Você está na página 1de 148

special features

instrumentation & control


rotating equipment
refining
gas processing
petrochemicals
PETROLEUM TECHNOLOGY QUARTERLY
ptq
Q3 2011
cover and spine copy 8.indd 1 9/6/11 11:29:52
Are you looking to step up plant performance?
Topses next generation BRIM catalysts offer refiners the opportunity to increase
performance through an increase in catalyst activity.
Using the original BRIM technology Topse has developed several new catalysts, resulting
in higher activity at lower flling densities.
The next generation BRIM catalysts display
- high dispersion
- high porostiy
- high activity
We look forward to stepping up your performance!
WWW. TOPSOE. COM
Stepping up performance
next generation BRIM

technology
www.ptqenquiry.com
for further information
haldor.indd 1 7/6/11 17:02:14
2011. The entire content of this publication is protected by copyright full details of which are available from the publishers. All rights
reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic,
mechanical, photocopying, recording or otherwise without the prior permission of the copyright owner.
The opinions and views expressed by the authors in this publication are not necessarily those of the editor or publisher and while every care
has been taken in the preparation of all material included in Petroleum Technology Quarterly the publisher cannot be held responsible for any
statements, opinions or views or for any inaccuracies.
3 Sea change in Gulf
ChrisCunningham
5 Processing Trends
15 ptq&a

21 Automating procedural operations for continuous processes
MauriceWilkinsandMarcusTennantYokogawa Electric Corporation
27 Naphtha to diesel via novel routes
DavidGibbonsFoster Wheeler
35 Optimising hydrogen production and use
RonaldLong,KathyPicioccioandAlanZagoria
UOP LLC, A Honeywell Company
49 Improving residue hydrocracking performance
JoniKunnasNeste Oil
LeeSmithHTI
59 Industrial-scale production of renewable diesel
RasmusEgebergandKimKnudsen Haldor Topse
StefanNystrm,EvaLindGrennfeltandKerstinEfraimsson Preem
67 Improving solids handling in a delayed coker
RandyHull,SteveStewartandJamesTeetsMarsulex
75 Refnery confgurations for maximising middle distillates
ArunAroraandUjjalMukherjeeChevron Lummus Global
85 Bitumen processing: crude unit revamps
MikeArmstrong,RobHendersonandJonMorettaJacobs Consultancy
97 Improving proftability despite lower utilisation
EricStreitandRobertOhmesKBC
109 Multi-stage reaction catalysts
JoesphMcLean,BramHoffer,GarySmith,DavidStockwellandAlexisShackleford
BASF
119 Preventing environmental excursions on the FCC unit
RayFletcherandMartinEvansIntercat
131 Catalytic oxidation of spent caustic
VivekRathore,ShaliniGupta,TSThorat,PVCRao,NVChoudaryandGBiju
Bharat Petroleum Corporation Limited, India
136 Technology in Action
138 Focus: rotating equipment
140 Industry News
87000b/dvacuumgasoilplantsuppliedbyUhdeatConocoPhillipsWilhelmshavenrefnery,Germany.
Photo: ThyssenKrupp Group

Q3 (Jul, Aug, Sep) 2011
www.eptq.com
p
t
q
Y L R E T R A U Q Y G O L O N H C E T M U E L O R T E P

contents/ed com copy 11.indt 1 13/6/11 10:21:47


Your Company +KBC Produces NextGen Performance
n
We collaborate with our clients to create unique solutions to their specic
challenges. Some of these challenges may include:
For 30 years, KBC consultants have provided independent advice and expertise to enable
leading companies in the global energy business and other processing industries manage
risk and achieve dramatic performance improvements.
For more information on how
KBC can help you achieve
NextGen Performance,
contact us at:
AMERICAS +1 281 293 8200
EMEA +44 (0)1932 242424
ASIA +65 6735 5488
answers@kbcat.com U www.kbcat.com
Where do You Want to be
on the Performance Curve?
P = People
M = Methodologies
T = Technologies
Strategic Challenges
U Creating Effective Business Strategy/Decisions
U Increasing Return on Investments
U Enhancing Returns on Acquisitions/Divestitures
U Reducing Strategic/Capital/Market/Investment Risk
U Enhancing Yields
U Creating Effective Response to Crude/
Feedstock/Product Markets
U Improving Financial Performance
Operating Challenges
U Improving Yield
U Increasing Availability
U Reducing Maintenance Costs
U Improving Safety Performance
U Implementing/Improving
U Managing Operational Risk
Behaviour-based Reliability
U Improving Supply Chain Performance
Capital Challenges
U Increasing Return on Capital Investment
U Rationalising/Optimising Environmental
Compliance Capital Expenditures
U Reducing Capital Risk
Organisational Challenges
U Increasing Organisational Effectiveness
U Improving Employee Competency/Capability
U Enhancing Employee Support Systems
U Improving Shift Team Function
Environmental Challenges
U Reducing Emissions
U Ensuring Compliance
U Reducing/Managing Environmental Liabilities
U Improving Energy Efciency
U Rationalising Compliance Expenditures
www.ptqenquiry.com
for further information
kbc.indd 1 9/3/10 14:03:08
T
he European Union has arguably
been the global leader in biodiesel
production and use, with overall
biodiesel production increasing from 1.9
million tonnes in 2004 to nearly 10.3 million
tonnes in 2007. Biodiesel production in the
US has also increased dramatically in the
past few years from 2 million gallons in
2000 to approximately 450 million gallons
in 2007. According to the National Biodiesel
Board, 171 companies own biodiesel
manufacturing plants and are actively
marketing biodiesel.1. The global biodiesel
market is estimated to reach 37 billion
gallons by 2016, with an average annual
growth rate of 42%. Europe will continue to
be the major biodiesel market for the next
decade, followed closely by the US market.
Although high energy prices,
increasing global demand, drought
and other factors are the primary
drivers for higher food prices, food
competitive feedstocks have long
been and will continue to be a major
concern for the development of biofu-
els. To compete, the industry has
responded by developing methods to
increase process effciency, utilise or
upgrade by-products and operate
with lower quality lipids as
feedstocks.
Feedstocks
Biodiesel refers to a diesel-equivalent
fuel consisting of short-chain alkyl
(methyl or ethyl) esters, made by the
transesterifcation of triglycerides,
commonly known as vegetable oils or
animal fats. The most common form
uses methanol, the cheapest alcohol
available, to produce methyl esters.
The molecules in biodiesel are pri-
marily fatty acid methyl esters
(FAME), usually created by trans-
esterifcation between fats and metha-
nol. Currently, biodiesel is produced
from various vegetable and plant oils.
First-generation food-based feedstocks
are straight vegetable oils such as
soybean oil and animal fats such as
tallow, lard, yellow grease, chicken fat
and the by-products of the production
of Omega-3 fatty acids from fsh oil.
Soybean oil and rapeseeds oil are the
common source for biodiesel produc-
tion in the US and Europe in quanti-
ties that can produce enough biodie-
sel to be used in a commercial market
with currently applicable
PTQ Q3 2011 3
Editor
Chris Cunningham
editor@petroleumtechnology.com
Production Editor
Rachel Zamorski
production@petroleumtechnology.com
Graphics Editor
Rob Fris
graphics@petroleumtechnology.com
Editorial
tel +44 844 5888 773
fax +44 844 5888 667
Advertising Sales Manager
Paul Mason
sales@petroleumtechnology.com

