass
t = 1/ghsv = hr or t = 3600/ghsv = sec
transfer, diffusion, adsorption and ln (1-Conversion eq) = -k * 3600 / GHSV,
surface reactions, it is possible to
Equilibrium conversion (ConverConversion of components is
simplify the expression to kinetic
activity. At extremes, other resist- sion eq) represents the amount of dependent on space velocity and
component a, which can be actually catalyst activity (see Figures 1 and
ances exert influence.
2). Equilibrium conversion compensates
for
the
effects
of
100
concentration, composition and
temperature. The resultant outlet
concentration of components is
90
determined by their inlet concentration, equilibrium and conversion.
The influence of space velocity on
80
conversion is explicit in the first
order reaction relationship. The
quotient of activity divided by
70
space velocity is proportional to the
SOR
logarithm of conversion. Higher
EOR
conversion is obtained with lower
60
1500
2000
2500
3000
3500
space velocity (greater catalyst
Space velocity, GHSV 1/h
volume) or higher catalyst activity.
Higher gas rates, higher temperaFigure 2 COS conversion vs space velocity
tures or lower pressures all increase
COS conversion, %

70

22 PTQ Q4 2014

criterion.indd 2

www.eptq.com

10/09/2014 12:48

Comprimo Sulfur Solutions

Gas Treating & Sulfur Technology

With more than 500 units built during the
last 40 years, we are a global leader in the
gas treating and sulfur recovery technology

Total Project Solutions

We are one of the worlds largest and most diverse providers of
professional technical services. Our global network includes more
than 200 offices in more than 25 countries
From our gas treating & sulfur technology centers of expertise in
Leiden, The Netherlands and Calgary, Canada, we provide engineering,
procurement and construction services to our clients around the world

Visit our webpage to learn more about all our products >

gas jacobs.indd 1

www.jacobs.com/comprimo

04/03/2014 10:46

90

80

70

50
500

cosh 300C act.

cosh 280C act.
cosh 260C act.

cosh 300C conv.

cosh 280C conv.
cosh 260C conv.

60

1000

1500

2000

2500

3000

3500

1
0
4000

First order activity co-efficient

(k)

Conversion, %

100

aGHSV, 1/h

Figure 3 Standard feed C-534 COS hydrolysis conversion and activity vs space velocity

underlying equilibrium constant

expression, which is a function of
temperature.
The activity coefficient, k, represents catalyst activity. It must be
determined experimentally, or may
be determined for a specific operating plant from a plant test. Its
relationship to temperature is
represented by the Arrhenius
relationship.
The temperature dependence of
the catalyst activity coefficient, k, is
represented
by
the
classical
Arrhenius relationship:
Reaction rate constant, k = A e-Ea/RT

Where A = activity coefficient

Ea = activation energy exponent
R = gas constant
T = absolute temperature

100

90

80

70

2
1

60
50
250

cosh 3500 act.

cosh 2500 act.

cosh 3500 conv.

cosh 2500 conv.
260

270

280

290

300

310

320

0
330

First order activity co-efficient

(k)

Conversion, %

actual GHSV for a given standard

volumetric gas flow rate.
The influence of concentration is
scalar on reactor outlet concentration; whereas higher concentration
increases rate of disappearance and
conversion remains constant, thus
outlet increases in proportion to
inlet concentration. If it is necessary
to obtain a target outlet value for a
given component, say to meet an
environmental target, as an inlet
concentration of that component
becomes higher, a higher conversion is required.
The influence of tail gas compositional change is handled through
the equilibrium value of CO or
COS with their related reaction
components and the equilibrium
constant, Kp. The influence of
temperature on equilibrium value
for CO or COS is handled in the

Temperature, C
Figure 4 Standard feed C-534 COS hydrolysis conversion and activity vs temperature

24 PTQ Q4 2014

criterion.indd 3

An alternative expression is: k =

A*T*e-(Ea/RT), but it is more complex
to model and provides only slight
improvement in correlation.
Parameters for the Arrhenius coefficients for COS hydrolysis were
established from published data for
alumina and fitted against plant
data and experimental data on a
SCOT catalyst. The CO function was
fitted to SCOT catalyst data. This
handles the important temperature
dependency function for the activity
and provides the basis for a model
for reactor performance. The
temperature dependency activity
functions for CO and COS are
generally proprietary to the respective catalyst manufacturers.
Chemical reaction equilibrium
and kinetics are influenced by
temperature as the following examples show.
The chart of experimental data in
Figure 3 shows conversion decreasing as space velocity increases. This
is entirely consistent with the first
order reaction model. Note that in
the reaction model, catalyst activity
is a function of temperature, not
space velocity. In Figure 3, the first
order activity coefficient, k, is
derived from values calculated
from conversion and equilibrium
and space velocity. The activities
are almost constant for each
temperature. The decline at low
space velocity is attributed to low
gas velocity in the reactor bed with
diffusion limitations in mass transfer as the flow regime approaches a
laminar state.
Temperature has a strong and
direct influence on activity (see
Figures 4 and 5). For COS, with an
activity on the order of 1.5 to 2.5,
an increase of factor 2 was observed
for a 70C increase. This relates to a
1% increase in activity for every
degree Centigrade increase. A 1%
improvement when compounded
over 70 degrees is a doubling: 2 =
(1.01)70.
Over the range 250-320C, the k
cosh catalyst activity for COS
destruction doubles and equilibrium COS, for example, increases
from 2 ppm to 15 ppm. Over the
range 250-320C, the k wgs catalyst
activity
for
CO
destruction
increases 50%. Because the activity

www.eptq.com

10/09/2014 12:48

Optimize your plant for a safe, efficient start-up

Through our detailed engineering and expert field execution
When considering mechanical pre-commissioning and commissioning for your plant start-up consider this.
Innovative engineering and service excellence results in greater quality. And THAT reduces costs. Were an oil and gas
services company committed to excellence. Everything we do in the field we engineer first to meet your specific needs.
Demand Professionals. Demand FourQuest Energy.
For a complete list of services, visit our website.

www.fourquest.com

Connect with us on:

SCAN THIS CODE TO
VIEW OUR WEBSITE

four quest.indd 1

04/03/2014 11:04

90

80

70

2
1

60
wgs 3500 act.
wgs 2500 act.

wgs 3500 conv.

wgs 2500 conv.

50
250

260

270

280

290

300

310

320

0
330

First order activity co-efficient

(k)

Conversion, %

100

Temperature, C
Figure 5 Standard feed C-534 CO conversion and activity vs temperature

Concentration, ppm

CS2
SO2
H2S

CO
COS
C1SH
CH4

10000

1000

100

10

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Fraction reactor bed depth

Figure 6 Standard feed C-534 reactor concentration profile 250C RIT

Concentration, ppm

10000

CO
COS
C1SH

1000

CH4
CS2
SO2

100

10

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Fraction reactor bed depth

Figure 7 Standard feed C-534 reactor concentration profile 270C RIT

level is higher, the relative increase

is a smaller percentage. Equilibrium
CO for a standard feed example
increases from 50 ppm to 250 ppm.
Temperature can be increased to
improve CO and COS destruction

26 PTQ Q4 2014

criterion.indd 4

performance for a catalyst that has

lost activity, but aging accelerates
with increased temperature. This is
usually an end-of-run tactic to
extend a run to scheduled
turnaround not a fountain of

youth because the rate of activity

decline increases:
The equilibrium back pressure,
which governs how low a concentration can possibly be achieved, is
favoured by lower temperatures
At low temperature, catalytic
activity declines and conversion
performance is dominated by kinetics; equilibrium does not matter
very much since all equilibrium
values are very low
At conventional temperatures,
catalyst kinetics are stronger, hence
catalyst kinetic property is important, although equilibrium back
pressure sets a lower boundary for
performance
At high temperatures, performance is set by equilibrium, if
catalyst activity is fair and space
velocity is reasonable.
Figure 6 shows the concentration
of feed and product components
through the reactor bed for
Criterion C-534. This demonstrates
the performance range for fresh
catalyst in the conventional temperature range. Early bed effects are
not measured, such as likely rise of
COS concentration from early CS2
conversion or the possibility of
COS from sour shift with CO and
H2S. Note that conversion occurs
for methyl mercaptan to methane.
The bed is effectively shortened
by a loss of activity or an increase
in space velocity. With a decrease
in conversion outlet, the concentration moves away from equilibrium.
Conversely, at higher temperatures,
the catalyst has higher activity and
reactions proceed faster. This is
indicated by the shift in concentration profiles for a 20C higher inlet
temperature (see Figure 7).
The first order kinetic model
provides a tool for quantification of
design and evaluation of catalyst
performance from operational data.
This also is useful for forecasting
performance for different conditions, such as increasing tail gas
load (space velocity) or temperature. Activity and space velocity
are paired, which means that, at
constant catalyst activity, an
increase in tail gas loading (space
velocity) reduces conversion.
Figure 8 shows conversion with
illustrative activity lines for start-of-

www.eptq.com

11/09/2014 11:22

Activity required, basis first

order reaction

Conversion, %
99.95
99
98
95
93
85
75
60

Table 1

www.eptq.com

criterion.indd 5

aGHSV, 1/hr
1000
2000
1/hr
1/hr
2.11
4.22
1.28
2.56
1.09
2.17
0.83
1.66
0.74
1.48
0.53
1.05
0.39
0.77
0.25
0.51

3000
1/hr
6.33
3.84
3.26
2.50
2.22
1.58
1.16
0.76

100

CO conversion, %

run, mid-run, and end-of-run

conditions as a function of space
velocity. If the condition of the
catalyst is known, projections can
be made about future performance
capacity. A catalyst with good
activity has the potential for a long
cycle life and continued use,
whereas a low activity catalyst
probably needs replacement:
In Figure 8, the relativity between
SOR and EOR is half. This is illustrated at the 93% conversion level,
with SOR achieving this conversion
at 3000 GHSV and EOR achieving
this conversion at 1500 GHSV.
With respect to determining the
state of the catalysts health, note
that conversion of 93% can be
achieved by a catalyst at all three
health stages depicted, depending
on space velocity (tail gas loading).
Load matters because the same bed
performance comes from a catalyst
with good activity and plenty of
life remaining, mid-life catalyst at
GHSV = 2100 1/hr, or from a
ready-for-replacement EOR catalyst
loaded at GHSV = 1500 1/hr.
Alternatively, at 2100 1/hr
GHSV, conversion ranges from 87%
to 98% depending on activity state.
Table 5 shows the rst order activity coefcients needed to achieve the
indicated conversion values at various space velocities. The required
activity coefcients are a function of
catalyst property and temperature.
As the table shows, space velocity
impacts unit performance, a higher
space velocity requiring a proportionally higher activity for a give
conversion. Lower space velocity
may be achieved by adding more

90

80

70
1500

SOR
Mid-run
EOR
2000

2500

3000

3500

Space velocity, GHSV 1/h

Figure 8 Reactor conversion vs space velocity (actual)

catalyst or lowering the gas rate at

end-of-run.
This is a closing point on the
issue of a combination impact of
equilibrium and conversion. When
CO conversion declines, be it from
catalyst activity decline or space
velocity increase, the outlet CO
concentration rises; it is a sum of
equilibrium residual plus unconverted material. This means that
CO2 and CO are not in equilibrium,
and the COS equilibrium coupled
to those compounds will not be
equal. The ultimate COS level is
determined by parallel and sometimes competing pathways, such as
hydrolysis and sour shift (hydrogenation).
A
higher
than
equilibrium CO level elevates the
COS equilibrium as expressed via
sour shift. The COS equilibrium
value the system feels will be
between that expressed by hydrolysis with CO2 and sour shift from
CO. Therefore, there is a knock-on

Delivered CoO
oxide form MoO3

Co
Mo

Catalyst sulphiding

In order for a catalyst to realise its

full capacity for activity, it is essential that the catalyst be properly
loaded and activated. The TGU
reactor catalyst, consisting of
molybdenum and cobalt supported
on an alumina substrate, is manufactured in a metal oxide form.
Before the catalyst can be utilised
in the TGU reactor, it must be
made active by converting the
metal oxide sites to metal sulphide
sites. Figure 9 depicts the sulphiding and reduction pathways. A
sulphiding pretreatment step is
required to activate the catalyst.
During this step, the catalyst is
heated in a controlled fashion to
600F (315C) in an H2 and H2S
environment.
The
sulphiding

Sulphiding
200C-315C (390F-600F)
in H2 and H2S environment

Reduction >200C (390F)

in H2 environment
with no H2S

Reduced
bad S/U

effect on COS due to an elevated

CO resulting from a weak conversion of CO. Thus, as goes CO, so
goes COS.

CoS
MoS2

Active
form

Not possible due

to metals
agglomeration

Inactive form

Figure 9 CoMo catalyst states

PTQ Q4 2014 27

10/09/2014 12:49

KALDAIR

jzink.indd 1

10/09/2014 11:49

pretreatment may be performed in

either an in situ or ex situ fashion. It
is important to perform sulphiding
properly because the activity of
hydrotreating catalysts is known to
be sensitive to the activation procedure.
During
the
initial
presulphiding of the TGU catalyst,
it is of particular importance that
the reactor circuit be air-free in the
presence of H2 and H2S. An inuence on activity is also observed for
the temperature at which the catalyst is sulphided.

TGU catalyst deactivation

mechanisms

Hydrothermal aging, the loss of a

catalysts surface area and metals
activity, caused by exposure to
water vapour at normal TGU operating temperatures, is the primary
deactivation mechanism in a stably
operated TGU. In typical tail gas
units, the decline in catalyst activity
due to hydrothermal aging is gradual. TGU in service with air-blown
Claus typically achieves eight years
of catalyst life, and units have experienced as long as 12 to 14 years.
Units in service with high-basis,
oxygen-blown Claus may be
expected to provide six to eight
years life.
Misoperation can result in severe
and rapid losses in catalyst activity.
Primary TGU catalyst poisons are
carbon and oxygen. Catalyst
coking, primarily from poor burner
operation, is one of the two
common causes of catalyst poisoning. Catalyst activity is reduced by
coke blocking access to and covering the active sites on the catalyst.
Additionally, catalyst coking can
cause high pressure drop across the
catalyst by reducing the void fraction in the catalyst bed. Small
quantities of oxygen over a period
of time can deactivate the catalyst
by blocking the active sites through
a sulphation reaction. Larger quantities of oxygen that enter the TGU
reactor can seriously damage the
catalyst due to temperature excursions that can sinter the catalyst
support. Exposure to hydrogen
without H2S for extended periods
of time at temperatures greater
than 400F (200C) can result in
reduction of catalyst to base metals.

www.eptq.com

criterion.indd 6

Tin

Layer
depth
75mm

Inert support
T1
T1
Main bed
Main bed
T1
T1
T1
TGU catalyst
TGU catalyst
T1
T1
T1
Inert support
T1

950mm

75mm

Support grid
Tout

Figure 10 TGU reactor diagram

Determining TGU catalyst health

TGU catalyst health can be determined in several ways. Indicators

from which we can infer performance include:
Temperature prole observed in
the reactor bed
Changes in conversion, indicated
by increased incinerator emissions
(SO2, maybe CO)
Activity evaluation from reactor
inlet and outlet stream composition, determination usually by unit
testing and analytical evaluation
results
Measuring physical properties
and/or activity testing for an actual
catalyst sample.
The temperature prole in the
reactor bed can yield a simple
answer: if temperature rise has
shifted into the mid and lower
portion of the bed, the catalyst has
poor activity and it is time to
replace it. The incinerator has a
likewise simple answer: if emissions are high but vent gas H2S
level and the pit are normal, then
likely the catalyst has poor activity
and it is time to replace it. A catalyst assay activity test gives a
denitive answer, but it is usually
not available until the catalyst has
already been removed from the
reactor. The challenge is establishing activity for the not obviously
dead cases before replacement is
under way.
Catalyst activity evaluation from
unit performance testing is done
when the unit is still running, so
assessment is timely. However, the
activity determined for the catalyst
may be less certain, as is the need

to replace it. Past practices that

attempt to assess activity with
online test data do not provide
quantitative measures for catalyst
activity, in part because many of
them do not include tail gas
loading.
Maximising service life from the
catalyst and obtaining best tail gas
unit performance means avoiding
conditions that deactivate the catalyst. Conditions to avoid are
excessive temperature (over 700F),
or a reducing atmosphere before
sulphiding without H2S, exposure
to oxygen, accumulation of soot or
particulates, or exposure to heavy
hydrocarbons or aromatics. High
percentage water in tail gas (50%)
accelerates hydrothermal aging and
inhibits catalytic activity by occupying much of the catalytic surface.
The kinetic reaction model
provides a measure of good quality, granular catalyst activity from
compositional data provided by
online performance test data. This
approach explicitly includes tail gas
loading and its impact on catalyst
bed performance. Real performance
can be discerned and, more importantly,
future
performance
capability can be anticipated. A
timely and appropriate catalyst
change can be planned. A meaningful comparison can be made across
time for tail gas reactor performance if all tests are conducted at
the same temperature.

Temperature profile example

In TGU reactors with two or more

levels of temperature indication
(see Figure 10), the temperature

PTQ Q4 2014 29

10/09/2014 12:49

choose wisely.
CRI Catalyst Companys global resources in research and development (R&D),
manufacturing, and surface and materials science establishes CRI as a desirable choice
for custom catalyst development. CRI provides the tools to develop and progress your
custom catalyst project from lab scale to full commercialization. Our customers are
continuously striving to improve their production processes, provide better products,
develop leading-edge technologies and maintain consistent operating results. In many
cases, a key to achieving these goals is the development of custom catalyst products,
designed specifically for individual customer applications/processes. CRI works with
customers to create and/or identify specific catalysts to help achieve their program goals.

It is all part of our commitment to delivering innovation.

cricatalyst.com

cri.indd 1

10/09/2014 11:39

80
70

Temperature, F

60
50
40
30
20
10

0
40
0
50
0
60
0
70
0
80
0
90
0
10
00
11
00
12
00
13
00
14
00

30

20

10

prole provides a simple but telling

answer: if the majority of the reaction exotherm has shifted into the
middle or lower portion of the TGU
reactor bed, this indicates catalyst
activity has declined and it may be
time to replace it. In many TGU
reactors, catalyst temperatures are
measured in three zones: top
(~25%), middle (50%) and bottom
(~75%). In addition, reactor inlet
and
outlet
temperatures
are
measured.
Twenty-four hour averages of the
reactor inlet and outlet temperatures, as well as the temperatures
within the reactor, are useful to
construct two assessment plots. The
rst plot is used to track the total
temperature differential across the
reactor. The preferred manner to
calculate this differential is to calculate the difference between the
average top layer temperature and
the reactor inlet temperature, the
difference between the average
middle layer temperature and the
average top layer temperature, and
the difference between the average
bottom layer temperature and the
average middle layer temperature.
These three values are summed to
calculate the total temperature
differential across the reactor. The
outlet temperature is usually not
used as it is often lower than the
bottom zone temperatures in the
hydrogenation reactor due to
outside path heat losses and recombination at the outlet:

Days on stream
Figure 11 Total reaction exotherm, Customer 1

Criterion recommends that unit

engineers or process specialists
periodically record and plot the
total temperature differential across
the reactor as well as the percentage contribution to the total
reaction exotherm by reactor catalyst zone top, middle and bottom
from the beginning of the cycle.
This technique is often helpful to
determine effect on catalyst health
after a unit upset or an unplanned
shutdown.
Figure 11 shows a plot of the total
reaction exotherm for a TGU in
operation for 1350 days (44
months). The unit engineer who
provided the plot wanted to know
if the TGU catalyst could provide

DT = (Avg Top TempReactor Tin)+(Avg Mid

TempAvg Top Temp)+(Avg Btm TempAvg
Mid Temp)

an additional two to three years of

cycle life, hopeful that she could
report to management the turnaround planned for six months
ahead could be postponed for at
least another couple of years.
When more detailed temperature
information was received, a plot
showing the percentage contribution per reaction zone to the total
reactor exotherm was generated
(see Figure 12,). This plot indicates
that over the rst 450 days of operation, activity in the top catalyst
zone declined by approximately
half. During this same period, the
middle zone catalysts contribution
to the total reaction exotherm
doubled. Around day 950, the

Bottom zone

80

Middle zone

Top zone

50
40
30
20
10

00
13
00
14
00

00

12

11

00

10

90

80

70

60

50

40

30

20

Middle Zone % = (Avg Mid Temp Avg Top

Temp) / DT

60

10

Top Zone% = (Avg Top Temp Reactor Tin) /

DT

Temperature, F

70

The second plot tracks the change

in percentage contribution of the
various reaction zones to the total
reactor temperature differential
throughout the cycle. It is calculated as follows:

Days on stream
Bottom Zone% = (Avg Btm Temp Avg Mid
Temp) / DT

www.eptq.com

criterion.indd 7

Figure 12 DT contribution per reaction zone, Customer 1

PTQ Q4 2014 31

10/09/2014 12:49

Were Not Exaggerating. Our Boilers

are something to brag about.
We custom design and custom build boilers to perform efficiently, safely and cleanly.
Your RENTECH boiler will lower operating costs, reduce emissions, and provide faster start-up
and cool-down. Youll find satisfied customers on six continents with specialty boilers, HRSGs,
wasteheat boilers and fired packaged watertube boilers from RENTECH. Weve been designing
and building boilers for people who know and care since 1996.
WWW.RENTECHBOILERS.COM

rentech.indd 1

10/09/2014 11:54

www.eptq.com

criterion.indd 8

Zone 1

90

Zone 2

Zone 3

80
70

Rx, %DT

60
50
40
30
20
10

0
10
0
20
0
30
0
40
0
50
0
60
0
70
0
80
0
90
0
10
00
11
00
12
00
11
00
12
00

Days on stream
Figure 13 DT contribution per reaction zone, Customer 2

reactor performance. By employing

this practice, an operator of a large
TGU was able to gain condence to
extend a 10-year cycle length to 13
years and counting. In doing so,
they have deferred several million
dollars in catalyst and associated
maintenance turnaround costs.
Analysis of catalyst samples is an
effective gauge of the health of the
TGU catalyst. The primary tests
performed on the sampled catalyst
include: surface area, carbon on
catalyst, crush strength and, if the
catalyst sample is collected and
stored inertly, X-ray photoelectron
spectroscopy (XPS). Comparing the
results of the used catalyst to fresh
catalyst helps predict remaining

in the midst of an operating cycle.

Periodically
collecting
and
analysing hydrogenation reactor
feed and efuent streams is also a
useful way to gauge catalyst health.
Collecting and analysing representative feed and outlet gases from the
hydrogenation reactor can be quite
challenging and is not often done
by TGU operators. Companies such
as Brimstone have the expertise to
provide this service. The total cost
associated for a maintenance shutdown for large tail gas units is
often several times the cost of the
TGU catalyst. Some TGU operators
bring in a company like Brimstone
several times during the TGU catalyst cycle life to benchmark TGU

Fresh
West top
West middle
West bottom

East top
East middle
East bottom

Intensity

second zone starts to decline and

the third zone contribution starts to
increase. Figure 12, demonstrating
the change in reaction exotherm
distribution within the TGU reactor, provides a deeper level of
discernment than Figure 11 which
showed only the total reaction
exotherm. A cursory examination
of Figure 11 might have led to an
incorrect determination of the TGU
catalyst health. The majority of the
reaction exotherm is generated
from the hydrogenation of SO2 and
Sx and sometimes the water gas
shift reaction of CO (CO heat
release is dependent on the CO
feed concentration to the TGU reactor). The hydrogenation of SO2 and
Sx proceeds to completion in a
much faster manner than the
slower hydrolysis reactions. By the
time hydrogenation reactions have
shifted lower into the catalyst bed,
conversion of the slower hydrolysis
reactions, particularly COS conversion, has typically undergone a step
change reduction.
In a stably operated TGU, it is
not an unreasonable expectation
that the majority of the reaction
exotherm remains in the upper
portion of the TGU catalyst bed for
the majority of the operating cycle.
This is demonstrated in Figure 13.
The reaction exotherm shifting
lower into the reaction bed is indicative of diminution of catalyst
activity either through normal
hydrothermal aging or a sudden
operational upset. For example, a
three-foot catalyst bed essentially
becomes a two-foot catalyst bed.
The upper portion of the catalyst
bed no longer catalytically contributes to the reactions, resulting in
effectively higher GHSV. Thus, a
2000 GHSV bed design becomes a
3000 GHSV bed and conversion is
correspondingly lower.
Of the four indicators listed
above, activity evaluation and
measuring
physical
properties
potentially provide the most
conclusive information with regard
to TGU catalyst health. Due to cost
and/or difculty, most TGU operators avoid both online sampling/
analysis of the TGU reactor feed
and outlet streams and collecting
TGU catalyst samples for analysis

Binding energy, eV
Figure 14 XPS analysis of spent TGU catalyst

PTQ Q4 2014 33

10/09/2014 12:49

C-534 properties fresh and end-of-run

Properties
Fresh catalyst End-of-run
Surface area, m2/g
300
120
Crush strength, N/bead 100+
<30
Carbon, wt%
0
0.9

Table 2

catalyst life/activity as well as

providing clues to potential unit
upsets or misoperation that may
have occurred during the cycle.
Table 2 shows fresh and end-of-run
(EOR) characteristics for Criterion
C-534 conventional TGU catalyst.
An example of an application of
XPS is shown in Figure 14. A
customer with Criterion TGU catalyst, ex situ activated via Eurecats
Totsucat G process, inertly loaded
the fully activated catalyst. The
customer put the bed under a nitrogen blanket for three weeks prior to
starting the TGU. Upon introduction
of SRU tail gas, the unit experienced
SO2 breakthrough with the quench
water pH dropping below 3. The H2
concentration in the reactor outlet
gas stream was approximately 5
mol%. After making several operational adjustments, the TGU was
shut down and it was decided that
the catalyst would be replaced.
During the subsequent investigation, two events that might have
impaired catalyst activity were
identied. First, it was discovered
that, at the midpoint between catalyst loading and start-up, readings
from the thermocouples in the TGU
reactor increased from ambient

Figure 15 Oxide, sulphide and sulphate forms of TGU catalyst

temperature to 250F (120C).

Second, upon introduction of SRU
tail gas, the in-line heater operation
may not have been sub-stoichiometric. One or both of the events
could have impaired catalyst activity by sulphation.
Spent catalyst samples collected
during inert unloading of the TGU
reactor were shipped to Criterion in
air-tight sample containers for testing. The unit engineer also
provided a retain sample of the
Totsucat G processed catalyst
loaded in the TGU reactor. Figure
14 shows the S 2p region. The location and intensity of the right peak
for the fresh catalyst and west
bottom samples indicate that these
samples had signicant amounts of

sulphur present in the sulphide

form. The location and intensity of
the peaks for the east top, east
middle, east bottom, west top, and
west middle samples indicate that
the sulphur is almost exclusively
present in a sulphate form.
Figure 15 shows the oxide,
sulphide and sulphate forms of
Criterion C-534 TGU catalyst: the
blue coloured material at the top is
the oxide form; the black coloured
material is the sulphide form; and
the grey coloured material is the
sulphate form. The customer
mentioned that the appearance of
the Totsucat G processed TGU catalyst loaded into the reactor was
consistent with the sulphide form
shown in Figure 15. The appearance
of the catalyst that was unloaded
from the reactor was consistent
with the sulphate form in Figure 15.
A photograph of the customers
catalyst is shown in Figure 16.
Additionally,
TGU
catalyst
samples can undergo activity testing
versus fresh catalyst in a TGU pilot
plant. During TGU shutdowns,
especially planned shutdowns, it is
recommended to collect catalyst
samples for analysis. Even if the rst
instance that catalyst samples may
be caught is when the catalyst is
being replaced, useful benchmarking information may be obtained.

Conclusions
Figure 16 Sulphated TGU catalyst

34 PTQ Q4 2014

criterion.indd 9

Equilibrium and kinetic effects can

www.eptq.com

10/09/2014 12:49

be summarised as follows:
At low temperatures, performance is dominated by kinetics;
equilibrium does not matter very
much since all equilibrium values
are very low
At conventional temperatures,
kinetics is a stronger inuence, and
hence catalyst kinetic property is
important, although equilibrium
back pressure sets a lower boundary for performance
At high temperatures, performance is set by equilibrium if
catalyst activity is fair and space
velocity is reasonable
Catalyst performance is critical to
achieving high sulphur recovery
and the performance demanded by
todays environmental regulations
Conversion of sulphur species to
H2S in the reactor is affected by
reactor performance and catalyst
activity required
Conversion required is established by the inlet concentration of
contaminants and environmental
performance requirements
Space velocity (gas load) and

catalyst inventory impact conversion

Temperature (absolute temperature) affects equilibrium back
pressure and catalyst activity
COS is handicapped by high CO
via sour gas shift, elevating outlet
concentration
To cope with damaged catalyst,
the temperature can be increased or
the space velocity reduced

Analysis of spent TGU

catalyst can provide
useful information on
TGU operation and
benchmarking
Plotting TGU reactor temperature
proles can provide an indication
of catalyst health while the TGU is
in operation
Activity evaluation from analysis
of reactor inlet and outlet stream
composition can provide information to defer catalyst change-outs

Elemental Analysis
of Fuels and Oils

Analysis of spent TGU catalyst

can provide useful information on
TGU operation and benchmarking.
Michael A Huffmaster is a process expert
and consultant to industry for gas processing
and treating, refining operation, CO2 capture,
and related research. His activities regarding
sulphur recovery include amine treating,
Claus, tail gas treating, and tail gas treating
catalyst development, design and operation.
He retired from Shell Oil in 2005 with 36 years
of experience. He holds a bachelor of science
degree in chemical engineering from Georgia
Institute of Technology and is a registered
professional engineer in Texas.
Email: michael.huffmaster@att.net
Fernando Maldonado is the Business
Development Manager Gas Treating Catalysts
for Criterion Catalysts and Technologies,
located in Houston. He has global responsibility
for Criterions gas treating catalyst business.
Prior to joining Criterion Catalysts &
Technologies in 2001, he held positions as a
process design engineer, unit contact engineer,
and an operations superintendent in two US
Gulf Coast refineries. He holds a bachelor of
science degree in chemical engineering from
Texas A&M University.
Email: Fernando.Maldonado@CRI-Criterion

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.

Determination of Sulfur and other

elements at-line and in the laboratory
Discover more exciting details at
www.spectro.com/fuels

www.eptq.com

criterion.indd 10

PTQ Q4 2014 35

10/09/2014 12:50

j matthey.indd 1

11/09/2014 10:02

Predicting future FCC operations

via analytics
Data analysis enables a company to take advantage of patent information to unveil
underlining development trends for formulating technology strategies for future markets
PATRICK J CHRISTENSEN, TOUSEEF HABIB, THOMAS B GARRETT and THOMAS W YEUNG
Hydrocarbon Publishing Company

merican
writer
Mark
Twains famous misquotation, Reports of my death
have been greatly exaggerated
may well be applied to the current
status of refinery fluid catalytic
cracking technology and certain
gasoline-centric FCC units operated
in parts of the world.
Declining
gasoline
demand
caused by bioethanol mandates and
improving vehicle fuel efficiency,
poor fuel demand because of a
weak European economy, rising
shale/tight oil processing in the US
resulting in growing production of
gasoline and naphtha, and diminishing outlets for naphtha as more
steam crackers source cheaper
ethane and propane feedstock are
the major reasons for concerns
among refiners, technology holders,
and catalyst producers. On the
other hand, the FCC unit has been
continuing to perform miracles for
refiners in light of increasingly
stringent fuel standards, changing
market conditions, and competing
technologies. Its role has expanded
from a gasoline machine to an
olefins maker, a sulphur remover, a
residue upgrader, and a ULSD
feedstock contributor by maximising LCO output. Furthermore, it is
lending itself to two additional
roles biofeeds user and refinery
CO2 emissions reducer to alleviate
growing concerns over energy
security and global warming.
So, the question is: what is the
future of FCC operations? One way
to predict is by looking at where
technology companies have been
investing in research and development, especially patent applications.
This article analyses the patenting

www.eptq.com

hpc.indd 1

trends of FCC technology around

the world during the years 2008 to
2013, reviews the significance of the
data, and explains how the analytics can be used to predict the future
business opportunities and challenges for refiners and technology
developers alike.

Big data analytics

Analytics, which is a term referring

to the search for and use of patterns
in data, has been applied in a variety of industries to help businesses

Analytics has
become a powerful
management tool
for refiners and
vendors to establish a
competitive edge
gain a competitive edge. Marketing
agencies, sales companies, and even
sports teams have turned to
analytics.
In the case of analysing patent
data, the authors of a management
handbook entitled Strategic and
Competitive Analysis: Methods and
Techniques for Analyzing Business
Competition said, Patent analysis is
a unique management tool for
addressing the strategic management of the firms technology and
product or service development
process. Translating patent data
into competitive intelligence allows
the firm to gauge its current
competitiveness, to forecast technology trends, and to plan for

potential competition based on new

technologies.
Analytics has become a powerful
management tool for refiners and
vendors to establish a competitive
edge. A generalised combination of
external and internal patent analysis
can help companies assess their
technology portfolios and directions
in the context of the marketplace
and enable them to strategically
position their technologies, particularly at the time of changing crude
slates and shift in product demand.
Also, it is possible to identify areas
that are not profitable and focus
R&D investment elsewhere.
In analysing refining technologies, FCC seems to be a good place
to start, because it produces high
volumes of fuels and petrochemical
feedstocks. In addition, the FCC
unit is considered to be the workhorse of the refinery.

Trends and focus of FCC

patent activities

According
to
Hydrocarbon
Publishings database of patent
literature for FCC there are 496
unique patents and patent applications with issue dates during the
period 2008-2013. There are a
number of different dates that can
be associated with a given patent:
filing date, issue date, priority date,
and expiration date. The issue date
denotes when the patent or patent
application was published by the
patent office.

Annual global patent counts

Figure 1 shows the annual global

counts for FCC patents issued
during the period 2008-2013. We
see that this count dropped sharply

PTQ Q4 2014 37

10/09/2014 12:55

160

Regional differences

Figure 2 shows a regional breakdown of the global FCC patents

issued from 2008-2013. The five
regions shown provided 491 of the
496 FCC patents issued during the
period. Four patents had insufficient information to establish their
origins and one patent was from
Russia. Based on this data, North
American companies had 42% of
the issued patents, and Asian
companies were somewhat lower at
38%. European, South American,
and Middle Eastern companies
contributed 12%, 4%, and 3%,
respectively.
Almost all of the Asian patents
(169 out of 186) originated from
Chinese or Japanese companies,
while US companies provided 206
of the North American patents, the
other two coming from Mexico. As
for South America, 20 of the patents
have Brazilian origin, with two
coming from Colombia. The majority (36) of the European patents
was issued to companies in the
Netherlands, and almost all (15) of
the Middle Eastern patents are
from Saudi Arabia.

146

Number of patents

140
120

104

100

91
76

80
60

41

38

2012

2013

40
20
0

2008

2009

2010

2011

Figure 1 Annual global counts of issued FCC patents, 2008-2013

250

Number of patents

208

200

186

150
100
58

50
22

North
America

South
America

17

Europe

Middle
East

Asia

Company comparison
Figure 2 Regional counts of issued FCC patents for 2008-2013

from 2008 to 2009,

somewhat in 2010
through 2011 to
the count in 2013
of that in 2008.

then recovered
but fell again
2013. Overall,
was just 26%
Therefore, the

general picture for FCC patents

over the period 2008-2013 is that of
a much lower level of activity at
the end of the period than at the
beginning.

100

96

Number of patents

90
80

75

70
60
50
40
30
20

16
13 13 15

24

27 29

Id To
em ta
its l
u
I
C
In J FP
hi
d
na C ia GC
o
U smn O
ni o il
v. O
P il
et
ro
l
B
N
P
ip S P
et
po he
ro
ch
n ll
P
O
in
et
a KB il
ro
C R
lE
om
ne In
rg ter p.
y ca
C t
e
In
B nt
de A AS .
r
pe am F
P nd co
e e
A tro nt
Ex lbe br s
xo m as
nMarl
e
o
G b
Si ra il
no ce
pe
U c
O
P

10 10 10 11 12
10 5 5 7 8 8 9 9 9
0

20

Figure 3 Company counts of issued FCC patents for 2008-2013

38 PTQ Q4 2014

hpc.indd 2

Globally, 67 companies received

issued patents for FCC technologies
during 2008-2013. Figure 3 shows
the identities and the counts for
those companies that received five
or more patents. Twenty-three
companies are in this group, and it
is easily seen that two of these
UOP and Sinopec readily stand
out from the rest in terms of the
number of patents awarded.
Clearly, these two companies by
themselves contribute heavily to
the regional counts for North
America and Asia that are shown
in Figure 2. Also, their combined
total of 171 patents is more than
one-third of the global total of 496.

Specific applications

Figure 4 shows the number of global

FCC patents with issue dates during
2008-2013 that are found in each of
the 17 application categories. These
cover almost all of the subjects to
which FCC patents apply. It must
be noted that a patent may fall into
more than one of these categories.

www.eptq.com

10/09/2014 12:55

Merichem Company-FY14-MID.pdf 1 2/24/2014 8:19:32 AM

Midstream Solutions
Proven Results

Midstream Solutions with Industry-Leading Speed to Market

To win a race you need not only speed, but a timely start. Merichem Company has a proven track record
of providing hydrocarbon treating solutions that meet the stringent timelines of our customers. Get your
treating project off to a timely start by choosing Merichem Company as your technology licensor.
C

CM

Merichem Companys portfolio of midstream treating technologies

MY

includes our patented FIBER FILM Contactor that offers a variety of

CY

caustic, amine and acid treating processes. Merichem also offers

CMY

solutions for gas treating, including our patented LO-CAT wet scrubbing,

liquid redox system that converts H2S to innocuous, elemental sulfur. In

addition to our large portfolio of licensed treating technologies,
Merichem offers industry leading spent caustic management services that
utilize Beneficial Reuse options to create a non-waste solution for your
spent caustics. For fast, effective treating solutions, choose Merichem.

Think Fast...Think Merichem

www.merichem.com | 5455 Old Spanish Trail | Houston, Texas 77023 | 713.428.5000

Copyright 2013 Merichem Company

merichem.indd 1

28/02/2014 10:28

17 2013

30

29

28

27

26

25

24

23

22

21

20

19

18

17

16

15

14

13

Big changes

SOMETIMES START
SMALL.
CATALYST SOLUTIONS THAT DELIVER VALUE:
CLARIANT CATALYSTS.

Our products are small, but they deliver big value. Use our
high-performance catalysts to make more of what you want
and less of what you dont, all with less energy. We are ready
for your big challenges.

12

11

10

Trust our long experience

from predecessor company
Sd-Chemie.

WWW.CATALYSTS.CLARIANT.COM
5

centimeter 1:1

claraint.indd 1
3088_Catalysts_A4_en.indd 1

10/09/2014 11:36
03.06.14 14:37

www.eptq.com

hpc.indd 3

140
123

Number of patents

120

123

100
80
64

60

50

40
27

20

53

50
28

33

31

36
26

17

16

23
10

17

0
LC Ga
O so Ga
yi lin so
el e li
d re ne
En Lig and for yi
er ht /o ma eld
gy o r ti (1
l q o
m ef u n )
C
G a ins ali (1
at
n
t
H
al
S G ag yi y ( )
ys
R NO em em eld 1)
tf
es x
or
C
id emiss ent (1)
at
m
. p i io (
al
ul
ro ss ns 2)
ys
a
ce ion (2
t m tio
n
C
/p Bi ssi s ( )
an
at
r
ag
e ofe ng 2)
C al
em par ed (3
P ata yst
A a s )
ro ly s
du st ep F en dd tio (3
ct re ar ee t/a itiv n ( )
fr ge ati d dd es 4)
ac n o in it (
tio era n/s jec ion 4)
na tio tri tio (5
t n p n )
P ion /c pin (5
ro / oo g )
ce re li (5
ss co ng )
co ver (5)
nt y (
ro 5)
l(
5)

Consequently, the sum of the

numbers shown at the tops of the
columns in Figure 4 is greater than
the actual number of patents that
are represented. These categories
can be sorted into five general
groups. Where an application is
placed in these groups is indicated
by the number in parentheses at the
end of the category name.
Group 1 concerns FCC product
yields and qualities, and the four
application categories here appear
207 times in the FCC patents issued
during 2008-2013. The foremost
category at 123 counts is that of
light olefins yield. Please note that
a small number of patents deal
with producing aromatics (BTX)
from non-conventional catalytic
cracking processes, and these
patents are included in the sections
that cover light olefins yield. The
three remaining categories in this
group, which cover fuels, have a
combined total of 105 counts, with
gasoline reformulation making up
almost half of them.
Group 2 covers energy and the
environment. The three categories
here show up 112 times in the 20082013 FCC patents. The leading
focus for patents here has been
SNOx emissions. This was a subject
of 2008-2013 FCC patents almost
four times as often as was greenhouse gas (GHG) emissions, and
about twice as often as was energy
management. Note that the SNOx
application covers emissions of
SOX, NOX, and particulate matter
(PM).
Group 3 is for FCC alternative
feed sources. There are only two
categories here, and they appear 59
times in the patents for 2008-2013.
The category of resid processing
appears about one-fourth more
often in the 2008-2013 FCC patents
than does that of biofuels.
Group 4, catalysts and additives,
includes patents in which formulations and preparation schemes for
these materials are covered. Of
course, most if not all of the patents
that can be included here will fall
into other groups as well. Together,
additives and catalyst formulation/
preparation are cited as applications 172 times in the 2008-2013
FCC patents.

Figure 4 Applications for 2008-2013 issue dates

Finally, Group 5, FCC operations,

contains categories for FCC process
stages and control. The process
control application is made up of
advanced process control, monitoring, and simulation/modelling.
Together, the six categories in this
Group appear 155 times in the
collection of FCC patents issued
during 2008-2013.

One can uncover

competitors
strategies and
future technology
deployment by
looking at where a
company is applying
its patent efforts
Patent mining

A total of 496 unique patents offers

a good size for data analysis. There
are many ways to delve into the
materials and use different metrics
to bring out the intrinsic values for
the research works. For illustration
purposes, the following describes
three of several metrics (or so-called

levels of mining) that have been

explored.
The first metric explores technology competition. One can uncover
competitors strategies and future
technology deployment by looking
at where a company is applying its
patent efforts. Often, R&D fundings
are allocated to experimental works
that satisfy customers needs (for
example, iron trap to reduce catalyst poisoning when processing
shale/tight oil in the US), capitalise
business potentials recognised by
market development executives
(such as increasing benzene yield in
FCC product to anticipate benzene
supply shortfall), and complete the
companys product portfolio (such
as cracking biomass).
For example in the Group 1
applications, FCC product yields
and qualities, 35 companies have
patents with issue dates in 20082013. The 23 that have two or more
patents are shown in Figure 5. UOP
and Sinopec are dominant players
with respectively 37 and 34 issued
patents. They are followed by
Petrobras (14), Grace (13), and
ExxonMobil and Saudi Aramco
(12). Despite FCC being considered
a developed technology, vendors
like Grace and UOP still find
opportunities in innovation. The
same can be said of refining companies such as Petrobras, Saudi

PTQ Q4 2014 41

10/09/2014 12:55

50

Number of patents

45
40

37
34

35
30
25
20
15
10
5

10

14
12 12 13

6 6 6
4 4 4 4 4 5 5
2 2 2 2 3 3

ShEas
an t C
gh h
ai ina
D U
on ni
gh v.
ua
IF
Sh P
el
EN l
R J I
SK eli GC
a
En nc
C P
hi et
er e
N
na ro
ip gy
c
p
U h
ni in T on
v. a o
P Co ta
et m l
ro p
P
le .
et
um
ro
IN
le
um
D
IO
En A
C
l
K
b
er e B
gy m R
C arl
e
A ent
ra re
m
c
XO o
P Gr M
et a
ro c
Si br e
no as
pe
U c
O
P

Figure 5 Company patent counts for Group 1 applications with issue dates during
2008-2013

Aramco and Sinopec. These companies outperformed their peers in

investing FCC R&D efforts in
boosting product yield and quality,
and data also show their strategic
focus and direction in the future.
(Note: the label IND in Figure 5
represents patents issued to independent inventors.) So, what does
this mean to the competitors of
Grace,
ExxonMobil,
Petrobras,
Saudi Aramco, and Sinopec?
A second metric goes further into
innovation concentration. The goal
is to find out development trends
and which companies are involved.

To go deeper into the analysis,

Figure 6 shows for companies in
Figure 5 the distribution of patent
coverage over the four application
categories in Group 1 gasoline
yield, reformulated gasoline, LCO
yield and/or quality, and light
olefin yield. The most popular category appears to be light olefins
yield. It dominates as the subject in
the patents issued to UOP, Sinopec,
Petrobras, Saudi Aramco, KBR, and
others. On the other hand,
ExxonMobil, Albemarle, Eni, and
Shell lean towards LCO yield and/
or quality, and it is of some signifi-

40

Category count

35
30
25

Light olefins yield

LCO yield and/or quality
Reformulated gasoline
Gasoline yield

20
15
10
5
S U
P ino OP
et p
ro ec
b
G ras
P
ra
et
ro
c
le A XO e
um ra M
m
A Ene co
lb r
em gy
C
hi
ar
na
K le
U
B
ni
R
v.
IO
P
et I C
P ro ND
et le
ro u
ch m
in
T a
SK Nip ota
l
E po
R ne n
el rg
ia y
nc
JG e
Sh
C
an
EN
gh
S
Ea a
he I
i
St st Do I ll
on Ch n FP
e in gh
& a ua
W Un
eb iv
st .
er

Figure 6 Company distributions of patent coverage over Group 1 application categories

issue dates 2008-2013

42 PTQ Q4 2014

hpc.indd 4

cance for Petrobras and Total.

Gasoline yield gets most of the
patent attention at Petroleum
Energy Center and JGC, both of
which are Japanese organisations.
Finally, reformulated gasoline is the
major application category for
Grace and PetroChina, but it is also
a very important part of Sinopecs
portfolio of patents for Group 1
applications.
A third metric breaks down the
applications into the specific technology areas or themes where
patent coverage is being pursued.
The purpose of the analysis here is
twofold: providing an overview of
technology innovations in individual applications; and identifying
potential areas of patent infringement or collaboration opportunities.
For illustration purposes, Table 1
summarises the technology themes
that appear in the four applications
from Group 1 Application FCC
product yields and qualities.

Who can benefit?

R&D analytics presents tremendous

resources for refiners and vendors
alike, particularly R&D scientists
and engineers, commercial development
executives,
corporate
strategic planners, purchasers of
products and services, and technology licensees. The information
generated can be translated into
technology strategies for refiners
better preparing for the future, and
business opportunities for vendors
to help their clients meet operational and financial goals.
For refiners, this could provide
something of a deeper level of
analysis beyond what will come
from the licensing companies that
is mostly driven from a marketing
perspective. A broad knowledge
of patent directions and activities
helps refiners understand technology
trends,
availability,
and
details in order to make better decisions in selecting the right
technologies for their operations.
As for technology licensors, this
could be another way to track
competition.
For
example,
Company X is the leading commercial provider of dual riser FCC
technology, but Company Y with
no market position in the technol-

www.eptq.com

10/09/2014 12:56

UOP-Uniflex-Refinement-Redefined-Ad-A4_UOP Ad A4 size UOP6463 9/9/14 10:44 AM Page 1

refinement redefined

UOP Uniflex residue-upgrading technology yields 25% more

clean fuel, turning bottom of the barrel into top of the line.
Get the most from every barrel. The UOP Uniflex Process can double the diesel yield
of other residue-upgrading
technologies,
turning
bottom
of the
barrel
intoofmore
clean fuel, turning
bottom
ofthe
the
barrel
into
top
the

line.

black on your bottom line. The Uniflex Process is a high-conversion, commercially

proven technology that processes low-quality residue streams like vacuum residue into

of distillate
other residue-upgrading
technologies,
turning
theofbottom
of the barrel into more
high-quality
products leading to
refinery margin
increases
up to 100%.
Simply put, with the Uniflex Process from UOP, youll maximize production and profits
from everyproven
barrel. technology that processes low-quality residue streams like vacuum residue into

1914 - 2014

A Century of Innovation
in the Oil and Gas Industry

Looking for exclusive information on UOP Uniflex technology?

Scan the code to register for the UOP Portal.
2014 Honeywell International, Inc. All rights reserved.

uop refinement.indd 1

10/09/2014 12:11

Specific research areas in Group 1 applications FCC product yields and qualities
Application
Technology theme
Process
Short contact time
Gasoline yield

Dual riser or multiple reaction zones


Catalysts
Composition can reduce coke yield or

reduce dry gas and LPG

Patent assignees*
ExxonMobil, UOP
KBR, Luoyang Huazhi Petroleum Engineering, Nippon Oil,
PetroChina, China Univ. of Petroleum (CUP), Sinopec, UOP
Grace, JGC Catalysts and Chemicals, Nanjing Petrochemical,
Petroleum Energy Center Japan (PECJ)/Cosmo Oil, Reliance
Industries

Gasoline reformulation Process

Various innovations to improve octane, increase

propylene and BTX yield, and reduce gasoline

Converting and/or removing undesirable

components from cracked naphtha

Catalyst
Reducing gasoline sulphur

Improve gasoline octane

Sulphur-reducing additives

CDTech, Indian Oil, KBR, Nippon Oil, PetroChina, Saudi

Aramco, Sinopec, SK Energy, UOP
ExxonMobil, Sinopec, Luoyang Petrochemical

LCO yield and quality Process

Catalyst

CUP, ExxonMobil, Grace, Idemitsu Kosan, PECJ, Sinopec/RIPP

PECJ/Cosmo
Bharat Petroleum, East China Normal Univ. (ECNU), Grace,
Saudi Aramco

Multi-stage or multi-reactor process configurations ExxonMobil, Petrobras, Stone & Webster,

Integrated process to boost yield
Eni, Sinopec/RIPP
Using pre-coked catalyst to raise output
Total
LCO with low aromaticity
Albemarle, Petrobras
Yield of LCO > HCO
ECNU, Eni

Light olefins yield

Process
Multi-stage, multi-reactor or downstream reactor CUP, Indian Oil, KBR, Petrobras, Reliance Industries,

and hardware
Saudi Aramco, Nippon Oil, Shell, Sinopec/RIPP, Total, UOP

Co-processing heavier olefins
IFP, Sinopec, UOP

Unconventional approaches to make olefins
Indian Oil, Petrobras, Saudi Aramco, Sinopec/RIPP,

and BTX
SK Innovation, UOP

Catalyst management
UOP

Feed injection scheme
Slavneft, UOP

Enhanced recovery
Lummus Technology, Reliance Industries, Shanghai Donghua

Environment, SK Energy/Korea Energy Research Institute, UOP

Catalysts and Modifed ZSM-5 zeolite

additives
Catalyst mixture



New zeolite materials

*Partial list

CUP, Petrobras, Reliance Industries, Sinopec/RIPP, Total

Asahi Chemical, BASF, Grace, Petrobras, Saudi Aramco,
Sinopec-RIPP, UOP
Grace, Petrobras, Sinopec-RIPP, Shanghao Research Institute
of Petrochemical Technology, Univ. Valencia Politecnica, UOP

Table 1

ogy has been issued a number of

patents in this area over the past
several years. As part of competition analysis, one can correlate
R&D work with commercial offerings and product developments to
monitor and predict the readiness
of Company Y to enter the market.
Patent analytics on a global basis
allows technology companies to
focus on competitors located within
the same regional market but also
with an eye on external market
competition.
Most importantly, R&D personnel can track research trends based
on big data analytics, size up
competition, and compare technology development direction. One
can identify potential infringement
either by external companies on his
or her technology, or vice versa by
evaluating similar developments

44 PTQ Q4 2014

hpc.indd 5

being patented by other companies.

On the other hand, there may be
opportunities for potential collaborations if similar work is being
done.
Finally, one can identify emerging
trends to see where commercial
practices line up with recent
R&D undertakings. Furthermore,
incorporating some recent market
fundamentals in terms of fuel
and
petrochemical
feedstock
supply/demand trends and also
future
business
opportunities
could
provide
guidance
to
companies in formulating future
strategies.

Analytics: prerequisite for gap

analysis and SWOT analysis

The investments in R&D by refiners

and vendors reflect the current
refining business environment and

companies views of the future.

Technology
developments
are
being driven by numerous factors.
The first major factor is the need to
satisfy changing market conditions,
including declining demand for
gasoline in developed nations and
rising demand for diesel in developing nations. With increasing
NGL and tight oil used in the US,
supplies of propylene, butadiene,
and BTX from steam crackers are
expected to be in short supply.
Global fuel oil demand is declining
due to increasingly stringent
sulphur specifications of bunker
fuels.
The second major factor is
mandated environmental legislation requirements such as Tier 3
gasoline standards in the US to
lower gasoline sulphur from 30
ppm to 10 ppm nationwide and

www.eptq.com

10/09/2014 12:56

tighter controls on emissions of

NOx, SOx, PM, and perhaps GHGs.
Also included here is minimising
waste discharges.
The third major factor is the
global crude supply. This ranges
from opportunity crudes like
Canadian oil sands bitumen, extra
heavy oil from the Venezuelan
Orinoco Belt, and high TAN crudes
to light, sweet, and highly
paraffinic tight oils. In general,
technology suppliers seek to offer
better performance in catalysts and
process designs for FCC units to
improve yield and selectivity,
reduce energy usage, and increase
reliability and safety. All of this is
with the ultimate goal of achieving
superior operational efficiency and
economic benefits.
Some people have looked at
spending for R&D as being similar
to playing a slot machine in a
casino. If you continued to put
money into the machine, sooner or
later some payout would occur,
and one could try to increase the
frequency by putting money into

multiple
machines.
Throwing
money blindly at a problem is not a
wise strategy. Although R&D
investment
is
risky,
unlike
gambling on slots the risk here can
be managed. One way is to have a
comprehensive view of recent
patent activities and of the competitive
landscape
via
analytics.
Another is to identify the differences between whats patented and
what the industry needs or wants
through a gap analysis. Both of
these can help a company to undertake
a
SWOT
(strengths,
weaknesses, opportunities, and
threats) analysis and then formulate a R&D strategy that allies with
the company objectives in achieving financial goals and business
sustainability in light of fast-changing
environmental
legislation,
market dynamics, and global
competition.
This article is an adapted excerpt of a report,
Strategic Roles of Fluid Catalytic Cracking in
Refinery Operations: Predictive Analytics and
Gap Analysis to Identify Technology Challenges

Multi-CLIENT
STRATEGIC
REPORT

and Explore Future Business Opportunities, by

Hydrocarbon Publishing Company.

Reference
1 Fleischer CS, Bensoussan B, Strategic and
Competitive Analysis. Methods and Techniques
for Analyzing Business Competition; Prentice
Hall: Upper Saddle River, US, 2002.

Patrick J Christensen is Project Manager with a

BS degree in chemical engineering from Drexel
University.
Touseef Habib is Technology Analyst with a
BS degree in chemical engineering from the
University of Delaware.
Thomas B Garrett is Senior Consultant with a
BS degree in chemistry from Carnegie Mellon
University and a PhD degree in chemical
physics from Lehigh University.
Thomas W Yeung is Principal and Managing
Consultant He is a licensed professional
engineer in New York State and holds a BS
degree in chemical engineering from the
University of Wisconsin-Madison, a MS degree
in chemical engineering from the University of
Connecticut-Storrs, and a MBA degree from
New York University.
Email: info@hydrocarbonpublishing.com

STRATEGIC ROLES OF

FLUID CATALYTIC
CRACKING

IN REFINERY OPERATIONS

www.hydrocarbonpublishing.com/fccanalytics

PATENT ANALYSIS
GAP ANALYSIS
SWOT ANALYSIS

PREDICTIVE ANALYTICS

CAN IDENTIFY TECHNOLOGY

CHALLENGES & EXPLORE FUTURE
BUSINESS OPPORTUNITIES

HYDROCARBON PUBLISHING CO.

Translating Knowledge into Profitability

PO Box 661 Southeastern, PA 19399 (USA)
Phone: 610-408-0117 | mcreports@hydrocarbonpublishing.com

www.eptq.com

hpc.indd 6

PTQ Q4 2014 45

10/09/2014 12:56

CREATING ENGINEERING SOLUTIONS SINCE 1953

PARIS - BEIJING - BUCHAREST - BUZAU - DUBAI - HOUSTON - JOHANNESBURG - KUALA LUMPUR

LOS ANGELES - MUMBAI - RIO DE JANEIRO - SEOUL - ST PETERSBURG - TULSA

heurtey.indd 1

06/06/2014 12:31

Predicting reactive heavy oil process

operation
Characterisation of feed and product yields through component structures for
better understanding and prediction of operations
GLEN A HAY, HERBERT LORIA and MARCO A SATYRO Virtual Materials Group, Inc
HIDEKI NAGATA Fuji Oil Company Ltd

o understand and optimise

reactive heavy oil processes
encountered in refineries
there must be a strong knowledge
of the product yields and their
physical properties. Sometimes
property predictions within models
of the reacting material are important due to operational constraints.
The coke induction point,1 or point
at which solid coke begins to form
in heavy hydrocarbon mixtures, is
an example of the importance of
property predictions since many
unit designs need to take into
account solid precipitation if it
occurs. The simulation modelling
application discussed in this article
applies to the Eureka process where
fluidisation of reactor materials and
inhibiting the coke induction point
is essential.2
When catalytic or thermal cracking
simulation
models
are
developed, enough physics must be
encoded into the mathematical
development to accommodate for
the significant changes in both the
normal boiling point of produced
material and the associated molecular structures. Many different
models for the reactive chemistry
mechanisms required to model this
class of processes have been
proposed.
These
mechanisms
include the use of hydrogen donor
components, cyclic ring breaking,
dehydrogenation of saturated rings,
and cracking versus oligomerised
propagation of small to large molecules, to name a few. The required
basis for the development of a reliable
simulation
environment
designed to handle the type of
detail associated with chemical
reaction
mechanisms
requires

www.eptq.com

hay.indd 1

flexible component chemical structures to represent the products

from different chemical reaction
pathways. In order to capture this
type of behaviour directly within
the requirements of industrial
process simulation software a new
PIONA (n-Paraffin, Iso-paraffin,
Olefin, Naphthene, and Aromatic)
pure component basis environment
was developed to both characterise
the feedstocks and the resulting
products
estimated
chemical
make-up and yields.
The PIONA technique3 consists of
using constant groups also known
as slates of predefined compounds
required to cover the carbon
number ranges for feeds and products necessary for the modelling of
different refinery reactors, such as
the Eureka process thermal cracking
vessels.
The
different
combinations of these component
slates and the compositions of the
components within allows for the
matching of the experimental distillation curve of a given feed and the
calculation of its chemical characteristics, ranging from simple
properties such as molecular
weight and standard liquid density
all the way to more complex
physical properties such as heating
values, liquid viscosities, and pour
points. The key advantage in using
this method is its ability to capture
the essential chemistry of the feedstock and product mixtures and
how the changing compositions
upon reaction affect property
calculations. The number of components used in the simulation is
kept constant and consistency
is
enforced
throughout
the
simulations.

The PIONA structure group classification was found to introduce

an unacceptable property estimation error in studies when
modelling feeds with an average
carbon number higher than ten,
where larger aromatic content was
encountered. Further investigation
showed that a single aromatic
structure group was not enough
to
differentiate
multi-paraffin
branched aromatic components
against
those
more
reacted
compounds that were stripped of
straight carbon branches. Therefore,
an extra chemical type defined as
dehydrated aromatic is included
in the PIONA technique.3 From a
molecular structure configuration,
these dehydrated aromatics replace
branched contributions to a base
aromatic ring with additional dehydrated aromatic rings.

Heavy oil reaction models

Models using detailed specific reaction pathways are commonly found

in the technical literature on
thermal cracking or pyrolysis of
lighter gases such as ethane or even
naphtha feedstocks.4,5 This is possible because the overall number of
pure component and radical species
are still manageable within a simulation environment (typically less
than 150 species), assuming good
numerical techniques for the integration
of
the
differential
component material and energy
balances are employed. Even then,
many of these models have to be
linearised to help achieve faster
computational speeds.6 Once heavy
residual oil feedstocks are introduced, a change in mathematical
solution methods for the compo-

PTQ Q4 2014 47

15/09/2014 13:06

100

Heavy oil vacuum residue

100
80

100

60
40
20
0

Gas Liq Gas Liq Gas Liq Gas Liq

TBP1

TBP2

TBP3

TBP4

Mass flow

Mass flow

Mass flow

Offgas

Liquid pitch

80
60
40
20
0

80

Gas Liq Gas Liq Gas Liq Gas Liq

TBP1

60

TBP2

TBP3

TBP4

40
20
0

Note: TBP = True boiling point

Gas Liq Gas Liq Gas Liq Gas Liq

TBP1

TBP2

TBP3

TBP4

Figure 1 Basic lumped component heavy residue thermal cracking representation

nent material balances is noticed in

solutions using generalised, or
lumped reaction pathways7,8 due to
the sheer complexity of the feed
material. In these models, the heavy
residue thermal cracking reactor
might use only eight lumped
components, a dozen reaction pathways, and resolution of the feed
and product material balances
would look similar to the
representation in Figure 1.
This type of modelling relies on
the availability of a substantial
amount of experimental data which
limits the quality of results extrapolated from the model. This lack of
predictive power is mainly driven
by the non-mechanistic approach to
the reactions kinetic parameter
tting and eventual increase of
error outside of the tted data
range. Conversely, this approach

also presents advantages such as

fast solution speeds and its ability
to predict specic properties of
the generalised yield cuts such as
the softening point of the pitch and
the contained volatile matter7
through the use of lumped components
and
effective
lumped
properties and mixing rules for
these properties.
In these lumped models, the
resulting
temperature
proles
calculated for the furnace tubes can
also be roughly estimated due to
the matched enthalpy of formation
for each lumped component,
assuming the chemical structure
shifts stay consistent and that
enthalpies of combustion are available. Signicant error would be
introduced in these models if
hydrogenation versus dehydrogenation occurred since these

reaction pathways are exothermic

instead of endothermic. In that
respect, average boiling point
lumping used by these models is
completely
non-predictive
and
would
need
to
be
rebuilt
for specic processes and eventually even specic plants and
equipment.
The PIONA slate technique was
applied in the commercial process
simulator software VMGSim, to
represent multiple hydrocarbon
feedstocks in lighter and heavier
cut ranges.3,9 The focus was on
property calculations and characterisation of multiple feeds using
mixtures based on the same basis
component slate. The ability of this
PIONA system to model reactive
systems is best illustrated in Figure
2 when compared with the lumped,
linearised systems. In Figure 2, the

ISO paraffin Paraffin Olefin Naphthalene Aromatic Aromatic

(dehyd)

Light gas

60

rxn1
rxn4

Gasoil

rxn2

C1

rxn3
rxn5

Distillate

Naphtha
rxn7

FeedCut 2

rxn10

rxn11

Isomerisation

iC4

10
0

nC4

rxn12

Coke

C8

Cracked C8 distribution

Dehydrogenation

nC6

oC6

Hydrogenation

C7
Pitch

Isomerisation

C5
C6

20

he
ot

FeedCut 1

C4

nC3

30

ic
rm
he
ic
ot
rm

rxn9

40

Cracking
(distribution-based)

Ex

rxn8

C3

Propagation

d
En

rxn6

C2

50

nC2

Cyclisation

nC9

N8

Cyclisation crack

C9

Cyclic dehydrogenation

N9

A9

Adehy9

Cyclic hydrogenation

C10

Figure 2 Example of lumped kinetic versus PIONA kinetic thermal cracking reactive pathways

48 PTQ Q4 2014

hay.indd 2

www.eptq.com

10/09/2014 13:00

combustion leader

Callidus Technologies, LLC is an established world leader in

process burner, flare and thermal oxidizer technology.
We have earned that reputation by providing solutions that set new
standards in performance and reliability. Our approach to combustion
science continues to break new ground every day with innovative new
products and solutions for the worlds most difficult combustion
challenges. This commitment to technical excellence is driven by our
focus to provide a custom designed solution for every project based on each customers specific
requirements. One example is our flare gas recovery system, designed to achieve zero flaring, reducing
process fuel costs while eliminating visible flame, odors and auxiliary flare utilities.
Having one of the largest installed bases of process burners, production flares and thermal oxidizers in
refining and petrochemical facilities around the world. Callidus should be your choice for next generation
combustion equipment and solutions for new construction or retrofit operations with installation
capabilities anywhere in the world.

7130 South Lewis, Suite 335, Tulsa, OK 74136, call 1-918-496-7599 or visit our website www.callidus.com
2014 Honeywell International Inc. All rights reserved.

uop callidus.indd 1

10/03/2014 11:13

ability to model the transition

between carbon number and molecular structure types caused by
chemical
reactions
modelled
through reaction pathways is
explained as well as the overall
heat of reaction effects.

product. Figure 4 shows how these

product cuts details would look
when shown in a similar manner to
the simple lumped model represented in Figure 1.

TT

Pour point calculation

considerations

Characterising process feedstocks

and predicting product yields

The key step for the correct modelling of a thermal cracking process
such as Eureka is the denition of a
correct mixture of PIONA based
components needed to characterise
the feedstock. Laboratory analysis
of hydrocarbon feed material is
used to provide the necessary information to t the model of material
stream values against measured
properties. The reacted product
yields are similarly characterised.
The reaction kinetics of the thermal
cracking vessel at the desired operating conditions are then simulated
and the resulting material product
yields properties and ow rates
would be compared to known data.
At that point, the process is
repeated until the adjustable model
parameters are properly dened
and the errors between model and
experiment are minimised.
The experimental data was gathered through a bench scale batch
distillation apparatus shown in
Figure 3. This bench scale atmospheric thermal boiling/thermal
cracking vessel was meant to help
understand, characterise, and ne

Figure 3 Bench scale thermal cracking

experimental apparatus

tune the feedstock PIONA composition to be used in the simulation

before being used as a basis for the
construction of a complete Eureka
plant model. The study was done
using a Peace River vacuum residue sample representing material
from an atmospheric crude tower
followed by a vacuum tower in the
actual renery.
The feed sample was introduced
to the thermal cracking vessel with
the help of a carrier gas and heated
to a temperature of 430C through
electric heaters for approximately
45 minutes. An agitator was added
within the vessel to minimise
temperature
gradients,
which
helped mimic stripping steam
introduced to the full scale vessels
and the corresponding agitation in
actual operation.
Table 1 shows a comparison
between the bench scale experiments and VMGSim simulations
for Peace River bitumen feed and

The key physical property for the

quantication of pitch yield from
experimental data included the
liquid density, atomic gross analysis, heating value, and pour point of
the product. In this case, all properties were important for nal pitch
product specication, but the pour
point is a key indicator to ensure
proper uidisation of the material in
the reactor for trouble-free, continuous operation. Typical standard
methods to estimate pour points
such as ASTM D97 also accepted by
the American Petroleum Institute
(API), could not be used in a simulation environment due to their
inherent limitations. For example,
the ranges of the ASTM D97 methods equation are limited to
petroleum fractions of 140 to 800 g/
gmol and 13 to 50 API gravities10
and fell short in dealing with ranges
of pitch products with molecular
weights in the thousands of
g/gmol and negative API gravities.
A more rigorous and exible
approach to handle the pour point
calculation in the model was
devised using viscosity as a
correlating parameter. A value of
164 000 cP (164 Pa-s) was selected,

Peace River bitumen feed and product comparisons

API (60/60F)
H/C mass ratio
Molecular weight

Feed
0.86
0.115

Model9
0.98
0.115
1123

Naphtha
54.5
0.160

Model9
57.1
0.165
110

CLO
30.2
0.144

Model9
35.8
0.144
178

CHO
13.6
0.127
-

Model9
14.7
0.127
428

IBP, C
10%, C
50%, C
90%, C

D1160
455
-

477
541
716
857

D86
57
89
130
162

48
100
143
182

D86
192
211
250
288

194
200
233
301

D1160
330
368
439
526

334
372
458
526

FBP, C

892

179

194

310

322

569

7.0

1.3
0.0
37.2
61.5
7.0

46.8
13.9
26.5
12.8
2.8

47.0
14.0
24.4
14.6
2.9

4.9

27.9
12.4
26.7
33.0
4.9

5.6

5.0
5.4
42.3
47.3
5.6

Iso-Paraffin + n-Paraffin, wt%

Olefin, wt%
Naphthene, wt%
Aromatic, wt%
Sulphur, wt%

*CLO = cracked light oil, CHO = cracked heavy oil

Table 1

50 PTQ Q4 2014

hay.indd 3

www.eptq.com

10/09/2014 13:01

Mass flow

40
30
20
10

69
10
-1
5
C
16
-1
C 9
20
-2
9
C
30
-3
C 9
40
-4
C 9
50
-5
C 9
60
-6
C 9
70
-7
C 9
80
-8
C 9
90
C
10 99
015
0
C
15
0+
C

1C

2
35

Offgas

Heavy oil vacuum residue

Mass flow

40

Liquid pitch

30
20
10

P&I
O
N
A
A-dehy.
A-S, N, V

69
10
-1
5
C
16
-1
C 9
20
-2
9
C
30
-3
C 9
40
-4
C 9
50
-5
C 9
60
-6
C 9
70
-7
C 9
80
-8
C 9
90
C
10 99
015
0
C
15
0+
C

1C

2
35

Mass flow

40
30
20
10

-3
9
40
-4
C 9
50
-5
C 9
60
-6
C 9
70
-7
C 9
80
-8
C 9
90
C
10 99
015
0
C
15
0+
C

20

30

9
-2
9

16

-1

-1

10

6C

3-

1-

Figure 4 PIONA component slate heavy residue thermal cracking representation

after discussions with heavy oil

experimentalists, as a rough equivalent to a pour point based on the
observed behaviour of heavy oils
and bitumen at ambient temperature. From that reference point a
temperature could be found at
which the models pitch viscosity
matched and became the estimated
pour point temperature. In order
to use this type of solution, it would
then become important to accurately predict the viscosity of the
heavy oil mixtures, also a challenging problem when dealing with
heavy hydrocarbon feedstocks.
The viscosity prediction method
chosen in the model for the feed
and light to heavy product yields
was the expanded uid viscosity
model (VMG-EF).11 The general

www.eptq.com

hay.indd 4

principle of this model is that as a

uid expands there are greater
distances between molecules and
uidity (inverse of viscosity) of the
mixtures increases. The uidity is
assumed to be an exponential function of the expansion of the uid
from a near-solid state. An
additional benet of this method
was that binary interaction parameters that model non-idealities due
to different molecular sizes and
chemical types could be incorporated into the PIONA component
slate and characterisation procedure. At the end, the nal model
was tuned to within a few degrees
centigrade to the experimental
pour point temperatures together
with some actual plant conditions.
These results and other resulting

feed and product comparisons of

kinematic viscosity are shown in
Figure 5.
The Advanced Peng-Robinson
property package from VMGSim9
(VMG-APR) and API correlations
(VMG-API) were also used to
predict viscosities and are shown in
Figure 5. As expected, a correlation
like the API based on lighter hydrocarbon mixture data matched
experimental data very well for
lighter products, but was inadequate when applied to heavier cuts.

Application to Eureka process

operation

Complete
process
simulation
models were developed once the
experimental
models
property
predictions and reaction kinetics

PTQ Q4 2014 51

10/09/2014 13:01

Kinematic viscosity, mm2/s

1000
CHO
API = 15.2

Feedstock
API = 0.81

100
CLO
API = 34.3

10

EXP
1

VMG-EF
VMG-APR
VMG-API

Naphtha
API = 57.1

0.1
50

50

100

150

200

Temperature, C
Figure 5 PIONA based kinematic viscosity prediction comparisons

were fine-tuned and validated

against the available experimental
data. Some of the key points centre
on the coke formation trending in
the preheating units (not discussed
in this article) and operational optimisation like recycle ratio of reacted
heavy feed to fresh feed. Figure 6
shows a general Eureka process
setup and some of the resulting
product yield trends of different
bottom recycle oil ratios tested in
the model. With the product properties and reacting material pour
points being calculated for any
point in the operation plant a good
overall understanding of new feedstocks outside of previous running
conditions was reached.

Conclusions

Characterisation of a heavy feedstock used in the Eureka reactive

heavy oil thermal process was
presented based on the use of an
extended PIONA and carbon
number component slate. This
PIONA style simulation model
basis was also used for tracking
and estimation of hydrocarbon
mixture thermodynamic properties
before, after, and during reaction.
Although solution times were not
as fast as with conventional lumped
kinetic models (minutes compared
to seconds), this model showed
flexibility in its more rigorous
approach for process situations
where structural shifts across

Reactor

BFW

different boiling point ranges could

occur.
It was also shown that estimation
of focal properties within a PIONA
basis simulation, such as the pour
point temperature used to monitor
the fluidisation inside the vessel,
could be estimated with a specially
developed empirical correlation,
more adequate to heavy feedstocks
than available published methods.
Estimation methods for properties
like viscosity at different temperatures
was
reviewed
with
comparisons between the API
method and a full range expanded
fluid method. The latter allowed for
proper temperature dependent
viscosity trending when compared
with measurements taken from
reacted product cut samples.
The model successfully predicted
shifts in component representation
of feed to reacted products using a
PIONA driven reactive kinetic pathway.
This
more
generalised
approach still contained all key reaction pathways allowing for the
appropriate overall reactive shift to
product mixtures in the model,
which was further confirmed with
resulting accuracy of property
prediction comparisons and overall
heat of reaction balances. Since
the mixture properties calculated
were directly tied to the component
types
and
carbon
numbers
contributions
in
the
model

Gas
G/C

100

CHO
Pitch
stabiliser
WWRS

Vacuum
residue
Preheater

Steam
superheater

Fractionator
Pitch
flaker

Waste
water

Product yield, wt%

CLO
Cracking
heater

CG
AG
CLO

90
80
70
60

CHO

50
40
30

ASP

20
10

Petroleum
pitch

0.20

0.15

Recycle ratio

Figure 6 Eureka process and Piece River predicted product yields vs bottom recycle oil ratio

52 PTQ Q4 2014

hay.indd 5

www.eptq.com

12/09/2014 13:45

there was accuracy shown in

comparisons across a wide range of
boiling point temperature product
cuts. One could imagine catalyst
driven reaction pathways also being
potentially modelled using PIONA
basis with the reaction kinetic rates
properly altered to compensate for
catalysed pathways.
VMGSim is a mark of Virtual Materials Group,
Inc.
Acknowledgement
The authors are grateful to Fuji Oil Corporation
Ltd and Virtual Materials Group, Inc. for the
permission to publish this work. The authors
would also like to thank Dr Yarranton and
Mr Schoeggl at the University of Calgary for
information provided regarding laboratory
pour point measurements.
References
1 Rahimi P M, Teclemariam A, Taylor E,
deBruijn T, Wiehe I A, Determination of coking
onset of petroleum feedstocks using solubility
parameters, Fuel Chemistry Division Preprints,
48(1), 103, 2003.
2 Wiehe I, Process Chemistry of Petroleum
Macromolecules, CRC Press, 2008.
3 Hay G, Loria H, Satyro M, Thermodynamic
modeling and process simulation through

PIONA characterization, Energy & Fuels, 27,

3578-3584, 2013.
4 Bennett C, Klein M, Using mechanistically
informed pathways to control the automated
growth of reaction networks, Energy & Fuels,
26, 41-51, 2012.
5 Sedighi M, Keyvanloo K, Towfighi Darian
J, Olefin production from heavy liquid
hydrocarbon thermal cracking: kinetics and
product distribution, Iran. J. Chem. Chem. Eng.,
Vol 29, 4, 2010.
6 van Goethem M, Kleinendorst F, van
Leeuwen C, van Velzen N, Equation-based
SPYRO model and solver for the simulation of
the steam cracking process, Comp. & Chem.
Eng., 25, 905-911, 2001.
7 Mosby J, Buttke R, Cox J, Nikolaides C, Process
characterization of expanded-bed reactors in
series, Chem. Eng. Sci., 41, 989-995, 1986.
8 Takatsuka T, Kajiyama R, Hashimoto H,
Matsuo I, Miwa S, A practical model of thermal
cracking of residual oil, J. Chem. Eng. Japan, Vol.
22, 3, 1989.
9 Virtual Materials Group, Inc. VMGSim
Process Simulator, Version 8.0, Virtual Materials
Group, Inc., Calgary, Alberta, Canada, 2014.
10 Riazi M R, Characterization and properties
of petroleum fractions, ASTM International,
West Conshohocken, PA, 2005.
11 Loria H, Motahhari H, Satyro M, Yarranton
H W, Process simulation using the expanded
fluid model for viscosity calculations, Chemical

Engineering Research and Design,

10.1016/j.cherd.2014.06.019, 2014.

DOI:

Glen Hay is Vice President of Business

Development with Virtual Materials Group
Inc., Alberta, Calgary, Canada. His experience
is focused on reactors, heat transfer units,
and overall plant modelling and optimisation.
He holds a bachelors degree in chemical
engineering from the University of Calgary and
a masters in advanced process control.
Herbert Loria is a Process Simulation Software
Developer with Virtual Materials Group Inc.
He specialises in the development of physical
property packages, estimation methods for
heavy oil physical properties, oil characterisation
schemes and upstream applications for
VMGSim. He holds a PhD in chemical
engineering from the University of Calgary.
Marco Satyro is a Senior Fellow with Virtual
Materials Group Inc. He is one of VMGs
founders and helped design the property
package system VMGThermo. He graduated
from the Polytechnic School of the University
of Sao Paulo as a chemical engineer and holds
a PhD from the University of Calgary.
Hideki Nagata is Manager of the Operations
Management Group at Fuji Oil Company Ltd,
Chiba, Japan. He holds a bachelors degree in
chemical engineering from the University
of Kagoshima and a masters in reaction
engineering.

MOD-8000 Process NMR Analyzer

revolutionizes on-line crude oil analysis

Continuous monitoring of
critical crude oil properties
TBP (True Boiling Point)
Distillation Curve IBP, T(10), T(50), T(90), FBP
API Gravity
Water
Viscosity
Sulfur
TAN (Total Acid Number)
Aromaticity
and many others...

Modcon Systems Ltd.

145-147 St John Street, London EC1V 4 PW UK Tel: +44-207-5043626

www.modcon-systems.com analyzer@modcon-systems.com

www.eptq.com

hay.indd 6

PTQ Q4 2014 53

10/09/2014 13:01

No other hydrogen comp

company
pany
supplies so much.

2014 Air Products and Chemicals, Inc.

air prod.indd 1

10/09/2014 11:23

Integrated hydrogen management

Redesigning an existing hydrogen system leads to an integrated, reliable and
flexible supply
SAA POLOVINA, DANIJELA HARMINA and ANA GRANIC ARAC
INA Rijeka refinery

ydrogen has become one of

the most important refining
energy media and its efficient
use is of the highest priority.
Therefore, refineries are forced to
exploit their existing hydrogen
sources to the maximum and
continually increase their hydrogen
sources.
This article describes the integration of two hydrogen systems at
INA Rijeka refinery: one system
linked to a catalytic reforming unit
as a source of hydrogen and the
other linked to a hydrogen generation unit based on steam reforming.
The integration of these two hydrogen systems involves three different
purities of hydrogen-rich gas.
Hydrogen-rich gas produced by
catalytic reforming is used in the
refinery processes naphtha hydrotreating
(NHT),
isomerisation,
kerosene hydrotreating (KHT1) and
gasoil hydrotreating (GHT2). The
hydrocracking units requirement
for make-up gas of high purity in
large amounts relies on the hydrogen generation unit. The two
sources of hydrogen and two
systems of purification for hydrogen-rich gases give rise to three
purity levels for hydrogen-rich gas:
Gas from the catalytic reforming
unit of 73-75 vol% purity, depending on the catalyst cycle stage
(start-of-run or end-of-run)

Gas after absorption with a
hydrogen content of 83-85 vol%
Gas after purification in the pressure swing absorption (PSA) unit,
with a hydrogen content of 99.99
vol%.

The current situation

Make-up hydrogen for the hydro-

www.eptq.com

ina.indd 1

cracking unit comes from the

hydrogen generation unit which
uses natural gas as feed. Make-up
hydrogen for isomerisation, NHT,
KHT1 and GHT2 comes from catalytic reforming.
In order to integrate these two
hydrogen systems, and to achieve
better utilisation of the produced
hydrogen,
the
systems
are
connected via two pipelines with
two manual valves. One pipeline
supplies hydrogen produced in
catalytic reforming (after obligatory
purification in the PSA unit) as

The amount of
hydrogen-rich gas
produced depends on
feed composition and
process conditions in
the reactor section
make-up for the hydrocracking
plant; the other pipeline supplies
hydrogen from the hydrogen
generation unit as make-up for the
isomerisation, NHT, KHT1 and
GHT2 plants.
Shut-off for both pipelines relies
on
ordinary
manual
valves.
Problems in the operation of one
plant can lead to shutdowns of
other plants connected to this integrated hydrogen system.
The hydrogen system in the catalytic reforming plant starts with the
high pressure (HP) separator where
hydrogen-rich gas is separated
from unstabilised gasoline. The
amount of hydrogen-rich gas

produced depends on feed composition and process conditions in the

reactor section. The purity of the
produced gas depends on the pressure and temperature at the HP
separator and on ambient conditions, especially temperature.
After physical separation, hydrogen-rich gas enters the suction
vessel of the booster compressors
where it achieves the required pressure and is distributed to the
consumers.
A PSA unit is located between
the HP separator and the booster
compressors. This is used for purifying the reformers hydrogen-rich
gas. Pressures at the HP separator,
at the inlet of the PSA unit, and at
the suction line of the booster
compressors are regulated with the
same pressure controller.
In addition to pressure control of
these positions, the control valve
has a safety role. In case of problems in the catalytic reforming
process, fast depressurisation of the
HP section in the refinery fuel gas
system would be crucial.
This type of pressure control has
significant disadvantages. Finelevel flow control of hydrogen-rich
gas discharges to the fuel system
cannot be implemented. At 1%
valve opening, the flow rate reaches
around 2000 Nm3/h. If the process
is operated in a mode in which the
regulator valve is opened or closed,
large amounts of hydrogen are
discharged to the refinery fuel
system.
Booster compressors are the main
part of the hydrogen system since
they maintain the pressure required
for normal operation of hydrogen
consumers and high pressure

PTQ Q4 2014 55

10/09/2014 13:35

 Blue line: H2 rich gas from reformer

Red line: returned streams to PSA HGU and PSA REF
Pink line: stream line to fuel gas system
Green line: H2 from HGU to HCU
Dark line: stream line to blowdown system

HGU

Booster
compressor

Absorber

PSA_HGU

REF
Compressor
suction drum
Returned
from NHT

PSA_REF
Isomerisation NHT GHT1 or KHT1
GHT2

HCU

Returned
from NHT ISO
Fuel gas
system

PC to
blowdown

Figure 1 Existing scheme of the hydrogen system in Rijeka refinery

absorption. Hydrogen at two different levels of purity is fed to the

suction side of the compressors. If
the PSA unit is operating, hydrogen
purity is 99.99%; if not, hydrogen-rich gas is fed from the
reforming HP separator.
After
cooling
and
passing
through the knockout drum, one
part of the hydrogen-rich gas is
distributed to the isomerisation and
NHT plants and the rest goes
through the absorption column.
After purification, hydrogen-rich
gas goes to the KHT and GHT units
at a purity of 83-85%.
Make-up gas for the hydrocracking unit must be at 99.99% purity,
which means that it can only come
from the PSA unit. If the PSA unit
of the reforming plant is not in
service, excess hydrogen-rich gas
cannot be sent to the hydrocracker
due to insufficient purity. Excess
produced gas is discharged to the
fuel system.
If the PSA unit is operating and
the produced hydrogen is at 99.99%
purity, this is distributed through
booster compressors to the consumers. Excess hydrogen from the PSA,
together with hydrogen from the
hydrogen generating unit, becomes
make-up gas for the hydrocracker.
The economic impact of this solution can be measured in the
reduction of demand for expensive
natural gas feedstock in the hydrogen generating plant.

56 PTQ Q4 2014

ina.indd 2

Since two separate hydrogen

systems are connected by a manual
valve, this solution carries significant risks to the normal operation
of both hydrogen systems. The risk
is much higher for operation of the
hydrocracker; in the event of a PSA
shutdown, the connection between
these hydrogen systems must be
closed immediately, so the hydrocracker is without substantial
amounts of make-up hydrogen. In

Make-up gas for the

hydrocracking unit
must be at 99.99%
purity, which means
that it can only come
from the PSA unit
order to compensate for the
reduced volume of make-up gas,
the hydrogen generating unit must
significantly raise its capacity. Thus,
sudden loss of the required amount
of make-up hydrogen may lead to
shutdown of the hydrocracker.
If operational problems in the
hydrocracker lead to a shutdown of
the PSA unit, problems in the old
hydrogen system are less challenging. At the shutdown of the PSA
unit a bypass is automatically
opened to ensure the necessary

quantity of gas for compressor

operation. In this way, continuous
supply of make-up hydrogen is
ensured to all consumers connected
to the pressure side of the compressor. By increasing the flow to the
absorption column, a higher level
of make-up hydrogen purity can be
ensured. To ensure continuous
hydrogen supply to all consumers,
integration of two separated hydrogen systems is vital. The existing
scheme of the hydrogen system in
Rijeka refinery is shown in Figure 1.

Reconstruction of the existing

hydrogen system
High pressure absorption and
operation with one PSA unit

After separation in the reforming

HP separator, hydrogen-rich gas
goes to the suction vessel of the
booster compressors. Pressure in
the HP section, as well as in the
PSA unit and the suction vessel of
the compressors, is maintained by a
pressure controller (PC).
In order to achieve the best possible pressure regulation, and to
increase hydrogen utilisation and
minimise discharge into the fuel
gas system, an additional smaller
control valve should be installed
that would work in a split range
configuration with the specified PC.
This new, smaller PC would
discharge in the normal operational
range (at valve opening of 30-40%)
around 100-150 Nm3/h hydrogen-

www.eptq.com

10/09/2014 13:36

YOUR BENEFIT:
LOWEST LIFE CYCLE COSTS

FULL RANGE:
Rod load up to 1500 kN / 335000 Ibs
Power up to 31000 kW / 42100 hp

API 618
RELIABLE SWISS QUALITY
YOU GET MORE THAN JUST A
PROCESS GAS COMPRESSOR
Lubricated up to 1000 bara,
non-lubricated up to 300 bara

burckhardt.indd 1

For highest availability: We recommand 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

10/09/2014 11:34

Actions speak louder than words

High pressure CO2 gas compression

Marine Engines & Systems Power Plants Turbomachinery After Sales

MAN Diesel & Turbo has unique compression solutions in its portfolio for the complete range of CO2, N2,
propylene and vapour related applications, with single machines or complete train solutions. Others talk

about CO2 compression. We have the credentials! Over 200 RG Gas compressor units are in operation

or on order. We are world`s number one in CO2 high pressure applications, thanks to sophisticated test-

ing facilities and proven track records in the eld. With an RG Gas compressor from MAN Diesel & Turbo
you will gain an exceptionally exible and optimized solution to maintain your business economic.
Find out more at www.mandieselturbo.com

man.indd
1
Typ 02 - Actions
speak louder than words (RG Gas) - 297x210.indd 1

10/09/201411:59:29
11:51
2014-01-27

rich gas into the fuel gas system,

which is a significant flow decrease
compared to regulation by the
previous PC. The current PC operates at a valve opening of 1-2%,
discharging about 2000 Nm3/h of
hydrogen-rich gas to the fuel gas
system. This old valve is usually
closed, which means that when the
valve is opened, in half an hour
1000 Nm3 of hydrogen-rich gas is
discharged. At 80% valve opening
of the new PC, the old PC would
take over regulation to ensure operational safety in the plant.
After the suction vessel, the
booster compressors raise the pressure to the required 32 bar.
Hydrogen-rich gas goes through
the water cooler and knockout
drum to the absorption column.
Stabilised gasoline from the bottom
of the debutaniser column of the
reforming plant is used as absorbent. To achieve better absorption,
the temperature of the absorbent is
kept as low as possible. Therefore,
stabilised gasoline from the bottom
of the debutaniser column is first
cooled in air and water coolers.
In order to supply make-up gas
with a higher hydrogen content to
all consumers, the pipeline for
supplying the isomerisation and
NHT units should be moved after
the absorption column (see Figures
1 and 3). Thus all consumers in the
old part of the refinery have a
supply of make-up gas of sufficient
quantity and quality. Once the
needs of consumers of make-up gas
are satisfied, the full volume of the
excess hydrogen-rich gas could be
sent to the PSA at the hydrogen
generation unit. To further protect
this PSA, any large amounts of
liquid need to be removed by integration of an additional water
cooler and knockout drum before
the PSA (see Figure 3).
Hydrogen-rich gas purified in the
absorption column has a similar
chemical composition as gas from
the HP section of the hydrocracker.
This means that gas from the
absorption column could go to the
hydrogen generation units PSA
unit. This kind of operation would
eliminate the need for two PSA
units. Both plants would benefit
from this solution; a tail gas

www.eptq.com

ina.indd 3

Catalytic
reforming
20 barg
73-74%vol H2

Isomerisation

Absorber
32 barg
84%vol H2
10%vol CH4

Booster
compressor
32 barg

To NHT

To KHT

NHT isomerisation

GHT

From NHT isom.

26.5 bara
86%vol H2
8.4%vol CH4
From NHT
24.5 bara
91%vol H2
7%vol CH4
From KHT
to fuel gas
86%vol H2
9.7%vol CH4

HGU steam
reforming

PSA HGU

From GHT
75%vol H2
23%vol CH4
NHT heavy
naphtha

HCU

KHT

Red line: H2 rich gas from reformer

Blue line: cascade H2 line
Green line: stream line to fuel gas system
Dark line: H2 from HGU to HCU

Figure 2 Cascade use of make-up hydrogen

compressor would be eliminated

from reformings PSA unit, and
consumption of expensive feed
(natural gas) at the hydrogen
generation unit would decrease.
In order to increase overall utilisation of hydrogen and to reduce
consumption of feed for hydrogen
production, online analysis of the
hydrogen-rich streams, and reuse
of these streams in refinery
processes, should be implemented.
Thus the integrated hydrogen

system could be further improved.

Cascade mode for hydrogen
streams could be applied (see
Figure 2).
The basis of this system is that
each plant has its own direct line of
make-up hydrogen. Off-gases from
the HP sections of the isomerisation
and NHT plants, containing a minimum 86 vol% hydrogen (see Table
1) also can be used as make-up gas
for other plants. The limiting factor
for this kind of usage is the

Composition of high pressure off-gases at hydrogen consuming plants

HP gas, vol% HDS section of Isomerisation
H2
86.24
CH4
8.4
C2H6
2
C3H8
0.5
i-C4H10
0.2
n-C4H10
0.23
i-C5H12
1.1
n-C5H12
0.8
C6+
0.4
H2S
0.13

NHT
91.1
7.2
1.3
0.25
0.05
0
0
0
0
0.15

HDS1
86.43
9.65
1.4
0.7
0.5
0.2
0.2
0.1
0.1
0.8

HDS2
74.5
22.55
2.32
0.14
0.1
0.1
0.1
0.1
0.1
0.01

Table 1

PTQ Q4 2014

59

10/09/2014 13:36

Maximize Reliability, Availability & Profitability

with Ariel API 618 Process Compressors
With more than 19,000 possible frame and cylinder
configurations for API 618 process service, Ariel compressors
can be designed and built to maximize your unique
process operation. Ariel has been producing durable, lowmaintenance reciprocating compressors since 1966 and
API 618 compressors since 1999. Ariel offers high quality
WORLD STANDARD COMPRESSORS

compressors that ensure long continuous run-times.

Visit www.arielcorp.com to find an Ariel Process distributor
in your region.

To learn more about Ariel Compressors, please visit www.arielcorp.com

ariel.indd 1

09/06/2014 14:57

Booster
compressor

Valve
open

Absorber

REF
Valve
closed

Isomerisation

Compressor
suction drum

FC

FC

FC

FC

PSA_REF

NHT

Isomerisation
Refinery
fuel gas
system

FC control valve
for flow regulation
to PSA_HGU

NHT

GHT1 GHT2
or
KHT1

PC control valve on
pressure side of compressor

Valve
closed

Blue line: H2 rich gas from reformer

Dark blue line: cascade H2 line
Red line: returned streams to PSA HGU
and PSA REF
Pink line: stream line to fuel gas system
Green line: H2 from HGU to HCU or to
old H2 line
Dark line: stream line to blowdown
system

PSA_HGU
Cooler

HCU

Knock out
drum

Blow down

Figure 3 Hydrogen system, absorber and hydrogen generations PSA unit in operation

required
pressure
difference
between the plants.
The NHT isomerisation section
with a pressure of 26.5 bar is the
first unit in this cascade sequence.
For the installed capacity of the
plant, make-up of about 1400-2000
Nm3/h of hydrogen-rich gas is
required.
At the NHT plant, pressure at the
HP section is 24.5 bar so off-gas
from the NHT isomerisation section
can be directed to the NHT as
make-up gas. The quantity available at 1400-2000 Nm3/h is more
than enough for this process.
From the HP section of the NHT
plant, off-gas can be directed to
unit KHT1. In this way, three
plants can work with 1400-2000
Nm3/h of hydrogen-rich gas.
Off-gas from the KHT 1 unit goes
to the fuel gas system.
For operation of the GHT2 unit,
higher amounts of hydrogen-rich
gas are needed so an independent
line of make-up gas must be used.

www.eptq.com

ina.indd 4

Purity at 84 vol% hydrogen is

enough for the operation of the
GHT2 unit at maximum capacity.
This off-gas contains high amounts
of hydrogen so it is desirable to use
it in some way. One possibility is to
send it, after amine washing, to the
inlet vessel of reformings PSA unit
for
purification
and
reuse.
Compositions of these off-gases are
shown in Table 1. Figure 2 illustrates cascade use of make-up
hydrogen.
The hydrogen generation units
PSA unit works at a pressure of
22.4 bar so all hydrogen-rich gas
streams at higher pressures can be
directed to this unit. In this solution
there is no cascade mode of operation and all off-gases from HP
sections can be directed to the
hydrogen generation units PSA
unit.
Pressure at this PSA unit is
controlled electronically (see Figure
3) by a pressure controller located
at its outlet. However, with a new

connection to the PSA unit it is

necessary to re-regulate it in order
to keep the hydrogen yield at 85%
and avoid any release of new
hydrogen to the fuel gas system.
In this way consumption of natural gas for hydrogen production
can be reduced by an amount
equivalent to the recovered hydrogen. Additionally, by reducing
hydrogen generation
capacity,
significant savings can be delivered
by the operation of only one PSA
unit (that is, the operating costs of
the other unit and of the tail gas
compressor) because total tail gas
from this PSA unit is used up in
the steam reforming furnaces. It is
important to emphasise that these
two systems are connected in such
a way that hydrogen from the
hydrogen generation plant goes to
the suction vessel of the booster
compressors (see Figure 1), which
means that pure hydrogen can be
distributed to other plants in the
refinery, even if the reforming plant

PTQ Q4 2014

61

15/09/2014 12:14

HP gas composition at hydrogen consuming plants if the make-up hydrogen comes

from the PSA unit
HP gas
Isomerisation HDS section
composition, vol %
of Isomerisation
H2
95.5
96.5
CH4
2.64
0.81
0.11
0.25
C2H6
C3H8
0.04
0.09
i-C4H10
0.02
0.02
n-C4H10
0.01
0.4
i-C5H12
0.49
0.89
0.61
0.82
n-C5H12
C6+
0.23
0.01
H2S
0
0.15

NHT

HDS1

HDS2

97.5
1.6
0.8
0
0
0
0
0
0
0.15

97.5
1.45
0.8
0
0
0
0
0
0
0.15

95
4
0.52
0.17
0.25
0.04
0.03
0
0
0.15

Table 2

is not in service. Figure 3 shows

operation of the refinery with
hydrogen generations PSA unit
and absorption column.

Refinery operation with two

PSA units

The integrated refinery hydrogen

system must be able to operate
with two PSA units.
After reformings HP separator,
hydrogen-rich gas goes to reformings
PSA
unit.
Following
purification, gas with 99.99 vol% of
hydrogen goes to the suction vessel
of the booster compressors; after
compression, it is distributed to the
consumers (see Figure 1). When
working in this mode, the installation of a smaller pressure control
valve is even more important and
the savings achieved are even
greater.
This mode of operation requires
changes in the operation of the
plants. Now, when make-up gas is
pure hydrogen, lower volumes of
make-up gas are required, and
additional savings can be achieved
by reduction of total pressure in the
plants. As the yield of hydrogen at
the PSA unit is 85%, it is necessary
to reduce pressure on the suction
side of the booster compressor and
therefore the total pressure in all
plants. With this make-up gas, the
hydrogen content of recycle gas in
all plants increases, which allows a
reduction in total pressure. The
reduction in total pressure is
defined by the required hydrogen
partial pressure in each process.
Operation of the plant is adjusted
in such a way that all control valves

62 PTQ Q4 2014

ina.indd 5

that maintain pressure in the HP

sections are working so that minimal amounts of hydrogen are
discharged to the fuel gas system.
(Purge of the system should be
avoided in any event.)
In this operational mode, a pressure of 30.5-31 bar at the suction
side of the booster compressor is
sufficient to meet the needs of all
hydrogen consumers. From the
suction side of the booster compressor through the line, hydrogen is
distributed to the consumers (see
Figure 1). Because of higher utilisa-

Following
purification, gas
with 99.99 vol% of
hydrogen goes to the
suction vessel of the
booster compressors
tion of the produced hydrogen, all
hydrogen-rich streams can be
connected to the input of the PSA
unit. Given the required pressure,
streams from all HP sections of the
refining process can be connected.
However, the return of hydrogen-rich streams to the PSA unit
can cause big problems for the PSA
compressor; while the PSA unit
produces pure hydrogen, the
returning flows from the HP
sections have 95-96 vol% hydrogen
content. Table 2 shows the composition of returning gas from the HP
sections of the plants if the

make-up gas is pure hydrogen.

Catalytic reformings PSA unit is
designed for purification of gas
with 73-75 vol% hydrogen content;
the cycles of the PSA unit are also
adapted for gas of this composition.
However, since the PSA cycles are
adjusted to purification of hydrogen-rich
gas
from
catalytic
reforming, some of the new returning hydrogen will end up in the tail
gas and will be lost in the refinery
fuel gas system. In addition, an
increase in hydrogen content in the
tail gas will reduce its molecular
weight; this will apply an additional burden to the compressor
and the entire tail gas system.
Therefore, before deciding on
further utilisation of these streams,
the supplier of the PSA unit must
be consulted, in order to readjust
the cycles so that gas with higher
hydrogen content can be purified
and to ensure normal operation of
the tail gas compressor.
As with the previous mode, these
two, now integrated hydrogen
systems are connected and, if
necessary, it is possible to open the
valve for supply from hydrogen
generation to the suction side of the
booster compressor and satisfy
demand for hydrogen from one of
the plants, without increasing the
capacity of reforming (see Figure 1).
In this case, the valve on the
make-up hydrogen line between
the absorber and hydrocracker is
closed.
Hydrogen recovery can be
achieved, as in the previous case,
with cascade connections between
the plants. Cascade mode has an
even greater significance than in
the previous operational mode. As
the make-up gas is pure hydrogen,
for operation at average plant
capacity much smaller amounts of
hydrogen are needed. For example,
processing 16-18 t/h of light gasoline at the HDS section of the
isomerisation plant requires around
1200 Nm3/h of 99.99 vol% make-up
hydrogen (once through reactor).
This quantity is sufficient for
processing about 58 t/h of heavy
naphtha at the NHT plant, and the
off-gas from this plant is sufficient
for processing 22 t/h of feedstock
in the hydrodesulphurisation unit

www.eptq.com

10/09/2014 13:36

USA

THINK GERMAN,
ACT LOCAL.
LOOKING FOR A RECIP PACKAGE
AS UNIQUE AS YOUR
SPECIFICATIONS?

BLUESTROKE
COMPRESSOR
SYSTEMS

WE ARE NOT JUST A PACKAGER,

WE ARE YOUR ENGINEERING EXPERTS.
As the compressor OEM, our packages are engineered
and customized to meet your specs. Our reciprocating
compressors are specially designed in a wide range of
sizes and configurations while operating up to 1,200 rpm
across all types of gases. Also, 3D compressor package
models are reviewed with you to ensure the best design
for optimal operation and maintenance.

NEUMAN & ESSER USA, Inc.

Located in Katy, Texas
www.neuman-esser.com

Contact me for North America:

Roy Jacobs
Sales Manager for Upstream & Midstream
jacobsr@neuman-esser.com
Direct Phone: +1 713-554-9643

NEA GROUP Headquarters in Germany

PTQ A4-Jacobs
nueman.indd
1 2014-09.indd 1

05.09.14 13:17
15:13
10/09/2014

Booster
compressor

Valve
closed

Absorber

REF
Valve
open

Isomerisation

Compressor
suction drum

FC

FC

FC

FC

PSA_REF
Valve
closed

NHT

Isomerisation
Refinery
fuel gas
system

FC control valve for

flow regulation to
PSA_HGU closed

NHT

GHT1 GHT2
or
KHT1

PC control valve on
pressure side of compressor
full open

Valve
open

Blue line: H2 rich gas from reformer

Dark blue line: cascade H2 line
Red line: returned streams to PSA HGU and
PSA REF
Pink line: stream line to fuel gas system
Green line: H2 from HGU to HCU or to old
H2 line
Dark line: stream line to blowdown system

PSA_HGU
Cooler

HCU

Knock out
drum

Blow down

Figure 4 Operation with two PSA units

(KHT1). In this way, 1200-1300

Nm3/h of hydrogen is sufficient for
the operation of three plants. For
no cascade connection between the
plants, a much greater amount of
hydrogen must be used. For example, at the capacities mentioned, the
required hydrogen streams are 1200
Nm3/h for the isomerisation plant,
900 Nm3/h for NHT and 900
Nm3/h for KHT1, indicating a 1800
Nm3/h increase in hydrogen
demand. This amount of hydrogen
could be used in the hydrocracker
and so further reduce consumption
of natural gas in hydrogen generation. Figure 4 shows operation with
two PSA units.