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

Petroleum Technology Quarterly (USPS 0014-781)
is published quarterly plus annual Catalysis edition
by Crambeth Allen Publishing Ltd and is distributed
in the USA by SPP, 75 Aberdeen Rd, Emigsville, PA
17318. Periodicals postage paid at Emigsville PA.
Postmaster: send address changes to Petroleum
Technology Quarterly c/o PO
Box 437, Emigsville, PA 17318-0437
Back numbers available from the Publisher
at $30 per copy inc postage.
Vol 16 No 4
Q3 (Jul, Aug, Sep) 2011
Sea change in
the Gulf
A
plant that has failed to achieve serious production in its 20-year
lifetime would seem an unpromising starting point for highly
signifcant developments in the international downstream industry.
Relatively modest as its scale is, Cubas Cienfuegos refnery is at the heart of
potential investment on a grand scale, which may be a big step forward in the
development of the Gulfs oil producing and refning industries.
The refnery was completed in 1991 with an installed capacity of 76 000
b/d, but with already outmoded Soviet technology that rendered it and the
rest of Cubas petroleum industry hugely ineffcient. Apart from its technol-
ogy, the Soviet Union was also the main supplier of crude to Cuba. The island
has its own resources, but so far these have been limited and of a quality that
inferior technology could not be expected to process on commercial terms.
After the collapse of the Soviet bloc, crude supplies fell drastically and, as a
result, so did the Cuban standard of living, until the early years of the last
decade when Venezuelan oil came to the rescue. On the same favourable
terms that other Caribbean states enjoy, Cuba was able to raise its imports of
crude to the point where indigenous supplies currently account for about 30%
of the total refned. Its processing relies on two other oil refneries: ico Lopez
in Havana and Hermanos Diaz in Santiago de Cuba.
Then, in 2004, contracted exploration companies detected potential new
reserves of good-quality crude off Cubas north-western (Gulf) coast. These
are estimated at 5 billion barrels minimum, perhaps as much as 20 billion bar-
rels, according to local optimism. At once, Cuba had the potential to switch
roles from a highly dependent importer of crude, to a self-suffcient refner, to
a potential large-scale exporter of crude.
That level of transformation is no easy task for a couple of reasons. The new
crude reserves are in deepwater that Cuban resources alone cannot hope to
exploit. And the means to raise the countrys economic game to fnance such
operations are limited by a raddled refning industry.
This is where Cienfuegos comes in. Industry estimates are that it needs a
catalytic cracker, vacuum distillation unit and a raft of upgrades at a cost of $1
billion simply to achieve rated capacity. No-one was going to pour such a sum
into Cubas current economic system. No-one, that is, until a high-level
Chinese delegation arrived in Havana in June. In view of proposed sweeping
reforms in Cuba a crucial condition for technology suppliers everywhere
China has co-signed a letter of intent towards the eventual doubling of
Cienfuegoss capacity through an agreement between Cuvenpetrol, the Cuban-
Venezuelan joint venture that owns the refnery, and Chinas CNPC and
France-based contractor Technip. Other components of the deal are expected
to be a regasifcation plant handling Venezuelan LNG and development of the
port that serves Cienfuegos, all with Chinese fnancial backing.
Meanwhile, test drilling offshore Cuba is set to begin later in the year, with
fve wells to be drilled by Repsol YPF from a state-of-the-art Chinese-built
platform. While it is not yet clear whether it will lease blocks among Cubas
offshore riches, China is certain to play a signifcant role in the development
of Cuban oil, upstream and downstream. In the meantime, Russia is back in
the Caribbean picture and in talks to build a brand-new Cuban refnery, possi-
bly fed with Venezuelan crude, pending the arrival of home-produced goods.
CHRIS CUNNINGHAM
p
t
q
Y L R E T R A U Q Y G O L O N H C E T M U E L O R T E P
contents/ed com copy 11.indt 2 10/6/11 13:26:12
KNOW-HOW DELIVERED
We put tested refning technologies
and processes to work in your
world. From hydroprocessing to
fuid catalytic cracking and
Veba Combi-Cracking to resid processing, we offer
proven know-how. So you can improve productivity and
lives. KBR Technology licenses deliver for greenfeld
and existing refneries of virtually every type and size.
See HOW we can help you meet mission-critical goals.
Click kbr.com/PTQ
REFI NI NG
2011 KBR
All Rights Reserved
K11027 03/11
www.ptqenquiry.com
for further information
kbr.indd 1 10/3/11 10:01:03
Trends in opportunity crude processing
Claire Weber and Serena Yeung
Hydrocarbon Publishing Company
Opportunity crudes offer attractive values for refners
who can take advantage of low-quality, inexpensive
grades when oil prices are volatile and discount differ-
entials vis--vis light sweet crudes are wide. By
defnition, opportunity crudes include heavy sour
crudes, oil sands/bitumen, extra-heavy oil, high-acid
crudes and oil shale. Except for a handful of instances in
the past few years, refners of opportunity crudes have
recorded better margins than those processing light
sweet crudes. According to the International Energy
Agency, coking margins for Maya crude have been
higher than cracking margins for Brent crude from 2005
to Q1 2011, except during the third week of August
2008, May 2009, the second and third weeks of
September 2009, and the frst week of 2009.
Of course, past experience cannot guarantee the future
advantages of opportunity crudes. Furthermore, there
exist impending market and legislative challenges that
refners must overcome when processing them. These
challenges include, but are not limited to: crude supply
uncertainty due to crises in the Middle East and North
Africa (MENA); insatiable oil demand by developing
nations; global supply/demand imbalances in diesel/
gas oil, gasoline, fuel oil and propylene; environmental
regulations requiring ultra-clean diesel and gasoline,
lower sulphur specifcations for bunker fuel and reduc-
tion in carbon footprint; and competition from shale oil,
NG condensates and biofuels.
Hydrocarbon Publishing Company initiated a study to
seek technical solutions and operational strategies to
help refners tackle upcoming market and legislative
obstacles. Through a direct survey of oil companies
around the world, conducted from October 2010
to February 2011, and an extensive search of
publicly available information, strategies under-
taken by oil companies have been revealed. This
article summarises the survey and search
results.
Direct survey
Forty-three refneries processing more than
25 000 b/d of crude oil responded to the
survey, which contained 14 questions. (For
a full list of these questions, visit
http://opportunitycrudes.com/survey/oppor-
tunitycrudessurvey.pdf.) The surveys results
were evaluated based on overall responses, a
regional breakdown (US refneries and non-US
refneries) and type of oil company (indepen-
dent or integrated). The results presented
exclude those respondents who do not process
opportunity crudes, unless otherwise noted.
Reduce gasoline cutpoint
Regional differences
Around two-thirds of US refners and two-thirds of non-
US refners indicated plans to boost their processing of
opportunity crudes. However, there appear to be differ-
ences in the major driving factors behind their decisions.
Of the US refners who reported plans to increase the
volume of opportunity crudes processed, 100% cited the
price discount as a major reason for their plan. Only
42% of non-US refners provided the same reason. The
difference between US and non-US refners planning to
increase the quantity of opportunity crudes processed
because of the price discount was determined to be
statistically signifcant using Fishers Exact Test (p-value
of 0.0084). Crude accessibility and energy security was
another popular reason for processing opportunity
crudes, as cited by both US (42%) and non-US (47%)
refners.
Not surprisingly, an examination of the correlation
between regions and the types of opportunity crudes
processed reveals that DilBit, SynBit, SynDilBit and
synthetic crude oil (SCO) are more commonly processed
in the US than the rest of the world (see Figure 1). The
close proximity of the US to Canadas oil sands provides
US refners with easy access to these bitumen-derived
crudes. Although plans are in the works to increase the
availability of Canadas crude oil to Asia, the majority of
the countrys bitumen-derived crude is currently
exported to the US.
Another difference between US refners and non-US
refners is the primary method for processing opportu-
nity crudes. Of the US refners participating in the
survey, 82% cited blending as their primary approach,
while only 45% of non-US refners reported the same.
Non-US refners expressed a greater willingness to
install new process units in order to accommodate more
www.eptq.com PTQ Q3 2011 5
Processing Trends
60
100
90
80
70
50
40
30
20
10
H
i
g
h