Directing reformings hydrogen-rich

gas to hydrogen generations
PSA unit
Full utilisation of hydrogen-rich gas
can be achieved by direct connection
between
reforming
and
hydrogen generation in two ways.
The first is to send gas from

64 PTQ Q4 2014

ina.indd 6

reforming as feed to hydrogen

generation (with a pre-reforming
section). Considering that this gas
has a high hydrogen content, it
would place an unnecessary burden
on hydrogen generations steam
reforming reactor and lead to
unnecessary energy consumption.
The second, better solution
regarding energy consumption is to
direct reformings hydrogen-rich
gas to hydrogen generations PSA
unit. To achieve this, pressure must
be raised at reformings HP separator to a value that will provide
smooth transport to the PSA unit.
The composition of this gas is similar to that of gas from the HP
section of the hydrocracker. Since
hydrogen from hydrocracking offgas is usually recovered at hydrogen generations PSA unit, there
should be no problem purifying
reformings hydrogen-rich gas at
the specified PSA unit.
Because gas from reforming
contains some C3, C4 and C5 hydro-

carbons, for a PSA units protection

and in order to remove any large
amounts of liquids, it is necessary
to install an additional water cooler
and knockout drum.
After purification, the produced
pure hydrogen goes into two main
streams, one stream to the hydrocracker and the second to the
suction vessel of the booster
compressors. Subsequently, after
reaching the required pressure pure
hydrogen is distributed to all
consumers. Flow of hydrogen to
the compressor suction vessel is
controlled by a flow controller
which ensures that the compressors
always distribute exactly the
amount of hydrogen necessary for
normal plant operation. (The exact
amount of hydrogen is considered
to be the quantity of hydrogen that
provides the necessary make-up for
all consuming units.) The pressure
controller that maintains pressure
at the inlet of the booster compressor must be minimally opened.

www.eptq.com

10/09/2014 13:37

Booster
compressor

Valve
closed

Absorber

REF
Valve
closed

Compressor
suction drum

FC

FC

FC

FC

PSA_REF
Valve
closed

PC-B
PC-A

Isomerisation
Refinery
fuel gas
system

FC control valve for

flow regulation to
PSA_HGU closed

NHT

GHT1 GHT2
or
KHT1

PC control valve on
pressure side of compressor
closed

Valve
closed

PSA_HGU
Cooler
Isomerisation

HCU

Knock out
drum
Valve
open

NHT
Blue line: H2 rich gas from reformer
Dark blue line: cascade H2 line
Orange line: direct line from reformings H2 rich gas to PSA HGU
Red line: returned streams to direct H2 rich gas line from reformer to PSA HGU and from HCU
to PSA HGU
Pink line: stream line to fuel gas system
Green line: H2 from HGU to HCU or to old H2 line
Dark line: stream line to blowdown system

Blow down

Figure 5 Directing reformings hydrogen-rich gas to hydrogen generations PSA unit

In this way, an integrated hydrogen system is created, which

ensures the supply to consumers of
hydrogen of maximum purity,
reduces consumption of natural gas
for production of hydrogen at
hydrogen generation and minimises hydrogen content in the fuel
gas system. This mode of operation
also ensures highly reliable and
stable operation of this hydrogen
system. The operation is shown in
Figure 5.
However, there are restrictions.
There is a point at which it is not
possible to send additional hydrogen-rich gas from the catalytic
reformer to hydrogen generations
PSA because the PSA tail gas has
too high a fuel value and the steam
furnace
reformer
temperature

www.eptq.com

ina.indd 7

cannot be regulated since no

make-up natural gas is needed.
The primary fuel for the steam
reformer furnace is tail gas from
hydrogen generations PSA unit,
and temperature regulation is
maintained by the addition of a
secondary fuel, natural gas. Around
25% of the steam reformer duty is
maintained through make-up natural gas. (This value is related to the
plant without addition of gas from
the hydrocracker.)
To illustrate this, the hydrogen
generation capacity in Rijeka is
76 000 Nm3/h, with a standard
yield from the PSA of 85%; hence
the reformer gas flow would be
about 120 700 Nm3/h. The overall
reformer duty would be 100 million
Kcal/h and the PSA tail gas would

give 82 million Kcal/h. So we need

a further 18 million Kcal/h which
would require the consumption of
around 2140 Nm3/h of natural gas.
The plant is designed to accept
5000 Nm3/h of hydrocracker off
gas, (composition around 83% H2,
11% CH4, 6% CO2). In this case, the
quantity of reformed gas drops to
about 115 000 Nm3/h, which is
about 95% capacity. If we assume
that the furnace duty reduces linearly with the reformer load, the
furnace duty drops to 95 million
Kcal/h. However, this hydrocracker off-gas contains a large
amount of hydrocarbon components with a higher calorific value.
Therefore, the make-up natural gas
required drops to 760 Nm3/h. (The
calorific value of the PSA tail gas

PTQ Q4 2014

65

10/09/2014 13:37

has increased from 1830 Kcal/Nm3

to 2010 Kcal/Nm3.)
With the addition of another 3000
Nm3/h of gas from the catalytic
reformer (after
absorption, the
composition and heat value of the
gas is similar to the return gas from
the HCU), we arrive at the point
where the steam reforming furnace
could function with only the tail
gas from the PSA unit and there is
no need for the addition of
make-up natural gas. At this point

AdvaSulf

we would lose the temperature

regulation in the steam reformer
and the system could not operate
safely under these conditions.
This problem could be solved by
installing an additional vessel
before the hydrogen generation
PSA unit, with the amount of
hydrogen-rich gas that goes to this
unit being regulated according to
one ow controller. The amount of
gas that goes to the PSA unit would
be controlled, and any excess gas

TM

HySWEET
COSWEET

TM

SweetSulf

TM

Sprex
AdvAmine

TM

A UNIQUE TASTE OF SWEET FOR YOUR GAS

50 YEARS OF EXPERIENCE IN GAS SWEETENING AND
SULPHUR RECOVERY PROCESSES
With its unique and complete proprietary technologies portfolio,
PROSERNAT offers optimized solutions to bring on specs any
type of gas contaminated with CO2, H2S, COS and organic
sulphur species, while producing sulphur with the most stringent
emissions standards.
www.prosernat.com

66 PTQ Q4 2014

ina.indd 8

would be sent to either the renery

fuel gas system or to another gas
consumer.

Conclusions

By reconstructing the existing

hydrogen system, an integrated,
reliable and exible hydrogen
system would be created. This new
system would ensure the most
favourable renery operation.
With the described conguration
of the system, the renery could
signicantly reduce the consumption of natural gas for hydrogen
generation.
Application of the cascade operational mode could solve two major
renery problems. It would reduce
hydrogen content in the fuel gas
system resulting in signicant
savings. If total hydrogen (taking
1600 Nm3/h of 99.99% H2) that can
be saved by using the cascade operational mode is directed to the
hydrocracking plant, the renery
can save up to 1 million per year,
because the volume of natural gas
used as feed for hydrogen generation is reduced.
By integrating the two systems of
hydrogen purication (absorber
and PSA), sufcient amounts of
hydrogen of appropriate purity for
the hydrogen consuming plants
would be ensured. Loss of hydrogen by absorption purication is
not signicant.
Operation of a single PSA unit (in
hydrogen
generation)
could
provide signicant energy savings
because the other PSA unit (reforming) with the associated tail gas
compressor would be in standby
condition.
The biggest investment would be
the construction of a pre-reformer
at the hydrogen generation unit,
but the benets of this investment
are huge. As total hydrogen ends
up at hydrogen generations PSA
unit, release of hydrogen-rich gas
to the fuel system would be eliminated, and thereby the hydrogen
content of renery fuel gas would
be reduced. Only hydrogen from
low pressure sections would be
present in the fuel gas system.
As hydrogen consumers receive
pure hydrogen from hydrogen
generations PSA unit, consumption

www.eptq.com

10/09/2014 13:37

of hydrogen in all secondary plants

would be reduced. Also, tail gas
from HGUs PSA unit now contains
C3-C4 hydrocarbon fractions from
reformings hydrogen-rich gas, and
has a higher caloric value which
means that less make-up natural
gas is needed for steam reforming
furnace operation. The cost of
reconstruction of the existing
system is negligible compared to
benets of the described integrated
renery hydrogen system.
Further reading
1 Adamic Z, Sertic-Bionda K, Adamic
T, Beic S, Catalytic reforming increasing
purity of produced hydrogen with physical
absorption, Chemistry and technology of fuels
and oils (0009-3092), 2010, vol 46, no 3, 170177.
2 Ahmad M I, Jobson M, Zhang N, Multiperiod hydrogen management, Chemical
Engineering Transactions, 2006, vol 18, 743748, DOI:10.3303/CET0918121.
3 Alhajri I, Integration of Hydrogen and CO2
Management within Refinery Planning, PhD
thesis, University of Waterloo, Ontario, Canada,
2008.
4 Ceric E, Nafta Procesi i Proizvodi, First
edition, INA Industrija nafte, Zagreb, 2006.
5 Davis R A, Patel N M, Refinery hydrogen
management, PTQ Spring 2004, 29-35, on 15
Feb 2013, www.c2es.org/docUploads/Air%20
Products%20Hydrogen%20Article.pdf.
6 Hallale N, Moore I, Vauk D, Hydrogen
optimisation at minimal investment, PTQ,
Spring 2003, 83-90, on 10 Mar 2013, www.
aspentech.us/publication_files/PTQ_
Spring_2003_Hydrogen_Optimization.pdf.
7 Hofer W, Moore I, Robinson R P, Hitting ULS
targets through hydrogen management, PTQ,
Spring 2004, 1-6, on 1 Mar 2013, www.hysys.
com/publication_files/PTQ_Spring_2004_
Hydrogen_Management.pdf.
8 Jia N, Refinery Hydrogen Network
Optimisation With Improved Hydroprocessor
Modelling, 2010, PhD thesis, University
of Manchester, UK, on 10 Nov 2012,
w w w. e s c h o l a r. m a n c h e s t e r. a c . u k / a p i /
datastream?publicationPid=uk-ac-manscw:118621&datastreamId=FULL-TEXT.PDF
9 Liu N, Refinery Hydrogen Management,
2004,
MSc thesis, Delft University of
Technology, Department of Chemical
Technology, Process System Engineering,
Delft, on 5 Feb 2013, repository.tudelft.nl/
view/ir/uuid%3Ad2e5bc2c-c14f-4321-9fba23b6614a0bd3/.
10 Rabiei Z, Hydrogen management in
refineries, Petroleum & Coal, 2012, vol 54, no
4, 357-368, on 15 Mar 2013, www.vurup.sk/
volume-54-2012-issue-4.
11 Saleh M, Jahantighy Z F, Gooyavar A
S, Samipourgiry M, Majidian N, Hydrogen

integration in refinery using MINLP method,

International Journal of Modeling and
Optimization, 2012, vol 2, no 2, 83-86, on 2 Feb
2013, www.ijmo.org/papers/90-JQ099.pdf.

a reformate splitter, a kerosene sweetening unit

and a PSA unit. She graduated from the Faculty
of Chemical Engineering and Technology at the
University of Zagreb.

Saa Polovina is an Area Manager in INAs

Rijeka refinery, Croatia, and is responsible for
several processing plants as a technologist.
He holds a MSc in chemical engineering and
technology from the University of Zagreb.

Ana Granic arac is a Process Engineer

providing technical support for processing
plants at Rijeka refinery including catalytic
reforming, hydrogen management and
purification, isomerisation, LPG/gasoline Merox
and gasoline hydrotreating. She joined INA as
a participant of the Growww programme and
graduated as a chemical engineer from the
University of Zagreb.

Danijela Harmina is a Production Engineer

at Area 2 of Rijeka refinery, which includes
catalytic reforming, isomerisation, Merox units,

150

PERFORMANCE 3.
EXCEEDING EXPECTATIONS:
THE NEW EFFICIENCY IN
PROCESSED AIR.

Pneumatic transport of bulk materials, in large volumes,

quickly, gently, and with minimal energy expenditure
a job for real specialists. Oil-free rotary lobe hybrid
blowers, rotary lobe compressors and screw compressors from AERZEN can handle any kind of transport
need. With unparalleled efficiency and reliability.
www.aerzen.com

117x190_1man_DE-EN.indd 1

www.eptq.com

ina.indd 9

YEARS

15.09.14 16:59

PTQ Q4 2014

67

15/09/2014 16:20

I NNOVATION
High performance

SUPERFRAC

TM

trays for
difficult separations

8-pass SUPERFRAC high performance tray for 10 m column

YOU CAN RELY ON US.

Highest combined CAPACITY and EFFICIENCY

single-pass cross-flow tray tested at FRI.
The SUPERFRAC tray has been industry validated for more than 25 years in new
tower designs and revamping of existing trays. Suitable for all tray operating regimes,
the SUPERFRAC tray provides:
Increased capacity AND efficiency when revamping conventional trays
Cross-flow enhancement and optimized flow distribution that promotes high tray
efficiency for superfractionator applications
Complete access for maintenance inspections without needing to remove the trays

United States (316) 828-5110 | Canada (905) 852-3381 | Italy +39 039 6386010 | Singapore +65-6831-6500
For a complete list of our offices, visit our Web site.

www.koch-glitsch.com
K KOCH-GLITSCH and SUPERFRAC are trademarks of Koch-Glitsch, LP and are registered in the USA and
various other countries. YOU CAN RELY ON US is a trademark of Koch-Glitsch, LP. SUPERFRAC technology is
protected by patents in the USA and various other countries; other patents pending.

koch.indd 1

10/09/2014 11:50

Structured packing in a CO2 absorber

Solvent regeneration is the biggest energy consumer in an amine treating unit
but efforts to minimise energy consumption should be made cautiously
RALPH WEILAND and NATHAN HATCHER
Optimized Gas Treating, Inc.

tructured packing in deep CO2

removal applications such as
LNG production can offer
significant advantages, including
higher throughput, over other
internals. Piperazine promoted
MDEA is the most common choice
of solvent for this application.
However, tight designs, as typically
specified in offshore situations, can
make such units hard to operate.
This article presents a case study in
which
absorber
performance
seemed to be disproportionately
affected by the reboiler duty of the
amine regenerator.
Liquefying natural gas enables it
to be transported economically over
immense distances to end users
remote from the gas source. Shale
gas is a large resource for natural
gas liquids (NGLs), but it is also
the source of enormous amounts of
gas. The so-called shale gas revolution is what is in large part
responsible
for
driving
the
construction of new LNG plants
and specialised shipping.
Most shale gas is sweet in that it
contains little or only nuisance
amounts of hydrogen sulphide and
other
sulphur
compounds.
However, significant concentrations
of CO2 are normal. Before gas can
enter liquefaction, CO2 must be
removed to less than 50 ppmv (and
sometimes even lower), referred to
as deep CO2 removal, and it must
then be dehydrated to an almost
moisture-free state, usually using
molecular sieves. This article is
concerned with the CO2 removal
step and focuses on amine treating
as the most commonly used
process.
Although a number of amines are

www.eptq.com

ogrt.indd 1

used in the CO2 removal step in

LNG production, by far the most
common solvent is based on
N-methyldiethanolamine (MDEA)
with the CO2 reaction kinetics
promoted through use of modest
concentrations of the activator,
piperazine. The reason is that, as a
tertiary amine, MDEA does not
form a carbamate by reacting with
CO2 and therefore it has a much

Before gas can enter

liquefaction, CO2
must be removed to
less than 50 ppmv,
referred to as deep
CO2 removal
lower heat of absorption than
primary and secondary amines. This
translates into lower solvent regeneration
energy
consumption;
however, the down side is that
MDEA by itself is unsuited to deep
CO2 removal. The CO2 absorption
rate is too slow and the phase equiis
often
librium
with
CO2
unfavourable.
To get to 50 ppmv CO2 in a
Raw gas
Temperature, C
Pressure, barg

20
60

Composition
CO2, mol%
Methane, mol%
C2+,mol%

2
85
13

Table 1

column of reasonable physical

height it is necessary to speed up
the absorption process. This is done
using a few weight percent piperazine (typically 3-9 wt%). The rate
constant for the piperazine-CO2
reaction has the extraordinarily
high value of 50 000 L.gmol1.s1
even at room temperature, making
this amine an obvious choice as a
CO2 absorption rate promoter. All
major solvent vendors offer at least
one or two formulations of piperazine with MDEA for deep CO2
removal applications, and some
include such solvents as part of a
licensed process.

Case study

The study is of an LNG related CO2

removal unit using piperazine
promoted MDEA in which the
absorber contains structured packing and the regenerator is trayed.
The packing specific (dry physical)
area is nominally 250 m2/m3. Table
1 is a simplified summary of the
conditions and composition of the
raw gas. The solvent is 50 wt% of a
proprietary, piperazine promoted
MDEA contaminated with about
mol% of heat stable salts. The gas is
a typical LNG unit feed.
The treating process uses a
completely conventional gas treating flow sheet with absorber and
regenerator tied together through
the usual rich amine flash for
hydrocarbon
recovery,
cross
exchanger, trim cooler, and pumps.
Before embarking on a plant optimisation, it was decided to
determine sensible operating ranges
for various parameters and to find
out how sensitive overall treating
performance was to the most

PTQ Q4 2014 69

10/09/2014 13:43

10000

Lean pinched

Bulge pinched

CO2 in gas, pmmv

1000
100
10
1
0.1
0.01

Treated
Equilibrium
0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

Lean solvent loading, mol/mol

Figure 1 How CO2 concentrations in the treated gas and in equilibrium with lean solvent
depend on solvent lean loading

important ones. This exercise

uncovered a rather surprising
sensitivity. It also demonstrated the
importance of simulating the entire
treating plant, not just the absorber
or regenerator as isolated pieces of
equipment, and of validating data
before an optimisation study is
started.
The first simulation run was
based on what was thought to be
good process information regarding
compositions, flows, thermal duties,
and so on. A thermal imaging scan
of the absorber was also available,
taken on the same day the process
information was recorded on the
plants DCS. The big surprise was
that the simulated peak temperature in the absorber was found to
be 40C hotter than the thermal
scan indicated, and the simulated
bulge temperature covered most
of the towers packed height.

The simulated temperature profile

could not have been further
away from the thermally imaged
one! But the main red flag was that
the simulated solvent lean loading
was nearly 0.07 moles CO2 per
mole of total amine. This is a ridiculously high value obtained by
simulating the entire plant (less
than 0.010.02 is more normal for a
well maintained solvent). These
observations indicated that bad
data had likely been input into the
simulation. However, the broad
temperature bulge made it worthwhile first to isolate the absorber to
assess the sensitivity to lean
loading.
Figure 1 shows that the residual
CO2 in the treated gas (blue line in
the figure) steadily climbs with
increasing lean loading, but it
suddenly escalates explosively as
the loading passes 0.033 mol/mol.

Distance from top, m

0.002
0.005
0.010
0.020
0.030
0.032
0.033
0.035
0.040
0.060

4
8
12
16
20
24
30

40

50

60

70

80

90

Temperature, C
Figure 2 Changing temperature profiles with lean solvent loading

70 PTQ Q4 2014

ogrt.indd 2

100

On the other hand, the CO2 concentration in equilibrium with the lean
solvent climbs continually and
steadily throughout the loading
range (as it should). Below the transition point at 0.033 mole loading,
the CO2 leak closely follows the
lowest achievable level consistent
with a given lean solvent. Because
treating very closely tracks CO2
equilibrium over the lean solvent, it
can be said that treating in this
region is thoroughly lean end
pinched. As will soon become
evident, this term means that the
CO2 concentration in the treated
gas is determined entirely by the
solvent lean loading. Nevertheless,
the transition itself is abrupt. The
reason is subtle but revealing.
Figure 2 shows a series of absorber
gas-phase temperature profiles for
several solvent lean loading values.
The curves at the low loading end
are what one might expect for CO2
absorption by a fast reacting solvent
such as piperazine promoted
MDEA; the curves at the high loading end, however, are not. The
observation that the peak temperature at the bulge increases with
increasing loading provides the first
clue to an explanation. There are at
least two reasons the bulge temperature itself increases with loading:
heat capacity decreases as loading
goes up that is, the same heat
release results in higher temperature; and the heat of absorption
itself increases with temperature
and this exacerbates the effect.
Somewhere between a lean loading of 0.032 and 0.033 mol/mol, a
bulge temperature is reached at
which the partial pressure of CO2
in the gas right at the bulge is equal
to the partial pressure in equilibrium with the solvent there. In
other words, at the transition loading, the driving force for absorption
becomes zero. At a lean loading
only slightly above this point the
zero driving force explodes across
most of the upper part of the
column until the cold lean solvent
draws the temperature down near
the top of the packing and absorption
resumes.
The
curves
represented by the solid lines all
correspond to lean-end pinch
conditions. The dashed curves are

www.eptq.com

15/09/2014 11:54

INNOVATIVE SOLUTIONS FOR THE

HYDROCARBON PROCESSING INDUSTRY

Bilfingers engineering department constantly works to offer the best and innovative solutions to the Hydrocarbon Processing Industry:
JOHNSON SCREENS SHAPED SUPPORT GRID (SSG)
designed to be installed into the bottom head of hydroprocessing
or gas dehydration vessels, allowing better liquid and gas flow,
bed utilization, distribution and an overall more efficient process
than traditional flat surface grid assemblies. Patented design.

JOHNSON SCREENS INLET DIFFUSER BASKET

designed to control velocities of gas or liquid distribution over
media, providing improved performance over traditional plate disc
type distributor designs as well as even distribution and minimal
scouring at the top of the bed. Patented design.

BILFINGER WATER TECHNOLOGIES

www.water.bilfinger.com
Australia - Asia Pacific
Phone +61 7 3867 5555
Fax +61 7 3265 2768
asiapacific.water@bilfinger.com

biflinger.indd
1
PTQ_Issue Q2.indd
1

France
Phone +33 5 4902 1600
Fax +33 5 49021616
france.water@bilfinger.com

North & South America

Phone +1 651 636 3900
Fax +1 651 638 3171
usa.water@bilfinger.com

04/03/2014
10:57
03/03/2014
17:50:54

Distance from top, m

0
4

0.032 actual
0.032 equilibrium
0.035 actual
0.035 equilibrium

8
12
16
20
24
10-6

10-5

10-4

10-3

10-2

10-1

Mole fraction CO2 in gas

Figure 3 Example of lean-end pinch and bulge pinching

all bulge pinched, meaning that

along the flat part of these curves
there is essentially no driving force
for CO2 transfer in either direction.
As the bulge spreads further across
the column at higher lean loadings,
less and less gas is absorbed and
this causes the peak temperature to
decrease slowly.
Lean-end and bulge pinches can
be more easily understood from
plots of composition across the
absorber. Figure 3 shows equilibrium (dashed lines) and actual
(solid lines) mole fractions of CO2
in the gas for the 0.032 and 0.035
lean loading cases. At the lower of
the two loadings, below the transition loading, only the bottom
section of the absorber has a
concentration difference driving
force for absorption. In the top
section (the lean solvent end) the
driving force is zero; hence, the

operation is termed lean-end

pinched. At the higher loading in
the bulge pinched region, fully the
middle three quarters of the
column has no driving force. The
driving force is pinched exactly
along the temperature bulge.
Consequently, here the column
operation is called bulge pinched.
Gas treating engineers are often
surprised to learn that systems
using piperazine promoted MDEA
in LNG production, and other deep
CO2 removal applications such as
ammonia and hydrogen production, can undergo a sudden
behavioural change near the transition point where the operation goes
from a normal lean-end pinched
condition to a bulge pinched state.
Indeed, the existence of a bulge
pinch is a new concept to many
engineers who are otherwise well
versed in acid gas removal.
2.0

0.08

1.5

0.06
0.05

1.0

0.04
0.03

0.5

0.02
0.01
0
0.5

Reflux ratio

Lean loading

0.07

Lean
Reflux ratio
0.6

0.7

0.8

0.9

1.0

Relative reboiler duty

Figure 4 How lean loading and reflux ratio respond to reboiler duty; a relative duty of
unity is recommended normal operation

72 PTQ Q4 2014

ogrt.indd 3

However, piperazine as a promoter

for MDEA has come into general
patent-free use only relatively
recently, so exposure to the details
of this technology has been limited.
In addition, mass transfer ratebased simulation is really the only
way to identify such behaviour,
and rate based simulation is a relatively new technology, albeit a very
powerful one.
It may be worth noting that, in
this instance, the 50 ppmv CO2
target can be met with a lean loading of around 0.03 mol/mol. This is
an
unusually
high
value
for piperazine-MDEA solvents.
However, because it would require
only very low energy inputs, it
might be tempting to operate there.
Unfortunately, the absorber could
prove to be quite difficult to control
in that part of the loading range. To
keep away from a hard-to-control
region of the operating map, one is
forced to use higher reboiler energy
inputs and to strip to quite a bit
lower loadings, resulting in far
better than the 50 ppmv CO2
specification.

Regeneration

The two main parameters that are

used to control solvent loading are
the solvent flow rate and the
reboiler duty. Solvent flow is
normally used to limit the rich
loading so that, for example,
corrosion rates remain reasonably
low. It is not usually used to
control lean loading. Instead,
lean loading control is done by
manipulating the flow rate or
temperature of the steam or hot oil
entering the reboiler. As Figure 4
shows, lean loading is fairly
responsive to duty at low reboiler
energy inputs although the solvent
becomes gradually harder to strip if
only because piperazine holds CO2
more tenaciously than MDEA does.
A reflux ratio of 0.7 (right side
scale) is clearly insufficient to reach
a lean loading consistent with the
treating specification. The specification can be readily met at a reflux
ratio of 0.8 or higher, and reflux
ratios in the range of 0.80.9 are
normal for piperazine promoted
MDEA solvents in deep CO2
removal applications.

www.eptq.com

10/09/2014 13:43

Conclusion

The possibility of a bulge pinch is a

fact of life when piperazine
promoted MDEA solvents are used
for deep CO2 removal in LNG,
ammonia, and hydrogen production. A bulge pinch starts to form at
the temperature bulge when the
CO2 concentration in the gas being
treated reaches a value equal to the
equilibrium value at the bulge. At
that juncture, further absorption is
not possible. Absorbing just a
minuscule amount more CO2 in the
immediate vicinity of the bulge
raises the temperature there to the
bulge value too. Thus the bulge
spreads. Perhaps the surprise is
how rapidly the bulge can spread
with seemingly minuscule increases
in lean solvent CO2 loading.
Solvent regeneration is the
biggest energy consumer in an
amine treating unit. Efforts to minimise
energy
consumption,
however,
should
be
made
cautiously. Non-rate based simula-

tion is oblivious to bulge pinches,

so reboiler energy reduction efforts
should be made with the aid of a
real mass transfer rate based simulation tool, such as the ProTreat
simulator. This will ensure that
absorber operation is kept well

The possibility of a
bulge pinch is a fact of
life when piperazine
promoted MDEA
solvents are used for
deep CO2 removal
away from the bulge pinched
region and that the unit will function as intended.
The culprit responsible for causing this study to be undertaken in
the rst place was an incorrect
reboiler duty. Perhaps the main

lesson of this article is that it really

pays to validate all the data entered
into a simulation, including the
units, although recognising the
possibility of bulge pinches could
have saved a number of deep CO2
removal plants from the failures
that occurred on start-up.
ProTreat is a mark of Optimized Gas Treating,
Inc.

Ralph H Weiland is a co-founder of Optimized

Gas Treating with offices in Clarita, OK,
Houston, TX and Buda, TX. He holds BASc,
MASc and PhD degrees in chemical engineering
from the University of Toronto.
Email: ralph.weiland@ogtrt.com
Nathan A Hatcher joined Optimized Gas
Treating, Buda, Texas, as Vice-President,
Technology Development, in 2009. 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

Real zero leakage

from your

reciprocating compressors.

The unique pressure packing solution

XperSEAL achieves leakage-free operation
of your compressor, using a pressurized oil
barrier.

Meet environmental requirements

Increase reliability and efficiency

Optimize workplace health and safety

For more information:

info-hkth@hoerbiger.com

www.hoerbiger.com
ptq_XperSEAL_2014.indd 1

www.eptq.com

ogrt.indd 4

09.09.2014 11:10:47

PTQ Q4 2014 73

10/09/2014 13:43

enersul.indd 1

10/09/2014 15:57

Reflux in a gas dehydration plant

Gas dehydration by adsorbent processes may lead to the damaging regeneration
reflux phenomenon during adsorbent regeneration
SAJAD MIRIAN and HOSSEIN ANISI Nitel Pars Co (Fateh Group)
XIANG YU Hengye Chemical Co
SEPEHR SADIGHI Research Institute of Petroleum Industry

ehydration of natural gas

entails the removal of water
that is associated with natural
gases in vapour form. The natural
gas industry has recognised that
dehydration is necessary to ensure
smooth operation of gas transmission lines. This pretreatment
prevents the formation of gas
hydrates and reduces corrosion.
The three major methods of dehydration are
direct cooling,
adsorption
and
absorption.
Adsorption-based processes for
separation of multi-component
gaseous mixtures are becoming
increasingly popular. The new
generation of synthetic and more
selective adsorbents developed in
recent years has enabled adsorption-based technology to compete
successfully with traditional gas
separation techniques.
Any adsorption-based separation
process requires two essential steps:
adsorption during which one or
more components are preferentially
adsorbed/separated; and regeneration
during
which
these
components are removed from the
adsorbent bed. The adsorbent is
repeatedly used in cycles by carrying out these two steps. When a
regeneration step is carried out
through reduction of the total pressure, the process is called pressure
swing
adsorption
(PSA).
Temperature swing adsorption
(TSA) is another technique used for
regenerating a bed of adsorbent
that is loaded with the targeted
impurity gas. This technology
began commercially in the 1960s
and continues today for drying
continuous air and natural gas
as well as other purification

www.eptq.com

ripi.indd 1

Solid caked layer

of adsorbent and
salt fused together
Effective
bed diameter
Original
bed diameter

Figure 1 Schematic of a bed faced with

regeneration reflux

applications such as carbon dioxide

stripping from air. TSA exploits the
capacity of certain adsorbent materials, such as activated alumina,
silica gel and zeolites, to adsorb
gases at moderate temperatures
(40C, 100F) and later release them
when the temperature rises above
120C (250F).
Natural gas treating units using
molecular sieves and TSA technology are usually optimised by
manipulating both the adsorption
and the regeneration time. By reducing the adsorption time, both the
vessel size and the amount of adsorbent used are reduced. Therefore,
the total cycle time is usually
designed such that at the end of the
adsorption a short time is available
for appropriate regeneration of the

adsorbent. Hence, the inlet section

of the adsorption bed is faced immediately with a high temperature
from the start of the regeneration
without any heating ramp. Heating
up the adsorber without using a
heating ramp causes a strong
temperature difference in the bed.
So, at the bottom, the molecular
sieve is very hot and desorbs the
adsorbed water while the top layers
are still at adsorption (low) temperature. Therefore, water desorbed in
the bottom layer condenses in the
top layer. This phenomenon is
called refluxing or retro-condensation. A schematic diagram of an
adsorber with regeneration refluxing
is shown in Figure 1. To prevent this
catastrophic phenomenon, a good
molecular sieve formulation (binder
and zeolite) or improvement in the
regeneration condition is inevitably
required.
In this article, modelling of the
regeneration reflux phenomenon
during regeneration is performed
and the effects of it on the adsorption
process
are
reviewed.
Recommendations to prevent this
phenomenon in a commercial scale
dehydration unit (as a case study)
are presented.

Process description

The purpose of a natural gas dehydration package is to reduce the

water content of the natural gas to
avoid freezing and hydrate formation in the pipeline. In order to
utilise natural gas for urban
consumption, the water dew point
should be reduced to below -10C,
accomplished by using a molecular
sieve
adsorption
unit
which
adsorbs water from the inlet gas.

PTQ Q4 2014 75

10/09/2014 13:49

Adsorption
mode

Regeneration
mode

Figure 2 Schematic diagram of the

dehydration unit studied

To perform such a process, water

saturated natural gas from the
upstream unit is sent to the molecular sieve dehydration plant where
the gas stream passes through a
separator to retain any free water
carry-over from the upstream facilities. It is then routed to the
molecular sieve dryers. A dehydration package consists of four dryers
loaded with a special type of
molecular sieve 4A; at any time
three dryers are in adsorption and
one in regeneration. The feed
stream is split into three identical
streams, each of which passes
downward through one of the beds
that are in adsorption mode (see
Figure 2).
Dry gas streams leaving the
adsorption beds are joined and
passed through a lter to retain any
solid particles coming from the
dryers. Finally, dry and ltered gas
is sent to the municipal gas station
via a transmission pipeline.
Each adsorption cycle takes eight
hours. After that, the dryer is
switched to regeneration mode for
removing the residual water. At
once, that bed which has completed
Adsorption and regeneration
operating conditions
Specifications
Adsorption temperature, C
Adsorption pressure, kPa
Adsorption mass flow, kg/h
Regeneration temperature, C
Regeneration pressure, kPa
Regeneration mass flow, kg/h

Table 1

76 PTQ Q4 2014

ripi.indd 2

Value
47
9101
2.409e+05
270
7929
4.751e+04

the regeneration step is replaced.

During the regeneration process, a
regenerative gas stream is passed
through a heater where it is heated
to approximately 270C. This hot
gas passes upwards through the
ofine saturated dryer heating the
molecular sieves. As the sieves are
heated up, adsorbed water begins
to desorb and is carried away by
the hot gas. The operating conditions of the target adsorption and
regeneration processes and specications of their feeds are shown in
Table 1 and Table 2, respectively.

Mathematical modelling
of regeneration

A computational uid dynamic

modelling technique was used to
model the momentum, heat content
and mass transfer of uid through
porous media, and also to investigate the reuxing phenomenon in
the regeneration process studied.
To solve these set of equations,
commercial
software
(Comsol
Multiphysics
Ver.
4.2)
was
employed that utilises the nite
element method to discretise partial
differential equations to ordinary
differential equations and nally
solve them. The following assumptions are considered during the
mathematical procedure:
To reduce computation time, 2D
axisymmetric mode is assumed
The gaseous phase is an ideal gas
Entrance and exit effects are
negligible
There is no slip condition near
the dryer wall.

Governing equations

Mathematical modelling of the

target regeneration process is
obtained by coupling a set of
general equations (including continuity, momentum, energy and
mass balances), and particular
equations such as physical properties, adsorption and desorption
isotherms and equation of state as
follows:

Energy equation:
( C p ) eq

T
+ C p u T = (keq T ) + Q
t

Mass equation:
ci
+ .( Di ci ) + u ci = Ri
t

In these equations, (kg/m3) is

the density of the uid; t (s) is the
time; u (m/s) is the velocity vector;
Qbr (kg/m3s) is the mass source or
mass sink; p is the porosity of bed;
P (Pa) is the pressure; (kg/ms) is
the dynamic viscosity of the uid;
(m2) is the permeability tensor of
the porous medium; F (kg/m4) is
Forchheimer drag option; F (kg/
m2s2) is the inuence of gravity
and other volume forces; (Cp)eq is
the equivalent volumetric heat
capacity at constant pressure; T (K)
is the bed temperature; Cp is the
uid heat capacity at constant pressure; keq is the equivalent thermal
conductivity (a scalar or a tensor if
the thermal conductivity is anisotropic); Q is the heat source (or
sink); c is the concentration of the
species (mol/m3); D is the diffusion
coefcient (m2/s), and R is the reaction rate expression for the species
(mol/m3s). Furthermore, the major
particular
equations
are
the
Langmuir adsorption isotherm and
ideal gas law. The proposed equations in 2D axisymmetric mode
have been solved using the
required initial and boundary
conditions.
Feed and regeneration gas
compositions
Components
Adsorption Regeneration
Methane, wt%
72.95
73.1
Ethane, wt%
8.13
8.14
Propane, wt%
4.1
4.11
i-Butane, wt%
1.22
1.22
n-Butane, wt%
1.56
1.56
i-Pentane, wt%
0.00
0.00
n-Pentane, wt%
0.140
0.141
n-Hexane, wt%
1.73
1.73
n-Heptane, wt% 0.1897
0.19
n-Octane, wt%
0.1622
0.1625
n-Nonane, wt% 0.0337
0.0338
CO2, wt%
3.79
3.8
Nitrogen, wt%
4.57
4.58
H2O, wt%
0.1573
0.000

Continuity equation:

+ ( u ) = Qbr
Table 2
t
Momentum equation:
2

u
u

( + (u ) ) = [ PI + (u + (u )T )
( u ) I ] ( + f u + Qbr )u + F
p t
p
p
3 p
kbr

www.eptq.com

10/09/2014 13:49

Sulzer Chemtech
Tower Technical Bulletin

Design and Installation of Cartridge Trays

Introduction
Cartridge trays, also known as package trays, are generally
used for tower diameters in the range from 12 (300mm) up to
36 (900mm). For tower diameters below 36, the installation of
segmental trays is difficult and packing is often preferable over
trays. Ultimately, the decision on what technology to use comes
down to process requirements and economics.
As can be seen below, the design and construction of cartridge
trays is unique and a bit complex. Cartridge trays typically
consist of one or more bundles of 6 to 10 trays stacked together
and connected with several tie rods running through the bundle.
This can be a challenge; the trays must be assembled with
near perfect alignment to ensure trouble-free installation. The
resistance of the tray seal rings increases the force required
to install and remove the tray bundles so proper design and
correct dimensions are critical. Alternatively, Sulzer also offers
Slit TraysTM for smaller column diameters. These trays are
installed individually to help minimize installation issues.

The mechanical design of the cartridge trays should also be

stronger than segmented trays since as they will be transported
in assembled condition. The tray thickness should be increased
(e.g 12ga or 2.5mm when referring to stainless steel) to maintain
rigidity and ensure a tight fit. Stronger tie-rods and Schedule 80
spacer pipes should be specified as well. Since the gaskets are
more prone to distortion, it is preferable to install them after the
trays arrive on site. The selection of gasket material should be
based on temperature and service. Metal gaskets of a suitable
material are often preferred for their mechanical durability.

Cartridge Tray Cross-Sectional Layout

Important Tips
During installation, the orientation of the trays with respect to
nozzles should be fixed prior to the bundle insertion as it will
be more difficult to rotate afterwards. Also, access around the
outside of the column must be properly allocated during the
design process to ensure that there is no external interference
with the bundle during insertion or removal.

Bundles of Cartridge Trays

Design Considerations
The unique aspects of cartridge tray designs are both mechanical
and process related. The trays must have a perimeter deck seal
that maintains its integrity while the bundle is installed. The
downcomers, which cannot seal to the column wall, must use
an envelope design which results in some wasted area behind
the downcomer (shown on the sketch to the right). In order to
properly rate these trays hydraulically, the wasted area must be
accounted for to ensure that it is not inadvertently counted as
active area, AA.
The trays must be fixed together in a bundle form along with a
mechanism to support the bundle within the column. The trays
should be partitioned to maintain a maximum bundle length of
13ft (4m) for ease of handling.

Standard pipe sizes are typically used for columns with

cartridge trays. Care must be taken during construction not
to compromise the diameter and roundness of the column
to ensure that the tray bundles will pass through the without
interference.
The Sulzer Applications Group
Sulzer has over 150 years of in-house operating
and design experience in process applications. We
understand your process and your economic drivers.
Sulzer has the know-how and the technology to design
internals with reliable, high performance.

Sulzer Chemtech, USA, Inc.

8505 E. North Belt Drive | Humble, TX 77396
Phone: (281) 604-4100 | Fax: (281) 540-2777
TowerTech.CTUS@sulzer.com
www.sulzer.com

Legal Notice: The information contained in this publication is believed to be accurate and reliable, but is not to be construed as implying any warranty or guarantee of performance.
Sulzer Chemtech waives any liability and indemnity for effects resulting from its application.

sulzer.indd 1

10/09/2014 12:04

6500
6000
5500
5000
4500
4000
3500
3000
2500
2000
1500
1000
500
0

Temperature, K
543.15

500

Regeneration reflux
zone
450

400

Desorption
zone

350

Starts at 120C-140C

4000 3000 2000 1000

320.14

1000 2000

Figure 3 Temperature distribution during the regeneration process

Results and discussions

Figure 3 shows the temperature

distribution of the adsorption bed
at an early stage in the regeneration
process. As is apparent in this
figure, a high regeneration gas
temperature
(without
enough
ramp-up) leads to a large temperature gradient along the bed, and
creates reflux at the early stages of
the regeneration cycle.
At these operating conditions, due
to the high pressure of the regeneration gas, high moisture concentration
and a large temperature gradient are
inevitable. For the design case, the
licensor charged a molecular sieve

with enough strength against reflux

which could work more than four
years without any malfunction. But
for the next loading, a regular molecular sieve, manufactured by another
company, could not withstand those
conditions. It was observed that,
only three months from the start of
run, the loaded molecular sieve was
ruined due to the reflux phenomenon. It also increased the pressure
drop of the dryers. Therefore, it can
be concluded that the molecular
sieve, especially the binder and additives, should be made of appropriate
raw materials to be capable of resisting the reflux phenomenon and

Recommendations and consequences to prevent reflux phenomena

Recommendation
1 Decreasing the regeneration gas pressure

Consequence
Needs compressor
Higher operating cost
2 Regeneration gas temperature ramp-up
Hot oil system modification (if applicable)
Higher regeneration cycle time
Adsorption cycle time limitation
3 Layer of activated alumina at the top of the bed This approach may minimise the rolling boil
but cannot fix the problem. Based on Figure 3,
the reflux happens through the bed because of
a high temperature gradient, so it can only
reduce the reflux. We can consider it
a modification.
4 Change the heating gas flow direction from the This is costly. Co-current regeneration
top to the bottom of the bed
requires more gas for stripping the
bed completely.
The downward flow pushes heavy liquid
contaminants, and possibly increases fouling
rate.
5 Try to reduce the heat loss through the top of
This can only reduce temperature gradient
the bed by adding extra insulation and even
between the top and bottom of the vessel.
installing a steam tracer
6 Reverse all flows
Bed fluidisation (lifting)
7 Using a special molecular sieve
The bed can possibly operate without any
problem.

Table 3

78 PTQ Q4 2014

ripi.indd 3

preventing operational malfunctions.

As Figure 3 shows, for our case
study liquid water moved downward until it encountered the
heating zone. At this point, boiling
water created a reflux which ground
the molecular sieve into a powder.
Since certain components of the
binder were somewhat soluble in
boiling water, the molecular sieve
subsequently became a wet cake
(mud) which was then baked by the
rising hot gas. These soluble components could ion exchange with the
zeolite and/or combine with anions
in water to form solid salts (Na2CO3,
CaCO3, MgCO3, NaNO3, and so on).
These solid salts could then paste
the remaining pellets or beads
together to form a solid mass. This

A high regeneration
gas temperature
(without enough
ramp-up) leads to a
large temperature
gradient along the bed
solid mass, formed in an annulus
shape with a centre opening of less
than one foot, did not allow gas to
pass through, and consequently
reduced the effective diameter of the
bed (see Figure 1).
Therefore, boiling water destroyed
the molecular sieve such that the
severity of the operating conditions
should be greatly reduced to extend
the replacement period of the adsorbent. The regeneration reflux
showed some undesirable effects on
the adsorption process which can be
summarised as follows:
Molecular sieve particle break-up
Increasing pressure drop
Gas channelling
Premature water breakthrough
which all lead to poor adsorber
performance.
As a consequence, these effects
increased the reflux phenomenon
with the following malfunctions:
High pressure regeneration gas
High moisture concentrations
Large temperature gradients
High degree of solubility of

www.eptq.com

10/09/2014 13:49

material. Tungsten and molybde- Conclusion

Nitel Pars
a insubsidiary
of Fateh The
molecular
sieve and howof
to prevent
it, paper
development
a recycling
num
are Company,
present
the purified
60
250
ppm
CO
1000
ppm
CO
Group,
for
technical
assistance
and
financial
presented
at
the
Gas
Processors
Association
solution which is mixed with process for NiW spent catalysts by
high-fidelity
support.
ppm
COFebthat
Europe,
2003,a2000
Paris.
ppm CO model
50precipitate calcium500responses
Eurecat,
LElectrolyse
and Valdi
reagents
Green-fieldtoprojects
share many of the
with
control &
5 T Wa J non-complementary
Crittenden B, Adsorption Technology
tungstate
or a mixture
of
calcium was subsidised by the European
same
characteristics
as
OTS
projects
Further reading 40
Design,imported
Heinemannfrom
Publication,
1998.
layer
the operating
and was
awarded Best
of
molybdate
and tungstate.
The
yield Union
for existing
plants,
but they
1 built
Agarwal
A, Advanced
Strategies
fortend
Optimal plant.
6 NieldUse
D, Bejan
Convection in Porous
Media,
the A,high-fidelity
control
the
Best
project
status
in
2009.
of
tungsten
recovered
one shot
is emulation,
to
be less
challenging
Design
and technically
Operation
ofinPressure
Swing
3rd
ed.,
Springer,
2006.
but be tolerant of the
30
After
several
step
developments
more
than
85%.driven.
These
and more
schedule
This ismetallic
Adsorption
Processes,
Ph.D.
Thesis,
Carnegie process
7 Le Bars
M,
Worster
M G,
Interfacial
model
and
accept conditions
thatin
the
process,
in
2013
Valdi
successconcentrates
because
thereare
is2010.
notcalcined
the same and
wealthsold some
Mellon
University,
between
a
pure
fluid
and
a
porous
medium:
controller tuning is likely to
20
fully
treated
morealloy
than
800 tonnes
to
steelmakers,
moreabout
particularly
to be
2 of
Serbezov
Sotirchos
S V,the
Particle-bed
implications
for binary
solidification,
J. of
detailedA,knowledge
beneficial.
Fluid
Mechanics,
550,
2006, 149-173.
of
spent
NiWcontinue
catalysts.
New
investcompanies
looking
for tungsten
operating plant
and assumptions
and
OTSs
will
to
become
10
8 Bearaccurate,
J, Bachmat
Y,
Introduction
Modeling
ments
made
inbut
2011
(a will
newto
roaster)
units,
for become
example
the EuroW more
estimates
more acceptable.
this
not
be
of
Transport
Phenomena
in
Porous
Media,
Data
are
usually
more
readily
available
and
2013
(hydrometallurgy
capaccompany which
produces
tungsten
achieved
through
software
devel0
Kluwer
Academic
Publisher,
1990. to reach
and reasonableness
becarbides,
judged
by or260opments
ity)
enabled
the
company
carbide
and cemented
200 will220
240
280
300
320
alone.
The
more 340
detailed
Sajad Mirian is a Project Manager in the
less
stringent
criteria.
The
project
team
3000simulation,
tonnes
With
its and
pyroErasteel which produces steel and
Temperature,
C capacity.
the
the more
data
Adsorbents Department of Nitel Pars Co.,
generallyalloys.
recognises
value is more
metallurgical and
tungsten
Thethat
hydrometallurunderstanding
that hydrometallurgiit holds, and,
Tehran, Iran. He holds a MSc in chemical
closely
linkedequilibrium
to is
the delivery
date
cal
the
company
is able to
gical
process
protected
thems002@nitelpars.com.
more engineering
Figure
6 COS
sour
gasby
shifta1%consequently,
H
O,process,
2% H2 Email:
engineering.
2
(to maximise training time) than it is
effort
required.
AsAdsorbent
with
any
engirecycle
aAnisi
large
range
of spent
matepatent.
Hossein
is an
Expert
in the
to accuracy.
Indeed, theisincremental
activity,
the time
and
effort
Michael
A Huffmaster
a process expert neering
Fernando
Maldonado
isNitel
thePars
Business
rials,
including
NiW,
NiMo,
Adsorbents
Department
of
Co. He
benefit
of a high-fidelity
OTS
forprocessing should
and
consultant
to industry
for gas
Development
engineering.
Gas
Treating Catalysts
beand
assessed
against
the
The
pyrometallurgy
process
NiCoMo
CoMo
catalysts.
holds
a BSc
in Manager
chemical
training
is
small

much
can
be
and
treating,
refining
operation,
CO
for
Criterion
Catalysts
and
Technologies,
capture,
value.
The
best
OTS
will
not
be
the
The leaching residues are melted
at
2
Email: ms003@nitelpars.com
achieved
with
relatively
simple, stable
located
in
Houston.
He
has
global
responsibility
and
related
research.
His
activities
regarding
best
choice
for
everyone.
Xiang
Yu
is
an
Application
Engineer
of
Hengye
a high temperature in a submerged
and robust
models.
is China.
the catalyst
Technical
for Criterions
gas treating
business.
sulphur
include adsorption-based
amine treating, Jean-Pierre
model
forrecovery
multi-component
Chemical
Co.,Dufour
Shanghai,
He holds
aand
BSc
electrode
arc furnace.
Nickel
alloys
Industrial
Director
of
Valdi.
He
holds
a
PhD
in&
Prior
to
joining
Criterion
Catalysts
Claus,
tail
gas
treating,
and
tail
gas
treating
separations:
application
to
pressure
swing
in
chemical
engineering.
are obtained from this operation;
metallurgy
and
has
worked
for
20
years
in
the
Technologies
in
2001,
he
held
positions
as
a
catalyst
development,
design
and
operation.
adsorption,
Chemical
Engineering Science,
54, Email: hyms@foxmail.com
they
are from
sold
to Oil
steelmakers
while
Martin
Sneesby
is an independent
dynamic
metallurgical
industries.
process
design
engineer,
unit
contact
engineer,
He
retired
Shell
in
2005
with
36
years
1999,
5647-5666.
Sepehr
Sadighi
is
Assistant
Professor
with
system
is complementary
to
a plant simulation consultant with more than 20
the
silico-aluminate
produced
Sophie
isResearch
the superintendent
Laboratory
Manager
with
and Catalysis
anComte
operations
two
US
experience.
HeAdsorption
holds A
a slag
bachelor
of science
3of
Dabrowski
A,
low-fidelity
from
theory years
the
Division
ofinResearch
or
simulation.
ofCoast
experience
inwith
process
simulation
and
isdegree
usedin in
roadworks,
while
dust Valdi.
She
started
the
company
seven
Gulf
refineries.
He
holds
a
bachelor
of
chemical
engineering
from
Georgia
to
practice,model
Advances
in Colloid
and Interface modelling,
Institute of
Petroleum
Industry
(RIPI),
Tehran,
process
with
a complemenincluding
many
operator
training
produced
from
the
melting
stage
is
years
ago
as
R&D
project
manager
and
holds
a
science
degree
in
chemical
engineering
from
Institute
of
Technology
and
is
a
registered
Science,
93, 2001,tuned)
135-224.control layer simulators
Iran. He holds
a PhD inengineering
chemical engineering.
tary (well
and detailed
studies.
in A&M
environmental
recycled.
Texas
University.
in Texas. damaging of PhD
4professional
Meyer P provide
Bengineer
Chr, Hydrothermal
Email:
Sadighis
@ripi.irchemistry.
martin.sneesby@apess.co.uk
might
better overall Email:
COS wet, ppm

dust
obtained
after
the second
and
water
gas shift.
As noted,
hydrogen
binder
materials
in water
third
filtration
stages
is sold to the
is held
constant.
Choosing
an mixture
inappropriate
flow
paper
so the
If theindustry,
entire
is at process
equiliboften need in
adjustment
to work and
well
direction
adsorption
delivers
zero
waste.
The
roaster
rium,
then
COS
would
be the
same
with
even
the
most
accurate
simuregeneration.
and
filtration
system are
the
aslation
ifitsformodels.
COS hydrolysis
or sour
Failure to allow
for
subject
of patents.outlet mixtures are
shift.
However,
controller
tuning
on
the
simulator
Recommendations
and
not
at equilibrium
because of to
space
system
may be detrimental
the
consequences
The
hydrometallurgy
process
velocity
(limited
catalyst
invenoverall projects success.
The
in
The
calcine
thatCOwas
produced
tory).recommendations
Typically,
is proposed
two
or three
Table
3
can
decrease
the
reflux
during
the roasting
is
times
equilibrium
evenprocess
with very
Prognostications
phenomena
which are
reviewed
in
treated
by
hydrometallurgy
in three
high
conversion
because
inlet
High-fidelity
process
simulation
brief
for
the
target
gas
dehydration
steps:
firstarecalcine
is
leached
toif
concentrations
are fairly
high. This
models
already
very
accurate,
unit.
produce
a slurry
containing
tungmeans
that
thethe
sour
for
built well
with
rightshift,
data
and
According
to recommendation
7
sten
andengineering
molybdenum,
depending
good
assumptions,
and
instance,
would
express
a higher
in
Table
3, a special molecular sieve
on
theequilibrium
composition
ofparticularly
the
catalysts
offer
a lot of value,
for
COS
value
than
would
4A
(with
high pathway.
resistance
against
and
their
contaminants
(which
are
OTSs.
However,
not everything
can
the
hydrolysis
As
the
reflux
phenomena),
manufactured
mainly
phosphorus
and
arsenic).
be modelled
Modelling
expression
says,perfectly.
as
goes
CO, In
so
by
Shanghai
Hengye
Chemical
Co.,
toolsCOS.
and Figure
principles
can
still
be
agoes
step
using
decantation
andCOS
filtra6 shows
for
was
into the
target
dryers
improved.
There
is still
room
for
tion,
aloaded
solid
concentrate
various
concentrations
ofcontaining
CO.
about
one
year
ago. To
date,
the
further
development
to
reduce
the
alumina
and
silica,
along
with
dehydration unit has shown a good
granularity
of
models
(such
oxides
of nickel, is produced. Thisas
Part two
performance
and
malfunction
smaller
smaller
timematerial
is volumes,
reserved
for
the pyroThe second
part no
of
this
article
has
been
observed.
steps,
more
detailed
unit
operation
metallurgical
process.
develops reactor
modelling with
models),
although
there
are slurry
diminsoluble
fraction
ofmodel,
the
aThe
kinetic
reaction
the
Acknowledgment
ishing
returns.
iseffects
purified
to
remove phosphorus
of
temperature
and
space
We would
like
to express
great appreciation
The
increase
inourfidelity
of DCS
and
arsenic.
These
contaminants
velocity,
catalyst
activation,
to
Mr F Noorbakhsh
and
Mr M A
Fatemi
for
emulation
that
is
achieved
by
using
are
not
recycled
and and
are disposed
catalyst
deactivation,
determintheir
valuable
and constructive
suggestions
real
hardware
and control
configof
as TGU
hazardous
waste;
they from
repreing
catalyst
health
during
the planning
and
development
of thisa
urations
is laudable,
but it should
sent
lesswork.
than
1%
ofalso
the
starting
commercial
unit
temperature
research
We
would
like
to
thank
be remembered that a control
profile example.

To date, the
dehydration unit
has shown a good
performance and no
malfunction has been
observed

Email: michael.huffmaster@att.net

Email: Fernando.Maldonado@CRI-Criterion.com

HIGH STANDARD VALVES

FOR NON-STANDARD CONDITIONS.
WWW.ZWICK-ARMATUREN.DE

124 PTQ
Q2 2014
ww.eptq.com
www.eptq.com
www.eptq.com

cat
valdi.indd 38
criterion.indd
ripi.indd 4

www.eptq.com
PTQ
Q1
2014
145 29
PTQ
Q3 2014
PTQ Q4 2014 79

12/03/2014
09/06/2014 12:07
12:54
10/09/2014 13:49

EMPOWERING PROCESS
SIMULATION

Improve the past.

Optimize the present.
Design the future.

As a process engineer you need innovative tools

to see your vision come to life.
VMGSim is a comprehensive steady state and
dynamic process simulator that provides flexibility
and control to help design new processes and
improve your existing plant's efficiency and
bottom line.

VISIT

Our

virtualmaterials.com/GetVMGSim
to find out more.

PIONA

molecular

structure

based

oil

characterization system allows high fidelity

modelling of light to heavy hydrocarbons from the
gathering system to refinery products and
distribution.

U.S. Canada Japan Europe Australasia Egypt China Middle East Brazil

vmg.indd 1

10/09/2014 12:18

Override control of fuel gas

When natural gas was to be added to a refinerys fuel gas system, detailed dynamic
simulation of the new system ensured stable operation and a shorter start-up period

Refinery and natural gas

import project

MiRO
(Mineraloelrafnerie
Oberrhein) is one of the largest oil
reneries in Germany. The renery
is located in Karlsruhe and consists
of two sites which are interconnected in multiple ways (feeds,
products and utilities). In the past,
the fuel gas system of the renery
was based on liqueed petroleum
gas (LPG) and fuel oil as the external make-up energy source. Price
changes have made natural gas
nancially attractive as an additional energy source, and thus
MiRO decided to include natural
gas as an alternative make-up gas.
A simplied structure of the new
integrated fuel gas system with the
energy sources renery off-gas,
FCC gas, coker gas, LPG and natural gas is shown in Figure 1.

www.eptq.com

hps.indd 1

Refinery off gas 2

Coker gas

LPG

FCC gas

Refinery off gas 1

Plant 1

Plant 2

Pressure:
8.5 barg

Pressure:
3.8 barg

New

Natural gas

iRO oil renery in Germany

has expanded its fuel gas
system using natural gas as
an additional energy source. The
new natural gas system, which
includes a complex override control
structure, had to be integrated in
the historically evolved fuel gas
system outside of a shutdown.
Therefore, a detailed dynamic
simulation study (DSS) of the new
and integrated overall fuel gas
system was carried out using
UniSim Design. The objectives of
the DSS include ensuring stable
and safe operation of the new
system as well as shortening the
start-up time.
This article gives an overview of
the project and shows initial results
after successful start-up of the new
fuel gas system.

LPG

RAINER SCHEURING Cologne University of Applied Sciences

ALBRECHT MINGES and SIMON GRIESBAUM MiRO Mineraloelraffinerie Oberrhein
MICHAEL BRODKORB Honeywell Process Solutions

Figure 1 Structure of MiRO fuel gas

system (simplified)

In order to meet the varying

requirements of the renery, MiRO
has developed a complex override
control structure. In addition,
piping and control of the new natu-

Price changes have

made natural gas
financially attractive
as an additional
energy source
ral gas system had to be integrated
in the historically evolved fuel gas
system outside of a shutdown.
Therefore a dynamic simulation
study of the overall fuel gas system
was carried out with the following
main objectives:

Ensuring the integrated fuel gas

system is stable in all operating
conditions and transition phases
Ensuring there are no oscillations
or other dynamic problems
Testing the control conguration
and pre-tuning the controllers prior
to installation
Ensuring additional control objectives (min/max ow rates, and so
on) are met
Testing that transitions from LPG
to natural gas and back, as primary
fuel, are easy to handle
Safe commissioning
Reducing commissioning time.

Dynamic simulation of fuel

gas system

Dynamic Process Simulators such

as Honeywell UniSim Design,
Invensys Dynsym, or AspenTech
Aspen Hysys Dynamics are based
on rst principle process modelling
engines that allow realistic modelling of the transient behaviour of
processes typically found in the oil,
gas and chemical industries. In
order to create a process model, the
user selects readily available
components and thermodynamic
packages to dene physical properties and phase equilibria for the
system and then creates a owsheet by adding and linking generic
unit operation models (pipes,
vessels,
pumps,
distillation
columns, for instance) and control
equipment (valves, PIDs, and so
on). The resulting model can be
initialised to a specic initial condition and run through different
predened scenarios as part of a
dynamic simulation study.
Dynamic simulation studies are a
standard tool in the process indus-

PTQ Q4 2014 81

10/09/2014 13:58

Figure 2 Small section of the UniSim model

tries for analysing and optimising

transient
process
behaviour.
Application examples for operability or safety studies include
dynamic are load estimation in
reneries1 or onshore gas elds,2
and compressor studies.3 For the
DSS carried out at MiRO renery,
the authors selected UniSim Design
for its modelling speed, stability
and its capability to model complex
control systems.
The new natural gas system
had to be integrated into the fuel
gas system using existing piping
and vessels as it was not possible
to add new equipment outside
of a shutdown (as explained
above). This existing equipment is
not optimised with respect to
capacity and pressure drop, and
could lead to problems in the transient behaviour of the integrated
system. Figure 2 shows a small
section of the overall UniSim
model, where piping of the real
plant is modelled with great
accuracy.

Override control

Override control is a control strategy

where
one
manipulated
variable is adjusted by two or more
Override
controlled
variables.4
control typically uses a PI/PID
algorithm for each controlled varia-

82 PTQ Q4 2014

hps.indd 2

ble. A low or high selector (LS or

HS) chooses between the PI/PID
outputs at a given time. Integral
parts of the PI/PID controllers

Override control
is a control
strategy where one
manipulated variable
is adjusted by two
or more controlled
variables
which are not selected are at risk of
integral windup, therefore an
anti-windup strategy is required.
One option is external reset feedSP1
PV1

C1

Simulation analysis of
many scenarios

SP2
OP1

LS

OP2

C2

PV2

OP
ERF

back, which prevents windup and

ensures that the outputs of all
controllers are equal.5 Figure 3
shows a possible realisation of an
over-ride controller with two
controllers (C1 and C2), two
controlled variables (PV1 and PV2),
a low selector, external reset feedback (ERF), and one manipulated
variable (OP).
The override control system of
MiROs fuel gas system is far more
complex and has to manage:
A large number of controlled
variables and controllers
An elaborate structure
Various requirements (safety
limits, optimal operating point, and
so on).
As an example, a part of the
override control scheme is shown
in Figure 4.

ERF

Plant

Figure 3 Override control with external

reset feedback

In order to ensure that the objectives shown above are fullled,

extensive simulation analyses of a
wide range of realistic scenarios
were performed. For instance, in
the case of a sudden shutdown of a
100 MW red heater at ve minutes
simulation time the fuel gas system
pressure stayed in the stipulated
range as shown in Figure 5. The red
curve represents the pressure. The

www.eptq.com

10/09/2014 13:58

Providing value through

performance
At Spirax Sarco we help our customers
to reduce emissions and improve energy
efficiency.
We are the first for steam solutions in the oil and
petrochemical industry:
38 Operating Companies across 6
continents
A global network of over 5,000
steam experts
9 manufacturing sites worldwide
34 customer training centres across
the globe

For more information visit

www.spiraxsarco.com/opc

spirax.indd 1

10/09/2014 11:58

205
PI
061

205
PC
001

205
PI
034

MID OF THREE
205
PI
001

Vessel
105m3
SV=
17.5 bar

8.50 bar

205
PC
260

205
PC
01B

HGP
260
P4
LOSEL

205
PI
259

HGP
260
P1
HISEL

HGP
061
P6

8.75 bar

HGP
260
P3

TARGET

Ps=19.5 bar

MAX

HGP
260
P2

HGP
061
P2

HGP
034
P1
205
PC
02A

HISEL
HGP
061
P1

Vessel

12 bar

HGP
002
I1

Flare

205
PC
60C

HISEL
HGP
061
P4

HGF
241
P1

13.8 bar

205
PC
034

MIN

LOSEL

205
FC
259

7.7 bar

8.30 bar

205
PC
061

LOSEL

HGF
259
I1

241
PI
257

8.70 bar

HGP
061
P3

HGF
259
P1

205
TI
259

205
PI
035

205
PC
60B
205
PI
060

205
HC
061

Evaporator

GDR
2
PC

52m3

Evaporator

205
PC
60A

205
PC
002

10 bar

10 bar

Figure 4 Part of override control scheme of MiRO refinery

PC101-PV
PC101-OP

PC101-PV, barg

9.5

84
70

9.0
56
8.5
42

PC101-OP, %

10.0

8.0
28
0

10 12 14 16 18 20 22 24 26 28 30

Time, minutes

Figure 5 Trend of fuel gas system pressure in the case of a 100 MW oven shutdown

pressure set point is 8.5 bar(g), and

the pressure rises up to 9.1 bar(g).
The blue curve shows the active
OP, which directly controls the
valve connected to HGP 260 P4 in
Figure 4.

84 PTQ Q4 2014

hps.indd 3

Because many PID controllers are

involved, which may interact and
cause oscillations, PID parameter
tuning was a central point of the
simulation analysis.
As a result of the investigations, a

number of issues were adjusted

such as:
Pressure limits for flare gas relief
Valve sizing
Control parameters.
On the whole, simulation analysis
has verified that the design of the
fuel gas system, including the natural gas system, and the complex
override control system is fully
functional and error free.

Commissioning of natural gas system

In December 2013, the natural gas

system was integrated into the
existing fuel gas system. Standard
control engineering tasks, such as
linearisation of non-linear valve
curves, were pursued where necessary. The control parameters which
had been designed in the simulation study provided good starting
points for detailed parameter

www.eptq.com

15/09/2014 12:00

Prdicted viscosity

93.0
tion study helped
to Validating
develop a fully
Ivelina
Shishkova
is R&D
Department
functional and error
free
system.
As
Training
92.5 Neftohim
Manager
with
Lukoil
Burgas. She
a consequence, commissioning
Capacity (KBPD) and
holds a MS in300
organic chemistry engineering
start-up of the expanded
fuel gas
DMCplus applications
92.0
and a PhD in petroleum
refining from Sofia
system went
smoothly
without
APC score
index any
Chemical and250
Technological and Metallurgical
significant
problems.
220 more than 20
91.5authored
University, and
has

APC score index

(Pace Setters)

4 Sawarkar A, Joshi J, Pandit A, Kataria K,

Kulkarni K, Tandon D, Ram Y, Petroleum residue
tuning.
the control
number In
of many
APC cases,
applications
andJ.
upgrading
via
visbreaking:
a review,
Can
higher
FCC of
Cthe
yield
led tosimulahigher
4 manipulated
parameters
dynamic
the
number
variaChem.
Eng., 85, of
2007,
1-24.
production
ofused
alkylate,
which
tion
could
be
without
any
bles
could
still
increase,
which
5
Sadighi
S, Ahmad
A, An
optimisation
resulted
in
the
production
of
changes.
would produce
a significant
gain
ina
approach
for increasing
the profit
of2%
more
grade
gasoline.
commercial
VGO of
hydrocracking
Can.
As a premium
result
the
careful
preparabenefits,
approaching
or process,
exceeding
J.tion,
chem.commissioning
Eng.,
13,top
2013,
1077-1091.
went smoothly
those
of
the
performers.
6
Soufi
H.Al,
Savaya
Z , Moahmmed
H, Alwithout any significant
problems.
It
References
Azami
I,
Thermal
conversion
of
heavy
Iraqi
Conclusions
is
noteworthy
to
emphasise
the
1 Watanbe
K, Nagai1714-1715.
K, Aratani N, Saka Y,
residue,
Fuel, 67,
Through
the1988,
joint
commitment
of
short
period
of time
that octane
was
7 Chiyoda
KrishnaN,R,Mizutani
KuchhalH,Y,Techniques
Sarna G,forSingh
I,
Isab
and AspenTech,
the 20th
Adaptive
required
forin FCC
commissioning
and
enhancement
gasoline,
Annual
Visbreaking studies on Aghajari long residue,
Process
Control
technology
made
startup
of Symposium,
the expanded
gasa
Saudi-Japan
Dhahran,fuel
December
Fuel,
67, 1988,
379-383.
2010.
17.
Montgomery
J
A,
Guide
to
difference
at
two
major
sites.
system.
8 Carlo S.Di, Janis B, Composition Fluid
and
Catalytic Cracking,
Partsuccesses
1, 1993.
Moreover,
these
proved
visbreakability
of petroleum
residues,
Chem.
Conclusion
thatSci.,
Adaptive
Process Control will
Eng.
47, 1992, 2695-2700.
MiRO
had
expand
change
the
new
9
Benitorefinery
A
M, way
Martinez
MtoTapplications
, FernandezitsI,
Ivan Chavdarov is a Chemical Engineer in
Miranda
J L, Visbreaking
ofprofit
anthe
asphaltenic
fuel
gas
system
usage coal
of
are built
and
APCfor
sustained
the Process Engineering department of Lukoil
residue,
74,around
1995.
natural
gas
as
an
additional
alternaatNeftohim
IsabFuel,
and
the
world.
Burgas, Bulgaria. His activities are
10
Delenergy
Bianco A, source.
Panartili N,AAnelli
M, natural
Beltrame
tive
new
focused
on guiding the operation
of the units
P,
Carniti
P,
Thermal
cracking
of
petroleum
References
gas
which
includes
a
of thesystem,
FCC complex,
troubleshooting
support
residues
1. Kinetic
analysisQ,of the
reaction,
Fuel,
1 and
Harmse
M, Zheng
Golightly
R,
An
optimisation
of the
performance
of the
complex
override
control
structure,
72,
1993, 75-80.
Enhanced
Iterative Process
APC
FCC complex.
was
integrated
into for
theMaintaining
historically
11
Trauth
D, -Yasar
M, Process
NeurockControl,
M, Nasigam
Applications
Adaptive
Aspen
Email:
Chavdarov.Ivan.S@neftochim.bg
evolved
gasS, system
outside
of
A,
Klein Stratiev
M,fuel
Kukes
Asphaltene
and resid
Technology,
Inc.
Dicho
is Chief
Process
Engineer
with
a
shutdown.
In
order
to
ensure
safe
pyrolysis:
effect
of reaction
environment,
Fuel
2 Lukoil
LodoloNeftohim
S, Harmse
M, Esposito
A, Autuori
A,
Burgas.
He
holds
a MS
in
andTechnol.
stable
operation
of the new
Sci.
Int.,
10(7),
1992,
1161-1179.
Use
adaptive
modeling
to
revamp
andamaintain
organic
chemistry
engineering,
and
PhD and
system,
well
asrefi
aProcessing,
smooth
startup,
12
F, Lababidi
M,
Al-Rabiah
H,
controllers,
Hydrocarbon
2012.
a AlHumaidan
DSc in as
petroleum
ning
from
the
Burgas
a University
dynamic
simulation
study
of
the
Thermal
cracking
kinetics
of
Kuwaiti
vacuum
Assen Zlatarov. He has authored
residues
in Eureka
process,
Fuel,103,
2013, 923fuel
gas
system
was
carried
out
moreVedernikov
than
130
papers.
Oleg
is Deputy General Manager
931.
using
UniSim
Design.
The
simulaEmail:
Stratiev.Dicho@neftochim.bg
Technical Director of ISAB Refinery, Siracusa,
13 Kataria K L, Kulkarni R P, Pandit AB, Joshi J
B, Kumar M, Kinetic studies of low severity

Rainer Scheuring is Professor of Automation

Technology
and Control Theory at Cologne
Today
Regular A-92
University of Applied
Sciences. He holds a
Premium A-95 300
masters in technical cybernetics and a PhD
A-98
from Super
Stuttgart
University, Germany.
Email: rainer.scheuring@fh-koeln.de
RVP = 60kPa
209

200
200
1%
technical papers.
200
186
Albrecht Minges
is a Senior Process Control
28%
Email: Shishkova.Ivelina.K@neftochim.bg
91.0
Engineer in the department for process
Rosen Dinkov150
is the Quality Manager in the
automation at the MiRO-Refinery in
UniSim Design is a mark of Honeywell Inc.
Process Engineering
117
90.5 department of Lukoil
114 Germany. He holds a BS degree
Karlsruhe,
100
97
Neftohim Burgas.
include
92
100 His research interests
in
process
engineering
from University of
82
79
76
crude oil characterisation,
bio/conventional
GAP
90.0
Applied Sciences, Mannheim.
References
fuels blends characterisation
and
modelling
of
90.0
90.5
91.0
91.5
92.0
92.5
93.0
50
Email:
Minges@miro-ka.de
71%
1refiGruber
D, Leipnitz
D-U, He
Sethuraman
26
nery distillation
processes.
a MSP,
viscosity
18
18 holds
16
16Measured
13
8
Alos
M A, Nogues
J M,
Brodkorb from
M, AreBurgas
there
in organic
chemistry
engineering
Simon Griesbaum is a Process Control
0
alternatives
to
an
expensive
overhaul
of
a
University and a PhD1 in the 2technology
3 of 4 Engineer
Isab
the for
department
for ofprocess
Figure
6 Comparison
of the measured
and predicted5datainAverage
points
the viscosity
bottlenecked
flare system?,
Q1 2010,
RVP =
50kPa
fossil and synthetic
fuels fromPTQ,
the University
automation at the MiRO-Refinery in
fuel
oil
93-95.
1%
of Chemical Technology and Metallurgy, Sofia.
Karlsruhe, Germany. He holds a masters in
2Email:
Panigrahy
P, Balmer
J, Alostoday
M A, Brodkorb
Figure
6 Overall
APC score:
26%
Dinkov.Rosen.K@neftochim.bg
chemical Engineering
engineering Department,
from University
of
Reaction
Research
visbreaking,
Ind.
Eng.
Res.,the
43,bottleneck,
2004.
M,
Marshall
B, Dynamics
Vladimir
Jegorov
is Chem.
thebreak
Sales
Development
Applied
Sciences,
Mannheim.
Institute of Petroleum Industry (RIPI), Tehran,
14
Bellosfor GGrace
D, in
Kallinikos
L2011,
E, Gounaris
Hydrocarbon
Engineering,
Sept
93-96.
Manager
the Lukoil
CIS region.
Prior to
Italy,
one of the
biggest
refineries.
He Email:
in Italy,
where he works with customers
griesbaum@miro-ka.de
He
holds a PhD in chemical engineering
C
E, Papayannakos
N
G, M,
Modeling
of How
the Iran.
3
Nugues
J
M,
Brodkorb
Feliu
J
A,
joining
Grace,
he
was
an
FCC
process
engineer
has around 20 years experience in refining, across Europe in advanced process control
Universiti Teknologi Malaysia.
performance
of
industrial
reactors
using
can
dynamic
process
simulation
used
for from
at the
Mazheikiai
refinery
inHDS
Lithuania.
including
Russian
(Perm)
andbeAmerican
and energy management. He has around 30
Email:
Sadighis
@ripi.ir
acentrifugal
hybrid
neural
network
approach,
Chem.
Eng.
Michael
Brodkorb
is Software Sales Support
compressor
systems,
Hydrocarbon
Petko Petkov
is a fullto professor
and rector
(Houston)
approaches
refining management.
years
of field experience in advanced
73%EMEA at Honeywell Process Solutions
Process,
44,
2005,
505515.
Leader
Engineering,
Aug
2012,
92-98.
of the
Burgas
University
Assen heavy
Zlatarov.
He
He
holds
five patents
covering
residue
process control in the refining, chemical
15
Abonyi
J,the
Babuska
R, Szeifert
F, Modified
Tarragona, Spain. He has a degree in
4
Luyben
L, social
Essentials
of Process
Control,
teaches
in M
science
department
inin in
conversion
processes,
delayed
coking
and petrochemical industries and has
Reza
Seif engineering
Mohaddecy from
is Project
Manager,
Gath-Geva
fuzzy
clustering
for122-125.
identification
chemical
University
of
McGraw-Hill,
New
York,
1997,
the
fi
eld
of
oil
refi
ning
and
lubricants,
and
has
particular, and a PhD in technical science.
implemented more than 100 MPC and other
Figure
3
Effect
of
changing
the
RVP
on
Catalysis
and
Nanotechnology
Division,
of
Tagaki
fuzzy
models,
IEEE
Trans.
Syst.
Dortmund,
Germany,
and
a
PhD
from
Bradford
5authored
Smith Sugeno
C
A,
Automated
Continuous
Process
more than 180 scientific papers and
automation projects. He holds a masters
nery gasoline
producedDepartment,
during
Catalytic
Reaction
Engineering
Man
Cybern,
32,Wiley
2002, 612.
University,
UK. grades
Control,
John
& Advisor
Sons, Hoboken
NJ, refi
five books.
Stefano
Lodolo
is Senior
and Advisory
degree in chemical engineering from Bologna
the
Resolution
RIPI.
He
holds catalyst
a MS inperiod
chemical engineering
Email:
michael.brodkorb@honeywell.com
2002,
88-92.
Email:
PST_Petkov@abv.bg
Business Consultant with Aspen Technology University, Italy.
Sepehr Sadighi is Assistant Professor, Catalysis from Sharif University of Technology.
and Nanotechnology Division, Catalytic Email: Seifsr @ripi.ir

QUICK, ACCURATE, PORTABLE

DISTILLATION
MINIDIS ADXpert

Portable, True Atmospheric D7344 Distillation

Excellent D86, ISO 3405, IP 123 Correlation
Fully Automated Measuring and Cleaning
6 mL of Sample
No Glassware, no Open Flame
Distillation of Gasoline, Jet Fuel, Diesel, Solvents,
Unknown Samples, Raffinates and Biodiesel
Specified for Use by US Pipelines

Grabner Instruments Messtechnik GmbH | A-1220 Vienna, Austria | Dr. Otto-Neurath-Gasse 1

Phone +43-1-282 16 27-0 | Fax +43-1-280 73 34 | info.grabner-instruments@ametek.at
ADXpert half page 1.indd 1

www.eptq.com
www.eptq.com
www.eptq.com

q2 grace.indd
ripi.indd 5 4
hps.indd
4
aspentech.indd
5

06.12.2013 11:46:37

PTQ
201491 121
PTQ
Q1Q2
2014
PTQ Q3
Q4 2014 47
85

10/03/2014
14:48
11/12/2013
12:49
10/09/2014
10/06/2014 13:58
13:37

Theres more
than one
way to skin the

CAT!

Spent catalyst
recyclers
use
them on their
vacuum truck load
out systems. The
flat rotating disc
renews the sealing
surfaces each time
the valve is cycled. No
other valve is similar, with vacuum through 10,000
psig (689b) and temperatures to 1500F (815C). See
more specifications on the reverse side.
Whether its fresh, spent, dry or slurry use
the Everlasting Rotating Disc Valve and see your
CAT valve problems disappear. There are many
ways of skinning the CAT, but one best valve for
catalyst slurries.

ryx GTL slurry phase distillate

(SPD) raw wax catalyst slurry is
the system where Everlasting Valves are
installed. This unique valve was designed
nearly 100 years ago for blowing down steam
locomotive boiler solids. The U.S. Department
of Energy in development of the clean coal
technologies selected this valve as state of the art
valve for dry solids and slurries.
A major licensor of refi ning processes now
specifies the Everlasting rotating disc metal seated
valve for Fluid Catalyst Crackers handling fresh
catalyst and for hot CAT withdrawal. It is also listed in
the Best Practices of a world class petroleum refi ner
for continuous catalyst reforming. A preeminent
catalyst additive supplier has standardized on this
valve concept for all their chemical injection units.

Call us today at 1-908-769-0700 or visit our website at (www.everlastingvalveusa.com)

Valve Manufacturers Since 1906

Boiler Blowoff Valves

Bulk Material Valves

Process Valves

American Boiler Manufacturers Association

everlasting.indd 1

Everlasting

VALVE COMPANY, INC.

09/12/2013 11:27

Snakes and ladders for maximising

propylene
Changes in process conditions in tandem with ZSM-5 additives widen the
potential for petrochemicals production from the FCC unit
BART DE GRAAF, MEHDI ALLAHVERDI, MARTIN EVANS and PAUL DIDDAMS
Johnson Matthey Process Technologies

he modern day equivalent of

turning lead into gold is
upgrading oil into petrochemicals. Whereas in some parts of the
world uid catalytic cracking (FCC)
units are being shut down because
of poor economics, new units are
still being built in growth regions
like Asia and the Middle East.
These new FCC units are typically
designed as petrochemical FCC
units, in particular to produce
propylene, often from heavy residual feedstocks. Nowadays, margins
for standard FCC units producing
mainly transportation fuels are
small or even negative, however
not so for petrochemical FCC units.
Demand for petrochemicals is
increasing, especially in Asia where
there is a shortfall in mixed xylenes
and propylene production. Mixed
xylenes exports from the US
account for 800 000 t/y or 25% of
the Asian markets consumption.1
Propylene is a key chemical for
plastics such as polypropylene,
acrylonitrile,
propylene
oxide
derivatives, and so on. Global
demand for polypropylene in 2000
was 25.1 million t/y, increasing to
42.3 million t/y in 2011 and is
expected to grow to an estimated
62.4 million t/y by 2020. This
growth is largely driven by increasing demand in Asia Pacic (to 62%
of global demand), the Middle East
and Africa.2 Growth in these
regions outpaces that in North
America. Future growth in the
European market is expected to be
very limited because of the
Eurozone crisis.
There is a clear growing demand
for petrochemicals, and the FCC
unit is well positioned to meet it.

www.eptq.com

j matthey.indd 1

Worldwide, 60% of the propylene

is supplied by steam crackers, 30%
by FCC and 10% by propylene
on-demand units. To boost propylene yield in the FCC unit, typically
ZSM-5 additives are used. The
widely assumed model of how
ZSM-5 additives work, that is
commonly used, is simple: ZSM-5
cracks olens out of the gasoline.
This article shows this model to be
incorrect. ZSM-5 does not just crack

There is a clear
growing demand for
petrochemicals, and
the FCC unit is well
positioned to meet it
gasoline range olens, it produces
these too via oligomerisation.
ZSM-5 acts both as a snake and
ladder, as in the board game, for
olens in the FCC unit. The interaction between hydrogen transfer (see
Hydrogen transfer: 3 olefins + naphthene
3 paraffins + aromatic
Naphthenes (cyclo-paraffins) donate
hydrogen to olefins)
-6H
Naphthene
Aromatic
(cyclo-paraffin)
Naphthene =
hydrogen donor
3 olefins

+6H

3 paraffins
Olefins = hydrogen
acceptors

Figure 1 Hydrogen transfer reactions are

the reactions between naphthenes (good
hydrogen donors) and olefins (good
hydrogen acceptors)

Figure 1) and snakes and ladders

for olens determines gasoline
composition in the FCC unit. With
this new model, gasoline compositions when using ZSM-5 can be
explained and the gasoline cracking model is left wanting.

Maximum propylene chemistry

Many reaction pathways play an

important role when cracking a
vacuum gas oil (VGO) or residue
feedstocks into products over an
FCC catalyst. The primary step is
cracking, that is catalytically
breaking carbon-carbon bonds (also
known as beta scission). In this
reaction, individual molecules react
with acid sites on the catalyst forming carbenium ion intermediates
that are readily able to crack.
Cracking results in olen formation
via a) cracking large parafns, b)
ring opening of naphthenes, c)
dealkylation of aromatic sidechains, and d) the subsequent
cracking of the olens formed in a)
to c):
a) Paraffin cracking: large paraffin Olefin
+ Smaller paraffin
b) Ring opening: Naphthene (ring) Olefin
c) Dealkylation: Alkylaromatic Olefin +
Aromatic
d) Olefin cracking: Large olefin Smaller
olefin + Smaller olefin

Concurring reactions include

thermal cracking, hydrogen transfer, dehydrogenation, cyclisation,
trans-alkylation,
oligomerisation
and polymerisation.
Thermal cracking and dehydrogenation are the primary sources of
dry gas (C1-C2 and H2), and polymerisation produces coke. These
reactions do not signicantly

PTQ Q4 2014 87

11/09/2014 14:22

directly contribute to the creation

of propylene (or other light olefins),
but may have an indirect contribution via their impact on FCC unit
constraints.
Propylene yields from the FCC
unit are improved by increasing
operation severity, conversion, and
through the use of ZSM-5 zeolite
containing additives. The higher the
conversion the higher the propylene
yield will be, especially when in the
naphtha overcracking region (typically overcracking begins at 70-75
wt% conversion depending on feed
properties). Operating parameters
are modified in such a way that
beta scission is maximised relative
to thermal cracking, dehydrogenation,
hydrogen
transfer
and
polymerisation. Various process
licensors have developed technologies for this purpose. Typically, the
following process parameters are
optimised to maximise conversion
within unit constraints:
Increase cracking temperature
Increase catalyst circulation/catalyst to oil ratio
Increase Ecat activity
Improve feed quality hydrotreating for example.
Other parameters that significantly help to increase propylene
selectivity are:

Minimise hydrocarbon partial
pressure. Reducing hydrocarbon
partial pressure via dilution with
riser steam shifts the reaction equilibrium towards light olefins by
reducing light olefin oligomerisation (recombination) and hydrogen
transfer reactions
Reduce hydrogen transfer reactions. These have the undesired
effect of saturating light olefins to
paraffins. This can be done by:

Reducing
contact
time:
advanced riser termination systems
provide rapid catalyst/oil disengagement
which
decreases
hydrogen transfer
Reducing backmixing: minimising catalyst slip in the riser reduces
hydrogen transfer

Reducing fresh catalyst rare

earth (RE) on ultrastable-Y (USY)
zeolite is of particular importance
in reducing hydrogen transfer. By
reducing hydrogen transfer reactions, the gasoline becomes more

88 PTQ Q4 2014

j matthey.indd 2

olefinic and more readily crackable.

Gasoline olefins are the prime
source for LPG olefins. A catalyst
with minimum rare earth helps to
maximise propylene selectivity
Naphtha recycle - to crack every
potential propylene precursor out
of the gasoline
Use a ZSM-5 additive.
All of these parameters help to
maximise the propylene yield, but
by far the greatest variable for
increasing propylene yield is the
use of ZSM-5 additives. ZSM-5
additives selectively crack gasoline
range molecules into LPG olefins
(C3 and C4), with highest selectivity
for propylene typically 50-60 wt%
of the incremental LPG is
propylene.
Maximum propylene FCC units
employ high levels of ZSM-5 additives, often 10% or more in the
circulating inventory. Some concern
has been expressed that at
extremely high inventory levels the
ZSM-5 additive can cause dilution
of base catalyst activity. However,
dilution effects are rarely observed
in practice because: target Ecat
activity is maintained by adjusting
the fresh catalyst make-up rate if
slightly higher addition rates are
required they effectively lower the
age of the inventory; and ZSM-5
additives do not significantly
contribute to delta-coke (about
20-30% of the delta coke of the base
catalyst), so as long as the FCC unit
has some capacity to increase catalyst
circulation
rate
ZSM-5
additives will not negatively affect
conversion.
ZSM-5 additives are designed not
to make coke; lowering delta-coke
reduces regenerator temperature.
The FCC unit responds to maintain
the heat balance by increasing catalyst circulation to sustain the
reactor temperature and restore
conversion to the additive free
level. For maximum propylene
resid applications the delta-coke
reduction by ZSM-5 additives is a
considerable benefit as it allows for
either a higher conversion or for
introducing a heavier feed.

Effects on gasoline composition

ZSM-5 additives crack gasoline into

LPG, mainly propylene. However,

introduction of ZSM-5 additives

into the FCC unit also changes the
gasoline composition substantially.
There is a great deal of literature
discussing whether ZSM-5 additives increase the aromatics content
in FCC gasoline specifically,
whether any increases in aromatics
in gasoline are simply due to
concentration effects, or whether
ZSM-5 can produce aromatic
components as well.3,4,5 Various
conclusions are reported, including
all permutations: gasoline aromatics may either increase, decrease, or
stay the same depending on the
study. This discussion on whether
or not FCC gasoline range aromatics are formed under FCC
conditions is remarkable considering that there are many industrial
applications in which light olefins
are used as feed, or appear as intermediates, in the formation of
aromatics. The first of these
olefin-to-aromatics
processes
became commercial in 1935.6,7 These
processes make it clear that propylene will react to form aromatics
over ZSM-5 at both high and low
pressure. However, the majority of
the studies mentioned previously
indicate that ZSM-5 does not significantly increase the yield of
gasoline range aromatics under
FCC conditions or ZSM-5 may
even reduce gasoline aromatics
slightly. This article seeks to resolve
this issue by studying in more
detail various reaction mechanisms
and the effect of ZSM-5 on gasoline
composition.

Pilot plant study

An ACE pilot plant study was

conducted to establish effects of
ZSM-5 additives on light olefin
yields and gasoline composition.
For this purpose, an FCC equilibrium catalyst (Ecat) with a rare
earth level of 1.2 wt% was selected.
The relatively low rare earth level
was chosen to avoid excessive
hydrogen transfer reactions, which
is important when maximising
propylene selectivity. This catalyst
was blended with various levels of
equilibrated SuperZ Excel, a leading, commercially available ZSM-5
additive. ZSM-5 is more stable
under
laboratory
deactivation

www.eptq.com

10/09/2014 14:05

BIG ENOUGH TO EXECUTE.

SMALL ENOUGH TO CARE.

GLOBAL CAPABILITIES
UTILITY FLARES
VAPOR COMBUSTORS

AIR-ASSIST FLARES

STEAM-ASSIST FLARES
CEB ULTRA-LOW EMISSIONS BURNERS

FLAME RING LOW BTU FLARES

VAPOR RECOVERY SYSTEMS

SERVICE AND RENTALS

www.aereon.com | sales@aereon.com

aereon.indd 1

10/09/2014 11:20

sabin.indd 1

27/02/2013 15:01

ZSM-5 additive effect on LPG

ZSM-5 additives selectively convert

gasoline range components into
LPG range olefins and have no
significant effect on coke, slurry or
LCO yields.8 Under extreme conditions, an increase in dry gas is
observed,
where
ZSM-5
can
produce small amounts of ethylene.
The observed LPG components
boosted by ZSM-5 usage are
propylene, mixed butylenes and
isobutane. However, the ZSM-5
additive itself selectively produces
light olefins: propylene and butylenes. It does not directly produce a
significant amount of isobutane.
The observed increase in isobutane
is due to secondary conversion of
isobutylene on the base catalyst via
hydrogen transfer. The rate of
hydrogen transfer is much higher
for iso-olefins compared with
n-olefins so there is minimal loss
of propylene and n-butylenes to
propane and n-butane respectively
via this mechanism. Higher rare
earth on base catalyst leads to a
greater conversion of isobutylene to
isobutane, whereas lower rare earth
leads to greater preservation of
isobutylene (the ratio of isobutane
to isobutylene may be used as a
hydrogen transfer index in Ecat
monitoring).
ZSM-5 is also known as an
oligomerisation catalyst for light
olefins.6 Propylene has been shown

www.eptq.com

j matthey.indd 3

14
12

C3=, wt%

10
8
6
4
2
0
55

Base
50% ZSM-5 additive
25% ZSM-5 additive
5% ZSM-5 additive
60

65

70

75

Conversion, wt%
Figure 2 Using ZSM-5 additives significantly boosts the propylene yield

to be oligomerised over ZSM-5

(and other acid catalysts) to form
gasoline range olefins. In the FCC
these oligomers can then re-crack,
forming more propylene and butylene. This has been termed the
snakes and ladders9 effect, and
has the result of dynamically redistributing the olefins formed in the
FCC from ethylene to very large
olefins. The final composition
depends on several variables,
including ZSM-5 concentration,
pressure and residence time. In this
way olefins in the FCC unit play
snakes and ladders, with a double
role for ZSM-5 as both snake and
ladder.
Overall, ZSM-5 additives substantially increase the propylene yield
in the FCC (see Figure 2). Increasing
the amount of ZSM-5 additive
boosts the propylene yield signifi-

cantly; however the effect on

butylene
yields
is
not
as
pronounced. The incremental effect
of ZSM-5 additives on butylene
yields strongly decreases when
ZSM-5 concentration reaches very
high levels. Increasing the ZSM-5
additive concentration from 25% to
50% in inventory continues to boost
the propylene yield, whereas for
butylenes no further increase was
observed

ZSM-5 effects on gasoline

composition

In addition to cracking gasoline

range molecules into light olefins,
ZSM-5 isomerises gasoline range
n-olefins into iso-olefins of the same
carbon number: these too are
subject to hydrogen transfer on the
base catalyst where they are readily
converted into iso-paraffins. Since

65

Gasoline aromatics, wt%

conditions than the REUSY containing FCC base catalyst. Because

ZSM-5 zeolite has a higher silica
alumina ratio (SAR) and smaller
micropores than REUSY it requires
a
more
severe
deactivation.
Deactivation conditions recommended for ZSM-5 additives are
815C for 20 hours in 100% steam, a
protocol that would be overly
destructive for the zeolite-Y present
in regular FCC base catalysts.
The study was carried out at
1060F (570C) to mimic petrochemical FCC operating conditions
(equivalent to a riser outlet temperature of about 550C) using a VGO
feed. In a parallel study, a full
range gasoline was used as feedstock and cracked at 1015F (546C)
using the same catalyst/additive
mixtures.

60
55
50
45
40
55

Base
50% ZSM-5 additive
25% ZSM-5 additive
5% ZSM-5 additive

60

65

70

75

Conversion, wt%
Figure 3 Using ZSM-5 additives when cracking VGO increases the aromatics content of
gasoline

PTQ Q4 2014 91

10/09/2014 14:05

iso-olefins and iso-paraffins have

higher octane numbers than their
straight chain counterparts the
gasoline octane improves partly
due to this isomerisation effect.
Reduction in gasoline yield is an
indication of ZSM-5 additive cracking activity. Gasoline is a mixture of
paraffins, olefins, naphthenes and
aromatics. Under FCC conditions,
gasoline range olefins crack very
readily
on
ZSM-5
additives.
Naphthenes and paraffins are
thought to crack under higher severity conditions, but aromatic rings do
not crack. The observed decrease in
gasoline volume therefore results in
an increase in concentration of the
least crackable parts of the gasoline,
gasoline aromatics (see Figure 3).
Gasoline aromatic cores (the C6
aromatic rings) cannot be cracked
so, as expected, they are preserved
in the final cracked gasoline.
However, C7 and higher gasoline
range aromatics contain substantial
paraffinic side-chains which can be
cracked. Therefore, although gasoline range aromatic cores cannot be
cracked, the aromatic compounds

Composition of gasoline components

when cracking VGO over Ecat and
Ecat + 25% ZSM-5 additive at 68%
conversion. All components reported
on a wt% feed basis
Ecat
Ecat +
Wt% feed basis (no ZSM-5) 25% ZSM-5
Paraffins
15.9%
11.2%
Aromatics
13.4%
12.3%
Naphtenes
3.3%
2.8%
Olefins
9.9%
7.4%
Total gasoline
42.5%
33.7%

Table 1

can be dealkylated. This results in a

reduction in the aromatic compound
content, at the same time as an
increase in the concentration of
aromatic cores in gasoline.
To further study this effect, the
carbon atoms in gasoline range
aromatic compounds were calculated as aromatic cores and
paraffinic side-chains for cases with
and without high levels of ZSM-5
additive. As an example for toluene, xylenes and so on, only the
aromatic core is counted as
aromatic, while side-chains are

Change in gasoline, FF wt%

5
0

5
10
15
20
25

Ecat
Ecat +
25% ZSM-5 additive
Paraffins

SC alkyl

Olefins

Naphthenes Aromatics

Figure 4 Ecat and Ecat with 25% ZSM-5 additive both readily crack olefins and
naphtenes. ZSM-5 cracks more paraffins, and makes slightly more aromatics. Both Ecat
and ZSM-5 dealkylate naphtenes and aromatics (SC Alkyls)
Composition of gasoline components as feed, and when cracking over Ecat and
Ecat + 25% ZSM-5 additive. All components reported on a wt% feed basis

Wt% feed basis

Paraffins, %
Aromatics, %
Naphtenes, %
Olefins, %
Total gasoline, %

Table 2

92 PTQ Q4 2014

j matthey.indd 4

Gasoline
feed
34.9
21.5
11.6
31.5
99.5

Ecat
(no ZSM-5)
32.4
22.7
6.3
14.1
75.5

Ecat +
25% ZSM-5
28.7
23.2
6.4
11.0
69.3

counted as paraffinic carbon atoms.

In this way toluene (CH3-C6H5) has
six aromatic carbons and one
paraffinic carbon atom. Figure 4
shows that the change in aromatic
core yield is small when adding
ZSM-5 additives. The change
however, appears to be negative
(see Table 1). How can gasoline
aromatic cores disappear when
using ZSM-5 additives?
When cracking gasoline over Ecat
with and without 25% ZSM-5 additive the gasoline composition
changes (see Table 2). Most obvious
and expected are the substantial
decreases in olefins and naphthenes,
more remarkable was the reduction
in paraffins. This effect is even more
pronounced when VGO is cracked
over an Ecat/ZSM-5 additive blend
(see Table 1). Whereas Ecat barely
affects gasoline range paraffins,
ZSM-5 additives with stronger acid
sites were able to convert a sizable
fraction of gasoline range paraffins.
In the case of Ecat (without ZSM-5
additive) the observed reduction in
paraffins was mostly via dealkylation of side-chains from naphthenes
and aromatics. ZSM-5 additives
showed a small effect of steric
hindrance, hence a slightly lower
amount
of
dealkylation
was
observed. It is interesting to note
that in this data the reduction in
paraffins was as large as that seen in
naphthenes. This study also showed
that the distribution of gasoline
range aromatics changed: there was
a decrease in heavier gasoline
aromatics (C10-C12 aromatics) and an
increase in light gasoline aromatics
(C6-C8) consistent with dealkylation (see Figure 5).
When cracking gasoline using
ZSM-5 additive, the distribution of
gasoline aromatic species shifted
towards smaller aromatics (fewer
or smaller alkyl groups attached) in
a similar way to that observed
when cracking VGO. However, in
gasoline cracking the total aromatic
cores increased, whereas in the
VGO cracking case total gasoline
aromatic cores decreased slightly
when ZSM-5 was present. In gasoline cracking, aromatic cores were
retained and additional aromatic
cores were formed. In VGO cracking this mechanism also applies,

www.eptq.com

10/09/2014 14:06

Delivering operational calm

Reducing costs and waste, while increasing process uptime is not
only stressful, its critical to operating a profitable refinery today. GE
can help boost your plants efficiency by identifying profit-robbing
problems, applying the most cost-effective treatments, as well as
monitoring and maintaining the results. Keep your refinery operating
safely and reliably, while lowering maintenance and energy costs
with GEs integrated solutions; encompassing water and process
chemicals, equipment, and services.
GEs offerings include:
Comprehensive water and process
chemical solutions
Real-time remote monitoring
and diagnostics
Expert technical assistance for
integrated solutions throughout your refinery
Let us help you manage the stress and
uncertainty of todays refining challenges . . .
and deliver operational calm.

For more information,

visit us at gewater.com or
scan the code above.

ge.indd 1

07/03/2014 13:19

EXPERT INSIGHTS.
RELIABLE SOLUTIONS.
SUPERIOR PRODUCTS.
A PROVEN TEMPERATURE MANAGEMENT PARTNER

TRUSTED PRODUCTS

With over 100 years of experience, Chromalox knows that a true temperature
management expert offers more than breakthrough products. In addition to a heat

INDUSTRY EXPERTISE

trace product line that includes cables and controls for a variety of tasks, we deliver
industry-leading expertise and a team of solutions specialists. Talk to us today and see

RISK REDUCTION
COST CONTROL

how Chromalox can develop an elegant solution for your most demanding pipe and tank
process temperature maintenance, freeze protection, long pipe line heating, foundation
heating, or snow and ice melting applications.

Please contact Chromalox to discuss your needs with a heating specialist today!
www.chromalox.com | sales @ chromalox.com
US Direct +01 412.967.3800

chromolox.indd 1
13124_CHR_PTQ-TMSMagAD_FINAL.indd
1

US & Canada 800.443.2640

UK +44 (0) 208 665 8900

11/09/2014
09:26
9/10/14 4:24
PM

Conclusion

There is huge scope for maximising

petrochemical yields in the FCC
unit. Both process conditions and
catalyst selectivities can be optimised to make this happen. Higher
severity FCC conditions and use of
ZSM-5 additives (ideally with low
rare earth base catalysts) favour
light olefin production. ZSM-5
additives are involved in many
reactions in the gasoline phase. In
addition to the well-known cracking and oligomerisation reactions,
ZSM-5 can promote the creation
of new aromatic species via
oligomerisation and subsequent
aromatisation. All of these reactions
appear to be in dynamic balance
under standard FCC conditions.
Changing reactor conditions can
help to push this to either side,

www.eptq.com

j matthey.indd 5

Ecat
Ecat +
25% ZSM-5 additive
Gasoline feed

Yield, FF wt%

but competes with cracking of

aromatic precursors into light
olefins. When cracking VGO, the
decrease in paraffins with ZSM-5
additive
present
was
more
pronounced than the shift in
olefins. Clearly, ZSM-5 additives do
not just interact with final gasoline
product. ZSM-5 is an active component in shaping the composition of
the gasoline range components via
cracking, aromatisation and oligomerisation (see Figure 6).
Gasoline chemistry over ZSM-5
additives and base catalyst involves
an intricate web of reactions.
Hydrogen transfer over the base
catalyst converts gasoline (and
LCO) range naphthenes into
aromatics. Cyclisation reactions of
higher carbon number (C6+) olefins
leads to naphthene formation,
which can subsequently form
aromatics via hydrogen transfer. In
addition, higher carbon number
naphthenes can be cracked open to
form olefins, which can subsequently be cracked into light
olefins. Light olefins can also
oligomerise to form higher carbon
number olefins which can either
re-crack or subsequently form
aromatics in ZSM-5 additives.
(Light) olefins should therefore be
seen as reactive intermediates shaping the gasoline and LPG
composition, rather than terminal
products in the FCC unit.

6
5
4
3
2
1
0

10

11

12

13

Number of carbon atoms in aromatics

Figure 5 Distribution of aromatic components in the gasoline on feed basis of the

gasoline used as feed and in the products obtained when cracking over Ecat and Ecat +
25% ZSM-5 additive

creating more or less aromatics,

improving or limiting propylene
yields.
*Also known as Chutes and Ladders in the US.
Super Z Excel is a trademark of IntercatJM.
References
1 Allen K, Special Report: Aromatics, The shale
revolution and its impacts on aromatics supply,
pricing and trade flows, Nov 2013, Platts.
2 Global Polypropylene Market Expected to
Grow at a Healthy Compound Annual Growth
Rate of 4.5% during the Forecast Period,
ASDReports, www.asdreports.com/news, 27
Jun 2013.
3 Triantafillidis C S, Evmiridis N P, Nalbandian
L, Vasalos I A, Performance of ZSM-5 as a fluid
catalytic cracking catalyst additive: effect of
the total number of acid sites and particle size,
Ind. Eng. Chem. Res. 1999, 38, 916-927.
4 Biswas J, Maxwell I E, Octane enhancement
in fluid catalytic cracking I. Role of ZSM-5
addition and reactor temperature, Applied
Catalysis, Vol 58, 1, 5 Feb 1990, 1-18.
5 Donnelly S P, Mizrahi S, Sparrell P T, Huss A Jr.,
Gasoline
paraffins

Gasoline
olefins

LPG
olefins

Gasoline
naphthenes

Gasoline
aromatics

Figure 6 LPG olefins are not terminal

products. They can oligomerise and form
naphthenes and aromatics, or be
hydrogenated into paraffins

Schipper P H, Herbst I A, How ZSM-5 works in

FCC, ACS meeting New Orleans, 1987.
6 Haag W O, Lago R M, Rodewald P G,
Aromatics, light olefins and gasoline from
methanol: mechanistic pathways with ZSM-5
zeolite catalyst, Journal of Molecular Catalysis,
17, 1982, 161-169.
7 Quann R J, Green L A, Tabak S A, Krambeck
F J, Chemistry of olefin oligomerization over
ZSM-5 catalyst, Ind. Eng. Chem. Res., 1988, 27,
565-570.
8 Radcliffe C, The FCC unit as a propylene
source, PTQ, Q3 2007.
9 en.wikipedia.org/wiki/Snakes_and_Ladders
Bart de Graaf is FCC R&D Director with
Johnson Matthey. Prior to joining the former
Intercat business in Savannah he joined
Johnson Matthey in the UK working on bio
feedstock conversion processes. He holds a
MSc in chemical engineering and a PhD In
heterogeneous catalysis and chemical processes.
Mehdi Allahverdi is FCC Applications Manager
with Johnson Matthey, Savannah, Georgia, US,
where he works on catalyst evaluation and
development. He holds a PhD in materials
science and engineering from McGill University,
Canada.
Martin Evans is Vice President of Engineering
Technical Services with IntercatJM, responsible
for providing technical assistance on the use
of FCC catalyst additives and for the design
and development of catalyst addition systems
technology. He holds a BSc in chemical
engineering from the University of Wales.
Paul Diddams is Vice President for FCC
Additives with Johnson Mattheys Refineries
Division, with over 25 years experience in
refining and catalysis, mainly in fluid catalytic
cracking. He holds a BSc in chemistry from the
University of Newcastle-upon-Tyne, UK, and a
PhD in physical chemistry from the University
of Cambridge, UK.

PTQ Q4 2014 95

10/09/2014 14:06

Process Insight:
The design and optimization of separation processes is
carried out using process simulators, which utilize various calculation approaches. Two techniques that are widely used for modeling
distillation are the ideal stage model and the mass transfer model.

Ideal Stage or Mass Transfer...