T
A
N
H
e
a
v
y

s
o
u
r
S
C
O
D
i
l
B
i
t
S
y
n
B
i
t
S
y
n
D
i
l
B
i
t
O
r
i
n
o
c
o
B
r
a
z
i
l
i
a
n
O
t
h
e
r
N
o
n
e
P
r
o
p
o
r
t
i
o
n

o
f

r
e
f
i
n
e
r
s
,

%
Type of opportunity crude
0
US refiners
Non-US refiners
Figure 1 Types of opportunity crudes processed in US and non-US refneries
proc trends copy 9.indd 1 8/6/11 12:59:43
refners have identifed as major issues or barri-
ers to processing more opportunity crudes.
Fouling, corrosion, crude compatibility and
accessibility are major areas of concern for both
US and non-US refners. The p-values calculated
using Fishers Exact Test indicate that there is no
statistically signifcant difference between the
responses from US and non-US refners.
Differences between integrated and
independent refners
The majority of independent refners (refneries
owned by oil companies with no upstream oper-
ations) and integrated refners (refneries owned
by oil companies with both upstream and down-
stream operations) reported that they plan to
process opportunity crudes at their refneries.
Both independent and integrated refners cited
the price discount and crude accessibility/
security as the two main reasons behind their
plans to increase opportunity crude processing.
Survey results presented in Figure 3 indicate
that the two types of refners both frequently
process high-TAN and heavy sour crudes.
Around 50% of both independent and integrated
refners reported that they process high-TAN
crudes, and about 60% of each indicated that
they process heavy sour crudes. A higher frac-
tion of integrated refners report processing
bitumen-derived crude including SynBit,
SynDilBit and synthetic crude oil (SCO)
compared to independent refners, but there was
not a statistically signifcant difference between
integrated and independent refners for any of
the crude types. However, only 50% of the inde-
pendent refners surveyed indicated that they
have performed a major revamp (or revamps) in
order to increase the refnerys ability to handle
opportunity crudes, compared to 79% of inte-
grated refners. This comparison seems to
indicate that integrated refners are willing to
make larger capital investments than indepen-
dent refners in order to process opportunity
crudes. However, it is important to note that the
results are not statistically signifcant and may
not refect the population as a whole. [Note: A
chi square test for independence indicates that
these results are not statistically signifcant at the 0.05
level. The p-value is 0.094 and the chi square value is
2.80.]
As far as greenhouse gas (GHG) emissions are
concerned, the surveys results indicate that indepen-
dent refners are more concerned than integrated refners
about regulations. Only 36% of independent refners
reported that GHG regulations would not affect their
ability to process opportunity crudes, compared to 58%
of integrated refners. Although independent refners
participating in the survey reported that they are more
concerned than integrated refners about GHG regula-
tions, independent refners appear to have implemented
a smaller variety of strategies to reduce CO
2
emissions,
6 PTQ Q3 2011 www.eptq.com
opportunity crudes in their crude slates (15%, compared
with 6% in the US). The difference may be partially due
to US refneries being more complex and better able to
handle opportunity crudes than those in other regions.
In particular, the US has signifcantly more coking
capacity compared to other regions, which is valuable
for handling heavy opportunity crudes. However, it is
important to note that a chi square test for independence
indicates that these results are not statistically signifcant
at the 0.05 level. The p-value is 0.069 and the chi square
value is 7.09.
One item that US and non-US refners do agree on is
that there exist major barriers to processing opportunity
crudes. Figure 2 illustrates which factors US and non-US
60
100
90
80
70
50
40
30
20
10
y
t
i
|
a
u
q

e
d
u
r
c

f
o

y
t
i
|
i
b
a
i
r
a
V
d
n
a
m
e
d

y
g
r
e
n
e

d
e
s
a
e
r
c
n
I
H

r
e
h
g
i
H
2
s
t
n
e
m
e
r
i
u
q
e
r

e
k
a
m

e
k
o
c

d
e
s
a
e
r
c
n
I
d
|
e
i
y

t
c
u
d
o
r
p

d
e
c
u
d
e
R
g
n
i
|
u
o
F
n
o
i
s
o
r
r
o
C
y
t
i
|
i
b
i
t
a
p
m
o
C
s
s
e
c
c
A
s
,

%
r
e
n
i
f
e
r

f
o

n
o
i
t
r
o
p
o
r
P
Primary issue/
barrier to processing more opportunity crude
0
US reners
Non-US reners
Figure 2 Primary issue/barrier to processing more opportunity crudes in US
and non-US refneries
60
100
90
80
70
50
40
30
20
10
H
i
g
h

T
A
N
H
e
a
v
y

s
o
u
r
S
C
O
D
i
l
B
i
t
S
y
n
B
i
t
S
y
n
D
i
l
B
i
t
O
r
i
n
o
c
o
B
r
a
z
i
l
i
a
n
O
t
h
e
r
N
o
n
e
P
r
o
p
o
r
t
i
o
n

o
f

r
e
f
i
n
e
r
s
,

%
Type of opportunity crude
0
Independent refiners
Integrated refiners
Figure 3 Types of opportunity crudes processed by independent and
integrated refners
proc trends copy 9.indd 2 8/6/11 12:59:53
Engineering Solutions . . . Delivering Results
CB&I has worked in more than 100 countries around the world, on all
seven continents. We have the global experience and local knowledge to
safely deliver superior results in all kinds of environments. From concept to
completion, CB&I gets the job done.
Safely. Reliably. Globally.
www.CBI.com
From Singapore to Saskatchewan
And everywhere in between.
C
M
Y
CM
MY
CY
CMY
K
PTQ_Ad_CBI-June.pdf 1 5/26/2011 4:39:02 PM
www.ptqenquiry.com
for further information
cbi.indd 1 7/6/11 17:07:12
8 PTQ Q3 2011 www.eptq.com
focusing mainly on proven techniques such as energy
benchmarking and computer-aided energy management
systems (see Figure 4). Although integrated refners also
utilise these techniques, 21% of integrated refners
reported that they were pursuing carbon capture and
sequestration (CCS), and 33% reported employing
cogeneration schemes to mitigate the increase in energy
intensity and CO
2
emissions associated with processing
opportunity crudes. Compared to independent refners,
a higher percentage of integrated refners also reported
using carbon offsets (for instance, planting trees), co-
processing biofeeds and supplementing conventional
electricity with renewable energy. However, there were
no statistically signifcant differences between the
responses for independent and integrated refners.
Upstream upgrading
Producers of unconventional crudes particu-
larly bitumen-derived crudes from Canada and
extra-heavy oils (EHO) from Venezuela are
increasingly looking to further upgrade their
crudes on site. Upgrading these unconventional
crudes to better-quality oils enables producers
(and the countries in which these reserves lie) to
realise more value from their crude resources.
From the refners perspective, the additional
upgrading will make it easier to process bitu-
men-derived crudes, but the upgrading will also
increase the price of the crude and may have a
negative effect on plant economics. Survey
respondents reported mixed opinions on
upgrading: 28% reported that their refnery(ies)
likely would process more of these unconven-
tional crudes; 28% reported that they would
process less; 16% reported that upgrading would
not change their plan for bitumen- or EHO-
derived crude processing, and 19% indicated
that carbon legislation would affect their deci-
sion to process upgraded crude. (9% did not
know how upgrading would affect their deci-
sion.) For refners concerned about carbon
legislation, particularly the potential for regula-
tions that consider life-cycle GHG emissions (for
instance, Californias Low Carbon Fuel Standard),
upgraders with carbon-capture capabilities may
persuade them to process more upgraded, bitumen-
derived crude. Both California (as part of Executive
Order S-1-07) and the European Union (as part of the
Fuel Quality Directives Article 7a) have been investigat-
ing the life-cycle GHG intensity of crude oil and may
place stricter regulations on refners in the future.
A closer look at the surveys results reveals that refn-
ers currently processing bitumen-derived crudes (DilBit,
SynBit, SynDilBit) or EHO from Venezuela have views
that differ from those of refners not processing these
types of unconventional crudes (see Figure 5). Of the
respondents, 44% who do not currently process these
The execution of functional
safety assessment and validation
must take place on every safety
project and must be carried out by
functional safety experts
30
50
45
40
35
25
20
15
10
5
O
f
f
s
e
t
s
C
C
S
C
o
g
e
n
e
r
a
t
i
o
n
C
o
m
p
u
t
e
r
-
a
i
d
e
d

e
n
e
r
g
y

m
a
n
a
g
e
m
e
n
t
B
i
o
f
e
e
d
s
E
m
i
s
s
i
o
n
s

c
r
e
d
i
t

t
r
a
d
i
n
g
E
n
e
r
g
y

b
e
n
c
h
m
a
r
k
i
n
g
R
e
n
e
w
a
b
l
e

e
n
e
r
g
y
O
t
h
e
r
N
o

r
e
s
p
o
n
s
e
P
r
o
p
o
r
t
i
o
n

o
f

r
e
f
i
n
e
r
s
,

%
Strategy to reduce CO
2
emissions
0
Independent refiners
Integrated refiners
Figure 4 Strategies by independent and integrated refners to reduce CO
2
Depends
on carbon
legislation
33%
Dont
know
11%
Dont
know
8%
Process
more
6%
Process
less
39%
Process
less
20%
No change
11%
Currently processing
bitumen-derived or Orinoco EHO
Not currently processing
bitumen-derived or Orinoco EHO
Depends
on carbon
legislation
8%
No change
20%
Process
more
44%
Figure 5 Effect of current crude slate on refners reactions to upgrading opportunity crudes upstream
proc trends copy 9.indd 3 8/6/11 13:00:03
To learn more about ISOMIX

-e and other
CLG hydroprocessing innovations, visit www.chevronlummus.com
ISOMIX

-e
Introducing ISOMIX

-e.
Hydroprocessing with an edge.
enhanced reactor internals.
exclusively from CLG.
8
4
0
2
0
-
1
0
4
8
2