Which Model Should Be Used?
balance requires kinetic rate expressions for all chemical reactions occurring in the system. As with equations for a non-reacting
system, an appropriate model for interface behavior must be used.
Mass transfer models require data necessary to calculate
interphase mass and heat transfer coefficients and interfacial area
based on correlations of the following transport and thermal properties: diffusivities, viscosities, densities, heat capacities, thermal
conductivities, etc. Furthermore, mass transfer models require
detailed information on the column internals. For trays, this includes
information such as weir heights and fraction active area. For packing, this includes surface area per unit volume and void fraction.
If the simulator allows the user to select from various
alternatives for these parameters, knowing the correct selection
may be problematic. Further, the prediction of multicomponent
mass transfer coefficients is of questionable accuracy. These facts
prompt the recommendation that columns modeled with the mass
transfer approach be checked against an ideal stage model with
an expected efficiency until sufficient experience with the particular
application is achieved.

IDEAL STAGE MODELS

The ideal stage model requires a minimum amount of
dataonly equilibrium relationships and enthalpy data for the heat
balance. The assumptions of the ideal stage approach are: 1) that
the vapor and liquid are both perfectly mixed so that the vapor and
liquid leaving a stage are at the same composition as the material
on the stage and 2) that thermodynamic equilibrium is obtained on
each stage. The equilibrium assumption also means liquid and vapor leaving a stage are at the same temperature. Ideal stage models can also account for non-ideal column performance through the
use of reaction kinetics as is done for amine sweetening columns.
Obviously, the main disadvantage of the ideal stage
approach is just thatthe use of ideal stages to model real trays
or packing depths. However, for most processes encountered in
gas processing and other industries, the overall efficiencies are
well established for proper operating conditions of the column. For
systems that are unavailable, similar systems often exist to allow for
efficiency estimation. If not, the mass transfer approach is available
as an option.

CONCLUSIONS
When performed properly, both the ideal stage and mass
transfer approach as implemented in ProMax 4.0 can calculate accurate results for a variety of separation processes with and without
reactions. The ideal stage approach can be used initially to determine appropriate equipment sizes and operating conditions. More
detailed studies can be performed using the ideal stage approach,
the mass transfer approach, or both. Although significant operating experience provides reasonable efficiency estimates for most
processes, the empiricism in scaling up from ideal to real stages or
ideal stages to real bed lengths can be a disadvantage when accurate overall efficiencies or HETPs are unavailable.
The mass transfer approach requires more equipment
design details and does not make use of overall efficiencies or
HETPs. More detailed composition and temperature profiles are
produced by this method at the expense of longer calculation time.
The mass transfer approach may appear more predictive in nature,
but is not necessarily more accurate. It relies on more parameters
that must be estimated, as both require thermodynamic data to
model equilibriumfor the tray composition in the ideal stage
approach and for the interface composition in the mass transfer
approach. Many of these mass transfer parameters are of limited
accuracy but also may be of limited sensitivity in some systems.
Both techniques are useful tools in process simulation.

MASS TRANSFER MODELS

For the end user, the notable feature made available via
the mass transfer approach is the ability to model a column with the
actual number of trays in the unit or the actual depth of packing.
However, there are still several assumptions that are made in this
approach that can have a significant impact on results. Two that are
worth mentioning include the mixing model for trayed columns and
the discretization of the packing depth for packed towers.
Application of the mass transfer model to random or
structured packing requires the column height to be discretized into
vertical segments or stages. For trayed columns, various mixing
models can be used for the liquid and vapor phases. The most
basic assumption is that of complete mixing in both the liquid and
vapor phases. However, the concentration gradients that develop
on a tray can significantly impact the predictions made by this model
since this gradient is the driving force for mass transfer. As the
column diameter becomes larger, the perfectly mixed flow model is
less applicable.
For modeling both liquid phase chemical reaction and
mass transfer, the use of the enhancement factor technique may be
considered. The enhancement factor describes the increased rate
of absorption due to the effect of a chemical reaction. The material

For more information about this study, see the full article at
www.bre.com/support/technical-articles.aspx.

ProMax process simulation software by Bryan Research & Engineering, Inc.

Engineering Solutions for the Oil, Gas, Refining & Chemical Industries
sales@bre.com www.bre.com 979 776-5220 US 800 776-5220

bre.indd 1

12/09/2013 11:21

Troubleshooting a C3 splitter tower

Part 1: evaluation
Distillation trays are prone to channelling and multi-pass maldistribution in large
diameter towers. Multichordal gamma scanning is key for solving such problems
HENRY Z KISTER Fluor
BRIAN CLANCY-JUNDT and RANDY MILLER PetroLogistics

he PetroLogistics giant C3
splitter (see Figures 1 and 2) is
a heat-pumped, 28ft (8.5m)
internal diameter tower operating
at 105 psig at the top. The tower
contains four-pass, equal-bubblingarea fixed valve trays with mod-arc
downcomers (MOAD) on the
outside panels. Open area on the
trays was 15% of the active area.
The tower started up in October
2010 and had experienced operational difficulties during its initial
eight-month run. Tray efficiency
appeared to be very low, about
40-50%, compared to a typical
80-90% tray efficiency experienced
with conventional trays in a C3
splitter. Due to the low tray efficiency it could not produce on-spec
polymer grade propylene. The
separation did not improve (if
anything, it had become worse)
upon turndown. Initial gamma
scans through the centre tray
panels indicated flooding.
PetroLogistics, Fluor (which was
not involved in the tower design),
and the tray supplier formed a task
force to conduct a troubleshooting
investigation to determine the root
cause of the poor performance and
to propose and engineer a fix. The
strategy was to conduct a
field
investigation
combining
PetroLogistics expertise in operating the C3 splitter, Fluors expertise
in distillation design and troubleshooting, and the tray suppliers
expertise in tray design and modification. Tracerco was later brought
in to provide diagnostic expertise
in anticipation of extensive use of
gamma scanning in identifying the
root cause.
The troubleshooting investigation

www.eptq.com

fluor.indd 1

Figure 1 PetroLogistics C3 splitter tower (left), 28ft (8.5m) wide and 309ft (94m) tangent
to tangent

combined hydraulic analysis and

detailed multi-pass distribution
calculations with the specialised
technique of multichordal gamma
scanning with quantitative analysis.7 The hydraulic analysis and
multi-pass calculations did not
identify a reason for the low tray
efficiencies, but confirmed that the
trays are prone to channelling and

maldistribution due to their large

open areas. The gamma scans
showed a maldistributed pattern on
the trays, with high L/V ratios on
the inside panels and low L/V
ratios on the outside panels. The
scans showed vapour cross flow
channelling (VCFC) on the outside
panels. Flooding was observed on
the inside panels well below the

PTQ Q4 2014 97

10/09/2014 14:16

Jeffrey J. Bolebruch Senior Market Manager

Our VectorWall is saving SRU

reaction furnace operators
millions of dollars
TM

This is not a checkerwall. This is not a

choke ring.
This is a VectorWallTM and it will change
the way you look at your reaction furnace.
With unmatched mechanical stability,
the VectorWallTM will increase your furnace
capacity and eliminate ammonia carryover,
saving your company millions of dollars.

For more information check out our new video

at petrochemical.blaschceramics.com
Or contact us: 1.518.436.1263
vectorwall@blaschceramics.com

blasch.indd 1

11/03/2014 16:31

calculated flood point. The scans

pointed at a combination of VCFC
and multi-pass maldistribution as
the root cause.
The investigation identified the
high open slot area (15% of the
active area) of the fixed valves to be
the prime factor inducing the channelling and maldistribution. A
likely initiator of the multi-pass
maldistribution was liquid preferentially flowing to the inside panels
from the false downcomers distributing the flashing reflux to the top
trays panels. This preferential flow
is believed to have occurred
through the gap at which the reflux
pipes entered the false downcomers. Another likely initiator was
channelled vapour blowing liquid
from the outside to inside panels
across the off-centre downcomers.
The high ratios of flow path length
to tray spacing (2.4 to 3.7), high
weir loads, and integral trusses
projecting a significant depth (4in)
into the vapour space were other
conditions that promoted the
channelling.
A short plant outage due to a
problem elsewhere provided the
opportunity for a quick fix. The key
modification was blanking about a
quarter of the valves on each tray
to reduce the tray open slot areas
from 15% to 11%. The gaps at the
reflux pipe entry to the false downcomers were sealed and the false
downcomer heights were raised to
ensure good reflux split to the top
tray panels. Anti-jump baffles were
added across the centre and
off-centre downcomers to prevent
the possibility of channelled vapour
from blowing liquid from the
outside to the inside panels,
towards the middle. Some downcomer blocks were installed to
improve liquid distribution. The
modified tower achieved tray
efficiencies
comparable
to
those obtained in well-operated,
smaller diameter, low pressure C3
splitters.
To the best of our knowledge,
this is the very first time that field
measurements demonstrated interaction
between
VCFC
and
inside-to-outside-pass maldistribution. A lesson learnt is that this
interaction must be considered

www.eptq.com

fluor.indd 2

105 psig
53F

Chemical grade
propylene
3% mole C3H8

PC

FC
Polymer
grade
propylene
<0.5% mole C3H8

Feed

FC

Propane
1% mole C3H6

Figure 2 Simplified process diagram of PetroLogistics heat-pumped C3 splitter and its

auxiliaries

when designing and operating

large diameter towers. Finally, the
investigation highlights that excessive open areas render trays prone
to channelling and maldistribution,
especially in large diameter towers
containing multi-pass trays.
The investigation is described in
two parts. Part 1 describes the
initial operation, as well as the
hydraulic analysis and how it
directed the investigation to focus
on the combination of VCFC and
multipass maldistribution as the
most likely root cause. Part 2 will
describe the application of the
specialised technique of multichordal gamma scanning with
quantitative analysis7 to validate
this theory, closely define and map
the channelling and maldistribution
patterns, and lead to the correct
solution.

Hydraulic evaluation at initial

operating conditions

Figure 2 is a simplified process

sketch of the C3 splitter tower and
its auxiliaries. Data for typical
initial operation were collected at
the highest rates at which operation
was stable, about 20-30% below
design. There is uncertainty about
the reflux flow rate due to a metering error that plagued the reflux

meter. The propylene product

contained 3.4% propane (by mole)
compared to the design 0.5%. The
propylene content of the bottom
stream was a little higher than
design. The tower temperatures
and pressures were similar to
design.
There was a question of whether
the trays in the tower were flooding or not. A typical pressure drop
for good operation is normally
about 0.1 psi per tray, while pressure drops exceeding about 0.2 psi
per tray indicate flooding. For the
C3 splitter, the pressure drop per
tray at operating conditions was
about 0.09 psi per tray, which
argues against flood. In contrast,
the gamma scans concluded that
many of the trays were flooded.
There was a need to reconcile the
two conflicting observations.
To determine whether the tower
was flooded, a plot of the measured
tower pressure drop against the
tower internal vapour traffic was
prepared (see Figure 3). The internal
vapour traffic is approximately the
sum of the reflux and product
meters. A point of inflection in such
a curve indicates the vapour load at
which liquid begins to accumulate
in the tower, and is a good indicator of flood.1

PTQ Q4 2014 99

10/09/2014 14:16

18
16
15
14

fied in the tower were low tray

efficiencies and premature flooding.

Top tower
Full tower
Last 24 full tower
Last 24 top tower
Full tower now
Top tower now

Possible theories

Several theories were advanced for

the low efficiency and premature
flooding. The task force performed
a preliminary review of the theories
to guide the field tests required to
narrow in on the root cause. The
following alternative theories were
proposed.

Tower dP

13.5

13
12
11
10
9

Multi-pass trays maldistribution

Flood initiation

7
6
1500

1600

1700

1800

1900

2000

2100

2200

2300

Vapour traffic (reflux + product)

Figure 3 C3 splitter pressure drop vs internal vapour load from plant operating data

In Figure 3, the upper curve is for

the entire tower, the lower curve is
for the trays between the propylene
side draw and the feed. Both curves
show a point of inflection at vapour
traffic just below 2000 mph, or a
total tower pressure drop of 13.5-14
psi. This suggests that the
initial operating loads in the tower
were right at incipient flooding.
While some flooding could have
started earlier, the significant accumulation of liquid started above
these loads.
With the loads at which flooding
initiated in the tower shown to be
well below the design loads, it was
concluded that the flooding was
premature.
A simulation was prepared based
on the operating conditions on 31
December 2010. Vapour and liquid
loadings from that simulation
provided the basis for hydraulic

calculations. Table 1 shows the

results of these hydraulic calculations. The values in Table 1 were
calculated by Fluor. The Fluor
values were more conservative than
the tray suppliers, but even Fluors
calculations do not indicate proximity to any flood limits. This
analysis verified the conclusion that
the flood observed in the tower
was premature.
The pressure drop values in Table
1 do not include the vapour static
heads, while the tower pressure
measurement does. The static
vapour head is about 2.5 psi, or
0.015 psi per tray. Subtracting the
static pressure drop from the total
tower pressure drop of 13.514 psi
gives 11-11.5 psi per tray of dynamic
pressure drop, or about 0.068-0.072
psi per tray. This is in good agreement with the Fluor calculation.
Overall, the main issues identi-

Hydraulic evaluation at operating conditions

Trays
CACTIVE, ft/s
GPM/in of outlet weir length
DC entrance velocity, ft/s
% Jet flood, FRI
% Froth in DC
% Downcomer choke
Dry pressure drop, in liq/tray
Pressure drop, psi/tray
Pass distribution ratio

Table 1

100 PTQ Q4 2014

fluor.indd 3

Top

Above feed

Bottom

0.236
6.9
0.21
73
73
67
1.01
0.075
1.09

0.237
7.4
0.22
74
75
68
1.01
0.074
1.09

0.227
8.6
0.26
73
78
69
0.92
0.074
1.09

The non-symmetry of multi-pass

trays makes them prone to maldistribution. In four-pass trays, the
inside panels are non-symmetrical
to the outside panels. For good
performance, the L/V ratio needs
to be the same for the inside and
outside panels. Maldistribution
among the panels of multi-pass
trays is a common source of tray
efficiency loss.3,11-14 Lockett and
Billingham14 show that the efficiency loss depends on the degree
of L/V unevenness, as well as the
proximity of the pinch.
Variations in the L/V ratio from
pass to pass also adversely affect
tray capacity. Regions where maldistribution increases the vapour or
liquid loads are pushed closer to
flooding, while other regions where
the loads are reduced have surplus
capacity.
Bolles12 defined the distribution
ratio as the ratio of the maximum
pass L/V to the minimum pass
L/V, and recommended keeping
this distribution ratio below 1.2 to
ascertain good tray efficiency.
Summers11 tightened Bolless criterion to 1.1. Summerss stricter
criterion also keeps the loss of
capacity due to maldistribution
minimal.3
To address this theory, we
applied the Kister et al3 and the
Summers11 multi-pass maldistribution models. These models are
among the most advanced and
most reliable in the industry. The
Kister et al method gave a distribution ratio of 1.09 while the
Summers method gave a distribution ratio of 1.02, so both were in
good agreement, giving distribution ratios below the stringent
value of 1.1 maximum. There were

www.eptq.com

15/09/2014 12:01

some concerns that, with MOADs,

it would be difcult to adequately
model the outlet weir length, ow
path length, and hydraulic gradient. We therefore re-ran the models
with several variations. All showed
that the distribution ratios for trays
in this tower were robust to
changes in these parameters and
remained small. There was nothing
in the calculations that would
explain the observed large drop in
tray efciency and severe premature ooding.
Pros: Large maldistribution would
explain the efciency loss and
premature ooding.
Cons:
Hydraulic
calculations
showed low distribution ratios. The
calculations veried that, by itself,
this theory cannot explain the
observations.
The theory of pass maldistribution was regarded as unlikely, but
it may be combined with other
theories.

vapour and liquid between the

passes, generating or aggravating
inside-to-outside-pass maldistribution. The large open slot area is
also conducive to maldistribution
between passes in multi-pass
trays.3, 15
Pros: This theory explains the
premature ooding. The theory
also agrees with the observation of
efciency loss without apparent
ooding below 13.5-14 psi pressure
drop. This theory explains the
inability to operate at lower loads.
Cons: No experiences have been
previously reported of interaction
between tray channelling and
multi-pass maldistribution. This

theory leaves unanswered questions regarding the nature of the

channelling and its propagation,
and these needed to be investigated
in the test programme.
This became by far the leading
theory, even though at this point
the details were not understood.

Downcomer unsealing

Downcomer unsealing was argued

to be caused by vapour entering
the off-centre and centre downcomers via large gaps where the
supports go through the downcomers. This theory can combine with
channelling on the trays.
Pros: This theory explains the

Channelling combined with

multi-pass maldistribution.

The large open slot areas (15% of

the active areas), provided as part
of the original design to keep pressure drop low for the heat pump
system, can render trays prone to
various forms of channelling such
as vapour cross ow channelling
(VCFC) at the high liquid loadings.
High ratios of ow path length to
tray spacing (2.4 to 3.7), high weir
loads (7-9 gpm/inch of outlet weir)
and integral trusses projecting a
signicant depth (4 inch) into the
vapour space are conditions that
when they come together with high
open areas lead to VCFC.2
VCFC is not the only form of
channelling previously experienced
on distillation trays. There are
reports7 of other forms of channelling, such as due to excessive
forward push (reverse vapour cross
ow channelling, RVCFC) or due to
vapour maldistribution. One thing
they have in common is that they
were only experienced at large tray
open areas, high ratios of ow path
lengths to tray spacing, and high
liquid loads conditions that apply
also for the current trays.
In multi-pass trays, channelling is
likely to interact with the split of

www.eptq.com

fluor.indd 4

PTQ Q4 2014 101

11/09/2014 14:38

premature flooding and low

efficiencies.
Cons:
Hydraulic
calculations
showed that to lose the downcomer
seal it would take a gap about one
square foot in area, so if the gaps
were properly welded this is
unlikely. The trays were thoroughly
inspected, and no gaps were seen,
let alone gaps of this magnitude.
This
theory
was
therefore
regarded as highly unlikely.

Excessive hydraulic loads and

poor metering

Excessive reflux and boil-up rates

due to incorrect metering can overload and flood the trays, giving low
efficiencies.
Pros:
Metering problems have
caused poor operation in many
towers, and there is such a case
reported for a C3 splitter.4 In the
current tower, the reflux flow was
measured by an annubar meter
with no independent check. There
is a meter on the heat pump
compressed gas, but that meter was
not working.
This theory explains the premature flood and low efficiencies.
Cons: The annubar was checked and
rechecked several times. While the
compressed gas meter was working,
a check found the annubar measurement to be within 1% of the value
inferred from the compressed gas
meter. During a crash shutdown the
gas meter was fixed, and again verified the reflux measurement. Also,
the annubar needed to be out by
quite a factor to explain the
observed poor performance. This
theory does not explain the inability
to operate at lower rates and the
poor operation below flood.
This
theory
was
therefore
regarded as highly unlikely.

Foaming

Foaming is known to induce

premature flood.
Pros: This explains the premature
flood. Also, some of the gamma
scan reports mentioned the possibility of foaming.
Cons: We are not aware of any
foaming cases in C3 splitters. Our
survey of tower failures reported in
the literature4 does not include a
single case of foaming in a C3 split-

102 PTQ Q4 2014

fluor.indd 5

ter. This theory does not explain

the poor efficiencies at low rates
and below flood. The tower feed
does not appear to contain foaming
components
in
significant
concentration.
This
theory
was
therefore
regarded as highly unlikely.

Damage

Damage may possibly induce

premature flood.
Pros: This explains the premature
flood. Although uncommon, tray
damage incidents may occur in
heat-pumped C3 splitters. The heat
pump starts up at near full rates,
which renders the tower start-up
bumpy. At start-up, base liquid
level is sometimes raised above
the reboiler return inlet in anticipa-

The theory of pass

maldistribution was
regarded as unlikely,
but it may be
combined with
other theories
tion of rapid boiling upon heat
pump start-up. Generally, base
liquid level rise above the reboiler
return inlet is a common cause
of tray damage, as the reboiler
vapour tends to travel through
the liquid as slugs that can uplift
trays.1
Cons: In C3 splitters, the potential
for slug formation is relatively low
due to the small reboiler temperature difference; as the liquid head
above the reboiler rises, so does the
boiling point, reducing the reboiler
temperature difference and the
boiling rate.
Other sources of damage may be
rapid pressurising or depressurising. We have not seen this kind of
damage in C3 splitters. The huge
volume in this tower is likely to
cushion the tower from this mechanism. Also, the high open area and
low dry pressure drop permit easy
spread of vapour.
Another source of damage is
flow-induced vibrations.5 This kind

of damage tends to occur near

weeping, which may be the case
here due to the high open area.
However, vibrations can be felt,
heard and measured. Also, the tray
supplier and the task force examined this possibility and determined
that it is unlikely in this tower.
Another argument against tray
damage is that the top to bottom
gamma scans did not show any
severe local variations. Usually,
damage shows local flooding or
disturbance (for instance, if there is
high liquid level damage, near the
bottom). In contrast, in the C3 splitter the problem appears to initiate
in every column section. In the
top-to-bottom scans, almost clear
vapour was reached near each
manhole, but the section above
showed much the same pattern as
the section below.
The damage theory was therefore
regarded as unlikely.

Unbolted manways

Unbolted manways are likely to

induce channelling and premature
flood.
Pros: This agrees with the premature
flood. Manways not installed, or
poorly bolted (and therefore lifted)
are common issues in all towers.
The number of manways in this
tower is well above 1000. This is
the largest number of manways we
have seen in a single tower, and a
huge number to be bolted up prior
to start-up. We often see people
leaving manways unbolted in a
20-tray single pass tower! If all, or
even some, of the manways at the
centre panels were left unbolted,
vapour would be channelled into
the unbolted region, initiating poor
efficiency and possible flooding.
Cons: The installers swore that all
manways were bolted adequately.
PetroLogistics personnel were well
aware of the risk and very closely
inspected the manways installation.
They too were sure that all, or at
least almost all, the manways were
properly installed.
The gamma scans show a consistent phenomenon throughout the
tower. This pattern may be consistent with unbolted manways
throughout the tower. Just a few
unbolted manways are unlikely to

www.eptq.com

10/09/2014 16:04

produce the scanning pattern

observed.
The unbolted manways theory
was therefore regarded as unlikely.

Likely theory

In summary, the channelling/maldistribution theory towered high

above the others. However, the
nature of the channelling and/or
maldistribution remained poorly
defined.
Part 2 of this article will describe
the application of the specialised
technique of multichordal gamma
scanning with quantitative analysis7
to validate this theory, closely
define and map the channelling
and maldistribution patterns, and
lead to the correct solution.
References
1 Kister H Z, Distillation Operation, McGrawHill, New York, 1990.
2 Kister H Z, Larson K F, Madsen P E, Vapor
cross flow channeling on sieve trays: fact or
myth?, Chem. Eng. Prog., 86, Nov 1992.
3 Kister H Z, Dionne R W, Stupin W J, Olsson
M, Preventing maldistribution in four-pass

trays (cover story), Chem. Eng. Prog, Apr 2010.

4 Kister H Z, Distillation Troubleshooting,
John Wiley & Sons, NJ, 2006.
5 Summers D R, Harmonic vibrations cause
tray damage, Paper 307g, presented at the
AIChE Annual Meeting, San Francisco, CA, 18
Nov 2003.
6 Harrison M E, Gamma scan evaluation for
distillation column debottlenecking, Chem.
Eng. Prog. 86 (3), 37-44, March 1990.
7 Kister H Z, Use quantitative gamma scans
to troubleshoot maldistribution on trays,
Chem. Eng. Prog., Feb 2013.
8 Kister H Z, Is the hydraulic gradient on sieve
and valve trays negligible?, Topical Conference
on Distillation, the AIChE Meeting, Houston,
TX, April 2012.
9 OBara J, Consultant Report, Carmagen,
April 2011.
10 Green D W and Perry R H, Perrys Chemical
Engineers Handbook, 8th Ed., McGraw Hill,
New York, 2008.
11 Summers D R, Designing four pass trays,
Chem. Eng. Prog, 26 April 2010.
12 Bolles W, Multipass flow distribution and
mass transfer efficiency for distillation plates,
AIChEJ, 22 (1), 153, 1976.
13 Jaguste S D, Kelkar J V, Optimize separation
efficiency for multipass tray, Hydroc. Proc., 85,
Mar 2006.
14 Lockett M J, Billingham, Trans. IChemE., 80,

OGT 128 ProTreat Half Page Horizontal PTQ Q3 2014.indd 1

www.eptq.com

fluor.indd 6

Part A, 373, May 2002; Trans. IChemE., 81, Part

A, 131, Jan 2003.
15 Summers D R, Tray stability at low vapor
load, Conference Proceedings of Distillation
and Absorption 2010, 611, Eindhoven, The
Netherlands, 12-15 Sept 2010.
Henry Z Kister is a Fluor Corp. Senior Fellow
and Director of Fractionation Technology. He
is the author of three books, 100 articles and
has taught the IChemE-sponsored Practical
Distillation Technology course more than 400
times. He holds BE and ME degrees from the
University of NSW in Australia, is a Fellow of
IChemE and AIChE, a Member of the NAE.
Email: henry.kister@fluor.com.
Brian Clancy-Jundt currently works in one of
the largest propane dehydrogenation plants
in the world with PetroLogistics and has had
direct engineering oversight over all aspects of
an olefins plant. He graduated from Texas Tech
University with a BS in chemical engineering.
Randy Miller served as Vice President,
Engineering for Petrologistics (2007-2014),
instrumental in the design and development
of the facility since the commencement of
front end engineering design. He has worked
in the petrochemical industry for over 20 years
and holds a BS in chemical engineering from
Texas A&M University, an MBA from University
of Houston at Clear Lake, and is a Registered
Engineer in Texas.

6/4/14 9:24 AM

PTQ Q4 2014 103

10/09/2014 16:04

TRAYS FOR EVERY PROCESS

TRAYS FOR EVERY PROCESS

For gas processing, rening and a wide range

rang
For gas processing, rening and a wide
of process applications. AMACS offers
off s a wide
offer
wid
range of process applications. AMACS
selection of alloys, sizes and congurations
conguration
offers a wide selection of alloys, sizes
sf requirement. As a
to meet any mass transfer
sfer
and congurations to meet any mass
in the
research and
leader
transfer
requirement.
As adesign
leaderof
intrays
the and

tower
internals,
design
to your
you
research
and AMACS
design ofcan
trays
and tower
speci
or improve
your process
cations
cations
internals,
AMACS
can design
to your with

latest cations
technologies.
ourspeci
or improve your process
with our
latest technologies.
(numerous
options)
Valve
Valve
or(numerous
perforatedoptions)
Sieve
Sievecap
or perforated
Bubble
Bubble cap
Cartridge trays
Cartridge trays
Dual Flow
Dual Flow
Distributor
Gallery
Gallery
DistributorTray
Tray

For all your process internal

For all your process internal requirements
requirements make one stop- AMACS.
make one stop- AMACS.

VALVE

SIEVE OR
PERFORATED

BUBBLE CAP

AMACS 24-7 Emergency Service Available at (281) 716 - 1179

AMACS 24-7 Emergency Service Available at (281) 716 - 1179
2013 Amacs Process Towers Internals. All Rights Reserved.
2013 Amacs Process Towers Internals. All Rights Reserved.
amacs gas.indd 1

CARTRIDGE
TRAYS

DUAL FLOW

GALLERY
DISTRIBUTOR

www.amacs.com
www.amacs.com
Member of Fractionation Research

11/03/2014 09:18

Overcoming tight emulsion problems

A refiners trials of a membrane coalescence mechanism were scaled up to plant
level and delivered significant product recovery from tight aqueous emulsions
HERNANDO SALGADO Cartagena Refinery, Ecopetrol
LUIS MARIO Ramgus S. A. Pall Corp.
ROSNGELA PACHECO Barrancabermeja Refinery, Ecopetrol

ne of the most challenging

problems in refining and
petrochemical processes is
the separation of liquid emulsions
and dispersions, which are often
formed during the purification of
products, especially when intimate
contact between hydrocarbon and
aqueous phases is involved. This is
the case in the treatment of hydrocarbon fuels such as fuel gas, LPG,
naphtha or jet fuel, when amines
and caustic solutions and water are
used to remove or wash contaminants such as H2S, CO2, mercaptans
or naphthenic acids.
In many cases, these liquid solutions or solvents can be entrained
by the main hydrocarbon product,
forming tight emulsions and
dispersions which are difficult to
separate and are often discarded as
effluents to the wastewater treatment unit. On the other hand,
entrainment of these liquid solvents
could potentially affect product
specifications or cause operational
upsets in downstream processes.
Nevertheless,
emulsions
and
dispersions can be separated by
coalescence, taking advantage of
the interfacial tension between
hydrocarbon and aqueous phases.
In the coalescence process, two
drops of a single phase and identical composition make contact with
each other, forming a single bigger
drop and in this way minimising
their specific surface (surface per
volume unit).
There are different types of
coalescing media, from fibre glass
of moderate performance through
to special polymer membranes
which can provide high performance in separating tight emulsions

www.eptq.com

ecopetrol.indd 1

Figure 1 Membrane coalescence steps

such as those with some content of

surfactant compounds.

Tight emulsions

In many refining and petrochemical

processes, formed emulsions are
very stable due to the presence of
small quantities of surfactant
compounds which can be additives
in regular use such as corrosion
inhibitors or anti-foaming and
anti-fouling additives, as well as
contaminants such as naphthenic
acids and sulphur compounds.
The tightness of an emulsion can
be measured in terms of its interfacial tension, which is the free energy
in the contact zone of two immiscible liquid phases. Interfacial tension
is a consequence of the superficial
tension of both liquid phases, and
can be expressed in units of force
per distance, generally in d/cm.
The presence of surfactants leads
to the formation of micelles (of
hydrocarbon in an aqueous phase)
or inverse micelles (of an aqueous
phase in a hydrocarbon), depending
on the concentration of each phase,

either the continuous or the

disperse phase. The internal energy
of an emulsion increases proportionally with surface and interfacial
tension; therefore, the lower the
interfacial tension is, the lower the
free energy. Consequently, the
emulsion is tighter, leading to the
formation of smaller drops which
are more difficult to separate.
The high coalescence and separation performance of polymer
membranes makes them suitable
for separate tight emulsions, which
cannot be treated by using conventional coalescing materials and
equipment.

Membrane coalescence mechanism

Membranes for coalescence are

polymer fibres which vary in diameter sizes and surface treatments,
depending on the application. The
linked structure of the fibres forms
very fine pores which allow droplets in the range 0.2-50 m to be
collected and transformed into a
dispersion of bigger drops of
500-5000 m diameter. However,

PTQ Q4 2014 105

10/09/2014 16:19

Figure 2 Membrane coalescence mechanism

due to the small pore size of the

bres structure, particulate matter
must be removed from the uid
before it is processed through the
membrane.
The membrane coalescence mechanism is dened by the following
steps (see Figures 1 and 2):
1. Removal of particulate matter in
the pre-lter
2. Adsorption of droplets on the
membrane bres
3. Movement
of
droplets
to
membrane bre intersections due to
entraining by process ow
4. Coalescence of two tiny droplets
to form a bigger one when a second
droplet reaches the same bre
intersection
5. Release of bigger drops from the
bre intersections due to accumulation of drops and entrainment by
process ow

6. Repetition of steps 2-5 with

progressively bigger drops and
larger pores in the bre structure.

Membrane coalescence on the

industrial scale

According to the mechanism

discussed above, membrane coalescence is generally designed on an
industrial scale with the following
processing arrangement: pre-ltration, coalescence and separation.
Generally, pre-ltration occurs in
a vessel with ltration cartridges,
with a mesh size depending on the
amount and size of particles previously measured or estimated. The
purpose of this step is to remove
the solid particles which could
increase emulsion stability and at
the same time to protect the
membranes
functionality
by
preventing clogging of pores.

Pre-conditioning
filters

Py-gas sump

Outlet
Coalescers

Inlet

Heavy oil sump

While pre-ltration is arranged in

a separate vessel, coalescence and
separation occur in the same piece
of equipment, although they can be
carried out in separated compartments.
A
typical
process
arrangement is shown in Figure 4.
Even though separation is not a
formal step in the coalescence
mechanism, it is required to achieve
the main goal of coalescence, which
is the separation of the two
liquid phases involved in the
emulsion. The selection of the
proper type of arrangement of the
coalescer-separator depends on the
particular application, requiring (or
not) an additional membrane
cartridge to assure the separation of
phases,
as
in
the
vertical
arrangement.
The vertical arrangement is
generally used to separate an aqueous contaminant dispersed in a
hydrocarbon continuous phase
with interfacial tension as low as 3
d/cm and a small difference in
density between the phases. In such
applications,
the
separation
membrane is made of a hydrophobic material to retain the aqueous
phase in a different compartment
and to allow the hydrocarbon to
ow through downstream in order
to facilitate and assure the separation process.
The horizontal arrangement (see
Figure 3) is generally used to separate
hydrocarbons
from
a
continuous aqueous phase with a
grater difference in density. In this
case, after coalescence is carried out,
the separation of phases is achieved
by settling the aqueous phase.
It should be taken into account
that the processes described above
correspond
to
general
rules;
however, lab or pilot testing is
recommended in order to determine
the appropriate membrane material
to be used, as well as key performance
parameters
such
as
separation efciency and ux
through the membrane, which is
helpful in designing an industrial
facility.

Overcoming tight emulsion problems

in a jet fuel treatment unit
Figure 3 Horizontal arrangement for membrane coalescence

106 PTQ Q4 2014

ecopetrol.indd 2

A jet fuel treatment unit has a main

goal of removing acidic compounds

www.eptq.com

10/09/2014 16:19

PSV Size
Relief load

25

350
300

20

Water
washing

Caustic
washing

250

15

200
150

10

100

50
0

35
Jet fuel to
salt30
bed filters

Jet-caustic
emulsion
5

10

LV
15

Jet-water
emulsion
20

Time, minutes

25

PSV size required, in2

Relief load, 1000 lb/h

500
Jet fuel
from crude
450
unit
400

Visit our new website at

www.amacs.com

Our new clean-fuels

plant is straining the
auxiliary units!

0
30
Spent
caustic

Figure
processrelief
scheme
fuel treatment
Figure 46 Typical
Reflux failure
loadsforofjet
dynamic
simulation at 55% initial volume

which
in the and
jet fuel
changesareonpresent
the trays
the
produced
mainly contribution
in crude distillasystem volumes
to the
tion
units,availability
for instance
hydrogen
transient
of cooling
and
naphthenic
sulphide
(H2S),also
heating. They
ignore the acids
time,
and
mercaptans.
general,dependjet fuel
temperature
andInpressure
treatment
consists
a removal
caustic
ency of heat
input orofheat
washing
NaOH
solution)
from the(5B
system.
In the
cases to
of
neutralise
compounds,
plant revampacidic
or debottlenecking,
followed
by a water
conservative
reliefwashing
loads where
from
the
traces of methods
caustic, will
previously
conventional
likely
entrained
by the of hydrocarbon
call for the addition
PSVs and/or
stream,
are removed.
Thereafter,
the modification
of the
existing
traces
of waterwhich
and surfactants
are
flare system,
adds substanremoved
adsorption
on salt
and
tial cost by
and
risk to the
project.
clay
beds respectively.
Dynamic
simulation simulates the
During
the caustic
water
actual
composition
andand
inventory
washing
are
changes steps,
insidestable
the emulsions
column and
generated
dueand
to intimate
accumulator
providescontact
more
between
hydrocarbon
accurate predictions
on and
reliefcaustic
loads.
as
well asthebetween
hydrocarbon
Typically,
relief loads
predicted
and
In order
to assure
by water.
dynamic
simulation
area safe
less
thickness
the emulsion
layer,
than thosein calculated
by convenand
this wayDepending
avoid excessive
tionalinmethods.
on the
water
in the
salt/clay
system,entrainment
the reduction
in calculated
filters,
the formed
relief loads
can beemulsions
significantneed
(by
to
be discarded
to revamp
an effluent
treat>50%).
For plant
or debotment
system,
such asthis
a wastewater
tlenecking
projects,
reduction
treatment
unit.can
Theresult
general
flow
in relief loads
in signifischeme
of theforprocess
is shown in
cant savings
the project.
Figure
This4.case study illustrated such an
According
to fielddynamic
observations
example.
Employing
simuand
tests deisobutaniser
performed in
lationseparation
for an existing
an
jet fuel
treatment unit,
has actual
reduced
the calculated
relief
the
emulsions
from
loadsdrained
by more
than 35%
for both
both
washing
(causticfailure
and water)
TPF and vessels
reflux pump
cases,
have
aboutto50thevol%
of each phase
compared
conventional
meth(hydrocarbon
andalso
aqueous).
In the
ods. This study
demonstrated
case
a flow
4 gal/min
that studied,
the initial
liquidoflevel
in the
was
measured
from each
overhead
accumulator
is vessel;
a key
4parameter
gal/min of
(equivain hydrocarbon
relief load calculation.
lent
Whentothe 137
initialb/d)
liquid were
level isbeing
set at
discarded
wastewater
70% volume to
(as it the
is in current
opersystem,
ation), thenegatively
relief loadsaffecting
predictedthe
by

performance
of the unit
due tothe
a
dynamic simulation
exceeded
high
content
of hydrocarbon
and
available
capacity
of the existing
caustic
in the wastewater.
PSV. Mitigation
approaches were
In order
to overcome
this situastudied
by lowering
the initial
liquid
tion,
separation
step
level ainfurther
the overhead
drum
to was
55%
proposed
hydrocarbon
volume. Astoa recover
result, the
relief loads
from
drained
emulsions;
andand
in
were the
further
reduced
by 26%
this
enhance
the reflux
wastewater
18% way
for the
TPF and
pump
systems
performance,
a
failure cases,
respectively. while
With the
valuable
hydrocarbon
stream
could
recommended
mitigation
approach,
be
improving
profittherecovered,
new predicted
relief the
loads
are
ability
refinery
but existing
at the
within of
the the
capacity
of the
same
its environmental
PSV. time
Therefore,
the risk and perforcost of
mance
and sustainability.
modifying
the existing relief and
flare systems are minimised and
Pilot
test avoided.
potentially
Following a rigorous selection
References covering the alternatives,
procedure
1 Chittibabu techniques
H, Valli A, Khanna
V, Calculating
separation
such
as gravColumn
Relief
Loads,
PTQ,
55-65,
Q2 2010.
ity settling or coalescence
with
2 API RP 520: Recommended
Practice for
the
conventional
materials
were
Design and Installation of Pressure Relieving
discarded, while membrane coalesSystems in Refineries, Part I (Sizing and
cence was the selected option due
Selection, 2008) and Part II (Installation, 2003),
to
the better results obtained from
American Petroleum Institute, Washington D.C.
lab
and pilot
testsF Y,
using
actual
3 Sengupta
M, Staats
A newthe
approach
to
process
The pilot
test for
relief valve fluid.
load calculations,
43rd Proceedings
demonstrating
theAmerican
membranes
of Refining Section of
Petroleum
performance
carried
Institute, Toronto,was
Canada,
1978. out in an
experimental
Figure
5),
Harry Z Ha is aset-up
Senior (see
Process
Engineer/
using
most
Specialistthewith
Fluorchallenging
Canada Ltd, emulCalgary,
sion
available
to treat
jet-caustic
Alberta,
Canada. He
holds athe
masters
degree
in environmental engineering from Hong
emulsion.
Kong
of Science set-up
and Technology
TheUniversity
experimental
was
and
a
PhD
in
chemical
engineering
from the
connected to the caustic washing
Universitytoof Alberta.
vessel
treat the actual process
Email: Harry.Ha@Fluor.com
fluid,
using a commercial fibreglass
Abdulla N Harji is an Executive Director of
Pall DFT Classic pre-filtration
Process Technology, at Fluor Canada. He holds
module
andin achemical
fluoropolymer
a BSc degree
engineering Pall
from
PhaseSep
membrane
Loughborough University, UK.module. The
main
goal
of the
waswith
to
Jonathan
Webber
is a pilot
Processtest
Engineer
prove
the concept
in ain challenging
Fluor Canada.
He holds a PhD
process control
environment
and to and
calculate
thein
from Dalhousie University
a masters
biotechnology from
McGill University.
separation
efficiency
and flux as

How can we boost their capacity

without major construction?
TO MEET CLEAN FUELS
requirements for gasoline, a mediumsized Midwestern renery added a lowsulfur fuels technology plant. Gasoline
throughput was unchanged. However,
the sharp increase in sulfur removal
required more hydrogen from the
hydrogen unit and sent more sulfur
gases to the amine treaters and
downstream sulfur units. These
auxiliary units became bottlenecks,
overdriven at the cost of product
purity and amine consumption.

On studying the hydrogen, amine,

and sulfur units, AMACS found many
opportunities for improving separation
efciency and capacity. The problems
were solved without major construction
by applying modern technology to mist
eliminators, liquid-liquid separators,
and tower trays and packing. Results
included haze-free product and
reduced amine consumption. Now a
diesel clean-fuels plant is being added.

Read more on this topic at

www.amacs.com

Phone: 713-434-0934 Fax: 713-433-6201

amacs@amacs.com
24-hr Emergency Service:
1-800-231-0077

www.eptq.com
www.eptq.com

ecopetrol.indd
fluor.indd 5 3

10/09/2014
12/03/2014 16:19
12:04

1
2
3
4
5

Pre-filter
Membrane coalescer
Flow meter
Pressure indicator
Valve
5

5
Hydrocarbon
phase
4

2
4

Emulsion

5
3
GPM

Aqueous
phase

Figure 5 Experimental set-up for membrane coalescence testing

main process parameters for further

scaling up.
As Figure 5 shows, the emulsion
was fed to the pre-lter and thereafter to the membrane module, which
is installed in the left-hand compartment of the coalescer; nally, the
phases were separated in the righthand compartment of the coalescer,
where the recovered jet fuel owed
upwards and the aqueous phase
settled downwards, as a result of
their difference in density.
During the pilot test, the
membrane module was tested at
different ow (ux) values of emulsion, while the quality of separated
hydrocarbon
(aqueous
phase
content) was monitored in order to
have an idea of the separation efciency. On the other hand, taking
into account that the test was

carried out for four weeks, it was

possible to test the mechanical and
chemical
resistance
of
the
membrane material to the actual
process conditions. The main results
obtained from the test can be seen
in Figure 6.

Implementing the solution on the

industrial scale

Based on the results shown in Figure

6, with a ux lower than 3.5 gal/
ft2min it is possible to treat an aqueous phase content near to saturation
(200-250 ppm of water in jet fuel),
proving that the membrane coalescence technique is a feasible
alternative for separating tight
hydrocarbon-aqueous
emulsions.
Therefore, a ux of 3.5 gal/ft2min of
jet fuel was taken as a calculation
basis to size the required equipment

Water in recovered jet fuel,

ppm

350
300
250
200
150
100
50
0
2.0

2.5

3.0

3.5

4.0

Emulsion flux, gal/ft2-min

Figure 6 Coalescence experimental results using process fluid

108 PTQ Q4 2014

ecopetrol.indd 4

4.5

5.0

to treat 8 gal/min of emulsion at

52 vol% hydrocarbon and 48 vol%
caustic solution (5B). This led to a
coalescer with a 40 inch (3.3 ft) long
membrane module, in a vessel of
0.75 ft diameter and 6.5 ft long.
A process scheme of the current
arrangement can be seen in Figure
7, where the pre-lter and the
membrane coalescer vessel are integrated in a typical process scheme
mounted in a single skid. In this
case, a horizontal coalescer was
selected due to the amount of aqueous phase present in the emulsion
(48 vol%), as well as an estimated
interfacial tension of about 0.5-1.0
d/cm.
Further analysis of the recovered
jet fuel from the coalescence skid
showed that the water content
varied between 200 and 250 ppmwt,
corresponding to water saturation of
the hydrocarbon, indicating acceptable water and caustic entraining in
the recovered jet fuel. Thus a performance similar to that obtained
during the pilot test was observed
in the industrial scale equipment,
with potential recovery of 140 b/d
of jet fuel and contributing to a
better environment performance
and sustainability of the renery.
The clear and bright appearance
of recovered jet fuel on visual
inspection conrmed its good quality for further processing in salt and
clay lters (for the jet fuel product
pool), or directly as a diluent for
fuel oil blending, or in other uses. A

www.eptq.com

10/09/2014 16:20

Jet fuel
from crude
unit

Jet fuel to
salt bed filters
Jet fuel to
diluent pool

Recovered
jet fuel
Flow meter
01

Water
washing

Caustic
washing

Pre-filter

Jet-caustic
emulsion

Jet-water
emulsion

Membrane
coalescer
Flow meter
02

LV
Spent
caustic

Coalescence
skid

Figure 7 Process scheme for recovering jet fuel from emulsions

comparison of the appearance of

recovered jet fuel and the original
emulsion can be seen in Figure 8.
Although the specic application
described in this study is still of a
relatively small size, it has demonstrated the suitability of membrane
coalescence to separate tight emulsions, as well as its straightforward
and safe scaling up after a few pilot
or lab tests. It should be noted that
other
successful
tests
were
conducted for different applications,
such as amine-hydrocarbon separation and sour water-hydrocarbon
separation, demonstrating again the
suitability of membrane coalescence
for multiple applications within the
rening industry.

Conclusions

Membrane coalescence was demonstrated

to
be
a
successful
technology for separating tight
emulsions, contributing to more
protable and more sustainable
operation of a renery.
Pilot and lab tests results are
easily and safely scalable to industrial applications, with a high
reproducibility of process parameters and performance. On the other
hand, these tests are helpful in
selecting the right materials for
both pre-ltration and membrane
modules.
The content of the aqueous phase
in recovered hydrocarbon is close to
its saturation value, indicating high
separation performance.

www.eptq.com

ecopetrol.indd 5

Figure 8 Comparison of recovered jet and original emulsion (left to right): Jet fuel-water/
caustic emulsion; recovered jet fuel; separated aqueous phase (spent caustic solution 2-3B)
The authors would like to acknowledge Ernesto
Gmez and Leonardo Snchez from the Crude
Distillation Department of Barrancabermeja
Refinery, for the logistic and operational
support which allowed to carry out the work
described in this article.
Further reading
1 Basu S, A study on effect of wetting on
mechanism of coalescence, J. of Colloid and
Interface Sci.,1993, 159, 68.
2 Hu S, Kintner R C, The fall of single liquid
drops through water, AIChE J.,1955, 42.
3 Brown R L, Wines T H, Improve suspended
water removal from fuels, Hydrocarbon
Processing, 72, 1993, No. 12, 95.
4 Sprow F B, Drop size distribution in strongly
coalescing agitated liquid-liquid systems,
AIChE J., 1967, 13, 995.
Hernando Salgado is Senior Process Engineer
at Ecopetrol SAs Cartagena Refinery (Reficar)
in Colombia and has worked in the refining and
petrochemical industry mainly for Ecopetrol

as a process engineer, in both refineries, in the

areas of FCC, crude distillation and energy
management. He holds a degree in chemical
engineering from the Industrial University
of Santander in Colombia and a professional
doctorate in process and equipment design
from Delft University of Technology, The
Netherlands.
Luis Marino is Applications Engineer at
Ramgus SA, the official representative of
Pall Corporation Technologies in Colombia,
with broad experience in filtration, fuels
and chemicals purification, and process
optimisation. He holds a degree in mechanical
engineering and a MSc in technology
management, both from the National
University of Colombia in Bogot.
Rosngela Pacheco is Process Engineer at
Ecopetrols Barrancabermeja refinery and
is a specialist in crude distillation and fuels
treatment (naphtha and jet fuel). She holds
a degree in chemical engineering from the
National University of Colombia, Medelln.

PTQ Q4 2014 109

10/09/2014 16:20

Leading-Edge Technologies for On-Purpose Olens

Medium and long-term forecasts expect to see a continuing growth in
demand for on-purpose olen production technologies (e.g. propylene,
butylenes) such as dehydrogenation of light parafns. Thanks to our
advanced, proven Uhde dehydrogenation technologies, STAR process
and STAR catalyst, we can supply, from a single source, complete,
optimized process routes to propylene and butylene derivatives, e.g.
Polypropylene, Propylene Oxide, ETBE and other high-value products.

Liquid hourly space velocity of

resulting in less catalyst

and lower reactor volume
Available with and without
oxydehydrogenation

ThyssenKrupp
Industrial Solutions
www.thyssenkrupp-industrial-solutions.com

t krupp.indd 1

10/09/2014 12:08

Enhancing bottoms cracking and

process flexibility
Catalyst designed with advanced zeolite stabilisation technology provides
selective conversion of heavy FCC feed molecules
YEE-YOUNG CHER, ROSANN SCHILLER and JEFF KOEBEL
Grace Catalysts Technologies

efiners require FCC catalyst

technology that delivers the
right selectivity at the right
time. In a world where fuel demand
is satisfied through a careful balance
of free trade, weather events or
refinery upsets could trigger price
volatility in product markets. The
ability to respond quickly to capture
short-term market opportunities is
critical. Amid declining gasoline
demand in mature regions, refiners
need to enhance distillate production. Graces premium bottoms
cracking family, the Midas catalyst
series, can be used to enhance FCC
process flexibility and capture incremental profit as opportunity arises.
These catalysts crack deep into the
bottom of the barrel, enhancing total
distillate and liquid yield, and have
been demonstrated in over 120
refineries that vary broadly in feed
composition and operating modes.
The flexibility that the catalysts
provide, used neat or as a component in a Genesis catalyst system,
can enhance the yield value by
$0.40-1.00/bbl of FCC feed.1
Midas is a moderate zeolite to
matrix ratio FCC catalyst that has
been successfully applied in half of
North Americas FCC unit capacity
as well as refineries in other parts of
the world. Its success is driven by
the fact that it effectively cracks all
feed types: heavy resids, severely
hydrotreated light feeds, and shale
oil-derived feed streams, via the
three-step bottoms cracking mechanism discovered by Zhao.2 The
catalyst design minimises the thermal and catalytic factors that result
in coke formation. The result is deep
bottoms conversion, regardless of
the starting feedstock.

www.eptq.com

grace.indd 1

Resid streams present the greatest

challenge in terms of deep bottoms
conversion. The dynamic molecular
dimensions
of
paraffins
and
aromatic species vary, based on
carbon number and molecular
configuration. Paraffins species present in the 700-1000F boiling point
fraction of FCC feed are typically in
the nC14 to nC34 range for normal
paraffins. The dynamic molecular
size of these compounds is 12-20
angstroms (). The heavy resid fraction also contains an abundance of
aromatic molecules (C14 to C60) in
the 700-1000F boiling range. The
range of molecular size for aromatics is 12-25 . Even aromatic carbon
molecules up to 60 carbon number
are still less than 30 in molecular
size.
Porphyrins are organic, cyclic
macromolecules that consist of a
ring of nine or more atoms.
Porphyrins are aromatic species
often present in resid fractions and
characterised by a central gap that
can bond to a metal atom, such as
nickel, vanadium, or iron. If a
porphyrin is complexed with vanadium, it is termed a vanadyl
porphyrin. The size of these metallic
complexes also varies with carbon
number, but is in the same dimen-

sional range as typical resid

hydrocarbons: 10-30.2
The relatively large molecules at
the bottom of the barrel that need to
be converted must first be cracked
by the catalysts matrix acidity. With
molecular sizes of 10-30 , the
hydrocarbons are too large to fit into
the zeolite pores, which are typically
below 7.5 . It is important that the
catalyst have the proper pore size
distribution to enable large feed
molecules to enter, crack into lighter
products, and diffuse out before
being over-cracked to coke and gas.
For free diffusion of resid molecules
(>1000F) to occur, the catalyst pore
diameter needs to be 10-20x the size
of the molecule, or 100-600 .2 The
desired pore volume should be in
the large mesopore region 100-600
. The benefit of mesoporosity for
bottoms cracking is well understood.5 However, not all the
measured pore volume is created
equal. Catalysts with similar total
pore volume measurements can
vary widely in pore size distribution. Midas is designed to have high
mesoporosity in the 100-600
range, typically twice as high as
competitive offerings (see Table 1).
Optimal porosity is required for
effective kinetic conversion of

Mesoporosity comparison
Catalyst

Micropores

Total
36-100
Midas
0.389
0.092
Midas
0.412
0.107
Cat 1
0.386
0.116
Cat 2
0.413
0.092

Hg-PV, cc/g
Mesopores
100-600
0.206
0.232
0.102
0.089

Macropores
600+
0.091
0.071
0.168
0.232

Table 1 Higher mesoporosity of Midas promotes bottoms conversion

PTQ Q4 2014 111

10/09/2014 16:30

Pore volume of commercial Ecats

0.45

Micropores result
in poor gas and
coke selectivity

Midas has the highest mesopores

plus best gas and coke-selective
bottoms cracking

0.40

Midas
Competitor A
Competitor B

0.35
0.30
0.25
0.20
0.15
0.10
0.05
0
10

Commercial experience

100

1000

10000

Pore diameter,
Figure 1 Pore volume comparison of commercial Ecats

bottoms. Midas catalysts crack

deeper into the bottoms.
Commercial examples of Midass
high mesoporosity, as measured by
Hg porosimetry of Ecat, are shown
in Figure 1. Note that Hg intrusion
measures the porosity greater than
36 , therefore the result specifies
the porosity associated with the
catalyst matrix only; N2 adsorption
or desorption must be used to
capture zeolite porosity. Graces
in-house manufacturing and quality
monitoring of the specialty alumina
used in Midas provides control over
the resulting porosity. It is generally
accepted that micropores (<100
diameter), though effective for
cracking, lead to poor coke and gas
selectivity as a result of poor diffusivity and over-cracking. Some
competitive benchmarks with high
surface area and activity are also
high in matrix microporosity, resulting in wet gas compressor
limitations that suppress feed rate
and ultimately profitability. In
contrast, Midas catalyst has the
lowest amount of small pores and
the highest amount of large mesopores. Optimal porosity guarantees
gas selectivity and coke-selective
bottoms conversion. High pore
volume also serves to enhance the
fluidisation characteristics. Several
units have observed substantial
improvement in the Ecat fluidisation
factor following a reformulation to
Midas or Genesis.1

112 PTQ Q4 2014

grace.indd 2

fresh zeolite surface area in a severe

regenerator, but the penalty for this
over-exchanged zeolite is poor coke
selectivity. The rare earth exchange
in Midas catalysts is optimised to
result in a formulation that lies in
the so-called sweet spot or between
24.28-24.32 for Ecat UCS to
deliver the best coke selectivity. The
right distribution of mesoporosity
coupled with optimal activity and
UCS gives the catalyst the coke
selectivity edge in commercial
cracking.