I
S
O
M
I
X
-
e

A
d

v
1
Title: ISOMIX

-e
Advertiser: Chevron Lummus Global
Client: Lori De Amaral, 1 510 242 3177, ldam@chevron.com
Designer: John Lind, 1 510 242 2131, lijr@chevron.com
Ad size: Full page (bleed), Trim: 210mm width x 297mm depth
Publication: PTQ Q# 2011
Representative:
Live
Trim
Bleed
L
i
v
e
T
r
i
m
B
l
e
e
d
Live
Trim
Bleed
L
i
v
e
T
r
i
m
B
l
e
e
d
www.ptqenquiry.com
for further information
clg.indd 1 7/6/11 17:09:54
unconventional crudes reported an interest in processing
more if the crude is further upgraded upstream. By
contrast, only 6% of respondents who already process
these unconventional crudes expressed an interest in
processing more if the crudes are further upgraded
upstream. [Note: A chi square test for independence
confrms that these results are statistically signifcant at
the 0.05 level. The p-value is 0.011 and the chi square
value is 11.05.]
Publicly available information
A search of company websites and news releases was
conducted to reveal trends in processing opportunity
crudes, producing middle distillates and reducing the
refnery carbon footprint among petroleum companies.
Among the different types of opportunity crudes, heavy
sour and high-TAN crudes are considered to be conven-
tional oils. These oils are produced by many state-owned
companies and are often off-limits to foreign companies.
On the other hand, international and national oil compa-
nies have gravitated towards the development of
unconventional oils (such as oil sands in Canada and
extra-heavy crudes in Venezuela) because of opportuni-
ties for participating in joint-venture exploration and
production projects.
Status of unconventional crude projects and competition
US-based companies import over three billion barrels of
crude oil per year, 40% of it as heavy crudes (~1.3 billion
b/y).
1
These companies also invest heavily in oil sands
projects, but have hardly any heavy crude projects in
South America; the convenience and lower business risk
of Canadian oil sands make them more appealing. In
fact, Chevron is the only US company to have active
projects in South America. However, despite the general
lack of involvement in South America, 65% of heavy
crudes imported into the US (878 million b/y) are from
Latin American countries.
Europe contains a small amount of the worlds recov-
erable heavy oil reserves and up to 8% of regional crude
oil imports are of heavy crudes (1 million b/d).
However, many companies including BP, Royal Dutch
Shell, Statoil and Total partake in oil sands projects in
Canada, while companies such as Eni, LUKoil, Repsol
YPF, Statoil and Total receive heavy crude from
Venezuela.
2
Recent discoveries of heavy oil and bitumen
felds in African nations, including Angola, Congo,
Egypt and Nigeria, have attracted companies to exam-
ine production possibilities in this region. The discovery
of bituminous sand in Congo has already attracted
Italian company Eni to develop two sites containing a
total of 2.5 billion barrels of oil.
Despite the lack of heavy oil reserves in Asia, compa-
nies such as China National Offshore Oil Corporation
(CNOOC), Korea National Oil Corporation (KNOC),
PetroChina and Sinopec have been trying to expand
their reserves by buying stakes in oil sands projects in
Canada. Also, companies including CNOOC, Indian Oil
Corp., KNOC and Sinopec hold stakes in heavy crude
projects in Venezuelas Orinoco Belt.
Current commitment to processing residual streams
and heavy oils
Despite the involvement of various international oil
companies in Canada and Venezuela, the heavy crudes
produced from these areas are mostly exported to the
US. In 2010, Canada exported 99% of its heavy crudes,
synthetic crudes and blended bitumen to the US.
3
Between 2000 and 2009, Venezuela exported an average
of 91.4% of its heavy crudes (API 22) to North
America and just 8.6% to Europe.
4
Since much of the worlds heavy crude is sent to the
US, it is not surprising to fnd that US refneries are
among the best-equipped plants to handle heavy resi-
dues. Table 1 summarises the total installed capacity for
key bottom-of-the-barrel processing units including
vacuum distillation units (VDUs), cokers, visbreakers
and solvent deasphalters (SDAs) in regions around the
world. In terms of VDU capacity as a percentage of
CDU capacity, the US, Latin America and the Caribbean,
and Western Europe lead the world with 44.4%, 44.2%
and 39.3%, respectively. Similarly, the US (13.6%) and
Latin America and the Caribbean (7.2%) hold the
number-one and number-two spots, respectively, for
coking capacity as a percentage of total CDU capacity.
The US also has the highest amount of solvent deas-
phalting (on both an absolute and a percentage basis).
The results of the direct survey support the fnding
that North American refners are better prepared to
handle large quantities of opportunity crudes. Of US
10 PTQ Q3 2011 www.eptq.com
www.ptqenquiry.com
for further information
-).)6!0/NLINE
NEW!
0rabner nstruments Messtechnik 0mbH / A-1220 Vienna/Austria / Dr. 0tto-Neurath-0asse 1
Phone +3/1/282 1 27-0 / Fax +3/1/280 73 3 / inIo.grabner-instrumentsDametek.at
4.-0.0#1130#
.0-!#11,*78#0
- VP oI gasoline, crude oil and LP0
- A5TM D511, 5188, 377, 378,
87, EN 1301-1+2, P3, 0, 81
- Highest accuracy Ior best
possible blending to oIIicial limits
- Up to 2 sample streams
- Automatic calibration
- Fast and easy maintenance
- Automatic lubrication
570002.indd 1 12/7/2010 10:30:28 AM
proc trends copy 9.indd 4 8/6/11 13:00:17
Before After
Effuent Water Quality
2011 Baker Hughes Incorporated. All Rights Reserved. 31714
solve processing
problems with
heavy crudes.
well share
our experience.
www.bakerhughes.com
Solving problems when you process heavy crudes means having a
partner with the right knowledge.
When an upgrader processing 100% heavy crudes had a signifcant
amount of desalter oil carryunder, we had the solutionBaker Petrolite
XERIC

7000 heavy oil demulsifer.


Baker Petrolite XERIC 7000 heavy oil demulsifer program
signifcantly reduced oil losses by more than 70% while maintaining
desalter operational stability and the desired salt, water, and solids
removal performance.
Visit us at www.bakerhughes.com/xeric
and fnd out how our unique combination
of experience, knowledge, and the new
Baker Petrolite XERIC heavy oil program
can improve your heavy crudes
processing performance.
Joe Nguyen, Research Scientist
31714.Xeric_ad_Joe_Nguyen_PTQ.0611.indd 1 6/3/11 10:32 AM
www.ptqenquiry.com
for further information
baker.indd 1 7/6/11 17:11:45
refners responding to the survey, 65% reported that
opportunity crudes make up more than 20% of their
crude slate. In comparison, only 30% of non-US refners
reported the same.
Although the Asia-Pacifc region is behind other
regions in terms of VDU, coking, visbreaking and
solvent deasphalting capacity, the region has the highest
capacity of residuum hydrocracking (RHC) and resid-
uum hydrotreating (RHT) in terms of both absolute
capacity and capacity as a percentage of total crude
distillation capacity. Asia-Pacifc is also probably leading
the world in resid FCC (RFCC) capacity, since much of
residuum hydrotreating capacity is used to pretreat
feeds for resid FCC units. In Asia, RFCC units are
frequently designed to produce ethylene and propylene
for petrochemicals. This trend may be partly due to the
fact that many Asian refneries are closely integrated
with petrochemical plants. In regard to the installed
capacity of RFCC processes, according to the Shaw
Group virtually every new FCC unit licensed in the last
fve years is designed to maximise the processing of
resid feeds, maximise the production of light olefns, or
a combination of the two.
Of the oil companies investigated, many independent
refners in the US (including LyondellBasell, Citgo
Petroleum, Frontier Oil and Valero Energy) led the way
in terms of coking capacity, while
integrated and national oil companies
had the greatest ratio of thermal
cracking capacity (see Table 2).
Refners with coking capacity are
generally better equipped to handle
heavy resids than refners with
visbreaking units and no coking
capacity. This scenario matches up
with crude import data reported by
the US Energy Information
Administration, which indicated that
independent refners imported more
heavy sour crude as a percentage of
their crude slate than did their inte-
grated counterparts. It also agrees
with the results of the direct survey,
in which 64% of respondents who
were independent refners reported that opportunity
crudes made up over 20% of their crude slate. In
comparison, only 38% of integrated refners reported the
same.
Although independent refners lead the way in terms
of coking capacity as a percentage of total crude capac-
ity, it is important to note that the major international oil
companies (including BP, ConocoPhillips, ExxonMobil,
Royal Dutch Shell and Total) have the largest capacity
in terms of absolute resid upgrading capacity. Among
the independent refners examined, Valero has the larg-
est absolute quantities of VDU, coking, thermal cracking
(visbreaking), RHC and RHT capacity.
Overall strategy and long-term outlook
For various political and business reasons, international
oil companies are more focused on Canadian oil sands
and national oil companies are mostly concentrated
in the production of EHO from Venezuela. Most
independent oil companies have refneries located in the
US; therefore, it is more convenient for them to gather
bitumen from Canada. However, there exist two obsta-
cles to increasing Canadian bitumen consumption in the
US. First, California is planning to give pass or fail
grades to crudes processed in the state based on their
carbon intensity. Preliminary results indicate that some
Canadian crudes are among those that may receive a
failing grade.
Second, the US State Department has not approved
TransCanadas Keystone XL pipeline extension and
other pipeline projects that would bring DilBit and
SynBit to the US Gulf Coast. If the US turns away from
Canadian bitumen, these heavy crudes will likely be
diverted to Asia, since Chinese, Japanese and South
Korean companies have recently been investing in
Canadian oil sands projects. Despite European indepen-
dents investments in unconventional oils in the western
hemisphere, European refners still rely heavily on
lighter grades from the Middle East-North Africa region.
Due to recent political tensions in this region, should
European refners (particularly those in the
Mediterranean region) explore increased processing of
unconventional crudes? The answer is not easy to reach,
12 PTQ Q3 2011 www.eptq.com
Region Residue upgrading capacity, K b/d (as a % of CDU capacity)
VDU Coking Visbreaking SDA
US 7720(44.4) 2419(13.6) 34(0.2) 382(2.2)
Canada 666(32.8) 57(2.8) 51(2.7)
39(0.8)
a
LatinAmericaandtheCaribbean 3601(44.2) 580(7.2) 252(3.1)
WesternEurope 5584(39.3) 343(2.4) 1249(9)
62(0.3)
b
CISandCentral/EasternEurope 3903(37.7) 319(3.1) 466(4.5)
MiddleEast 2186(28.2) 90(1.1) 514(6.5)
Africa 508(15.5) 66(2) 61(1.9) 165(0.2)
c
Asia-Pacifc 4389(19.3) 548(2.3) 364(1.5)
Worldwide 28557(33.3) 4422(5.1) 2991(3.4) 648(0.8)
aIncludesCanada,MexicoandVenezuela
bEurope
cWorldwidefgureexcludingUS,Canada,Mexico,VenezuelaandEurope
Regional refnery complexity
Table 1
Ranking* VDU Coking Thermal cracking
1 LyondellBasell LyondellBasell Eni
2 Chevron Citgo CEPSA
3 Eni FrontierOil&Refning Ecopetrol
4 Pemex ConocoPhillips RoyalDutchShell
5 Ecopetrol Valero Statoil
6 Tesoro Husky Total
7 Marathon Chevron RepsolYPF
8 ConocoPhillips Pemex IndianOil
9 Valero Tesoro ExxonMobil
10 Murphy RepsolYPF BP
*Basedonunitcapacityasapercentageoftotalcrudecapacity
Oil companies with the largest ratio of residue upgrading
capacity to crude distillation capacity
Table 2
proc trends copy 9.indd 5 8/6/11 13:00:28
since the move would require considerable investment
at a time when refners are already struggling with a
sluggish economy and increasingly strict GHG regula-
tions. Meanwhile, Latin American oil producers
(Ecopetrol, Pemex and Petrobras) have sizeable heavy
oil production, but limited upgrading capability. What
should be their strategies?
Finally, refners around the world considering oppor-
tunity crude supplies will need to overcome the
changing market and legislative environment and
address fve key factors: increasing middle distillates
yield and quality, displacing high-sulphur fuel oil, boost-
ing propylene make, mitigating fouling and corrosion
problems, and decreasing carbon footprint. Increasing
distillate yield and quality is the top priority, due to
diesel/gas oil defcits in Europe and Latin America.