The proprietary matrix in Midas

can withstand the most severe
applications, particularly those challenged by high levels of contaminant
iron and calcium. High alumina
content in FCC catalyst is known to
reduce the degradation of the catalyst surface due to iron and/or
calcium poisoning.4 Optimum distribution of mesoporosity also plays a
role in maintaining performance,
because diffusion to the active sites
remains unhindered despite the
high contaminant metals. Midas has
been successful in maintaining
bottoms conversion in units with
some of the highest levels of
contaminant iron on Ecat in the
industry.
A good bottoms cracking catalyst
requires high matrix surface area
(MSA). However the activity of a
high matrix catalyst needs to be
balanced with an appropriate level
of zeolite without compromising
attrition characteristics. Additionally, an appropriate rare earth
exchange level on the zeolite is critical to ensure optimal coke
selectivity. An optimum exists in the
relationship between zeolite unit cell
size (UCS) and coke selectivity.5 Too
often, high matrix catalysts also
have a high UCS, meaning they are
over-exchanged with rare earth.
Low zeolite input formulations with
high rare earth exchange (albeit low
total rare earth on catalyst) will
retain a higher percentage of the

Midass design features have led to

high performance in many commercial examples. The first example
comes from Refinery A that has an
FCC unit processing a mix of VGO
and light resid. The feed is moderately high in metals with good
cracking characteristics. The base
catalyst was a competitive high
MSA catalyst with high rare earth
exchange. The unit suffered from a
dry gas constraint, especially in the
summer. When the refiner switched
to Midas, a 5-10% drop in total dry
gas yield at equivalent riser temperature and metals loadings was
observed; this allowed the refiner to
maintain maximum feed rate
throughout summer. The better coke
selectivity also manifested itself in
lower bottoms yield by 15%.
At Refinery B, the FCC unit was
processing light resid using a high
matrix, high UCS catalyst. While
rare earth levels on the zeolite were
high, the catalyst itself had low total
rare earth because zeolite input was
very low. In spite of the high rare
earth exchange and resulting UCS,
the unit was constrained on total
wet gas and bottoms yield. There
wasnt enough zeolite present to
complete the second stage of the
bottoms
cracking
mechanism.2
Zeolite is critical to reduce the size
of the hydrocarbon via dealkylation.
Insufficient zeolites cause coke
precursors to condense on the catalyst surface and become coke before
they can be converted into transportation fuels. Poor coke selectivity
translates into subpar bottoms
conversion.
Introduction of Midas provided a
reduction in bottoms make on the

www.eptq.com

10/09/2014 16:30

End-to-End Combustion and Environmental

Solutions Are Just the Beginning.

Experience a breath of fresh air.

From our ultra low-NOx products to our ultra-knowledgeable people, every
ZEECO burner, flare, incinerator, and flare gas recovery system is backed
by a Zeeco team dedicated to you and your project. From site evaluation
and product design to manufacturing, testing, installation, commissioning,
and unparalleled aftermarket parts and service, we deliver end-to-end
answers to the industrys biggest challenges.
Onshore or offshore. Upstream, midstream, or downstream.
Zeeco engineers innovative combustion solutions to keep operators
in compliance, people safe, and plants running smoothly and we deliver
them on time and on budget.

Global experience. Local expertise.

ZEECO Ultra-Low NOx GLSF Free-Jet Burner

Experience the Power of Zeeco.

Burners Flares Thermal Oxidisers Vapour Control
Aftermarket: Parts, Service & Engineered Solutions
Explore our global locations at zeeco.com

Zeeco, Inc.
22151 E 91st St.
Broken Arrow, OK 74014 USA
+1-918-258-8551
sales@zeeco.com

Zeeco, Inc. 2014

zeeco.indd 1
Hydrocarbon Engineering_July2014_v2.indd 1

10/09/2014 11:53
9/5/14 2:11 PM

more feed was converted into gasoline and distillate, rather than LPG
Competitor (Base Fe+Ca)
and coke.
6
Midas (+25% Fe+Ca)
At Refinery C, the FCC unit
Midas (+60% Fe+Ca)
4
processes resid feedstock high in
grip-type
tube fittings
not backthe initial investment and on-going from
corrosion
(as mentioned
yield selectivity,
in will
particular
configuration
of ARDS
+
The average
composition
of RFCC,
NiW
iron
(Fe) and
calcium
(Ca). the
Over
2
To
To
offpropylene
with thermal
cycling
or
vibraearlier)
or
other
debris,
such
yield. The higher the catalyst replacement cost will be LVGO
fraction
of inthe
crudeas is
catalysts
received
2013
Spent NiW
time,
the
unit
suffered
from
classic
transmitter
transmitter
tion,
unlike
the threaded
fittingsthe much higher than for most hydro- burrs.
Burrsto area a VGO
byproduct
of the
hydrogen
the higher
routed
hydrocracker
catalysts
0 content,
4
As
symptoms
of iron
poisoning.
used
with
carbon
steel.
machining
process
during
manufacProduct
propylene
yield.
processing
units.
(HCU)
to
shift
the
yields
ina the
Company
in
the
UK,
I
designed
communicate
where
the
module
construction
of
refineries
are
being
promise
for
deriving
value
from
Composition
of
NiW(Mo)
spent
catalysts
2 objective maintaining
iron
nodules
built
on
the
surface,
The
third
turing.
A
critical
step
during
For
types ofoncrude
oil plan.
thatgas applied
The other
key construction
features of of
this sour
desired
direction.
Figure
2 shows a
water
stripper that
Characteristics
Wt%
Ammonia
will the
be installed
the plot
to the
smaller
accumulations
of eliminated
unconvenEqualisation
torange
sulphur
plantconfiguration
Ecat
activity,
unit
conversion
and
temperature
within
a certain
manufacturing
and
installation
Moisture
3.5 unit
have
been
considered
here,
the
to
maximise
the
schematic
of
the
refinery.
the
unnecessary
features
of
the
Connections
between
the
modules
distributed
GTL
plants.
The
GTL
tional
gas
that
would
otherwise
be

4
valve
Roasting
Filtration
Filtration
Filtration
Gas
Oilbottoms
13 refinery
typical
is
achieved
heating
the
should
be
the
removal
of
all
burrs
cracking
all
began
to
suffer.
requires
process
unit production
shown
in
Figure
2.
CCRincreased
content
of
atmospheric
of
petrochemicals
are:
The
key
features
of
this
are
designed
toby
be similar
in configprocess involves
two
operations:
the
left
underground,
such
as
shale
gas,
Isolation<50mgSO2 /m3/hr
C
14.8
6 so
impulse
lines.
insulate
your
from
all
surfaces.
Otherwise,
capabilities
inthat
acan
number
of13 PSIG
refining
Figure
shows
the essentials
of a higher
A switch
to bed
Midas
provided
uration
construction
is
conversion
ofvalves
natural gas
a tight
gas,3wetted
coal
methane
residues
is You
about
12-14
wt%,
and
Process straight-run
light tonaphare:
S configuration
4.9 and
190F
Dusts
Temp. 1
Temp. 2 Temp. 3 Temp.
1 Temp.
1dislodge
Temp.
2 Temp.
3
process
units,
particularity
hydroimpulse
lines
manually

field
they
can
when
manicorrect
sour
water
stripper
design.
relatively
mixture
of carbon
(CO) checkstranded
gasand
(gasimproved
fields located
tooselectivmetals
80-90 wtppm. Ventech
Thus, tha
along
with monoxide
light catalytic
A vacuum
tower
is the
installed
to
activity
coke
W
7.8
8are straightforward.
Sour The
90F
processing
assets.
magnitude
Feed
is
brought
in
at
ambient
back
tracing

or
purchase
tubing
that
fold
is
in
service
and
catch
intobottoms
orbe
),
known
as
estimates
that,
with
modularisation,
and
hydrogen
(H
far
from
existing
pipeline
infraMo
2.5
Slag
Pyro3 of water
2feed is 1
0 naphtha
1 (LCN)
2 the
3 cracker produce
pretreatment
the
required.
in
steam
a
diesel-type
cut
ity,
delivering
deeper
2
60
Solids
(15)
Roasted
Niprocessed
2.9
of
impact,
ofLeaching
course,
on
(70-100F,
21-38C)
Filtration
feed
syngas,
followed
by LPa Fischerapproximately
70%
a Relative
project
iscoke,
structure).
A in
small,
modularised
has
been
insulated
in isof
theadepends
factory
scratch
the
valve
stem
orfrom
seat, feed
HPmetallurgical
The
ARDS
unit
specially

Process
heavy straight-run
naph- conditions
the
distillate
hydrowt%
catalysts
conversion.
Eventually
the
P
0.5
process
the
particular
feed
processed,
the
the
sour
water
feed
tank.
To
Tropsch
(FT)
process
to
convert
the
already
complete
even
before
GTL
plant
has
the
flexibility
toheat
be
Alloy
and
encasedhydrotreater
in a polymeric
jacket.
aLVGO
positive
shut-off.
Burrs
designed
that pretreats
tha in a 50catalytic reforming
unit preventing
treater,
a
cut
to
be
processed
Asbecame
0.02250F
heavier
and
more
contamiinherent
flexibility
of the
facilitys
the
feed
from
(32C)
to
syngas
into
paraffinic
hydrocarbons
are RFCC
shipped
theirLiquid
installed
close
toalmost
themonoflanges
trapped
Figure
4 A manifold,
consisting
of three
Pre-insulated
bundled
tubing
may
also
be
an90F
issue
in
themodules
feed3 to the
unitfrom
tocomes
reduce
(CRU)
to
produce
reformate
a two-stage,
full-converFein
0.39
nated.
The
catalyst
never
lost its
Competitor
(Base
Fe+Ca)
operations
and
the
ability
to
(120C)
requires
about
16
wt%
that
can
be
further
refined
to
facility.
This
greatly
decreases
field
resource
and globe
then used
to 26.3
process
needle
valves, 40
enables the
technician to
ready
to install
inincluding
coiled
lengths.
It and
and
pattern
needle
Alsion
O3 other
contaminants
sulphur,
Process
hydrocracker
unit
oriented
2 Harrison
leads Lindes global Specialty
ured. In heavy
the US, catalytic
the EPA naphtha
has Stephen
prioritisation
of monitoring
offset
this
increased
processing
steam

ow,
or
about
1.3-1.4lb
of
selectivity
edge
over
the
competitive
Midas
(+25%
Fe+Ca)
produce
a
wide
range
of
construction
time
to
deliver
an
that
resource
locally.
Associated
gas
isolate
and
calibrate
the
transmitter
in
2 quantifying
cannitrogen,
be heated
with steam
or Purification
left
valves.
Gases
&
Specialty Equipment
business from of middle
organometallic
metals
and (HCN),
pygaslives and
reformate
towards
the production
defined shelf
for Garbage
protocol
emissions,
accuracy
requirement.
per
of
products,
includ- steam
operational
facility
(see
FigureFe+Ca)
1).
(gas
produced
withstripper
oil) is even
30
(8)
thehydrocarbon-based
field
high
MSA
base
formulation,
Midas
(+60%
Germany.
He isgallon
a along
British Chartered
AS
+caliP
unheated,
asreliability
the
gases as
between
six to 36complex
months Munich,
and
inapplication
measurement
Medium
content
CCR.In Additionally,
it
increases
the
through
the
aromatics
to
distillates
making
this
move,
proper
bottoms,
which
is
close
to a typical
Engineer
(MIChemE)
with
a
career
in
industrial
ing
clean-burning,
sulphur-free
These
methods
also
facilitate
easy
another
area
of
opportunity
forEcat Fe
NH
water
to

depending
on
the
gas
and
concenbration
is
critical.
The
demand
for
Table
1
3
after
increases
of
25-60%
in
Seat
seals
in
needle
valves
requires.
It
is
important
with
pre-in1
hydrogen
content
the atmosparaxylene
and benzene
The20stripping
HVGO
is have
combined
hydrotreaters
gases
spanning
years,
10 of which
been
20jet
selection
requires
a measurement
good
underdesign
steam
ratio
forwith
diesel
and
fuel.
Speciality
proddisassembly
and of
relocation,
ifthe produce
modularised
GTL
plants.
This gas
tration

and
similar
shelf
life
stable,
accurate
is
and
Ca.
Thewhy
high
transmitter
awith
standard
differenAnother
reason
valves
leak is in is
sulated
tubing
bundles
to followofthe
focused
in the
area
ofresidue
specialty
gases.
Hemesoporosity
has
sexist
relatively
high
pheric
residue,
the
The
C4in
vacuum
to
be
pretreated
(1)
standing
and
modelling
the
sour
water
strippers.
The
E-1byfeed
standards
in the
ISO framecornerstone
of improving
emissions
ucts
including
food-grade
waxes,
necessary,
at thus
some
point Precipitation
inanalysis.
is typically
disposed
of
either
retogether
with
spent
catalyst.
In
the
worked
in an design.
international
capacity
for both
tial
pressure
setup,
the
technician
poor
Most
needle
valves
manufacturers
instructions
for
sealparticularly
effective
in
preventing
a
Calcination
0
olefins
content
are
either
sold
or
crackability
and
selectivity
of
the
the
vacuum
residue
desulphurisaProduct
DN-3651
alternative
feed
DN-3551
alternative
feed
10
work and
for lubricants
reference can
materials.
calibration
facility However,
as
as
a strong
knowl-for solvents
preheater,
re
uxBOC,
pump
(P-2)
and
also
be
future.
Forwell
example,
a standards
remotely
injection,
at
considerable
expense,
Linde
Gases
and
previously
and
holds
a
CaWOcloses
roaster,
the
temperature
and
rate
of seen
4
the
two
isolation
valves
and
employ
a
metal
to
metal
seal.
The
ingresidue.
the
insulation
when
splicing
into
DN-3651
current
GO
DN-3551
current
GO
Notwithstanding
these
internareactive
mixtures,
typi- recycled
loss
in
cracking
activity
often
to from
the second
riser
of the
For
asources
typical
ARDS
HDS
tion
(VRDS)
unit
before
processing
edge oflow
thelevel
and
processing
the
re
ux
cooler
(E-2)
shown
in
masters
degree
in
chemical
engineering
from
produced
the
paraffinic
located
gas
processing
facility
could
back into the reservoir or by the
air
injection
are
in order
to loadBase
tional
protocols,
Lindesvalve
range
of Imperial
cally
those
with
levels
below
5 FCC
the
equalisation
(see
metal
of
the
stem
downor rate
cutting
into
the
pre-insulated
E-1
College,
190%,
(directly
or through
oligomeriof
the
CCR
reduction
is opens
it
in tip
the
RFCC
unit
requirements
of
the
available
Figure
2London.
are
alladjusted
eliminated.
How,
during
periods
of moves
high
metals
Hydrometallurgical
process
hydrocarbons.
be
easily
taken
apart
and
moved
to
wasteful
and
environmentally
HiQ
60
specialty
gas
products
have
ppm,
can
prove
to
be
unstable
over
remove
sulphur
(as
sulphur
dioxNaresh
Suchak
is
Seniorone
Project
Manager
with
Steam
Figure
4).
This
procedure
results
in
ward
to
seal
against
a
metal
seat
ofthe
tubing
bundle.
crudes.
With
the
continued
develthen,
does
know
that
the
sation)
to
maximise
propylene
about
65-70%
and
the
metals
reduc
Depending
on
the
quality
of
Large,
commercial-scale
GTL
a new natural
gas source if an existdamaging
of flaring,
which
ings.
Fepractice
and
can
eutectics
Current
feed
shelf life feed
a Linde
time and can result in incorrect an extended five-year Alternative
Gases
For
the Ca
past 20
yearsform
ide),
toDivision.
burn
hydrocarbons,
and
toand
theproduction
transmitter
being
zeroed
out
matching
shape.
The
tip
of
the
stem
opment
andrate)
increased
availability
design
shown
in
Figure
3
will
tion
(HDM
isdepleted
about
85-90
combined
vacuum
residue
plants,
including
the
Sasol
Oryx
and
ing
was
in
its
is
subject
to
increasing
regulation.
supply
2 measurements,
he has age
worked serve
in technology
development
lost productivity significant advance in the supply of
that
to essentially
melt over
Catalyst
Figure
3 Recycling
process
for
spentfeed
nickel-tungsten
catalysts
oxidise
the
metallic
elements.
Manifold
Ifplants
one
of the
two
may
beBecause
shaped
like
abuilt
ball
or way
athis
vee. a
of
such
feeds,
thisRFCC
knowledge
needs
work?
it to
was
this
thecalibration.
Shell Pearl
(both
located
current
location.
from
concept
tofeed
commercialistion
of
Modularised
GTL
plants
enable
Products
from
the
RFCC
unit
and
wt%.
The
HDT
will for
HVGO
the
VRDS
unit,
calibration
gases.
Previously,
gas
and
with
emissions
monitoring

the
catalyst
surface,
blocking
pores
A
stream
of
compressed
air
sends
innovative
processes
in
the
fields
of
chemical
to
be
continuously
updated
to
(at
the
Amoco
re
nery
in
Milford
The
manifold
consists
of
a
set
of
isolation
valves
is
leaking,
however,
In
either
case,
it
is
important
that
suppliers
offered
product
expirapotential
legislative
fines.
in
Qatar),
have
been
built
at
enorApplying
modularisation
to
refinotherwise
wasted
gas
to
be
250F of less
steam cracker are combined in a slip stream of vacuum residue
then have
watercontent
3LowaNHCCR
3
and
suppressing
conversion.
The
manufacturing
and
pollution
control.
He
to
produce
calcine,
which
provides
the
gas
into
a
post-combustion
tungsten
or
just
tungsten.
In
the
tobodies
desalters
tion 15
guarantees
generally
limited
to to HDS
To
keep metals
pace
with
increasingly
Figure
DN-3651
and
DN-3551
normalised
temperature
ensure
minimal
operational
Haven,
UK)
in
1970,
wherethe
valves
whose
are
machined
calibration
will
be
unsuccessful
thecould
tip Wales,
does
not
rotate
with
mous
capital
cost.
The
Oryx
plant,
ery
has
advantages
converted
into US
additional
revenue.
1.0
0.5
0 single
0.5
1.0
light
ends
recovery
section
than
6 construction
wt%
and
content
of the
and
with
has been
granted be
seventaken
patents
and blended
has
36
months,
with
many
products
stringent
legislative
requirements,
high
mesoporosity
of
Midas
ensures
feedstock
for
the
following
step.
chamber
which
treats
it
for
oxides
from
rstless
case
(including
the
presence
of
surprises.
Reviewing
TBP
curves
it
worked
just
as
well
as
the
designed
for
production
levels
of beenstem.
with
toblock
productivity,
prodP-1
In
the
picture,
a oil
a regard
single
of metal,
and
all subsequent
readings
will be
If larger
itforsuch
does,
may and
grind
into
responsible
the economic
development
produce
ethylene
and propylene,
than
10 wtppm.
diluents
asit RFCC
decanted
Relative
kinetic
conversion
available
with
12 orc;24 months
calibration
gas mixtures
are now
and quality
bulk
whole
crude
conventional
design
shown
in a to
that
high
diffusivity
isthrough
roaster
is only
speci
was commercial
ofMax
carbon.
Theiscapability
pass
molybdenum),
recycling
34Dekkers
000
b/d,
cost
around
$1.5
billion
uct
andthe
ensuring
the
safety
modular
GTL
can
bemaintained.
the
implementation
ofitLoTOx
NOx
usually
stainless
steel.
The
maniinaccurate.
the
seat,
scoring
and
creating
Ovchinnikov
afumes
Senior
Research
Chemist
2
C, Daane
R,
Oilproducts
& Gas
J.,it1999,
minimising
the
capital
investment
Due
to being
theproperties
feed
rate,of
nature
of
theThis
(DCO)
and
light
cycle
oil
(LCO)
of shelf
life.
Gas
with97,control
delivered
with
state-of-the-art
technology.
He Catalysts
holds
aMS
and
PhD
in the gas
Figure
4completed
Two-stage
sour
water
stripper
design
without
feed
preheat
or
the
individual
cut-point
Figure
2.the
to
build.
The
Shell
Pearl
with
of
construction
personnel.
Since
the
key
factor
that
enables
the
construcdeveloped
by
Valdi
toplant,
deliver
three
levels
of
ltration;
process
iseven
by
a further
with
Criterion
and
Technologies
145.
Figure
2feed,
Despite
higher
metals,
improved
bottoms
cracking
and
coke
It
is
preferred
catalyst
technolfold
mounts
to contaminant
the
side
of
Further,
after
calibration,
the
pathway
for
leakage.

Hydrogen
from
the
CRU
and
these more
limited
shelf
lives
can
and
processing
objectives,
athey
produce
some
fuel
oil.
The VRDS
packaging
technology
so the
that Midas
chemical
engineering from
Bombay
University.
ranges
does
not
based
in
Houston,
Texas.
He
is
primarily
an
ultimate
design
capacity
of
140
3
Puri
B
K,
Irgolic
K
J,
Environ.
Geochem.
Health,
modules
are
built
in
a
well-lit,
tion
of
upstream
projects
that
highly
effective
desulphurisation
rejected
is
devoid
of
particles
and
step
involving
the
removal
of
phosselectivity
at Refinery
C exceed
transmitter
and
serves
a and
critical
will
close
equalisation
There
are
two
design
impactcracker
measurement
accuracy,
as by
even
the
demanding
ogyis would
for
FCC
units
processing
high
Frank Fitch
Senior
Project
Manager
with
steam
is the
supplemented
relatively
high
pressure
low operator
unit
then
bemain
able
to
reduce
3
increased
water
infor environment,
the
sulphur
Figure
1. of
InGTL
other
words,
a lot120
of Linde
Two-stage
sour
water
stripper
necessarily
characterise
the
feedand
engaged
in otherwise
the
research
and
development
ofof
1989,
11,
95.
000gas
b/d
products
and
climate-controlled
would
be
cancelled
stability
in
terms
of
consistency
requirements
consistency
/hr
and
decarbonation,
good
producdoes
not
exceed
50
mg/m
phorus
and
arsenic.
Gases
Division
and
is
product
lifecycle
function,
enabling
calibration
or
valve
and
open
the
two
isolation
approaches
tothe
achieving
a nonFe
feedstocks.
The
selectivity
benefurther
hydrogen
production
from
space
velocity
design
is
required
the
CCR
to
the
levels
required
by
catalysts
for
hydroprocessing
applications
and
4
Nielse
B, Villadsen,
Appl.
Cat.,
1984,
11, 123.
Figure
shows
a sour
water
stripplant feed
and
more
condensaequipment
iscan
added
to
difficulty
indicate
the
and
quality
change
over
time.
stability
thus
down
to full
part-per-billion
b/d
of
natural
gas
liquids,
cost manager
work
continue
around
theimpact
clocktivity
because
of
poor
results
derived
for the4
Linde
NOx
control
technology
sulphur
dioxide.
three
stages
and
very
low
cost
ofgenerate
treatRecovery
oftoor
both
molybdenum
of can
the
transmitter.
valves,
placing
the
transmitter
back
rotating
stem
tip.
In
the
case
of shown
athe
steam
reforming
to
be
used
in isthe
compared
other
hydroprocessing
the
RFCC
feed,
especially
when
orderservice
ofregardless
8-10%
for weather
Refinery
B.NH5%
reduction
in
catalyst
additions
fits
realised
atThese
Refinery
C
are
has
12
years
of
experience
inworked
heterogeneous
5
(a)
Internal
communication,
LoTOx.
For
more
than
30
years
he
has
Where
consistency
or
purity
ofCriterion
the
levels.
per
with
a
side
draw-off.
The
partly
tion
in
the
sulphur
plant
feed
re
ux
when
no
fractionation
on
the
individual
processing
units.
around
$18-19
billion.
Conventional
of
conditions,
from
economic
models.
For
exam3
of

ltration
are
performed
by
sieves
ment.
It
was
designed
to
be
capable
andQuality
tungsten
from
spent
catalysts
is
especially
important
in
in
service.
If
the
equalisation
valve
vee
tip,
there
may
be
a
knuckle
for the
Even
the
cycle
CCR
of discoveries
the
feed
to the
catalysis
andcontent
refining
technologies.
He hasare
coCatalysts
Technologies;
(b)
R
N,in materials/technology
has&been
compromised,
thiseconocan
With
any
gasso,
used(and
for
calibration
Highertechnologies.
zeolite
and
optimal
UCS
a gas
lb/bbl
basis.
Incremental
resid
insome
Figure
2.development
gas
knock-out
drum
thuson hydrotreaters.
stripped
sour
water
is
extracted
required
between
theMerryfield
feed
and
GTL
plant
designs
rely
on
for agreater
productivity
and
easier
ple,
shale
gas
with
different
cut-off
points:
theto
of Gardner
treating
all
ofshut-off,
spent chemical
involves
combination
of pyrometindustry.
He
hasexceeds
been
granted
US stem
a length
manifold.
During
calibration
or
notLachieve
positive
joint
that
enables
the
upper
authored
over
20
technical
publications
and
E,
Gcategories
D,aCatalyst
Characterization
result
inParks
expensive
system
recalipurposes,
most
important
of
the
ARDS
unit
is typically
VRDS
unit
2434
wt%.
References
water
that
has
to the
be
recycled
backatdoes
from
tray
8
and
directed
to
the
overhead
product.
Again,
the
only patents,
improved
coke
selectivity,
and
the
lower
catalyst
additions
caused
Refinery
D
routinely
processes
mies
of
scale
to
drive
positive
quality
control.
Since
module
height
being
hampered
by
high
developisstage
the
author
of 36inpapers,
andchemistry
holds

rst
retains
sublimated
metal
catalyst.
The
conditions
of
roasting
allurgical
and
hydrometallurgical
bration
procedures,
additional
requirement
is
that
it
can
accurately
holds
a
PhD
degree
organic
from
Science,
ACS
Symposium
Series
1985,
1.
normal
operation,
at
leastcase
one
of high
pressure
will
leak
across
the
turn while theforlower
stem
remains
1
K year.
R, Arsenic,
Environmental
Chemistry,
configuration
with
only
one
For
the
evaluto
the
sour
water
stripper).
hydrotreaters
use
as
make-up
purpose
of
the
tower
isbyto
striponly
out a BSc
financial
returns
and are
viable
isHenke
restricted,
safety
isthe
enhanced,
asthe Complex
ment
costs,
whichinresult
in marginal
and
PhD inresid
chemistry
from
refinery
increased
not
only
fresh
metals
to
increase
1000
ppm,
heavy
theImperial
FCC
unit.
Similar
cylinder
changeovers
and
wasted
andFigure
repeatedly
values
ofEcat
Iowa
State University.
6 controlled
Bhan
Othe
K,
Arsenic
removal
catalyst
and particles,
the
second
and
third
are
and
the
gas
treatprocesses
(see
3). report
theated
valves
in
the
manifold
isoperating
inheights
the
seat
to
low
pressure
side,
stationary,
except
for
axial
(up
and
Health
Threats
and
Waste
Treatment,
Wiley,
Complex
configuration
with
HCU
+
VRDS
+
RFCC
and
ARDS
and
the
H
S.
HS)
removal
The
design
shown
in
Figure
2
water
in
the
salt
(NH
the
NH
College,
London.
where
there
are
large
supplies
of
workers
build
at
limited
economics
due
to
gas
prices
that
3 resource
2 US
4
human
time.
relevant
instrument
being
measfeed off
rate
but
also
the
amount
of
which
drove
a
corresponding
to
the
other
examples,
the
objectives
method
for
making
same,
Patent
6759364.
treat
the+ gas.
Dust
captured
is
also
closely
monitored.
2009, 186. Ifattheleast
position.
shut-off
less
rendering
the
differential
reading
down)
movement.
The
intention
of this
config-stages
SDA
+
DCU
HCU
+ RFCC
objectives,
step
of
the
reactor
ef
uent.
was
in two
theis reactor
1960s.ment
low-priced
natural
gas.refinery
within common
the fabrication
often
low.
These
projects
can
be
residThe
inroasting
the feedstock
byfacility.
increase
inat the
dry
gas.
theinareIn
were
to
reduce
and
improve
step
thetherefinery
rst
stage
ofball
gas
ltration
is
head
ofHowever,
the production
furnace
than
complete,
result
istoalmost
an
inacinaccurate.
case
of configurations
adry
tip,gas
a to
ball
Correct
stripper
design
water
from
However,
itthe
also
suffers
from
theScreening
The
illusuration
is to another
increase
the
trains,
each
with
three
five
very
However,
option
being Completely
enhanced
by stripped
converting
the
10%.First
Higher
zeolite
also
increased
balanced
zeolite
and
matrix
surface
the
yields
of
liquid
fuels
(gasoline
of
all,
a
new
roaster
that
began
treated
at
the
hydrometallurgy
and
at
its
output
enables
the
extraccurate
reading
from
the
transmitter.
Why
would
these
valves
leak?
floats
in
the
stem
tip,
like
in
a
ballIn
1969, while
for expense
the now
sour water
stripper
bottoms
is
same
heat
balance
and of
Modularising
GTL drawbacks
trated
above,
where
residue
middle
distillate
at the
developed
working
smaller-sized
and of the
large
and
thick-walled
reactors, will
higher-value
clean
fuels
produced
at activity,
Valdi
in 2011
is
used
with
roasted
calcine.
Thefeed
tion
of
materials
found
the operations
circulating
enabling
a athe
of ceramic
the
catalyst
meant
that
and
distillate),
without
compromisFor
example,
when
calibrating
One
cause
is
debris,
either
point
pen.
Among
ball
tips,
there
vanished
Amoco
International
sent
thethe
crude
desalter.
While
complications
asboth
seen
inareas
The
same
advantages
of
modular
distributed
GTL
plants

shows
in
theto
GTL
process.
hydrotreaters
reduce
the
petrochemicals.
Relative
toscaling
theOil
firststage
beneedless
required.
Consequently,
Relative coke, wt%

Normalised HDS temperature,

F

Relative bottoms, wrt%

ITW

Innovative
Technologies
Worldwide

HIRING EXPERIENCED PROFESSIONALS 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 ZLOOLQJWRWUDYHOQDWLRQZLGHDQGZRUOGZLGHDORQJZLWKEHLQJSURQHWRKDUGZRUNLQJDQGVDOHV
Having a technical degree and good market knowledge is also required.
The available positions will cover: technical sales, implementation of ITW technologies on the field,
sales and operations management.
Interesting compensation plans will be given along with serious career possibilities. Please contact:
ITW S.r.l. S.Cusumano 96011 Augusta - Italy
102 www.eptq.com
PTQ Q1 2013

www.eptq.com
www.eptq.com
20
PTQ Q4 2013

114 PTQ Q4 2014

lieberman.indd 3

velocys.indd
2
cat
valdi.indd
2 3
swagelok.indd
f wheeler.indd
3
cri.indd 10

www.itwtechnologies.com
PTQ Q2 2013 101
www.eptq.com

PTQ
Q2
PTQ
Q12014
2014123
149
www.eptq.com

www.eptq.com
www.eptq.com
www.eptq.com

grace.indd 3

jobs@itwtechnologies.com

linde.indd 5

PTQ Q3 2014
79 2013 83
PTQ Q3

08/03/2013 16:02
10/12/12
14:27:56
12/03/2014
12:07
11/12/2013
14:52

12/09/2013 13:40

09/06/2014 14:27

11/09/2014 14:33

10/06/2013 16:38

ITW

Innovative
Technologies
Worldwide

CHANGE YOUR MIND

ON FOULING REMOVAL

ARE THESE IMAGES FAMILIAR TO YOU ? FORGET ABOUT THEM !

Patented ITW Online Cleaning can remove any type of fouling from equipment, including the polymers, without the need to
open or enter hazardous process equipment.
An entire Process Unit can be cleaned by utilizing ITW Online Cleaning in as little as 24 hours on an oil-to-oil basis.
Online Cleaning can be applied at any time during the run of the Unit, in order to solve the problems when they start
appearing, rather than when they are no longer sustainable.
This will in turn avoid throughput reduction, giveaway and energy loss, associated with fouling.
The application of ITW Online Cleaning will be therefore driven by performance recovery and Opex improvement, rather than
the economics for placing a turnaround.

ITW Online Cleaning can be applied to all Refinery/Petrochemical/Gas Field/Oil Field production Units, as
well as storage tanks, and is the only technology which has an immediate Return On Investment (ROI).
Regular application of ITW Online Cleaning will target an increased run length under clean conditions, with a far higher ROI.
For Turnaround applications, ITW Online Cleaning can eliminate/dramatically reduce the need for mechanical cleaning,
thereby reducing downtime and improve operational HS&E.
In a Turnaround application, an additional ROI increase can be targeted by applying ITW Improved Degassing/
Decontamination to achieve quick and effective safe entry conditions.
Our patented chemistry does not create any emulsion and fluids can be easily handled by Waste Water Treatment Plant.

EVALUATE ITW ONLINE CLEANING AND ITW IMPROVED DEGASSING/DECONTAMINATION 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

Now hiring
Professionals
Worldwide

Join ITW Team worldwide and send your Curriculum Vitae to : jobs@itwtechnologies.com

itw.indd 1

05/06/2014 19:38

1
0

1
A

0.3
0

0.3
0.6
0.9

Midas

Dry gas shift SCFB

Midas

4
2

0
2
4

2.5
0

2,5
5.0
7.5

Midas

150
100
50
0

50
100

Slurry shift, vol%

5.0

Gasoline +
LCO shift, vol%

LPG yield shift, %

0.6

Feed CCR shift

Feed API shift

A = Competitor A

Midas

Midas

Midas

3.0
2.0
1.0

0
1.0
2.0

Figure 3 At Refinery D, Midas increased gasoline and distillate yield and reduced dry gas
at a similar feed rate

Relative LCO yield, vol%

7.5
5.0
2.5
0
2.5
5.0
7.5

15

Base
Base + 50% Midas
100% Midas

10

Maximum matrix surface area

0

10

Relative conversion, vol%

Relative slurry yield, vol%

7.5
5.0
2.5
0
2.5
5.0
7.5
15

Base
Base + 50% Midas
100% Midas
10

10

Figure 4 Refiner E observed higher distillate yield over a range of conversion levels

grace.indd 4

Midas Gold is the latest innovation

in this catalyst family, and is
designed to provide maximum
matrix surface area allowing refiners
to further upgrade the bottom of the
barrel while minimising coke. These
catalysts feature Graces Low Coke
Matrix (LCM) technology, which
renders nickel inactive for dehydrogenation
reactions.
The
coke
selectivity advantage of Midas Gold
delivers a 15% reduction in bottoms
yield over prior formulations.
Multiple commercial applications
are under way.

Conclusion

Relative conversion, vol%

116 PTQ Q4 2014

ing feed rate. A switch to Midas

catalyst met the objectives despite
deteriorating feed quality coming to
the FCC. Over time, the feedstock
became heavier, contained more
concarbon, and had higher metals
loadings. Lower dense bed temperatures minimise thermal cracking,
reducing dry gas yield. Slurry yield
was held constant (see Figure 3).
The primary operating objective of
the FCC unit at refinery E is to
maximise LCO yield. Reformulating
to a high matrix catalyst from the
starting
low
matrix
catalyst
presented a challenge for this refinery. A wholesale change in the
catalyst formulation was deemed to
be risky. Grace introduced Midas to
the FCC unit in a step-wise fashion.
Initially, Midas was blended with
the refiners base high zeolite catalyst (also a Grace catalyst) in a 50/50
ratio. The FCC unit realised a reduction in slurry yield of 12% (see
Figure 4). Refinery E later increased
the amount of Midas to 100% of the
fresh catalyst requirement. Again, a
step-change improvement in slurry
was observed, and the incremental
conversion was maintained as distillate as desired. Higher distillate
yield (+15%) was observed over a
wide range of conversion levels
with Midas over the original base
catalyst.

Midas is Graces premium high

matrix bottoms cracking catalyst
family. It is designed to possess the
key characteristics necessary for
selective conversion of heavy FCC
feed molecules. These characteristics

www.eptq.com

10/09/2014 16:31

are
controlled
during
proprietary
in that
it shows
a lower
average
manufacturing
of Albertas
specialty
recovery factorprocesses
to estimate
aluminas,
ultimately
crude oil zeolites
reservesand
than
is used the
for
whole
itself.orThe
net result
is:
either catalyst
Venezuela
Saudi
Arabia.
The
Maximum
matrix
mesoporosity
to
table also
uses
the highest
minimise
and of
gasreserves
formation
published coke
estimates
for
and
maximise
upgrading
Venezuela
andbottoms
Saudi Arabia.
Ideal
poreof size
diameters
to
The figure
848 billion
bbl13 for
Albertasfreeremaining
preserve
diffusion in crude
and out oil
of
reserves
does even
not include
natuthe
catalyst,
in theany
face
of
ral gascontamination
liquids in shale gas reserves
metals
in as yetrare
unmeasured
bitumen
and
Optimised
earth exchange
to
sands extending
into the Athabasca
deliver
best-in-industry
coke
Basin in the adjacent province of
selectivity
Saskatchewan.
Balanced zeolite-to-matrix activity
also does
not include
substanto Itmaximise
product
value from
the
tial developed, undeveloped and
FCC.
potential
crude oil
the
Midas contains
thereserves
highest in
mesorest of the
WCSB
including
porosity
in the
desired
range the
of
northernpart
of the Williston
Basin
100-600
and
most optimal
and inthe
Athabasca
Basin
in
UCS
the market,
providing
a clear
Saskatchewan,
the over
Intramontane
selectivity
advantage
competiand catalysts.
Pacific Basins
the province
tive
Severalofdistinct
grades
of British
Columbia,
the
are
available,
suitablebasins
for a inwide
Yukonof Territory,
the Mackenzie
range
feed and operating
objecBasins and
the Beaufort
Sea,
tives.
TheDelta,
catalysts
in-unit
Canadas Arctic
Islands
fluidisation
properties
are a archipelsolution
ago,FCCs
as yet
bitumen
for
withunmeasured
fluidisation or
circusands difficulties.
deposits indicated
by or
a in
60
lation
Applied neat
long catalyst
outcropsystem,
on Melville
amile
Genesis
Midas
Island in
Arctic, diffiand
provides
the Canadas
option of running
offshore
Eastern Canada.
cult
or opportunity
feedstocks to the
Estimating
often
as
FCC
without reserves
suffering isfrom
poor
muchand
artgas
as selectivity.
science and,
in the
the
coke
Finally,
case of OPEC
and the
evenflexibility
Alberta, can
catalyst
provides
to
be morethepolitics
thanindividual
science.
fine-tune
yield of the
Reservesstreams
estimates
product
to exploit generally
short- or
change over
time as
new informalong-term
economic
opportunities.
tion
becomes available.
There
are
In summary,
the catalyst
should
also considered
differences by
in definitions
of
be
any refiner
reserves, for example, technically
seeking:
versuscracking
proved reserves.
recoverable
Deeper bottoms
most certainly,
estiAnd
Enhanced
bottoms reserves
conversion
mates are
subject
to constraint
interpretation,
against
an air
blower
opinion,
agendas,
professional
Ability to process
high
Fe and and
Ca
politics.
feedstocks,
such as Bakken and
Eagle Ford
Incremental fuels yield, for
ReferencesLCO, or gasoline and distilinstance
1 See
ERCB/AGS Open File Report 2012-06,
late
maximisation
Oct
2012.
Relief against a wet gas compres2 ERCB
Report ST98-2013.
sor
constraint
when cracking resid
3 Canadian Association of Petroleum
Capability to process opportunity
Producers Jun 2013 report Crude Oil Forecast,
crudes
Marketing, and Transportation, www.capp.ca/
getdoc.aspx?DocId=227308&DT=NTV
Improved catalyst fluidisation
characteristics
4 Imperial Oil, www.imperialoil.ca/CanadaEnglish/files/News/N_S_Speech060608.pdf
Flexibility to capture the value of
opportunity
crudes
5 Alvarez J, Han
S, Current overview of cyclic
injection
process,
Journal of gasoline
Petroleum
steam
Ability
to shift
between
Science
Research,modes
Vol 2, 3,of
Juloperation
2013.
and
distillate
w w w. c n rcatalytic
l . c o m / o p e rresponse
a t i o n s / n o r tto
h6 Rapid

www.eptq.com
www.eptq.com

grace.indd
Priaro.indd 56

america/north-american-crude-oil-and-ngls/

capture
short-term
economic
thermal-insitu-oilsands/
opportunities.
7 Cenovus, Telephone Lake Project, Vol 1
Project Description, Dec 2011, 4-25, www.
cenovus.com/operations/docs/telephoneReferences
lake/Volume%201/V1_Sec4.pdf
18 Schiller
R K, Grace Davisons GENESIS
www.geoexpro.com/articles/2010/04/
catalyst
systems
provide refiners with
the-king-of-giant-fields
flexibility
to capture
economic
opportunities,
9 Macquarrie
Equity
Research,
Feb 2010,
Catalagram
107, 2010.
26, www.sunshineoilsands.com/uploads/files/
2macquarie_report_01_10.pdf
Zhao X, et al, FCC bottoms cracking
mechanisms
and Survey
implications
catalyst
10 US Geological
study, for
An Estimate
design
for resid applications,
NPRA 2002,
of Recoverable
Heavy Oil Resources
ofAMthe
02-53.
Orinoco Oil Belt, Venezuela, pubs.usgs.gov/
3fs/2009/3028/pdf/FS09-3028.pdf
Mitchell Jr. M M, et al, Fluid catalytic cracking:
science
technology,
Magee
J S, Mitchell
11 US and
Energy
Information
Administration,
M M Jr. (Eds.), Studies in Surface Science and

www.eia.gov/cfapps/ipdbproject/IEDIndex3.
Catalysis, Vol 76, Elsevier, Amsterdam, 1993,
cfm?tid=5&pid=57&aid=6
293.
12 OPEC Annual Statistical Bulletins, www.
4 Yaluris G, et al, The effects of iron poisoning
opec.org/opec_web/en/publications/202.htm
on FCC catalysts, NPRA 2000, AM-01-59.
13 Cumulative bitumen production of 10
5 Cheng W C, et al, Fluid catalytic cracking,
billion bbl to the end of 2013 was deducted.
Ertl G, et al (Eds.), Handbook of Heterogeneos
Catalysis, 2nd Edition, Wiley-VCH, 2008, 2763.
Yee-Young
is Regional
Technical
Service
Mike PriaroCher
has worked
in facilities,
production,
and
Businessand
Manager
for Grace
Catalysts
operations
reservoir
engineering,
as
Technologies,
based in Singapore.
engineering consultant,
area superintendent,
Rosann
is Marketing
Director
Grace
and in Schiller
engineering
management
infor
Albertas
Catalysts
in Columbia,
MD,
oil patchTechnologies,
for 25 years based
for companies
such
as
USA.
Sheand
holdsPetroCanada.
an MSE in chemical
engineering.
Amoco
He holds
a BEngSc
Jeff
Koebel is National
Technical
Manager
in chemical
engineering
from Sales
the University
for
Grace
Catalysts
Technologies,
working
with
of Western Ontario.
customers in the eastern US.

Side-feed entry,
center-feed results.
Extend coke drum life with DeltaValves new
retractable center-feed injection technology.

Side-feed entry has been shown to negatively impact traditional resid flow
patterns in coke drums. DeltaValve has responded with the introduction of an
innovative retractable center-feed injection device utilizing side-feed entry.
This technology transforms side-feed entry back to the center of the coke
drum, generating stream flow patterns similar to traditional bottom-feed
injection. Recent thermal and strain measurement data and subsequent
calculations indicate a significant increase
in coke drum life may be achieved when
utilizing DeltaValves new retractable
center-feed technology.
Contact us at +1.801.984.1000 or visit http://deltavalve.cwfc.com

PTQ
PTQQ4
Q3 2014
2014 117
113

11/09/2014
10/06/2014 14:33
12:41

Advanced Solutions for

the Worlds Toughest
Energy Challenges
KEY TECHNOLOGIES:

Premium Lubricants

Resid Upgrading

Methanol to Gasoline
Premium Diesel

Acid Gas Clean Up

ExxonMobil Global Leader in Fuels and

Lubes Process Technologies
Enabled by Proprietary Catalysts and Solvents
ExxonMobil Technologies are applied across our corporation and by licensees worldwide in
a growing list of process industry applications. Leverage our vast experience and ongoing
commitment to continuously improve our industry-leading technologies.

www.exxonmobil.com/tsl

FP.Technology_4PTQ.indd
1
exxon.indd
1

3/6/14 3:22
PM
10/03/2014
13:03

Preventing emissions in coke removal

A closed, automated coke removal system aims for gains in environmental,
economic and safety performance
ARTUR KRUEGER, BERND LANKERS and JOSEF WADLE
TriPlan AG

onventional delayed coker

units feature open pit/pad
coke handling methods, utilising a range of mechanical
equipment including bucket cranes,
front loaders, offsite crushers, dusty
loading facilities and blackened
railroad shunting or truck loading
areas.
In contrast, the Closed Coke
Slurry System (CCSS) developed by
TriPlan in Germany offers the operator a modern process that is
completely enclosed. It is environmentally friendly by eliminating
coke dust emissions, and it also
provides effective water management by cleaning and recycling the
water used in the coke quenching
and cutting process.
The CCSS has completely automated
operations,
the
plant
components are designed and fabricated with a view to high
reliability, resulting in significantly
enhanced overall plant reliability.
The safety aspects attained in this
design find a high acceptance level
by the operators as well as the
workforce. Streamlined operations
result in reduced cycle times for
coke drum unloading and dewatering, providing an interesting
pay-back to the operator.
In the hydrocarbon processing
industry, the conversion processes,
from crude to clean products, are
continuous. Residues from atmospheric or vacuum units are
converted by a number of different
processes into added-value products, to obtain maximum benefit
from the bottom of the barrel. One
of these conversion methods is the
delayed coker unit (DCU). The
product obtained from this unit is

www.eptq.com

triplan.indd 1

commonly known as petroleum

coke which serves as the feed for a
number of different applications.
Unlike the other processing units
in the refining business, delayed
coking is a discontinuous batch
operation. Full coker drums are
opened on a regular basis for
unloading. Challenges come from
the process itself, the metallurgy of
the equipment, and safety, health
and environmental issues in view
of the high degree of mechanical
handling steps that are typical for a
traditional
pit/pad
system.
Consequently, field operators face
more manual work in an uncomfortable, and sometimes risky, work
environment that is not known in
other refinery units. Staffing figures
for operators, maintenance staff
and
additional
housekeeping
contractors are traditionally high
and result in a substantially
increased workload.
These cost elements are considered unavoidable and are tolerated
by operating companies. However,
coker outages due to breakdown
and repair of mechanical equipment mean loss of coke production
capacity, and clean product uplift
value deteriorates dramatically,
eating into the units revenue
stream.

Traditional processing in coke drums

The process step from liquid coker
feed to the solid coke formed in the
coke drum is independent, whether
it involves the established pit/pad
design or CCS technology for
handling the coke after cutting.
The feed is heated to commence
cracking, ultimately to extinction to
carbon due to the long residence

time in a drum. Once a drum is

filled up, the feed stream is
switched to the next empty drum
and so on. Therefore a delayed
coker consists at least of two, sometimes of four or even six drums.
After switching, the full drum is
isolated from the rest of the unit for
further processing. With injection
of steam into the coke bed, light
hydrocarbons are stripped out, then
the bed is hardened and cooled.
Further cooling is achieved through
repeating quenching (filling and
emptying) with water. Once the
bed temperature falls below 100C,
the drum top and bottom flanges
are
manually
opened
(more
recently, remote operated slide
valves are commercially available
to replace the risky flange
operation).
Coke removal takes place with a
high pressure cutting water circuit
at 250-300 bar. First, a vertical hole
is drilled into the coke bed with a
downward water jet nozzle. This is
followed by the cutting operation
in which a horizontal water jet
slices top down to ream out the
drum up to the wall until the drum
is empty. The resulting coke chunks
are flushed through the drilled
hole.

Open coke handling is a burden on

the environment

Delayed coker operators with a

pit/pad system have made efforts,
together with mechanical equipment suppliers, to improve the
mechanical steps in the process, in
particular with installation of top
and bottom unheading equipment.
Unit on-time capacity has been
increased, contributing to the units

PTQ Q4 2014 119

12/09/2014 12:53

Vent
Coke
drum

Dewatering
bin

Water
settling tank

Clean water
tank
8

Make up
water

Chute
Crusher

Coke

Drain
water

Sludge
7

Drain water
basin

Slurry pit
Slurry pump

Drain water pump

Clean water pump

Key:
1. In-line crushing during the cutting operation
2. Water feeding for coke transport and further cooling
3. Forwarding coke slurry into the dewatering bin
4. Water diffusing through the voids of the coarse material into the drain water basin
5. Trapping and retaining of coke fines (99.5%) in the coke bed
6. Forwarding drain water into the water settling tank for separation of remaining coke
fines (<0.5%)
7. Discharging collected sludge into the slurry pit
8. Converting clean water into the clean water tank for reuse
9. Coke product unloading from the dewatering bin

Figure 1 How the closed coke slurry system works

cash flow. At the same time, environmental concerns have been

addressed, as well as the creation
of a healthier workplace.
In the traditional pit/pad process,
coke handling from the outlet of
the coke drum after cutting, to
loading
onto
trucks/railcars,
happens in open space. The free
stacking of hot coke during the
cutting operation and storage of the
coke from one coke drum (typically
1000 t/d) result in a highly visible
steam plume for several hours.
During this time, the coke stack
cools down and drains excess free
water to the ground or into a
receiving maze, usually uncovered.
It is known from investigative
reports that the steam contains
coke fines and aerosols, and also
aromatics and other polycyclic
hydrocarbons.
US federal and state regulations,
and European Community directives on environmental issues in the
hydrocarbon processing industry,
call for a critical review of all possible sources of air emissions and

120 PTQ Q4 2014

triplan.indd 2

ground pollution, such as the

mechanical equipment mentioned
above, including black water laden
with coke dust and fines, throughout all stages of coke handling
and storage.
Technical solutions to alleviate
these problems, for example the
enclosure of coke piles, water/oil
spraying, are all aimed at fighting
the problems at the back end rather
than at the source. They are costly
but can still be insufficient.