These defcits will create export opportunities for the US
and the Middle East, as well as export-oriented Reliance
Industries in India. However, accomplishing this feat
could be challenging because of the signifcantly fewer
straight-run distillate yields produced from opportunity
crudes compared with benchmark crudes. Additionally,
displacing high-sulphur fuel oil is important, as demand
for this product is diminishing due to the Annex VI
amendments to the MARPOL agreement, which require
signifcant sulphur reductions in bunker fuels. Boosting
propylene output remains the right strategy for refners
and petrochemical producers to beneft from the grow-
ing demand for polypropylene and propylene
derivatives worldwide. Also, refnery propylene is in
strong demand as the increasing use of ethane and
propane from shale gas diminishes propylene supply
from ethylene crackers. Mitigating fouling and corrosion
is important, as billions of dollars in operating budgets
are wasted each year to combat the effects of these
issues, particularly as more high-acid and heavy crudes
are processed. Lastly, minimising carbon footprint has
become a greater concern for refners due to stricter
legislation with regards to CO
2
emissions and the direct
relationship between GHG emissions and crudes with a
lower API gravity. Refners must also take into consider-
ation the difference in GHG emissions when opportunity
crudes are processed upstream rather than downstream.
(This analysis is based on Hydrocarbon Publishing
Companys Opcrudes Report II: Technologies & Strategies
for Meeting Evolving Market & Environmental Challenges,
May 2011.)
References
1 2009CompanyLevelImports,USEnergyInformationAdministration
website, http://www.eia.gov/pub/oil_gas/petroleum/data_publications/
company_level_imports/historical/2009/data/impa09d.xls.
2 EUCrudeOilImports2010,EuropeanCommission,http://ec.europa.
eu/energy/observatory/oil/doc/import/coi/eu-coi-2010-01-12.pdf.
3 Estimated Canadian Crude Oil Exports byType and Destination 2010
Q4, Q3, Q2, Q1, National Energy Board, www.neb-one.gc.ca/clf-nsi/
rnrgynfmtn/sttstc/crdlndptrlmprdct/stmtdcndncrdlxprttpdstn-eng.html.
4 Table22:IEAMemberCountryDestinationsofSelectedCrudeStreams
1994-2009,Annual Statistical Supplement for 2009 [2010 Edition, IEA
website.http://omrpublic.iea.org/omrarchive/sup2010.pdf.
www.eptq.com PTQ Q3 2011 13
Opportunity Crudes Report II:
Technologies & Strategies for Meeting Evolving Market & Environmental Challenges
2011 Hydrocarbon Publishing Company
Email: mcreports@hydrocarbonpublishing.com Telephone: +1 610-408-0117 www.hydrocarbonpublishing.com
Processing opportunity crudes gives a competitive advantage
at a time of volatile oil prices and supply uncertainty. This
techno-economic study focuses on five major forces essential
to the success of processing low-quality, discounted crudes.
x Increasing distillate yield and quality
x Displacing high sulfur fuel oil
x Boosting propylene output
x Mitigating fouling and corrosion
x Minimizing carbon footprint
Visit www.hydrocarbonpublishing.com/Report2011 to view the Prospectus and Table of Contents.
www.ptqenquiry.com
for further information
proc trends copy 9.indd 6 10/6/11 10:54:15
Sometimes it seems like the rare earth market has gone crazy. With prices surging to
new highs, the cost of many FCC catalysts and additives has gone through the roof.
there iS an alternative.
albemarles high-performance low rare earth technology (lrt) products are based
on proven technology and deliver outstanding results with around half of the rare
earth content.
albemarles lrt products:
GO LRT AMBER LRT UPGRADER LRT CORAL LRT BCMT-500 LRT
www.albemarle.com
the reSultS you need at a priCe youll like
3UPERIORPERFORMANCE
DOESNTHAVETOCOST
THE%ARTH
www.ptqenquiry.com
for further information
albemarle.indd 1 7/6/11 17:13:50
Q
The rocketing prices of rare earth metals are really
taking effect on our cost of catalyst replacement. Will the
price effect diminish soon, or is there a reliable alternative
to rare earth-containing catalysts?
A
Yen Yung, Global Knowledge Management, FCC,
Albemarle, yen.yung@albemarle.com
Over the past year, the price of lanthanum oxide has
increased by nearly 1500%, and leading rare earth
mineral companies predict that prices will continue to
rise. Previously uneconomic mines are being reopened
and plants for rare earth mineral recovery from
consumer electronic waste are being planned. These
events have brought Albemarles low rare earth tech-
nology(LRT)catalystintosharpfocus.
Creating high-performance catalysts that use a lower
proportion of rare earth minerals can help to protect
refners from these rising costs. LRT products have
been developed in anticipation of sustained increases
in rare earth mineral prices and to fulfll Albemarles
commitment to providing comprehensive economic
catalystsolutions.Asrareearthmineralpricessoar,the
valueprovidedcontinuestoincrease.
Although the standard line of FCC catalysts already
offers performance at some of the lowest rare earth
contents available, the LRT products reduce rare earth
contentbyuptoafactoroffour.Thetechnologiesused
in LRT products are fully commercialised and proven
in refneries. They provide the activity and stability of
catalysts with higher concentrations of rare earth
minerals, and their yield selectivities can be tailored.A
range of catalysts and additives is available, including
GO LRT and Amber LRT for the vacuum gas oil
market; Upgrader LRT and Coral LRT for the residue
market; and BCMT-500 LRT additive for enhanced
bottomscrackingandmetalstolerance.
Q
How can I maximise my visbreaker unit run before a
shutdown to deal with fouling becomes necessary?
A
Marcello Ferrara, Chairman, ITW, mferrara@itw.it
ITW online cleaning can be used effectively to clean a
visbreaker unit in preparation for a turnaround and to
recovervisbreakerunitperformanceduringtherun(ie,
recovery of quench temperature). Regular, proactive
use of online cleaning during the run will avoid any
foulingissuesandincreasetherunlength.
Q
Are there ways of optimising a hydrogen plant to
enable extra throughput in the refnery?
P
A
Henrik Olsson, Principal Engineer, Haldor Topse,
heo@topsoe.dk
There is a range of possibilities for developing hydro-
gen units to enable extra production. In old hydrogen
units, incremental capacity increases can be obtained
by replacing old catalysts, reformer tubes, purifcation
systems and so on with the latest developments on the
market. In newer hydrogen units, 25% capacity
increases can be achieved through the installation of
gasheatedreformerand/orpre-reformertechnology.
Q
Contaminants are limiting the life and cycle length of
my ULSD catalyst. Should I switch feedstock or can I miti-