Closed coke slurry technology

In Europe, coker operators have to

demonstrate that their current practice is in accordance with best
available control technique (BACT).
This is where TriPlan CCSS technology comes into play. It is a
completely enclosed system, avoiding the mechanical, environmental
and health drawbacks which are
inherent in conventional pit/pad
systems.
In all coke handling steps, from
the discharge point at the bottom of
the coke drum to the coke to load-

out, the water circuits are designed

for re-use of contaminated water in
the process after clean-up, and the
effective separation and removal of
coke fines. These steps have been
streamlined at all stages of the
process, making it no different
from any other typical process unit
in the refinery. This is achieved by
pumping the coke slurry stream in
a fluid state. With consistent
improvement of metallurgies in the
equipment that forms part of the
CCSS, the process is robust
and stable.
An installation has been operating for more than six years without
interruption for repair or maintenance work in service in a refinery
in Germany. The instrumentation
in this process permits a fully
controlled operation and allows the
console operator to view the status
of the process at any time.
In summary, the technology gives
the delayed coker operator an operationally enhanced process, with
ecological as well as economic
advantages. Figure 1 illustrates the
principle of the technology.
Mastering corrosion and abrasion
in hot, wet and harsh service were
the prevailing challenges in the
course of developing the CCS
process. These challenges were
compounded by the fact that some
equipment has to be totally newly
designed and constructed to suit
the special purpose.
To pump solids like coke, a
reduction in grain size by crushing
is required prior to entering the
pump. The crusher is located
immediately beneath the coke
drum. A remotely operated telescopic chute connects the crusher
and drum bottom flange prior to
the boring/cutting operation. A
slurry pipe routes the mix of
crushed coke and water into
the pit.
Crusher torque, materials, essential construction details like teeth
pattern, roll diameter, pull-in and
swallow capability are designed to
cope with large chunks and peak
loads, often simultaneously. All of
these parameters are aimed to meet
specific requirements for achieving
maximum run length.
Likewise, the construction and

www.eptq.com

12/09/2014 12:54

PERSONALLY VISITING THE CYCLONE FABRICATOR TO ENSURE THE PRECISE FIT OF EVERY LUG WOULD BE

At AltairStrickland, obsessing over the tiniest details is an enormous part

of who we areand a huge advantage for our clients. Our fanatical focus
on pre-planning to reduce project schedule helps ensure a safer, more costeffective turnaround every time. When a million pounds of steel is hoisted from
a million-dollar crane, thats not the time for costly surprises. Thats why we
use AutoCAD, 3D surveys, even plywood to build full scale models to explore
virtually every conceivable what-if scenario. Over the top? Maybebut we
believe your next FCCU revamp or turnaround deserves it.

How can we obsess for you?

emcor_info@emcor.net 866.890.7794 altairstrickland.com

TURNAROUND
WELDING SERVICES

TIGER TOWER
SERVICES

DIAMOND
REFRACTORY SERVICES

AltairStrickland
CRITICAL PROJECT EXECUTION FOR THE REFINERY AND PETROCHEMICAL INDUSTRIES

AS_PTQ_140528.indd
1
emcor.indd
1

6/2/14 9:59
AM
10/09/2014
11:41

Comparison of the steps in pit/pad design and CCS technology

PIT/PAD technology
CCS system
Initial capital cost
Base case (100%)
Approx. 20% higher
Dewatering efficiency
High on top (vaporisation);
Uniformly good

poor at the bottom of pile
Water loss/make-up
High
Negligible
Number of coke handling steps (mechanical)
3
0
Degree of sludge disposal efforts
High/mechanical
Nil/will be pumped
Reliability of mechanical equipment
Low
Very high
High
Low
Flexibility against outages
Manpower requirements field/control centre
High/not possible
Nil/low
Instrumentation and degree of automatisation
Limited
Full
Healthy and safe working conditions
Very low
High
Risk of fire due to self-ignition
Very high
Nil
Poor
Good
Access for maintenance/fire brigade

Table 1

materials of the slurry pump enable

stable and reliable transport of
abrasive slurry at lowest net
positive suction head (NPSH)
requirements.
An important objective is system
availability to handle various coke
loads and stream conditions from
water-only up to avalanchestyle coke bed break down which
is convincingly solved in the
design of crusher and pump
together with their sensible drive
concepts. In addition, a stable transport water rate is provided to
maintain pumpable slurry at all
times.
Another issue is the use of the
coarse coke in the dewatering bin
as a filtration bed for fines-laden
black water. Selective operational
steps provide an absolutely clear
water run-off with shorter residence times, to be collected in a
tank and re-used as quench
and cutting water for the next
cycle. The well known abrasion
effects in expensive multi-stage
cutting water pumps are basically
history.
The technological and economical
advantages of the system include:

Reduction in cycle time

The CCSS system permits a reduction of the total cycle time of up to

four hours (see Table 1).

Capacity gain

Shorter cycles allow the operator to

process more feed, which also
results in increased production of
clean products, depending on

122 PTQ Q4 2014

triplan.indd 3

downstream capacity for light

products.
A production gain of 400 tonnes
of coke can be achieved. Another
aspect is the avoidance of unit
downtime due to repair and
unscheduled maintenance. Feedback from existing delayed coker
units indicates an average loss of
5% of the optimum yearly production schedule due to damage or
wear and tear in the mechanical
equipment which amounts to
approximately 400 hours outage
each year.

Reliability

The equipment in the CCSS system

is robust and reliable due to its
design and to metallurgy selection
for each and every component. The
unit operates without trouble
between two regular scheduled
coker turnarounds. In the refinery
reference unit mentioned earlier, a
record of 99.9% unit availability
was achieved over a period of
six years.

Capital cost (capex)

In a grassroots/greenfield project
for a delayed coker, the design of
the CCSS downstream equipment
can be planned and executed at
incremental capital cost.
In the example of a two-drum
unit with a capacity for each drum
of 1000 t/d of coke, a payback time
of about 1.5 years can be achieved.
However,
the
design
varies
between individual delayed coker
units and a careful study is
required.

Operating cost (opex)

An average loss of 5% of the optimum yearly production schedule

due to damage or wear and tear to
mechanical equipment, resulting in
approximately 400 h/y outage,
costs 300 000-500 000 per annum.
However, the yearly loss of light
products, depending on their
market value, in the range of 10
million per hour of cycle time, is
significant in comparison.
Environmental issues are alleviated through the completely
enclosed process, offering an
advantage also when renewal of
existing permits or permitting of
new units is required.

Conclusion

The CCSS technology is a convincing solution when a new coker unit

is planned. A mix of gains that are
economical, ecological, technical,
and safety-orientated comes into
play. Investment decisions will
consider operational benefits, in
addition to pure capex considerations, when new coker units are
being
planned.
Furthermore,
increasing awareness of environmental
issues
by
permitting
authorities in many countries
would also provide strong support
for the technology described here.
Artur Krueger is the Houston, Texas-based
Senior Account Executive with TriPlan AG,
Germany, consulting to operators and
engineering firms on coker projects worldwide.
He holds a masters degree in business
from Columbia University, New York and a
betriebswirt graduate business degree from
Wirtschafts-Akademie, Bremen, Germany.
Email: artur.krueger@gmail.com
Bernd Lankers is a Project Manager with
engineering company TriPlan based in
Germany. He has over 40 years of refinery
and petrochemical experience, including
several major overseas project assignments.
He holds a BS degree in chemical engineering
and economics from the University of Applied
Science at Bingen/Germany.
Josef Wadle is a Senior Consultant with
engineering company TriPlan in Germany. He
has over 42 years of broad refinery background
with various project assignments as lead
mechanical engineer. He holds a BS degree
in process automation/ control and chemical
engineering from the University of Applied
Science at Mannheim/ Germany. He is also a
certified welding and metallurgical engineer.

www.eptq.com

15/09/2014 12:08

What you can do

with a

touch of blue.
You can

Improve your refinery profitability by maximizing the production of clean

transportation fuels with our leading residue upgrading technologies.
Deliver the best in refinery hydrogen production while reducing
your operating costs with our unique Terrace Wall reformer design.
Enhance the efficiency of your overall sulfur recovery to achieve peak
operating and environmental performance with our SRU technology.

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
fw.indd 1

11/06/2014 10:08
7/3/12 13:55:37

Trust the #1 online monitoring system for mission-critical equipment

Who monitors the most reciprocating compressors?

When you look at the numbers, it all adds up to PROGNOST Systems:
Largest installed base of online diagnostic systems on reciprocating compressors
The worlds top reners and HPI producers trust PROGNOST systems
Over 20 years of experience in monitoring reciprocating machinery
PROGNOST has one focus reciprocating machinery.
Our dedication ensures that your system remains best-in-class.
reply@prognost.com www.prognost.com

prognost.indd 1

05/06/2014 19:23

Troubleshooting steam ejectors

Some causes and practical solutions concerning problems with refinery steam
vacuum ejectors
NORMAN LIEBERMAN
Process Improvement Engineering

he majority of vacuum ejectors in refinery services do not

perform as per their design.
Steam vacuum ejectors, or jets, are
widely used in both condensing
steam turbine surface condensers
used to generate power and drive
large compressors. Also, steam jets
are employed in vacuum towers
which are used to produce paving
asphalt and lubrication oils.
Poor performance of a steam
turbine surface condenser will
typically increase the energy
requirements to generate electricity
by 10%. Poor performance of a
steam jet in a refinery may cost the
refiner $20 000 (US) per day in
reduced gas oil recovery from
visbreaker bottoms or from delayed
coker feed. Degraded vacuum may
preclude meeting the viscosity and
flash point specification for the
production of paving grade road
asphalt or industrial fuel oil.
Many refiners are able to sustain
a reasonable vacuum during the
winter, but find that their vacuum
breaks and their steam ejectors start
to make a surging or hunting
sound when summer arrives.
The problems that give rise to
these malfunctions number in the
hundreds. A book on this subject1
has been rightly criticised for omitting half of the common causes for
vacuum system malfunctions. The
following tabulation is far from
complete, and only represents the
most common malfunctions seen
most recently.

Erosion of steam nozzle threads

The steam nozzle (see Figure 1) is

typically machined from 316 (L)
stainless steel. It is not generally

www.eptq.com

liberman.indd 1

subject to damage to its threads.

The major problem is that the stainless nozzle screws into the carbon
steel ejector backing plate. The
female threads of the backing plate
corrode-erode, apparently due to
galvanic corrosion and/or erosion
due to wet motive steam. The
motive steam then partly by-passes
the steam nozzle and blows into
the mixing chamber. The moisture
in the steam acts like an electrolyte.
Superheated steam would reduce
this galvanic corrosion problem.
Another common cause of steam
nozzles becoming loose because of
erosion of their threads is because
the nozzles were never installed
properly. Either the threads were
dirty or the nozzle was not screwed
in tightly. This reportedly happens
when the steam nozzles are
replaced in the field. However, we
have noted this problem on original
ejectors too, with no history of
nozzle removal. Whatever the
cause, a small amount of leakage
around the threads will escalate
due to steam erosion. So, if steam

nozzles are removed for inspection,

clean off all the threads and reinstall the nozzles tightly using
Teflon tape.
If you notice an improvement in
vacuum by reducing the motive
steam pressure by 20% below the
design motive steam pressure,
eroded threads may be the problem. We have temporarily fixed this
failure with lots of Teflon tape.
Why do the ejector manufacturers
not use a stainless steel backing
plate, and weld the steam nozzle to
the ejector and avoid this problem?
Actually, several refiners do follow
this useful practice.

Plugging of steam nozzles

We have written about this subject

and published photos of partly
plugged steam nozzles extracted
from a steam turbine surface
condenser.2 The problem here is
silicates and other hardness deposits in the motive steam supply. This
is a consequence of poor level
control in upstream waste heat
steam boilers. This is such a

C.S. backing plate

S.S. steam nozzle
Mixing chamber

To condenser

Steam

Diverging
Diffuser
Converging

Gas inlet

Figure 1 Ejector erosion occurs where a stainless steel nozzle screws into a carbon steel
backing plate

PTQ Q4 2014 125

10/09/2014 16:37

Vapour to
ejector

CW

Pass partition
baffle

CW

Channel head
tubesheet

Drain leg
Condensate

Figure 2 A leak at the pass partition baffle gives a false impression of fouling

common problem that the manufacturers typically provide a small,

screwed, clean-out plug in the
backing plate of the ejector (see
Figure 1). As most vacuum systems
have two or three parallel sets of
jets, one could isolate one jet and
easily clean out the silicates while
still on-stream.
Incidentally, if one steam ejector
with a relatively clean nozzle is
operating in parallel with a jet with
a fouled nozzle, the good jet will
extract motive steam out of the
fouled ejector and greatly degrade
the efficiency and capacity of the
clean ejector.

Distance between steam nozzle and

diffuser throat set wrong

The distance between the steam

nozzle and the inlet to the diffuser
(see Figure 1) is adjustable. There
are spacers that are installed
between the ejector backing plate
and the steam nozzle that are used
to make this adjustment which is
supposed to be done in the factory.
This adjustment is, at least partly, a
function of the motive steam pressure. On several troubleshooting
projects, the author, having failed
to correct the ejector malfunction,
was told by the vendor that the
steam nozzle position had been
incorrectly set in the factory for the
current motive steam pressure and
temperature.

Moisture content of steam

The variable that provides the

energy to compress the gas to the

126 PTQ Q4 2014

liberman.indd 2

ejector is the kinetic energy of

the motive steam. This kinetic
energy is not derived from the
motive steam pressure but from
the motive steams enthalpy or
heat content. If the steam is
wet, the moisture flashes to steam
in the steam nozzle. The resulting
conversion of the sensible heat
of the steam to latent heat of evaporation of the entrained water
reduces the speed or velocity of
the steam entering the diffuser.
In one recent case, wrapping 25 ft
of a two-inch steam supply
line with loose insulation improved
the vacuum from 54 to 49 mm
Hg.
On the other hand, highly superheated steam is also marginally bad
for the ejectors performance.
Hence, steam condensate is injected
through a desuperheating station.
However, it is not uncommon for
the temperature controller to be set
a few degrees below the saturation
temperature of the motive steam.
The author has seen this done at
two refineries recently. The effect,
of course, is to inject uncontrolled
water into the motive steam.
Auto-refrigeration of the exterior
of the converging section of the
ejector will be observed. Also,
there will be erosion of the carbon
steel threads where the stainless
steel steam nozzle screws into the
ejector body as a consequence of
the water in the steam. In one refinery,
correcting
this
problem
improved vacuum from 28 to 15
mm Hg.

Air leaks that cause nitrogen in the

off-gas to exceed 10-20%

The easy way to find significant air

leaks is to look for moisture
condensing on the ejectors body,
barometric drain legs, or flanges.
Air, as it is drawn into the diffuser
or flange, expands and auto-refrigerates. In areas of high humidity,
atmospheric moisture will condense
out on the exterior of chilled metal
surfaces. The quite common solution for on-stream repair is duct
tape. This is a permanent and inexpensive repair technique. Air leaks
in a seal leg will cause poor drainage through the affected leg and
condensate back-up.
High concentrations of CO2, as
opposed to oxygen, in the vacuum
tower off-gas is an indication of an
air leak in the line connecting the
fired heater to the vacuum tower
obviously a potentially dangerous
situation which requires careful
inspection of the heaters transfer
line. Imagine what may (and has)
happened if the vacuum is suddenly
lost and the pressure in the transfer
line goes positive. Several catastrophic fires have resulted.

Condenser pass partition

baffle leakage

The heat transfer coefficient (U

value) of a surface condenser,
vacuum tower pre-condenser, or an
ejector after-condenser should be
about
110130
BTU/hr/ft2/F.
However, only too often the
observed U value is only 3040.
This would superficially indicate
fouling. Yet, frequently, both the
shell side and the tube side of the
condenser are found to be quite
clean. The problem is not fouling or
really even a low U value, but the
cooling water leaking internally
inside the condenser channel head
(see Figure 2). If the closure or
clearance between the channel head
pass partition baffle and the channel head tube-sheet is not tight
(and there is no way to pressure
check this closure), water will leak
from the inlet side of the channel
head, directly to the outlet side. If
there is sand in the cooling water,
the leakage rate will gradually
increase due to the scouring action
of the sand.

www.eptq.com

10/09/2014 16:37

Superficially, the problem seems

to be a large delta T between the
cooling water outlet and the
condenser vapour outlet temperature, hence the small calculated U
value. In reality, the U value is
normal. It is just that a large
percentage of the cooling water
supply is by-passing the condensers tubes.
One can identify this sort of
malfunction by comparing the
calculated water side pressure
drop, to the observed delta P. To
prevent this problem, use of a silicon sealer along the edge of the
channel head, pass partition baffle
and the channel head tube sheet is
recommended.

Vapour inlet
260F

Air baffle
Seal strips

Vapour
150F

Defective seal strips

The shell side seal strips are

intended to keep hot vapours in the
inlet to a condenser from blowing
out directly to the vapour outlet
(see Figure 3) and by-passing the air
baffle. This malfunction is indicated
by the condenser vapour outlet
being 20-40F hotter than the liquid
outlet. The reason for the defective
seal strips is that the seal strips are
too short, damaged on installation,
or made of the wrong materials. In
one installation observed recently
in Mexico, the refinery vacuum
towers first stage condensers seal
strips were made out of brass. As
the vacuum tower overhead vapour
contains NH3, HCN, HCl, H2S, CO2
and H2O, brass is not the correct
metallurgy. The destroyed seal
strips were replaced with 316 (L)
stainless steel, which did not help
either as they were not installed
wide enough to contact the interior
of the shell. It is best to extend the
seal strips by inch beyond on the
condenser shells inner diameter.

Fouled final condenser in a steam

turbine surface condenser

Quite often, the final condenser,

which is the last component of the
small two-stage ejector system used
on all such condensers, fouls on the
steam side. The correct solution to
this very common malfunction is
not to clean but to discard the final
condenser and vent the second
stage condenser vapour effluent
directly to the atmosphere. Why

www.eptq.com

liberman.indd 3

Condensate
100F

Figure 3 Broken seal strips permit internal vapour by-pass around baffle

then are surface condensers which

are serving condensing steam
turbines equipped with a final
condenser in the first place if it
serves no function? The answer is
that there is no answer, other than
historical precedent.

Steam turbine surface condenser

loss of vacuum

The two-stage ejector system that

provides the vacuum to a steam
turbine surface condenser is
connected
to
the
two-stage
condenser (see Figure 4). The very
small condenser shell (typically
12-18 inches in diameter) has an
internal segregation baffle that
divides the shell into two unequal
compartments. Failure of this segregation baffle is common and allows
the discharge of the second stage
condenser to leak back into the
suction to the second stage ejector.
An extremely large loss in vacuum
in the steam turbine surface
condenser results. This quite
common failure has reduced a very
large steam turbine horsepower
output by almost 10%. The solution
to this problem is to simply vent
the discharge from the second stage
ejector directly to the atmosphere.
As noted earlier, the second stage

condenser serves no function.

The author has by-passed this
component
of
the
standard
condensing steam turbine many
times and has never observed any
detrimental effect. That is, simply
vent the discharge of the second
stage ejector directly to the
atmosphere.

Condensate back-up in condenser

This common problem is also indicated by the condenser vapour

outlet being hotter than the liquid
drain line. Once the liquid backs up
inside the condenser shell and
covers the bottom edge of the air
baffle (see Figure 4), there will be a
step change reduction in vacuum,
as liquid is drawn up into the inlet
of the downstream ejector. Factors
observed in the past few years that
cause condensate back-up are:
Drain line plugged with wax
Flange leak on drain (barometric
seal leg) line, that draws air into
the seal leg
Corrosion of drain line inside seal
drum, above the seal drums liquid
level
Seal drum flooded due to faulty
level control
Sludge in seal drum above the
bottom level of the internal drain

PTQ Q4 2014 127

10/09/2014 16:38

1st ejector
discharge

200 mm Hg

ual performance characteristics. Its

performance does not often conform
to expectations, but more frequently
to the minimum performance that
one is willing to tolerate.

2nd ejector
discharge
770 mm Hg

Refrigerated cooling water

To

2nd

ejector

ATM vent

190 mm Hg

760 mm Hg

Leak in
segregation baffle

Condensate drains

Figure 4 Steam turbine surface condenser ejector system failure

leg. This is the most common

malfunction
High pressure in seal drum
Seal drum elevation too high
compared to the condensers
elevation.
This tabulation is not complete
but are problems experienced in
the last few years. It is a good
example, though, of the diversity of
the problem when troubleshooting
refinery vacuum tower overhead
systems.

Digression from theory

Consultation with ejector manufacturers does not typically lead to a

solution. The difficulty appears to
arise partly from the manufacturers assumption that the vapours
flowing to the inlet of the diffuser
(see Figure 1) are flowing above
their sonic velocity as they exit the
mixing chamber. Based on the exterior temperature profile (that is, a
gradual increase in temperature
along the converging section of the
diffuser), this appears to not
normally be the case. The reasons
for the subsonic velocity at the
diffuser inlet are most likely to be a
combination of:
Motive steam blowing around
the steam nozzle through eroded
threads in the backing plate

128 PTQ Q4 2014

liberman.indd 4

Hardness deposits in the steam

nozzle (mainly silicates)
Moisture in the motive steam
Excessive back pressure from the
downstream condenser
Overloading the ejector with
excessive air leaks or cracked gas
Unless the motive steam enters
the diffuser inlet above the speed of
sound (sonic velocity), then the
ejector cannot perform in the
manner anticipated by the manufacturer, based on the performance
tests that have been conducted at
the manufacturers research and
testing facilities.
Each ejector has its own individSeal drum off-gas from refinery
vacuum tower
Component No air
admitted,
Mol%
H2
7.8
O2
1.5
5.8
N1
C1
18.2
CO
1.5
CO2
2.0
C2
13.2
9.9
H2S
C3
14.2
15.1
C4
C5+
11.7

Table 1

Air added through

inch nipple,
Mol%
2.9
11.1
43.8
6.4
0.5
0.8
4.9
3.7
6.5
7.9
11.7

Some refiners employ emergency

rental package cooling water refrigeration units during the summer
months. These packaged units will
reduce the cooling water temperature to the condenser by typically
20F. On many if not most units,
this reduction in the cooling water
temperature will have a dramatic
effect on improved vacuum. A
packaged chilled water system can
be put into service quickly.
Naturally, this is a relatively expensive option compared to operating
the existing system properly. Still,
the refiner sometimes is forced to
adopt such expedients, particularly
when producing paving grade
asphalt during summer operations.

Determining non-condensible flow

rates without a flow meter

The main concern when troubleshooting an ejector system is to

determine if the vacuum system is
working on its vendor performance
curve. If there is no flow meter, a
series of actions can be followed to
obtain the non-condensible load to
the ejectors:
Obtain an off-gas sample from
the seal drum. Check for N2 content.
Assume N2 content is 5% for illustration purposes
Through a small (- inch)
screwed fitting, located at the
suction of the first or second stage
ejector, allow air to be drawn into
the ejector system
Wait five minutes and obtain a
second off-gas sample from the seal
drum. Check for N2. Assume the N2
content is now 25%
Assume that the air velocity
through the screwed fitting is 1100
ft/s, the sonic velocity consistent
with critical flow
Calculate the volume flow of N2
(78% of air), based on the open area
of the screwed fitting, and the 1100
ft/s air velocity
Calculate the density of air at the
current ambient conditions (about
0.075 lb/ft3)

www.eptq.com

10/09/2014 16:38

Indicated power, %

and microbiological aspects of biofilm-metal interface. Both aero100

seal
drum
off-gasrefinery
contains
10-40%
result
Multiply
the volume
of of
airfouling,
by its bic
in corrosion.
better
handling
which
provided
personnel
microbial
and
anaerobic
organisms
play
Step control
S.
The
fatal
concentration
oftoH2be
S
H
density,
to
obtain
pounds
of
air
and
at
the
same
time
increase
with
the
right
information
90
Microbial activitiesReverse
act asflow
a drivimportant
role
in
the
initiation,
control an2 Energy
savings
is
0.1%.
Fresh
air
equipment
flow
per
second
HVGO
directly
(less
gasoil
recycled
able
to
constantly
keep
the
unit
at
ing
force80 for Ideal
biocorrosion.
control system propagation, and inhibition of
oil).
Figure
17 shows
the corrosion
the best
process
severity
for their
any
content
has
increased
when
obtaining
these
samples
is
asAswash
the N
2
Microbiologically
induced
corroin
different
ways
by
70
trend
of
wash
oil
rate.
processed
feed.
by
20%
(25-5%),
you
may
assume
advisable.
sion
causing
organisms
are different
energy
deriving
60
decrease
in wash bacteria
oil
rate pathways.
that The
the air
flow
calculated
above
sulphate
reducing
50 of the optimised.
continuously
The Conclusions
iswas
about
20%
non-condensi(Desulphovibrio,
Desulphotomaculum,
Fungi and algae may also be
40 analysed
HVGO
was
on
a
daily
The examples
citeddeterioration.
here are a tiny
ble
gas
flow.
Therefore,
multiply
and Desulphomonas sp.), iron reduc- involved
in metal
In
basis
with
respect
to
the
level
of
segment
of
the
vast
number
of
the
pounds
of
air
per
second
30
ing bacteria (Gallionellea ferrugine fuel and oil storage tanks,
fungal
contaminants,
using
the
VCI
techproblems
that
beset
steam
ejector
calculated
above
by
approximately
20
and Ferrobacillus
sp.), acid produc- species such as aspergillus, penicilnique that measures the coke and surface condenser perforfour
ing
bacteria
(Pseudomonas, lium and
fusarium may grow on
10
entrainedweight
within
the mance.3,4,5 There is always an
particles
If the molecular
of and
the
Aerobacter,
and
Bacillus),
fuel
components
and produce
HVGO (thus
removedfrom
by wash
understandable
temptation
to
cracked
gas 0oxidising
isnotdifferent
the
sulphur
0
10
20 bacteria
30
40 carboxylic
50
60
70acids
80 which
90 100corrode
oil).
consult
the
equipment
vendor
for
molecular
weight
of
air,
multiply
(Thiobacillus sp.).
iron. %
Capacity,
Also,
the ofrate
of antifoulant
on advice and guidance, when trying
the
pounds
flow
the ratio in
of
Microorganisms
arebyubiquitous
the wash
oil was optimised
and to troubleshoot steam ejector
the
molecular
of rates
the
Role of biofilms
nature
and grow atweight
very rapid
increased
when
needed.
performance
problems.
Typically,
non-condensible
gas,
divided
by The
Figure
6 Comparison
the compressor
power
consumption
step
control
term between
biofilm
refers
toandthe
in
soil,
water andof
air.
They show
reverse
flow controlweight
as a function
of the
compressor
capacity
this
does
not
help.
Only
field
the
molecular
of
air
(29).
extreme tolerance to varying envi- development of microbial commuConclusions
investigation,
experience,
careful
See
Table
1
for
an
actual
set
of
test
on submerged surfaces in
ronmental conditions such as acidic nities
VisTec
is a mark of and
Baker
Hughes
Incorporated.
This
article
example
of of
measurements
experimentation
results
from describes
apH,
low low
sulphur
refinery
automation
allows
for
precise
of
the
compressor,
theanand
number
of
aqueous
environments.
Biofilm
and
alkaline
higher
high
conversion
visbreaking
with
appear
to
offer
any
hope
of ofa
vacuum
tower.
plant
control
and
fine-tuning
cylinders
per
stage
can
be
reduced.
the
physic-chemical
temperatures as shown in Table 1. influences
increased
runa calibration
length
when
successful
outcome.
The
ejectors
If
using
air
as
gas
is
process
parameters,
such
as
gas
Hence,
the
capital
costs
and
footMicroorganisms and their meta- interactions between metal and
compared
to
typical
visbreaker
run pressure,
should
and
will
perform
on
the
unacceptable,
argon
or
helium
may
hydrogen/feed
ratio
and
print
of
the
compressor
are
References
bolic products, including enzymes, environment, frequently enhances
lengths.
manufacturers
performance
curves.
be
employed,
but
this
is
a
time
gas
flow
rate.
minimised.
Furthermore,
fewer
1
Petralito
G,
Respini
M,
Achieving
optimal
exopolysaccharides, organic and corrosion, and leads to deterioraThis successful
result
was
made
visbreaking
severity,
PTQ,
Q1, 2010.
The
main
ingredient
needed
to
consuming
option.
But
take
care
to tion
Theof
following
example
explains
cylinders
mean
maintethe
metal.
Biofilm
induces
inorganic
acids, reduced
and
volatile
possible
by
coupling
Baker
Hughes
2 A Phase
Separation
Kinetic
Model
for Coke
achieve
this
result
is
the
dedication
select
a
screwed
fitting
size
that
the
impact
on
the
compressor
nance
costs,
shorter
shutdown
compounds,
such as ammoniatreator many changes in the type and
VisTec
anticoke/antifoulant
Formation,
Preprints
ACS, on
Div.of
Petthe
Chem,
38,
and
the determination
field
will
not greatly
impact
the
vacuum
arrangement
based
the
chosen
periods
for
overhauls,
and
fewer
concentration
of ions,
pH
values,
hydrogen
sulphide,
can
alter
elec428-433,
1993.
ments
with
monitoring
technology,
troubleshooter.
tower parts
operating
pressure.
Also,
the capacity
control
system. potential.
A reciprospare
to be
kept in stock.
and
oxidation
reduction
trochemical
processes
at The
the
higher efficiency reduces electric cating compressor shall compress
power costs, and the high degree process gas in two compression

A controlled decrease
in wash oil can result
in better handling of
fouling, and at the
same time increase
HVGO directly

stages
into a affecting
process that
typically
Parameters
the developReferences
3
Respini,
Jones,
Spanu,
Sesselego,
Avoiding
foul
runs
at
70-90%
load
of
the
rated
ment of biofilms include:
1play,
Troubleshooting
Vacuum
Systems,
Wiley
Hydrocarbon
Engineering,
Nov 2006.
capacity.
When
considering
pneu
Temperature of the system or
Publications.
matic
suction
valve
unloaders and
ambient
temperature
2Matteo
Process
Equipment
Malfunctions,
Virzi
is Senior
TechnologyaMcGrawManager
spillback
valve
control,
fourHill.

Water
past
with
ISAB flow
Priolo rate
refinery
in the
Italy.surface
He is an
cylinder
compressor
with
two
3
Lines
J
R,
Understanding
ejector
systems
expert
Nutrient
availability
in distillation
and thermal conversion
cylinders
per
compression
stage
necessary
to
troubleshoot
vacuum
distillation,
processes.
SurfaceWith
of the
substratum
more
than 20 years of
Graham
Technology
Bulletin,
Batavia,
would
be
tooperations
enable
at

pH ofCorp.
water
in the system
experience
in required
technology,
and
NY,
US.
least
25%
load
increments
to
run,
he holds
degree in chemical
automation,
Effectiveness
of abiofouling
reme4engineering
Putman Rfrom
E,efficiently,
Steam
Surface at
Condensers,
relatively
loads
the University
of Palermo.
dial
measures.
ASME
Press,
2001. and 75%. However,
Email:
mvirzi@isab.com
between
70%
Biofilm formation
is for
thechemical
result of
5 Applied process
design76%
&
between
100%
and
load,
the
an
accumulation
process
not
Marco
Respini
is
a
Senior
Technology
Expert
petrochemical plants, Ejector & Vacuum
dispensable
process
gas
needs
with Baker
Hughes
Downstream
Chemicals,
necessarily
uniform
in still
time
or
Systems,
218-222.
to
be
recycled
through
the
spillspecialising
in
refinery
and
petrochemical
space that starts immediately
back
In thisofinworst
case,
up
processvalve.
improvements
fouling
control.
after
immersion
metal
in
the
He
has
15
years
of
refining
experience
and
to
24%
of
the
rated
compressor
Norman
Lieberman
is
a
chemical
engineer
aqueous environment. The growth
is currently
in Additionally,
developing
new
who
specialises
inconsidered
troubleshooting
power
is involved
of
biofilm
iswasted.
to refinery
be a
technologies fixed
forprocess
improving
refinery conversion
non-catalytic
equipment
and in
installed
clearance
pockets
result
of complex
processes
involvprocesses.
With design.
extensive
experience
in
retrofit
process
He
has
taught his
would
improve
the
situation
ing
transport
of
organic
and
asphaltene relatedSeminar
problems since
in oil production
Troubleshooting
1983
to
slightly,
but
the control
logic
inorganic
molecules
microbial
and refining,
he
isengineers
an inventorand
of fiveexperienced
US patents
over
18 become
000
and
would
more
complicated
cells
to published
the surface,
of
and has
technical
papersbooks,
and
plant
operators.
The10
first
ofadsorption
his eight
and
the
compressor
for
molecules
totothe
surface
and initial
seven access
conference
papers
onOperations,
visbreakers
and
Troubleshooting
Process
has
maintenance
impeded.
However,
heavycontinuously
fuel oil stability
problems.
graduate
attachment
of
microbial
cells
been
in print
for 35A years.
He
two
cylinders
per
stage
are
still
of
Milan
University
with
a
degree
industrial
followed
by their
irreversible
adhepreviously worked
for Amoco
Oilinin
Whiting,
chemistry,
has
been
Research
Fellowand
in
required
asTexas
compared
with 1980,
stepless
Indiana
andhe
City,
sion
facilitated
bya through
production
of
the field
ofwas
organometallic
catalysts
andGood
is a
control.
until
1983
Refinery
Manager
for
the
extracellular polymeric substances
registered
professional
chemist
in Italy.
He
is
Therefinery
same
with
a stepHope
Louisiana.
He
graduated
from
(EPS).
Onceinapplication
attached,
the
organisms
also
a
member
of
ACS
and
NACE.
Cooper
Union
in New
York
City in 1964.
less
full-range
flow
control
system
begin
to produce
material
termed
Email: norm@lieberman-eng.com
marco.respini@bakerhughes.com
Email:
would
require
only
one
cylinder
extracellular biopolymer, or slime
per compression stage (two cylinders in total for the compressor)

alves
Best V 67
8
since 1

38 PTQ
PTQ
Q2 2014
2014
www.eptq.com
98
Q3
www.eptq.com

hoerbiger.indd
5
baker hughes.indd
8
bpcl6.indd
2
liberman.indd 5

www.eptq.com
PTQ Q1 2014 131
www.eptq.com
PTQ Q4 2014 129

10/03/2014 13:40
13/12/2013
10/06/2014 11:38
14:26
10/09/2014 16:38

LEWIS PUMPS
Vertical Chemical Pumps

Weir Minerals Lewis Pumps is known internationally in the

sulphur, sulphuric acid and phosphoric acid industries with
equipment installed in more than 120 countries worldwide.
With new product innovations and a dedicated group of
employees, Weir Minerals Lewis Pumps is the recognized
world leader for pumps and valves in difficult applications.

Weir Minerals Lewis Pumps

8625 Grant Rd.
St. Louis, MO 63123

T: +1 314 843-4437
F: +1 314 843-7964
Email: sales.lewis@weirminerals.com

To learn more about Lewis Pumps

or Weir Minerals products, visit:
www.weirminerals.com/Lewis

Copyright 2014, EnviroTech Pumpsystems, Inc. All rights reserved. LEWIS and LEWIS PUMPS are trademarks and/or registered trademarks of Envirotech Pumpsystems, Inc.; WEIR is a trademark and/or registered trademark of Weir Engineering Services Ltd.

weir.indd 1

06/03/2014 13:53

Asphalt quality prediction and control

With knowledge of heavy vacuum gasoil cut-point and asphalt density, asphalt
quality can be inferred
ZAK FRIEDMAN
Petrocontrol

uch has been written about

crude column product separation by advanced process
control (APC) but very little about
vacuum column APC. Why?
Vacuum distillation unit separation
is important, especially in asphalt
mode, when vacuum pitch is sold
as a premium product, and
controlling asphalt quality is high
on the list of economic priorities.
One guess is that such APC applications have not been reported
because of inability to infer asphalt
qualities, and without such inference vacuum column APC would
not
be
effective.
Recently,
Petrocontrol had the opportunity to
set up asphalt quality inferential
models at two refineries; this article
describes the inferential techniques
and shares the inferential performance data.
Both refineries are located in
North America. Refinery A is a
land-locked refinery and throughput
is limited by asphalt sales. The
product grades vary from shingles
to road asphalts. Refinery B runs
mostly road asphalt by blocks and
only from certain crudes. In
summer, asphalt is the most lucrative refinery B product.
Asphalt quality is typically measured by dynamic shear rheometer
(DSR) apparatus, where the number
reported as G*/Sin(delta) is a measure of viscosity at a given
temperature. Our DSR inference
relies on the knowledge that asphalt
viscosity correlates with average
boiling point and density. The average asphalt boiling point is
estimated from column measurements, but for density we need the
aid of a density analyser. Without

www.eptq.com

petrocontrol.indd 1

To ejectors
LVGO
pumpdown

TC

FC

FC
LC
HVGO
pumpdown

LVGO
pumparound
FC
LVGO

FC

FC
LC

HVGO
pumparound
FC
HVGO

Furnace

TC

LC

FC
Vacuum
wax

Density
analyser
Asphalt

Figure 1 Vacuum column configuration

online density measurement, DSR

inference is not plausible.
Refinery A had installed a density
meter and developed asphalt DSR
inference several years ago but the
model was less than perfect. It
would predict adequately for a few
days, and then suddenly would
have to be biased substantially, even
when the crude had not changed.
That had caught us by surprise.
What was going on? Operation has
not changed much but the lab test is
suddenly showing a different value.
Upon studying the DSR test we realised it is carried out at several
defined temperatures of 46, 52, 58,
64 or 70C, the specific test temperature being a part of the asphalt
grade specifications. During asphalt
runs the schedulers switch grades
often; carried out at different
temperatures, the lab test would
yield different results. The operator,

not being aware of the change of

DSR test temperature, views the
sudden labinference discrepancy as
a sign of problematic inference and
turns off the APC.

The vacuum column configuration

Figure 1 shows a typical vacuum

column configuration. Reduced
crude feed comes from the atmospheric crude column through a
furnace and into the flash zone.
There are two distillate products:
light vacuum gasoil (LVGO) and
heavy vacuum gasoil (HVGO).
LVGO is diesel range material,
going into the diesel pool. HVGO is
FCC or hydrocracker feedstock.
There is a possibility to also draw
vacuum wax, though normally wax
is circulated back to the furnace to
improve the separation. As is
common
in
vacuum
column
designs, the draws are from total

PTQ Q4 2014 131

10/09/2014 16:45

500

25.0

490

22.5

480

20.0

470

17.5

460

15.0

450

12.5

440

10.0

430

7.5

420

5.0

410

2.5

400
12

12

20
1/

0/

/1
27

28

/1

/2
/9
28

20

01

01
/2
/8
29

/2
/7
30

2
01

01
/2
30

31

/6

/5

/2

01

API at 60F

VBAPI_M
APIVB_L

Density measurement temperature,

F

APIVB_A
TVDEN

27.5

occasionally bias the density reading

and/or DSR correlation.

Figure 2 Correction of asphalt density from instrument temperature to 15C

draw trays. Part of the draw is

pumped around through heat
exchangers to cool the column,
another part is pumped down as
reux, and a third part is taken out
as product.

Density measurement

As Figure 1 shows, due to the limits

of density meter temperature
asphalt is cooled to a temperature of
about 215-240C (420-470F) before
being measured. Density is a function of temperature and we need to
correct the raw density reading to a
standard temperature of 15C (60F)
before using it in a DSR correlation.
The laboratory also measures

Vacuum gasoil 98% point, F

1070
1060
1050

asphalt density at 15C, and this

permits comparison of the analyser
corrected density against lab values.
Figure 2 shows a six-months trend of
density related measurements. The
orange line is raw density as measured
online
(VBAPI_A).
The
magenta line is the density meter
temperature (TVDEN) on the right
hand scale. The blue line is our
correction of the density meter to
15C (VBAPI_M), and the green
squares are lab tests of asphalt
density (APIVB_L). The corrected
density meter reading more or less
trends with lab density though that
agreement is not perfect. One might
expect slow drifts and a need to

HVG98_M
HVG98_L

1040
1030
1020
1010
1000
990
980
970
960

28
/1
2/
20
11
16
/2
/2
01
2
6/
4/
20
12
26
/5
/2
01
2
15
/7
/2
01
2
3/
9/
20
12
23
/1
0/
20
12
12
/1
2/
20
12

950

Figure 3 Inference of VGO 98% point for one year compared against lab data at refinery A

132 PTQ Q4 2014

petrocontrol.indd 2

The GCC inferential package

Generalised cut-point calculations

(GCC) is a well established inferential
package
for
wide
cut
fractionators such as the crude and
vacuum distillation units. GCC uses
column measurements to rst identify the true boiling point curve for
the feed, and then it estimates product cut-points. Being a rst
principles based model, GCC has
performed better than other methods, and has in addition the ability
to infer cut-points during crude
switches. Several GCC related
papers have been published,1-7 and
it is not the object of this article to
cover additional GCC sites beyond
saying that HVGO cut-point is
predicted well. That is important for
the asphalt inference because HVGO
back end cut-point is identical to
asphalt front end cut-point. Figure 3
shows a one-year trend of HVGO
98%
point
inference
model
(HVG98_M) against lab tests
(HVG98_L). HVGO 98% is predicted
from HVGO cut-point, and the high
delity of this prediction indicates
that the HVGO cut-point is well
estimated. That trend is for renery
A. We cannot show the equivalent
renery B trend because renery B
does not test the HVGO distillation
curve.
Is the front end asphalt cut-point
good enough for this inference?
Viscosity is, after all, a function of
asphalt average boiling point. In
calculating asphalt average boiling
point our model assumes the asphalt
effective back end cut-point to be
800C. Given that asphalt effective
endpoint is indeed around 800C the
magnitude of crude to crude inferential variation is fairly small.

DSR modelling

There is a way to consider DSR lab

tests carried out not only at one
temperature in isolation but also at
other test temperatures. Given the
sensitivity of viscosity to temperature, there is a certain temperature
called TE, such that if the DSR test
were to be carried out at TE the test
result would have a value of
precisely 1.00 KPA. The DSR-

www.eptq.com

10/09/2014 16:45

0HJDDLUVHSDUDWLRQXQLWVIRUWKHZRUOGV
ODUJHVWJDVWROLTXLGV*7/SURMHFWLQ4DWDU
4DWDU6KHOO*7//WGDPHPEHURIWKH5R\DO'XWFK6KHOO
*URXSDQG4DWDU3HWUROHXPFRXQWRQ/LQGH(QJLQHHULQJV
H[SHUWLVHIRU3HDUO*7/LQ4DWDU
,Q/LQGHZDVDZDUGHGZLWKWKHWXUQNH\FRQWUDFWWR
EXLOGHLJKWODUJHDLUVHSDUDWLRQSODQWVIRUWKH3HDUO*7/SODQW
LQ5DV/DIIDQ,QGXVWULDO&LW\4DWDU:KHQ3HDUO*7/WKH
ZRUOGVODUJHVW*7/IDFLOLW\ZHQWRQVWUHDPLQ4DWDU
EHFDPHWKHJOREDO*7/FDSLWDO

7KHHLJKWDLUVHSDUDWLRQSODQWVSURGXFHFXELF
PHWUHVRIR[\JHQSHUKRXUZKLFKDUHXVHGWRFRQYHUW
EDUUHOVRIQDWXUDOJDVLQWROLTXLGK\GURFDUERQV$V
VXFK3HDUO*7/LVWKHODUJHVWLQWHJUDWHGFRPSOH[RILWVNLQG
LQWKHZRUOG
:LWKWKLVSURMHFW/LQGHGHPRQVWUDWHVRQFHDJDLQWKDWLWLV
UHVSRQGLQJWRWKHFKDOOHQJHRISURGXFLQJFOHDQHUIXHOVDQG
HQHUJ\WRPHHWLQFUHDVLQJZRUOGZLGHGHPDQG)XUWKHUPRUH
WKHFRPSDQ\FRQILUPVLWVOHDGLQJSRVLWLRQDVDWHFKQRORJ\
SURYLGHUDQG(3&FRQWUDFWRU

/LQGH$*
(QJLQHHULQJ'LYLVLRQ'U&DUOYRQ/LQGH6WUDVVH3XOODFK*HUPDQ\
3KRQH)D[LQIR#OLQGHOHFRPZZZOLQGHHQJLQHHULQJFRP

linde.indd
1
00094_ADV_Quatar_EN.indd
1

11/06/2014
11.06.14 16:06
15:49

75

Temperature, F

70
65
60
55
50
45
40

DSR_L
TE52_L
TE58_L

TE64_L
TE46_L
TEVB_M2
20 days

Figure 4 Refinery A inference of TE (temperature at which DSR would be 1.00 KPA) versus
lab results

temperature model we are using is:

(1)

Where Ttest is the test temperature

in K and G*/Sin(delta)test is the
DSR outcome of that test.
With knowledge of coefcient B,
Equation 1 permits calculation of TE
from a DSR test result at any
temperature. Renery A tests the
same asphalt sample at different
temperatures and for a given test
sample the TEs calculated from test
results at different temperatures
should be identical. Thus, even if
the viscosity-temperature relationship is not precisely known the
many lab tests at different temperatures give us the opportunity to
calibrate the temperature inuence.
Figure 4 tests this concept, covering 20 days of TE calculations. The

G*/sin at 58C
asphalt API

G/sin at 58_L
VBAPI_A

DSR inferential performance

Figure 5 illustrates renery As inferential model performance over the

same 20 days compared to lab data
at 58C test temperature. The blue
G*/sin at 58_M2
CPVGO

550

500

450

400

350

300

250

200

150

20 days

Figure 5 Refinery A inference of G*/Sin(delta) @ 58C

134 PTQ Q4 2014

petrocontrol.indd 3

100

VGO cutpoint, C

ln[G*/Sin(delta)test] = B * [1/TE 1/Ttest]

lab tested samples at 46 (TE46_L), 52

(TE52_L), 58 (TE58_L) and 64C
(TE64_L). The reported DSR results
are very different at each temperature because viscosity is quite
sensitive to temperature, but the
calculated TE values coincide within
half a degree. The fth green square
lab value in Figure 4 (DSR_L) is a
laboratory calculation of TE based
on interpolation among the several
test results. Finally, the blue trend of
Figure 4 is a TE inference as a function of VGO cut-point and asphalt
density readings. While not perfect,
this inference tracks the lab well and
can reliably be used for control.

line and orange squares of Figure 5

are trends of DSR inference and
DSR lab test, both at 58C. The blue
line inputs are shown in brown
and green. Brown is the asphalt
density measurement after temperature correction to 15C. Green is
the HVGO cut-point inferred by
GCC (right scale), C. API and
cut-point often trend as a mirror
image, where the gravity changes
as a result of cut-point change, but
the mirror image is not perfect
because gravity can also change
with crude type, and renery A
continuously changes the crude mix.
Figure 6 shows a typical renery B
ve-day trend. Referring to the right
hand scale, the orange diamonds are
DSR test temperatures, the red triangles are temperature TE, calculated
from the lab DSR test result data,
whereas the heavy purple line is our
TE inference. On the left hand scale,
green squares are DSR test results,
and nally the blue line is our DSR
inference as calculated from our TE
model and lab test temperature.

DSR control performance at

refinery B

The complexity of renery Bs

multi-variable predictive controller
(MVPC) is beyond the scope of this
article. Sufce to say that there are
many product quality inferences, as
well as many constraints, and the
MVPC manipulates both the atmospheric crude column and vacuum
column together. Changes in the
crude column affect asphalt quality
and to achieve the desired DSR both
atmospheric gasoil (AGO) and
HVGO draw must be manipulated.
AGO is the lowest draw of the
atmospheric column and changes of
AGO affect vacuum column feed
and asphalt quality. Figure 7 shows a
ve-day trend illustrating how DSR
control is accomplished. During
these ve days, several grade
switches took place. DSR inference is
shown in green, which compares
well against lab DSR test results in
red diamonds. Upon grade switches,
both the inference and lab result
jump, not because of any process
change but simply because of the
change in test temperature. The TE
value itself actually does not change
until column conditions change.

www.eptq.com

10/09/2014 16:45

2.60

2.20

www.eptq.com

petrocontrol.indd 4

55

2.00
1.80

45

1.60
1.40

35

1.20
1.00

25

0.80
12
16

/1

1/

20

12
15

/1

1/

20

12
14

/1

1/

20

12
20
13

/1

1/

20
1/
/1
12

11

/1

1/

20

12

15

12

0.60

Figure 6 Refinery B inference of TE and DSR

HVGO_PA
AGO DRAW

Y Zak Friedman is Principal Consultant with

Petrocontrol, based in New York City.
Email: zak@petrocontrol.com

DIESEL DRAW
DSR ESTIMATOR
2

Grade
3

DRS LAB

10

0.5

0.5

28
/8
//2
10 01
:5 3
9

1.0

27
/8
//2
10 01
:5 3
9

15

26
/8
//2
10 01
:5 3
9

1.5

25
/8
//2
10 01
:5 3
9

20

24
/8
//2
10 01
:5 3
9

MBPD

product properties, O&G Journal, 19 Feb 2001.

6 Friedman Y Z, Crude unit advanced control
experience, Hydrocarbon Processing Journal,
Feb 1994.
7 Friedman, Y Z, Control of crude fractionator
product qualities during feedstock changes
by use of a simplified heat balance, American
Control Conference, 1985.

DSR, kPA

3 at Sinopec Gaoqiao (Shanghai) refinery,

Refining China Conference, Apr 2006.
3 Adnan A, Md. Sani N, Seung-Yun N, Friedman
Y Z, The use of first-principles inference models
for crude switching control, ERTC computer
conference, May 2004, later published in PTQ,
Autumn 2004.
4 Singh P, et al, Multivariable controller
implementation for a crude unit: a case study,
NPRA Computer Conference, Oct 2002, later
published in O&G Journal, 4 Nov 2002.
5 Schuler M G, Kesler M G, Friedman Y Z,
Refinery uses column data to infer and control

23
/8
//2
10 01
:5 3
9

References
1 Ochoa Fuentes J, et al, Implementation
of APC on Repsol Puertollano CDU1, ERTC
computer conference, May 2007.
2 Gang-Yun Z, Zhi-Qiang Z, Friedman Y Z,
Implementation of APC on CDU 1 and CDU

65

2.40

Conclusions

We have demonstrated that with

knowledge of HVGO cut-point and
asphalt density one can infer asphalt
DSR. In our case, HVGO cut-point is
a GCC inference whereas the
density is measured by an on-stream
analyser. Our inferential procedure
involves the following steps:
1. Obtain an inferential estimate of
HVGO cut-point, a GCC model
calculation
2. Obtain a current asphalt density
reading and asphalt analyser
temperature
3. Correct the analyser density reading to 60F
4. Estimate TE, the temperature at
which asphalt DSR would be
precisely 1.00 KPA
5. Obtain the lab DSR test temperature appropriate for the asphalt
grade
6. Use equation 1 to convert TE
into G*/Sin(delta) at the test
temperature.
While this procedure is not trivial
we have shown that good inferential precision is achievable.
Further, we have integrated the
DSR inference into a large MVPC,
manipulating both the crude and
vacuum columns. In addition to
controlling all distillate properties
at targets, subject to equipment
constraints, we have also achieved
effective control of asphalt DSR.
Asphalt is one of the most protable renery products and its precise
quality control has a high value.

DSR_TEMP

DSR_M1
TE_lab

TE and DSR test temperature, C

DSR LAB
ASPH_TE_M1

DSR, kPA

Anyway, upon that inference jump

the APC controller reacts quickly,
changing the main manipulated variable: AGO draw (orange) and HVGO
hot reux (blue), bringing DSR to
target. Vacuum diesel draw is shown
in magenta. It often moves as a
mirror image of AGO, but that
mirror image is not perfect, indicating that other changes have occurred,
for example crude quality drift due
to a non-uniform crude tank.

Figure 7 Refinery B DSR control trend

PTQ Q4 2014 135

11/09/2014 11:02

Attendee
company
type*

3%
15%

3%
11%

3%
12%

3%
14%

35%

29%

32%

37%

51%

53%

51%

48%

2010

2011

2012

2013

*Average attendees: 500

Academic

Refiner

EPC

Supplier

Attendee job function

Attendee seniority

Product & Technology

Sales & Marketing

Engineering

Other

Project
Management

Head/Manager

Director

Board/
C-Level

General
Manager

Other

57%

29%

9%

5%

56%

16%

11%

5%

12%

Geographical region
UK

13%

61%

Still the best conference

to get a good overview of
new technologies, contact
international experts and
renew friendships

Europe

Senior Advisor, OMV

2%

12%

Americas

8%
Middle East / Africa

4%
Asia

Russia
/ CIS

For many years ERTC has

been the best event in
Europe, where you can be
informed about the latest
developments in the oil
industry, meet interesting
people and make contact
with service providers
General Technical Manager,
Hellenic Petroleum

All the major companies within the industry have sent representatives to ERTC including: Bapco BP CEPSA Chevron Engen ENI Essar
ExxonMobil Galp Energia Grupa Lotos GS Caltex Gunvor INA MOL Murco Neste OMV Petrobras Petroineos Phillips66 Reliance
Rompetrol Rosneft Repsol Saudi Aramco Shell Sinopec Slovnaft Statoil Total Tupras ...to name just a few.

To find out more and to book your place visit gtforum.com/ertc-annual-meeting

ertc.indd
1
ERTC Infographic_v5.indd
1

11/09/2014
14/08/2014 13:55
14:00

Technology in Action
Maximising aromatic assets in Oman

Centre Amsterdam in the Netherlands. These transalkylation catalysts exhibit high activity and efciency
for the treatment of heavy feedstocks.
Shell Global Solutions worked closely with the Orpic
operations team to help meet Sohars goals and says
that it was condent that ATA-21 would meet the challenge at Sohar based on continual optimisation and
conrmation of pilot plant performance. The catalyst
delivers very low aromatics loss and reduced hydrogen requirements, and offers high benzene purity. Low
hydrogen requirements help to reduce utility or investment costs, and the catalysts high stability means that

Oman Oil Reneries and Petroleum Industries

Company (Orpic) approached Zeolyst Specialty
Catalysts (ZSC)* for assistance in enhancing the performance of its transalkylation unit. ZSC simulated the
transalkylation process in a pilot plant test unit
running Orpic feed and found that switching to ZSCs
ATA-21 catalyst would help with further process
optimisation.
The transalkylation process involves transferring an
alkyl group from one organic
compound to another and enables
Reformer
Benzene
operators to upgrade low-value
streams (including heavy aromatics
Catalyst X produces
C9+ and toluene) to high-value feednon-aromatics in
Aromatic
Naphtha
C6A-C7A boiling range.
stocks for paraxylene and benzene.
extraction
Transforming low-value feedDue to non-aromatics
stocks into valuable products is a
make, benzene purity
Valve
is off-specification.
major goal of the petrochemical
open
industry. Plant operators aim to
1.5kbd of benzene with
impurities has to be
maximise the production and purity
Toluene
drawn to aromatic
of the desired products while reducextraction unit to
TA-stabiliser
ing their overall costs.
C9+
maintain benzene final
product purity.
In
most
industrial
catalytic
processes, catalyst performance optiAromatic extraction unit
TA reactor
Benzene
utilisation is limited.
misation is a critical focus area. The
99.9%
study facilitated the decision to
Reformer and crude
distillation unit rate is limited.
perform an early catalyst changeout
Separation
justied by the improved perforcolumns
mance when using ATA-21 catalyst.
Orpic planned a catalyst changeout
Figure 1 Transaklyation upgrade project: plant operation with original catalyst
for the end of Q1 2013.
The main driver for catalyst
replacement was to improve Orpics
Reformer
More
economic benet signicantly by
benzene
using the latest technology for transalkylation, the company said. The
Aromatic
More
Catalyst ATA formulation
extraction
naphtha
challenge was to select a catalyst
produces low
that would improve productivity
non-aromatics in C6-C7
range.
and help to debottleneck a specic
Valve
unit. Orpic chose ZSCs ATA-21 cataclosed
Benzene purity is
improved.
lyst
on
the
basis
of
a
techno-commercial evaluation of the
Toluene
Side draw in stabiliser is
reduced to zero.
TA-stabiliser
proposed yields and the projected
C9+
improvements of various catalysts
Improved aromatic
(see Figure 1).
extraction unit utilisation.
ATA reactor
ATA-21 is the latest in the ATA
Benzene
Debottleneck reformer
99.9%
and
crude
distillation
unit.
(advanced transalkylation) range of
high-performance
transalkylation
Separation
catalysts developed and commercialcolumns
ised as a result of the partnership
between ZSC, South Koreas SK
Innovation and Shell Technology Figure 2 Transaklyation upgrade project: plant operation with ATA catalyst

www.eptq.com

case studies copy 9.indd 1

PTQ Q4 2014 137

12/09/2014 13:07

138 PTQ Q4 2014

case studies copy 9.indd 2

Non-aromatics in side draw,

wt%

Flow and NAs, ton/hr

they were pleased with the

performance.
The ZSC catalyst helped to
increase conversion and yields of
3
Old stabiliser side draw
selected products (paraxylene and
Stabiliser side draw
benzene) and reduced by-products
Non-aromatics in stabiliser side draw
such as heavy aromatics and
2
low-cost blend material (see Figure
3). The quality of the benzene being
produced improved substantially,
1
which, in turn, resulted in debottlenecking of the extraction unit.
Furthermore, the cost of olefin
0
treatment came down because the
0
500
1000
1500
2000
2500
products from the new catalyst
Time on stream, hrs
system contain less unsaturated
Figure 3 Performance of ATA-21 transalkylation/dealkylation catalyst at Orpic: stabiliser materials. After a few months of
side draw data
operation following the introduction
of the new catalyst, Orpic says that it
a typical cycle can run from four to eight years, could safely conclude that the performance of the
Orpic aromatics complex had shown a marked
depending on the operating conditions.
Orpic placed the commercial order for ATA-21 in improvement and enabled the company to achieve
December 2012; catalyst delivery to the Oman site was new production records for benzene.
slated for February 2013. Manufacturing and deliver*Zeolyst Specialty Catalysts is a division of Zeolyst International, a
ing 60 tonnes of novel catalyst technology within such
partnership between PQ Corporation and the Shell Global Solutions
a tight schedule was a technical and logistical chalaffiliate CRI/Criterion.
lenge, but the collaboration of Shell Technology Centre
Amsterdam, ZSC and SK Innovation enabled quick For more information: Richard.mauer@shell.com
delivery of the first commercial catalyst samples to
Amsterdam for catalyst evaluation and qualification.
This was ZSCs first user of ATA-21 catalyst and its
Upgraded control for amine and sour
first transalkylation customer in the Middle East, so
water acid gases
there was no room for error. When the catalyst was
ready for shipment, extra measures were taken to meet
the start-up schedule. Logistics were put in place to In 2008 Jacobs introduced its Advanced Burner Control
ensure that the catalyst would be on-site for installa- Plus (ABC+) for SRUs fed with amine acid gas (AAG).
This upgraded control system is based on the existing
tion and a successful start-up.
Selecting the ZSC catalyst and having it delivered to ABC system that has proven its value in decades of
site on time enabled Orpic to replace the old catalyst as operation in sulphur recovery units worldwide.
planned and to gain the benefits associated with the Whereas ABC leans heavily on feedback from the tail
new catalyst at the earliest opportunity, says the gas analyser in case of acid gas composition changes,
the ABC+ system adds a feed gas analyser to anticipate
Omani company.
There is often a stabilisation period following the changes in AAG composition. The readings from the
introduction of a new transalkylation catalyst system. AAG analyser are used by the ABC+ PLC to calculate
Typically, the high activity of the catalyst results in the air to AAG ratio as well as the AAG mole weight.
side reactions and affects the high benzene purity In a clients DCS these online variables are used to
target. For the first 12 hours at Sohar, process adjust- significantly improve the feed forward control
ments were implemented to achieve the high benzene response of ABC, thus forming ABC+. Jacobs has taken
purity requirement with the inclusion of a stabiliser the next step by including sour water acid gas (SWAG)
side stream. Unlike some other catalysts on the market, control into ABC+.
SWAG mainly differs from AAG because it contains
within a few days, the performance of ZSCs ATA-21
catalyst stabilised and the operator was able to shut a significant amount of ammonia. Furthermore the
down the stabiliser side stream. In addition, the oper- hydrogen sulphide as well as the carbon dioxide
ating temperature was 8090C lower than that content is much lower and the water content is much
achieved with the historical catalyst charge. This start- higher than in AAG. Last but not least, the hydrocarof-run temperature reduction with ATA-21 helped to bon content is often higher; it contains more
improve the energy efficiency of the aromatics unit hydrocarbon species which tend to fluctuate more than
in amine acid gas. Ammonia can be rapidly measured
further (see Figures 1 and 2).
During the first few weeks of operation, ZSC by the combined UV/IR optical cell that is part of the
received regular updates from Sohar. Two weeks into ABC+ feed forward analyser. Furthermore, the SWAG
the production cycle, the plant operators reported that analyser can be set up to also measure already familiar
12
11
10
9
8
7
6
5
4
3
2
1
0

www.eptq.com

12/09/2014 13:07

species from the AAG analyser Apart from SWAG

being different in composition compared to AAG, the
data received by the ABC+ PLC is processed in the
same way. In this way, the SWAG feed forward control
response can be significantly improved, providing the
same analytically driven control concept as already
available for AAG.
ABC+ has been shown to provide SRU control
robustness, enabling processing of varying quality sour
gas from refineries as well as gas fields. It is suited to
handling feeds that vary in flow rate, pressure,
temperature and composition. Often these variations
result from upstream operating complexities. By offering an effective feed forward control, the problems of
increased SO2 emissions, potential production losses
and regulator non-compliance can be avoided.
Considering sour gas composition changes, the usually
strong dependence of ABC+ on the tail gas analyser is
relieved somewhat since the focus now shifts more
towards the feed gas analysers.
The first ABC+ system was installed at Keyera
Simonette gas plant in Canada in 2008 and this set-up
has been running successfully ever since. Thanks to the
ABC+ system, the number of upsets as well as the
quantity of flared acid gas was significantly reduced.
Shortly after start-up, Keyera, then still belonging to
Suncor, reported an increase in capacity of about 10%
which directly increased revenues.
Since then, an improved ABC+ hardware concept has

been developed together with partner Analytical

Solutions and Products (ASaP, Amsterdam). The first
improved system, AAG ABC+, has been running
successfully since early 2012 at Valero Pembroke refinery in Wales, UK. At the same refinery, SWAG
ABC+ has been started up in July 2013 and, after some
extra commissioning attendance, also runs satisfactorily. This was merely due to some tuning in the
analytical set-up of the SWAG analyser, which could
be done online. Some aspects during the start of the
project had to be assumed, which could later be rectified with the aid of the vendor during actual
operation.
After experience at Valero Pembroke, very similar
AAG and SWAG ABC+ systems have been started up
in 2014 at Esso Augusta (Italy) as well as CPC Talin
(Taiwan), while a similar project at Takreer (UAE) is
halfway in this process. All locations report stable
operation leading to good unit performances.
In the case of the CPC Talin installation, low SO2
emission from the SRU is secured by application of a
tail gas scrubber. The primary benefit of the ABC+
control in this case is the prevention of bypass operations and trips, also when the feed gas composition
and/or quantity changes in AAG and/or SWAG.
Because of this, the use of caustic in the scrubber can
be kept low, which translates into lower operating
costs and a smaller bleed stream.
For more information: Jenny.deWolf-Rikkers@jacobs.com

POLYSTINGER
LED HAZ-LO
130 LUMENS
15,000 CANDELA

KNUCKLEHEAD
HAZ-LO
163 LUMENS
730 CANDELA

3AA
PROPOLYMER
HAZ-LO
120 LUMENS
14,000 CANDELA

When you have a job to do,

youre not thinking about
your flashlights engineer or
the design, functionality or
keen ingenuity that offers
a premium product for the
GREATEST VALUE.
In fact, you dont have to think
about your flashlight at all.
Because its a STREAMLIGHT.

3AA HAZ-LO
HEADLAMP

2014 STREAMLIGHT, INC.

120 LUMENS
6,800 CANDELA

www.eptq.com

case studies copy 9.indd 3

CONNECT WITH US WWW.STREAMLIGHT.COM

PTQ Q4 2014 139

12/09/2014 13:07

Online cleaning and decontamination

of a crude unit
A refinery wanted to validate ITWs Online Cleaning
and Improved Degassing/Decontamination technologies to use them strategically for future turnarounds/
slowdowns and during the run.
Online Cleaning was carried out on the entire crude
distillation unit (CDU, which has two preheat trains)
and took 24 hours. Subsequent to that, as the refiner
wanted to further recover downtime (and reduce steamout requirements), ITW applied its Improved
Degassing/Decontamination technology, which took 24
hours. By manways opening, the following results had
been achieved: LEL = 0%; HC = 0 ppm; H2S = 0 ppm;
benzene = 0 ppm; ammonia = 0 ppm; pyrophorics = nil.
Moreover, for the first time in the units history, there
was no need for columns wetting as pyrophorics were
completely eliminated. All of the maintenance operations were conducted under dry conditions.
Visual inspection of the bundles confirmed the
results of Online Cleaning. The refinery decided therefore not to extract some bundles during turnaround,
but to check their performance at unit start-up.
At CDU start-up, the results of Online Cleaning were
comparable to mechanical cleaning. The outlet
temperatures of the trains were the same and the fouling factor of the H.E. cleaned with ITW were
comparable to the mechanically cleaned ones. The
results of Online Cleaning technology exceeded the
refiners expectations, says ITW.
No waste was generated. All of the washing fluids
were sent to a slop oil tank and fully reprocessed at unit
start-up, with no problems at all during reprocessing.
For improved degassing/decontamination, the
condensate was routed directly to the oily sewage.
Again, no problems were recorded at the API separator/wastewater treatment plant.
Given the successful application, the refinery will
consider ITW Technologies applications as follows:
Avoid CDU slowdown to clean the preheat trains
(one day ITW Online Cleaning vs about 25-30 days at
half capacity for slowdown)
Regular application on the visbreaker to increase run
length
Rescheduling of future downtime during turn-

arounds by taking into consideration both the reduced

time for safe entry and the reduced need for cleaning
exchangers.
ITW technology can be used to clean storage tanks
on a closed-loop basis, with no operator entry.
For more information: mferrara@itwtechnologies.com

Alternative to stick-built sulphur plants

Prosernat, acting both as process licensor of a portfolio
of sulphur recovery technologies (AdvaSulph for
sulphur recoveries up to 99.9%+) and as specialist of
modular treatment units, has delivered modular
sulphur units which meet a variety of project
constraints. In general, the project execution scheme
where the same company carries out simultaneously
the process design and the modular fabrication and
delivery offers significant cost/delay savings compared
to a traditional execution scheme where basic design/
FEED is followed by EPC tendering with subsequent
stick-built fabrication. The modular design and supply
scheme also helps to solve specific site constraints such
as congested areas, ongoing operations, lack of
manpower or harsh climates. The following two projects illustrate those benefits of modular supply for
sulphur recovery units.
Prosernat delivered modular tail gas treatment to
Ecopetrols Barrancabermeja refinery in Colombia for
start-up in 2010. The design uses Prosernats proprietary
process Clauspol designed to achieve 99.7% sulphur
recovery downstream of a 110 t/d Claus unit. The unit
comprises a tail gas absorber (delivered loose) and
smaller process equipment items delivered totally skid
mounted in order to meet the refinerys plot space limitations and also to speed up the installation of the unit.
A modular gas sweetening and sulphur recovery
unit delivered by Prosernat to Total EPs Kharyaga site
in Russia consists of two main units:
A gas sweetening unit (GSU) to remove the H2S from
associated gas, designed with AdvAmine MDEAmax,
a proprietary selective MDEA process developed by
Total and licensed by Prosernat
A Claus unit with sulphur degassing (SRU) with a
sulphur production capacity of about 16 t/d.
The Kharyaga project had to cope with two main
constraints:

Figure 1 Modular gas sweetening and sulphur recovery units at Total EPs Kharyaga site

140 PTQ Q4 2014

case studies copy 9.indd 4

www.eptq.com

10/09/2014 17:05

 The field is located beyond the arctic circle and the

severe climatic conditions do not favour site works
Transportation from the nearest sea port to the site
had to be made by road and train which limited the
size and weight of each single transported element to
3400 mm width x 3460 mm height x 18 000 mm length
and 37 tonnes weight.
In order to meet these constraints, both the GSU and
the SRU were made of an assembly of several smaller
skids, each single skid being designed to meet the
limitations of transportation. The SRU is composed of
38 main equipment items installed in 11 skids, representing a total weight of 300 tonnes. The GSU is
composed of 43 main equipment items installed in 12
skids representing a total weight of 330 tonnes.
The whole gas treating plant is composed of 23 skids
and weighs 630 tonnes (see Figure 1).
Modular design is of course applicable to complete
sulphur blocks. At the end of 2013 Axion Energy
placed an order with Prosernat for its Campana refinery in Argentina to design and deliver a fully
modularised sulphur plant comprising: complete
amine fuel gases treatment system; double stage sour
water stripper; Claus plant; tail gas treatment
(with Prosernats licensed Sultimate process); and
incinerator and sulphur degassing. This complete
sulphur block is designed to produce 30 t/d with
sulphur recovery of 99.9% and it will be made of 24
different skids, each weighing between 50 tonnes and
100 tonnes.

Feed to the colum

Is your plant
ready for the future?

The feed to the

other units. The
11th tray by mea
tant source of di

Top of the colum

Reflux drum p
controller. This
columns operat
pressure change
amount of feed
almost no reflux
amount of reflu
valve which is ca
ler (LC). The t
controlled with
ture indicator. In
top temperature
with a reflux dr
this study is fo
developing the s

Separation Index

Reboiler duty or
the
column
liquid rate. The
sharpness of sep

As the dru
the gas pr
heavy end

For more information: pweber@prosernat.com

Hydroprocessing internals upgrade

Effective cut poin

Chevron Lummus Global (CLG) worked with a private
refiner to upgrade the reactor internals in a hydrocracking unit to improve the safety, operability and
profitability for this unit. This refiner was perpetually
struggling with high catalyst bed outlet temperature
spreads, severe cold and hot spots, poor quench gas
distribution, poor catalyst bed temperature control and
low volume expansion. The quench gas distribution
was so poor that lower bed hot spots could only be
indirectly controlled by a combination of moves
between the upper bed quench valves. This raised
significant operating safety and emergency response
concerns.
CLG reviewed the existing reactor internals design
and performance and determined that the temperature
control issues were due to inadequate mixing between
cold quench gas and hot process gas, compounded by
limited mixing and temperature equilibration between
the gas and liquid phases. CLG also compared plant
yields, chemical hydrogen consumption and volume
expansion to pilot plant results for the same feed and
catalyst system and determined that plant results were
below expectation. Further, this performance gap was
due to poor gas and liquid distribution to the catalyst
beds resulting in severe gas and liquid channelling and
overall low catalyst utilisation. The company

www.eptq.com

case studies copy 9.indd 5

Life-cycle optimalization

Proposed approa
Making critical plant information fully visible is just the
beginning of the vigilant cycle. Seeing clearly gives
you the knowledge necessary to anticipate the changes
required in your process. Knowing in advance brings you
the speed and flexibility to optimize your plant in real time.
And by acting with agility, you are able to adapt to the
ups and downs of your business environment. VigilantPlant
excels at bringing out the best in your plant and your
people - keeping them fully aware, well informed, and
ready to face the next challenge.

Please visit us at www.yokogawa.com/eu

or contact us at info@nl.yokogawa.com

Control_86x270_2014.indd
32 PTQ Q3 2014 1

tupras.indd 2

This feature is m
column. The m
concerned with
pressure decreas
ends and gas flo

In this section
explained in det
mended approa
similar processes

Base layer enhan

Here, the contro

fix the main o
solved and app
ters (Proportion
decided. In add
taking into accou
column.

Data collection

The data for thi

month before its
The control e
these control ele

06/03/14 11:36

PTQ Q4 2014 141

11/09/2014 14:34

Isomix internals installed

Radial temperature
spread (T), C

temperature sensing points (inlet

and outlet), with no hot or cold
Inlet radial T
spots even at high quench gas rates
50
Outlet radial T
Excellent catalyst bed temperature
40
prole control, with the ability to
30
operate completely at (equal average bed outlet temperatures), which
20
is desired to maximise product
10
selectivity and cycle length
0
Reduced catalyst bed radial
600
0
200
400
800
1000 1200 1400 1600 1800
temperature spreads to 1C maxiCalendar days from arbitrary start date
mum at the inlet and outlet of all
beds, even at high catalyst bed axial
Figure 1 Bed 3 radial temperature spread (T) before and after installation of Isomix internals temperature rise
Increased
volume
expansion
determined that the gas and liquid mixing and distri- including a 6% increase in recoverable middle distillate
bution issues could be resolved, and temperature (jet/diesel)
control and volume expansion improved by replacing Increased cycle length by 12 months, from three
years to four years (estimated after three-plus years
the reactor internals.
CLG proposed upgrading to its latest generation on-stream)
Figure 1 shows the catalyst bed inlet and outlet radial
Isomix internals which are comprised of two primary
process components: a sealed collector tray and central temperature spread over time for Bed 3. This bed has
mixing device; and a distributor tray based on high the highest axial temperature rise and quench gas ow
efciency ow nozzles, a new concept in multi-phase rate, and showed the worst radial temperature spread
and quench gas maldistribution prior to the upgrade.
ow distribution.
Isomix internals have signicant advantages over The improved performance is representative of all beds
prior generations. For instance, the mixing box following similar installations. Control console data are
achieves intimate gas-liquid mixing, promotes heat shown in Figure 1. Point by point eld surveys
transfer between the two phases, and provides stable conducted at start of run conrmed that the actual
performance over a broad range of gas and liquid ow radial temperature difference for this bed and all other
rates. The distributor tray ow nozzles promote high beds was 1C maximum spread at all levels.
catalyst utilisation by providing uniform gas-liquid ISOMIX is a mark of Chevron.
distribution with high liquid dispersion and superior For more information: BSRI@chevron.com
60

The distributor tray flow nozzles

promote high catalyst utilisation
by providing uniform gas-liquid
distribution
surface coverage. They also provide intimate mixing
and heat transfer between gas and liquid phases which
supplements mixing box performance. Flow nozzles
are tolerant to distributor tray out of levelness and
agitation of the liquid surface, so performance does not
degrade if solids accumulate or tray segments warp
over time.
This rener has historically taken a conservative
approach to the application of new technology. It was
important that CLG could demonstrate prior experience with similar installations. Isomix internals were
developed circa 2003, commercially proven in Chevron
units from 2003-2008 and commercially proven in
licensed units since 2008.
The unit performance post-upgrade to Isomix internals has exceeded expectations for this rener. The
benets achieved are:
Direct temperature control of all catalyst bed

142 PTQ Q4 2014

case studies copy 9.indd 6

New sour gas shift approach for coal

gasification
Recent cooperation between Clariant and Siemens Fuel
Gasication has resulted in a new sour gas shift (SGS)
process and catalyst for coal to chemical applications,
which present signicant end user advantages in terms
of cost, efciency, safety, and feedstock exibility.
One of the main challenges in SGS technology is to
avoid temperature run-away along the reactors. This is
especially true for dry feed gasication technologies
with carbon monoxide (CO) concentrations of more
than 60 vol% (dry) in the syngas. Traditionally, there
are two techniques used to avoid this. The rst is the
addition of excess steam to raise the steam to gas ratio
in the syngas. The second is partial condensation of
water to lower the steam to gas ratio. In either method,
adjusting the steam concentration in the gas reduces
the equilibrium temperature of the shift reaction,
thereby avoiding undesired and highly exothermic
methanation as a side reaction.
The new SGS technology developed by Clariant and
Siemens Fuel Gasication uses a completely different,
more efcient approach to avoid temperature
run-away reactions; the temperature is controlled by a

www.eptq.com

10/09/2014 17:05

DigitalRefining.com is already the most

extensive source of freely available
information on all aspects of the
refining, gas and petrochemical
processing industries, providing a

constantly growing database of

technical articles, company literature,
product brochures, videos, industry
news, industry events and company
information.

Visit DigitalRefining.com to see what content is relevant to you

dig ref.indd 1

15/09/2014 12:08

catalyst. Based on Clariants ShiftMax 821 catalyst, the

new technique enables a simple, once-through process,
which requires no further adjustment of the exit gas
from the gasifier for temperature control. The simplified layout also uses smaller and fewer reactors.
Furthermore, the system offers greater feedstock flexibility due to its entrained flow Siemens Fuel Gasifier,
which is able to produce syngas from a wide range of
fuels, even for low ranks of coal.
The new technique can handle various steam to gas
ratios as well as high carbon monoxide content in the
gas. This results in improved availability and reliability
in the whole process. Because of steam independent
control of the exothermal reaction, the process is inherently safe.
Through optimisation and simplification of total
plant concepts, the new SGS technology significantly
decreases total capital cost for coal to chemical and
IGCC (integrated gasification combined cycle) applications. Compared to conventional sour gas processing,
the new solution reduces capital expenditure for the
costs with up to 30% lower catalyst volume. Hence, it
offers substantial financial benefits, particularly for
large scale applications.
The ShiftMax 821 catalyst has undergone testing at
Clariants R&D laboratories in Louisville, Kentucky,
and its performance has been verified in a large scale
coal to methanol unit. After three years of operation,
the unit still demonstrates very high performance.
In addition to joint development of the new SGS
technology, the agreement between the two companies
appoints Clariant as the exclusive catalyst supplier for
all Siemens gasification integrated SGS projects. The
collaboration covers all global projects. However,
commercialisation will focus on China the region
with the highest growth rate of coal to chemical projects. The partners first large scale plant is expected to
go online in 2016.
For more information: HansGeorg.Anfang@clariant.com

Ceramic extends air distributor life

By switching from stainless steel to a ceramic construction based on refractory grade alumina bonded silicon
carbide for its FCC air distributor nozzles, a refiner
expects a significantly extended lifetime for the distributor nozzles in severe operating conditions.
Within a FCC unit, there are a number of different
types of air/steam rings. In each of these there are
small nozzles that distribute air over the grid, which
creates the fluidised bed. These FCC air distributor
nozzles are subjected to high temperature, high velocity, erosive environments. To function in this harsh
environment, FCC air distributor nozzles have historically been fabricated from a variety of erosion-resistant
metallic materials. While these materials are proven
effective at reducing rates of erosion, by their nature,
they do not perform well in sliding abrasion or at
elevated temperatures. As such, they pale in comparison to most ceramic bodies. Ceramic materials are

144 PTQ Q4 2014

case studies copy 9.indd 7

widely accepted and proven to be more resistant to

erosion than metallic materials at acute angles of
impingement. The characteristics that impart erosion
resistance also tend to make these materials suited for
nozzle applications that rely on consistent orifice diameters for performance and reliability.
In an FCC air grid, the nozzle diameters set the
velocity of air exiting the nozzle. Designers engineer
specific orifice diameters that minimise catalyst attrition while preventing catalyst from entering the
nozzle. If the orifice in the nozzle erodes and becomes
larger than designed, a scenario could develop where
the pressure drop decreases, causing catalyst fines to
re-enter the nozzle leading to accelerated wear and
ultimately grid failure. Nozzle length is also a critical
dimension in nozzle design, and it is set based on the
nozzle diameter. It is important for the nozzle to be
long enough to allow the air/steam passing through
the restriction orifice to spread across the entire grid. If
the tip of the nozzle erodes, this length is sacrificed
and air distribution across the bed is diminished. This
can ultimately lead to reduced catalyst recovery.

It is important that refineries

step away from traditional
metallic solutions and
embrace ceramics
Over the years, many metallic and ceramic materials
have been tried in grid nozzles, all with varying
results. The life of these various materials has been
quite inconsistent, from six months up to 20 years. As
turnarounds grow further apart and the need to stay in
operation becomes ever more vital, the median life of
the grid nozzles must be extended. It is therefore
important that refineries step away from traditional
metallic solutions and embrace ceramics.
One such refiner did that recently. Based on 10+
years of life that a sister plant experienced with
ceramic air grid nozzles, the refiner converted its
current 304H stainless steel nozzles over to ceramic, in
this case Blaschs Altron.
Altron is a refractory grade alumina bonded silicon
carbide that exhibits the wear resistance generally seen
in fully dense technical ceramics, which can be subject
to catastrophic failure from thermal shock due to
sudden swings in temperature. This material can be
cast into precise, intricate nozzle shapes that require no
finish machining. Coupled with an engineered welded
assembly, the ceramic/metal nozzle can be welded to
the grid with ease. Thermal expansion allowance, sealing and tolerance control have all been thought out
with each nozzle design. Once installed, the Altron
material can take the full brunt of the erosive wear and
high temperatures. Orifice diameters and nozzle length
maintain their geometry throughout the life of the FCC
campaign giving refiners the ability to hit their financial performance objectives.

www.eptq.com

15/09/2014 12:14

AFPM 2015
NAVIGATE NEW TRENDS
ATTEND AFPM MEETINGS
The global energy landscape is rapidly changing. Growth in production is leading
to changes in technology, planning and management at all levels. Gain a competitive
advantage. Come to 2015 AFPM Meetings to hear about the latest developments
impacting our industry.
Annual Meeting
San Antonio, TX
March 22 24
International
Petrochemical
Conference
San Antonio, TX
March 29 31
Security Conference
New Orleans, LA
April 13 15

afpm.indd
1
AFPM_GeneralMeetingsAd_PTQ_PRINTER.indd
1

Labor Relations/
Human Resources
Conference
New Orleans, LA
April 16 17
National Occupational
& Process Safety
Conference and
Exhibition
Austin, TX
May 12 13

Reliability &
Maintenance
Conference
and Exhibition
Austin, TX
May 19 22
Board of
Directors Meeting
(Registration by
invitation only)
Park City, UT
September 20 22

Q&A and
Technology Forum
New Orleans, LA
October 4 7
Environmental
Conference
Salt Lake City, UT
October 18 20
International
Lubricants &
Waxes Conference
Houston, TX
November 12 13

10/09/2014
11:22
7/18/14 10:53
AM

Once installed, Blaschs Altron FCC air distributor

nozzles will far exceed the turnaround time on the cat
cracker which will allow the reliability engineers to
focus on other areas of the plant.

For more information: LGicewicz@blaschceramics.com

Active steam trap management

An oil renery in Texas was experiencing a high failure
rate in its 4790-strong network of steam traps. The
rener was also experiencing water hammer problems
caused by the accumulation of condensate and freezing
sulphur lines, caused by ooding in the tracing system.
Eventually, losses were so high that the plant management team called upon Spirax Sarco to help them
reduce their costs.
A thorough steam trap audit carried out by our
Spirax Sarcos Energy Services Group revealed several
problem areas:
Improperly installed and leaking steam traps
Incorrectly specied traps, unsuitable for their
applications
Freezing sulphur product caused by poor drainage
of process condensate
Water hammer
Blowdown wastage (a potentially valuable source of
recoverable heat)
Steam turbine inefciencies.
By replacing failed steam traps and putting in place
a steam trap management programme, Spirax Sarco
resolved most of the problems experienced by the
customer, improving process efciency and reducing
energy losses.
According to the site operations manager, the audit
programme has been in effect for several years and
has resulted in signicant steam savings. In
addition, Spirax Sarco has carried out a detailed
steam leak survey and an air/nitrogen leak survey,
identifying additional annual savings of $160 000 and
$460 000 respectively and has managed the steam trap
population, leaving the customer to focus on the
process.
Key benets of the project include:
Savings of almost $800 000 in the rst year
alone after implementing a steam trap audit programme.
Within ve years this gure had reached $4.7 million
Resolved the problem of freezing sulphur lines by
replacing incorrectly specied steam traps with
versions more suitable for the application
Identied potential annual savings of over $900 000,
by recommending the replacement of the damaged
boiler blowdown heat recovery system (which was
dumping valuable, hot blowdown to the sewer) with a
packaged heat recovery system
Reduced the annual steam trap failure rate from
5.5% to 1.5% within ve years
Eliminated water hammer problems caused by a
leaking high pressure steam trap and by identifying
missing or cold (malfunctioning) steam traps.
For more information: Greg.Ferkin@uk.spiraxsarco.com

146 PTQ Q4 2014

case studies copy 9.indd 8

Optimised loading of naphtha

hydrotreaters
A US rener asked Criterion to help reduce the
frequency of catalyst changeout s and reduce overall
cost without sacricing unit throughput on their two
naphtha hydrotreating units (NHT). The unit cleans up
coker naphtha feed for the reforming unit so the ability
to stretch the catalyst cycle life beyond the current
eight months per reactor has a direct impact on protability. As the incumbent catalyst supplier for the unit,
Criterion focused on quickly reviewing the operating
data and current set-up.
Each hydrotreating unit has two single-bed reactors
in series. The unit conguration provides the exibility
to change which reactor is in the lead position, leaving
the alternate reactor as the lag. This exibility also
allows the unit to isolate a reactor while the other
remains in service as needed. The primary method of
deactivation for these reactors is silicon poisoning,
which is very common in this service.
The loading prole prior to Criterions optimisation
(see Figure 1) was to load each reactor in series with
silica trap catalyst (MaxTrap [Si] in this case) followed
by an active denitrication catalyst to balance activity
with poison control, thus setting the cycle life. After
about eight months, the lead reactor would be deactivated from poisoning. The operators would then
exchange the lead-lag positions of the reactors to facilitate a reload of the spent reactor. Operation would
continue in this manner until the reactor deactivated
on poisoning cycling lead/lag reactor positions again,
completing a full reload in a 16-month time frame.
Although this scenario provided quality naphtha feed,
the duration was such that the expense was not falling
into the capital spend 100% of the time. Additionally,
the operating scheme created a percentage of unit
operation loss that categorised the unit out of the top
quartile bracket.
Criterion evaluated several catalyst loading options
and concluded that obtaining the maximum cycle life
required the two reactors to be operated in series with
Lead
reactor

Lag
reactor

Trap Si
MaxTrap(Si)

Trap Si
MaxTrap(Si)

Si removal/
HDS/HDN

Si removal/
HDS/HDN

HDS/HDN

HDS/HDN

Figure 1 Current - load both reactors and alternate lead/lag

www.eptq.com

10/09/2014 17:06

epc.indd 1

10/09/2014 12:06

Lead
reactor

Lag
reactor

Si removal/
HDS/HDN
Trap Si
and protect
activity

HDS/HDN

Figure 2 Optimised - load lead reactor with 100% MaxTrap and

lag with HDS/HDN catalyst. Do not alternate lead/lag

an optimised loading scheme to facilitate longer cycles

and fewer changes. Conguration changes meant that
the lead reactor had a primary function of poison
control, leaving the lag reactor as the activity reactor for
denitrication. Compared to the existing loading
scheme, the new loading scheme provided substantially
more poison control catalyst in the lead reactor. With the
unit run length controlled by feed silicon poisoning,
using Criterions MaxTrap[Si] catalyst extended the
cycle life beyond 12 months. While the MaxTrap[Si] lead

3-6 November 2014, Marriott Rive Gauche, Paris, France

Sulphur2014
Sulphur
Conference

12TH WORLD

Oil | Gas | Fertilizers | Metallurgy | Industrial uses

The worlds leading industry event

See you in Paris to celebrate

our 30th Anniversary!

reactor is changed out, the conguration allows the lag

reactor to remain in service. Once reloaded, the reactor
repositions itself in the lead and operation continues.
In this new scenario, the lag reactor does not reach
maximum capacity on Si poison for several lead cycles.
The timing is synchronised to allow both reactors to be
shut down in sequence and the lag reactor catalyst is
changed along with the lead on the third or fourth
change, depending on poison levels in the feed slate.
When this occurs, the scheme provides again for uninterrupted operation while both reactors are reloaded.
The lead reactor containing MaxTrap[Si] is unloaded
rst and reloaded as the lag reactor. This conguration
consists of Criterion DN-140 followed by Criterions
Ascent DN-3531 while the current lag reactor remains
in service (see Figure 2). Once the lead reactor is
reloaded, it is then switched into service while the lag
reactor is reloaded as the lead reactor with
MaxTrap[Si] catalyst. When loaded, this new lead reactor is put into service to restart the cycle scheme. This
loading methodology ensures ow to the unit is not
interrupted and product quality is not compromised.
Utilising the new loading scheme, the rener benets by achieving less frequent catalyst changeout s and
lower ll costs. The lead reactor is expected to achieve
a cycle length of 12 months, while the lag reactor is
expected to achieve a cycle length of 36 months. The
majority of the catalyst purchased and loaded is
MaxTrap[Si] in the lead reactor, which has saved in ll

50%
Operator/
Producer
discount

Network with 550+ industry professionals

Set up meetings in advance using
the CRU app
40 technical papers
4 practical workshops for operators
CRU global updates on sulphur &
sulphuric acid markets
Site visit to Total Normandy Refinery

Register today at www.sulphurconference.com

Gold Sponsors:

148 PTQ Q4 2014

case studies copy 9.indd 9

Silver Sponsors:

www.eptq.com

10/09/2014 17:06

n
de
uam
rit,

costs while protecting higher activity catalysts and

prolonging the life of the lag reactor. When amortising
the catalyst changeout frequency and cost over a
10-year period, this loading methodology has resulted
in a reduction of about 16% in catalyst costs and a
reduction of about 11% in catalyst changeout
frequency. For the refiner described in this article, this
strategy specifically resulted in approximately $400 000
savings in catalyst cost per year and one less catalyst
changeout every five years. This also moved all of the
catalysts costs away from operating expense to more
favourable capital expense.
For more information: teresa.brod@cri-criterion.com

Valve control for reciprocating

compressors

Control
pressure
port

Pocket
Connection
bore

Cylinder
Valve elements

Figure 1 Design of the clearance pocket

manually operated piston for setting the clearance

pocket.
If this variable pocket solution is applied, it is possible to insert a mechanical drive and to connect it to a
process management system. However, such an
approach would be highly complex and could only be
implemented with great effort, says NEA.
The new method uses pressure control to set the
valve switching point. Pressure control enables the
activation of the clearance pocket at lower or higher
pressure. In both cases, the switching pressure of the
clearance pocket valve is depending on the control
chamber pressure or rather the control pressure.
Control pressure is the actuating variable for adjusting reciprocating compressor mass flow. Hence, control
pressure must be adjusted in order to change the operational point. If the operational point is to be
maintained, control pressure must be kept at a specified level. Adjustment or maintenance of control
pressure is effortless with process gas as the control
medium. The control chamber connects to the cylinders suction side or control side using solenoid
valves, depending on whether control pressure is to be
reduced, maintained or increased.
Standard approaches for a pressure monitoring system for the
Discharge
Maximum
control chamber include pneumatic
pressure
impact
or solenoid valves, a pressure sensor
(opening)
and a control circuit. The clearance
Suction
pocket valve is a novel development
pressure
whose requirements are basically
equivalent to those of normal
0
compressor valves:
Transient pressure
Fast operation and short reaction
Cylinder pressure
time
Maximum
impact
High operating life
Valve
(closing) Valve
Low pressure loss for higher
closing
opening
compressor efficiency
0
45
90
135
180
225
270
315
360
Low additional fixed clearance for
Crank position, degrees
high volumetric efficiency at maximum compressor load.
Figure 2 Cylinder pressure over the crank angle (at 100% load) including pressure
These requirements are contradicfluctuation; the valve switching points with the highest impact velocities are marked
tory in part because short reaction

In a pilot study, Neumann-Esser (NEA) demonstrated

that a fast-switching clearance pocket valve controlled
by process gas is a highly efficient solution for reciprocating compressor mass flow adjustment. The
company says that its Bluepocket design provides the
opportunity to fit or even retrofit existing compressors
relatively easily. The add-on system does not need any
auxiliary hydraulic equipment, which is not allowed in
some applications. It is also not necessary to measure
intricately the crank position. In the pilot study,
process gas was successfully used to control the
valves switching points under all operating conditions. Feedback from early installations reflects
significant energy savings.
Mass flow control using clearance pockets is one of
the most energy effective means of adjusting compressor performance. Solutions using fixed clearance
pockets are common in this context. However, these
standard options enable only a stepwise mass flow
control, whereby throttling the mass flow depends on
the compressor stage pressure ratio. Clearance pockets
are also used for stepless mass flow regulation using a

www.eptq.com

case studies copy 9.indd 10

PTQ Q4 2014 149

10/09/2014 17:06

times and high durability require low valve lifts and,

on the other hand, low pressure loss needs a large
cross-sectional valve area. The answer to these requirements is a valve design which is very similar to the
familiar poppet valves. Figure 1 shows the design of
the new clearance pocket and these valves.
The compressor design must be adapted to the new
conditions; rod loads, load reversals for the crossheads
and so on must be rechecked. The static and dynamic
stress distribution must be recalculated using the finite
element method (FEM) and simulated.
To ensure the durability of the valve elements, their
dynamics must be checked precisely. Here, low impact
velocities are essential in order to ensure a long service
life. Therefore, the permissible valve lift was determined by simulating valve dynamics.
Since there are different operating points, the simulation of dynamics must be implemented under the
respective conditions. The simulation displays that the
higher the pressure fluctuation in the cylinder, the
higher the impact velocity of the valve elements. Figure
2 shows a graph of the cylinder pressure over the
crank angle at 100% load including pressure fluctuation. The valves switching points which provide the
highest impact speed occur when opening and closing
the poppet valves. The valve elements consist of
compressor valve materials which have been proven
time and time again, in particular for manufacturing
poppet valves. The valve lift is designed in order to
avoid exceeding the permissible impact velocity range,
irrespective of the operating status.
For more information: martina.frenz@neuman-esser.de

Meeting gasoline sulphur regulations

In the mid-2000s, Japan committed to lower gasoline
sulphur. As early adopters of more stringent gasoline
quality regulations, Japanese refiners faced challenges
similar to those that US refiners face today in meeting
Tier 3. Since 2005, Japanese refiners have utilised Grace
GSR products to maintain compliance, observing
35-40% reduction in the gasoline sulphur/feed sulphur
ratios allowing them to meet the 10 ppm gasoline
specifications.
The new Tier 3 ultra-low sulphur gasoline regulations
have the potential to create further challenges for US
refiners. Tier 3 regulations require the reduction of the
average gasoline pool sulphur level to 10 ppm with an
80 ppm cap by January 2017, compared to the current
limit of 30ppm with an 80 ppm cap. Compliance will
require adjustments to operating strategies and possible
capital investment. With FCC gasoline making up
approximately 36% of the gasoline pool, the average
FCC gasoline will have to be controlled at or below 25
ppm. In-unit reduction of FCC gasoline sulphur with
GSR technologies create a variety of opportunities and
options for refiners to drive profitability while meeting
the new Tier 3 gasoline requirements, says Grace.
These clean fuels solutions create economic advantages around feedstock and gasoline blending and asset

150 PTQ Q4 2014

case studies copy 9.indd 11

optimisation, according to the developer. The refiner can

adjust operating conditions in order to preserve octane,
maximise throughput or extend pre-treatment and/or
post-treatment hydrotreater life. Many refiners have
utilised the technologies to create operating flexibility
during hydrotreater outages and generate gasoline
sulphur ABT credits to defer capital investment.
GSR technologies have been used in more than
100 FCC applications worldwide delivering 20-40%
sulphur reduction in FCC naphtha in both full and
partial burn operations. The additive technologies are
used at a 10-25% loading in inventory, whereas the catalyst are a customised 100% drop-in replacement for a
base catalyst. The longest continuous application is now
12 years.
For more information: julie.ellis@grace.com

Crude oil analyis by NMR

Online analysers based on nuclear magnetic resonance
(NMR) technology enable accurate correlation with
laboratory analyses, according to Modcon.
In order to reduce the cost of a crude oil blend, refineries are obliged to blend maximum quantities of low
cost crude oils with sweet crudes. Incorporating heavy
crudes has not only the benefit of reducing CDU feed
costs but also increases the capacity of heavier distillates like gasoil.
Crude blending can be performed by in-tank blending or in-line blending. Especially in the case of in-line
blending continuous measuring of the physical properties of the crude oil feeds provides a platform for
directly mixing different crudes feeds, resulting in a
blend which continuously conforms to the requested
specification.
NMR process analysers have the advantage of being
non-optical analysers and therefore applicable in transparent and opaque fluids alike. NMR analysers are
molecule specific and they make it possible to distinguish qualitatively and quantitatively between the
different chemical structures of the molecules.
Physical properties and NMR spectra are the
outcome of the presence of different molecules in a
crude blend. Correlating them enables a refiner to
quantify the properties of crude oil: API Gravity, true
boiling point yield, aromatic content, olefin content,
pour point, water, total acid number, sulphur and
other parameters.
High accuracy in the correlation between results
obtained by NMR process analyser and in the laboratory characterises the new generation of NMR process
analysers. NMR magnets are sensitive to temperature
differences, hence early generations of NMR process
analysers were sensitive to temperature differences due
to the accumulation of excessive heat by electronics and
heat conducting measuring probes. In the new generation of NMR process analysers, the overall design
excludes any accumulation of heat in the magnet or in
its surroundings by uncontrollable fluctuations, such as
heat transformation by electronics, the magnet itself and

www.eptq.com

10/09/2014 17:06

by the material, which is measured. This increases its

stability to heat fluctuation by some eight degrees. This
means that any required heating of the crude oil is
possible, without affecting the analytical results, as
long as a temperature deviation of 10C is
maintained.
Taking into account the time required for laboratory
analyses, the cost to perform crude oil assays, or the
purchase and maintenance of SimDist justify the use
of NMR process analysers for crude oil analysis and
in CDU feed monitoring processes, it enables precise
monitoring of the quality of blend production, or of
the crude oil entering the CDU.

Frankfurt am Main 15 19 June 2015

Thanks to Aspect imaging for their cooperation in NMR

analysing crude oils.
For more information: ronnym@modcon-systems.com

Pilot ignition for zero flaring

With ever-increasing environmental regulations and a
continued focus on safety by the oil and gas industry,
the criticality of highly reliable and effective pilot gas
ignition systems has never been more important, says
Aereon. Failure of the ignition system can cause environmental excursions with resultant fines, reduction
in production throughput, bad company publicity or,
even worse, a safety issue that causes injury or death
to the site workers.
Flare Industries, an Aereon company and supplier
of flare systems, has over the last 30 years developed
proprietary pilot ignition systems. One of these was
developed for zero flaring or for operators who
cannot, or prefer not to, burn continuous pilot gas.
The Continuum Ignition System requires no pilot
gas, instead utilising a slipstream of combustible
waste gas which is ignited with a high voltage
electric arc, creating the pilot flame which then
ignites the bulk of the waste gas like a conventional
pilot. This arc usually occurs every 20 seconds for a
two second duration (timing is adjustable), 24
hours per day, for continued waste gas combustion
without the need for additional and sometimes
costly continuous pilot gas. This helps not only
reduce the extra emissions generated from the pilot
gas, but also improves combustion integrity as the
pilot routinely ignites any gas that is present in the
flare system.
Flare Industries has added its Mechanical
Retractable Package to the system. This allows the
operator to move the pilot system from flare tip to
ground level, for ground level access to the key
system components when required. This can also save
costs through the elimination of fixed ladders and
platforms usually required with fixed pilot systems,
according to Aereon. With an optional solar package,
the ignition system can also be used in remote locations that do not have ready access to electrical
utilities.

World Forum and Leading Show

for the Process Industries

3,800 Exhibitors from 50 Countries

170,000 Attendees from 100 Countries

Be informed.
Be inspired.
Be there.
www.achema.de

For more information: tlyons@aereon.com

www.eptq.com

case studies copy 9.indd 12

15/09/2014 12:05

Alphabetical list of advertisers

ACHEMA 2015

151

Heurtey Petrochem

46

Aereon

89

Hoerbiger Kompressortechnik Holding

73

Aerzener

67

Hydrocarbon Publishing Company

45

AFPM

145

ITW Technologies

Air Liquide

IFC

Jacobs Comprimo Sulfur Solutions

23

Air Products and Chemicals

54

Johnson Matthey Process Technologies

36

John Zink Company

28

Albemarle Catalysts Company

AMACS Process Tower Internals
Ariel Corporation
Axens

IBC
104 &107

114 & 115

KBC Advanced Technologies

60

Koch-Glitsch

68

OBC

Kurita Europe

BASF Corporation, Catalysts Division

11

Linde

Bilfinger Water Technologies

71

MAN Diesel & Turbo

58

Blasch Precision Ceramics

98

Merichem Company

39

Bryan Research & Engineering

96

Modcon Systems

53

Burckhardt Compression

57

Neuman & Esser

63

133

CB&I

OHL Gutermuth Industrial Valves

129

Chevron Lummus Global

Optimized Gas Treating

103

Chromalox

94

Process Consulting Services

Clariant

40

Prognost Systems

CRI Catalyst

30

Prosernat

66

Rentech Boiler Systems

32

Criterion Catalyst & Technologies

12 & 14
124

Delta Valve

117

Sabin Metal Corporation

90

DigitalRefining.com

143

Sandvik Materials Technology

20

Emcor Industrial Services

121

SPECTRO Analytical Instruments

35

74

Spirax Sarco

83

ERTC 19th Annual Meeting

136

Streamlight

139

EuroPetroleum

147

Sulphur 2014

148

Enersul

Everlasting Valve Company

86

Sulzer Chemtech

ExxonMobil Research and Engineering Company 118

ThyssenKrupp Industrial Solutions

Foster Wheeler

UOP

123

77
110
19, 43 &49

Four Quest Energy

25

Virtual Materials Group

GE Water & Process Technologies

93

Weir Minerals Lewis Pumps

130

Grabner Instruments

85

Yokogawa Europe

141

Grace Catalysts Technologies

17

Zeeco

113

Graphite Metallizing Corporation

101

80

Zwick Armaturen

79

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

152

PTQ Q4 2014

ad index copy 4.indd 1

www.eptq.com

11/09/2014 14:48

diesel yield.

Albemarle has extensive experience in designing catalyst systems

to maximize volume swell. Commercial experiences validate the
benefits of using Nebula and STAX to realize economic gain
from ULSD hydrotreaters. Albemarles Nebula catalysts deliver the
highest denitrogenation and hydrogenation activity of any base metal
hydrotreating catalyst. When combined with our STAX technology to
balance unit objectives with capabilities, hydrogen addition over the
catalyst cycle can be optimized based on the limitations of the unit.

For more information on Albemarle catalysts for maximizing volume swell

or the many products and services Albemarle provides,

CALL (281) 480-4747 TODAY OR VISIT WWW.ALBEMARLE.COM

REFINERY CATALYST SOLUTIONS

albemarle.indd 1

10/09/2014 11:28

Stimulate the heart of

your hydroprocessing unit
ImpulseTM, the catalyst technology that combines
the stability you recognize with the activity you need
4JOHMF
TPVSDF
*40

t
*40

t
0)4"4

www.axens.net

Imp_PTQ.indd 1
axens.indd 1

04/12/2012 14:07:08
6/12/12 13:17:48