gate with the type of catalyst I choose?
A
Steven Mayo, Global Applications Technology Manager,
HPC, Albemarle, steven.mayo@albemarle.com
In nearly all cases, cycle length limitation due to
contaminantsinthefeedstockcanbemitigatedbycata-
lyst selection, in particular design of the guard catalyst
system. Switching feedstock is a viable, but potentially
costly, approach to solving a contaminant problem.
Economicsgenerallyfavouracatalystsolution.
Before embarking on a catalyst solution, it is impor-
tant to understand the type, source and quantity of
contaminants in the unit. Are the contaminants coming
in with the crude (As) as a result of up-stream process-
ing (Si), from corrosion (Fe), additives or bio-derived
feeds (Na, P, Ca, Mg), or from some other source? With
the possible exception of contaminants in the crude,
there may be process and operational actions that can
reduce the contaminant level fed to the ULSD unit. Due
to a low concentration and varying rate, determining
contaminantquantityinthefeedstockitselfmaybechal-
lenging. Analysing spent catalyst usually gives a better
picture of the problem and an idea of the effcacy of the
currentcatalystsystemathandlingthepoisons.
After establishing contaminant type, sources and
quantity, an effective guard bed catalyst system can be
designed. Catalysts designed to trap specifc contami-
nantsare usedinsomecases,whileinothersacatalyst
that can capture several different contaminants is
preferred.Anunderstandingofthekineticsofcontami-
nant removal is equally important to maximise cycle
life. Albemarle offers a full range of hydroprocessing
catalyststomitigatetheeffectsoffeedstockpoisons.
Q
As part of a proposed revamp, we aim to review our
tray and packing selection in the crude distillation tower.
www.eptq.com PTQ Q3 2011 15
ptq&a
Additional Q&A can be found at www.eptq.com/QandA
Q&A copy 7.indd 1 8/6/11 13:06:44
What are the issues in achieving optimum run length and
avoiding plugging ?
A
Mark Pilling, Manager of Technology, Sulzer Chemtech,
Mark.Pilling@sulzer.com
Crude distillation columns typically suffer fouling in
threesections:
Inthetoptraysorpackingofthemainfractionator
Inthewashsection(traysorpacking)
Inthebottomstrippingsection(trays).
Fouling in the top section is typically the result of
chlorides in the feed not being removed at the desalt-
ingstage,whichlaterhydrolysetohydrochloricacidin
the fred heater. The acid is then carried with the
vapourised feed from the fash section up to the top of
the column. Chlorides are responsible for potential
corrosion and salt depositions. Tray corrosion occurs
when operating conditions (temperature, pressure,
water and chloride concentration) approach the water
dew point temperature, creating concentrated hydro-
chloricacid,whichattacksthemetalinternals.
Dependingonoperatingconditionsandthepresence
of ammonia or amine components (typically used as
H
2
Sscavengers),chloridesaltsmaydeposit.Saltsaccu-
mulating on the top trays cause progressive plugging
of the tray valves, deteriorating effciency and capacity.
When dealing with plugging from salts, sludge or
corrosion products, a review of the process conditions
is an essential frst step:
Can you reduce the harmful upstream components
(chloride, ammonia, amine and so on) and water (in
crude,fromstrippers)thatmayreachthetopsectionof
thecolumn?
Operatingconditionssuchastemperatureplayarole
ininitiatingandcontinuingthedepositionofsalts.You
need to understand the effects of temperature on your
particular corrosive process and then properly control
thetemperature
Installation of a water wash system for the top three
to fve trays often helps
Last, but not least, the design of the tray or packing
internalsthatresistfoulingorcorrosionshouldhelp.
Lookingmorecloselyatthelasttwopointsabove:
Modifcation of the top section to accommodate a
water wash system including a water draw to dissolve
andremovegeneratedsaltsshouldbeconsidered
Adopting anti-fouling trays such as Sulzers VG AF
trays will help. Anti-fouling trays are specifcally
designedtooperateinfoulingconditionsandtherefore
improveperformanceandrunlengths.
Wash section fouling is the result of entrainment
from the fash zone (asphaltenes) and/or a low wash
rate. A combination of the two can cause dry condi-
tions in part of the trays or on the packing surface,
resultingintheprogressivebuild-upofcoke.
Potential solutions to mitigate or avoid plugging of
thewashsectionare:
Improved design of feed inlet distributor, to provide
uniform fow and to reduce entrainment
Understand the process parameters affecting the
wash rate vs overfash
Maintain the minimum allowable wetting rate at the
bottomtrayorpacking
Useanti-foulingtrays(VGAF)
Useasmooth(non-textured)packingtype(Mellagrid
andMellapak)
Better liquid distribution on top of the packed
section.
The wash section is also subject to additional
mechanical stress due to feed upsets, such as excessive
water from the desalter and overheating at the feed
heater. This requires internals to be designed with a
differentialupliftrequirementoftypically1to2psi.
In the bottom stripping section, fouling and/or plug-
ging is the result of two major issues:
Solids carried in with the feed (sand, rust, coke
particles and sludge) to the fash zone
Improperdesignoftrays.
The use of antifouling trays can help to mitigate or
avoid the problem, along with a progressive open area
to maximise stripping effciency. The progressive open
area allows the tray capacity to match that of the
rapidlychangingvapourratesinthestrippingsection.
The stripping section is also subject to additional
mechanicalstresscausedbyhighbottomsliquidlevels,
and/or liquid water entrained with the stripping
steam. These conditions require that the trays be
designedwithanupliftloadof2psi,aswellastheuse
of a special steam distributor to reduce the impact of
liquidwaterentrainment.
16 PTQ Q3 2011 www.eptq.com
www.ptqenquiry.com
for further information
Dew Point Humidity Oxygen
Michell Instruments Ltd
48 Lancaster Way Business Park
Ely, Cambridgeshire CB6 3NW
T: +44 (0) 1353 658000
info@michell.com
www.michell.com/uk
Your partner for moisture,
hydrocarbon dew point and
oxygen analysis
In natural gas production, transmission,
storage and conditioning
Discover more visit www.michell.com/uk
PTQ_86x125_jun11.indd 1 08/06/2011 09:48:55
Q&A copy 7.indd 2 8/6/11 13:07:00
www.mprservices.com
MPR Services, Inc. Introduces
New and Improved, Compact Technology
with Limiting Use of Resources and Energy
www.ptqenquiry.com
for further information
mpr.indd 1 8/3/11 14:07:58
A
Marcello Ferrara, Chairman, ITW, mferrara@itw.it
ITW online cleaning has been used to clean tower
internals, including bottom trays and packings.
Optimumrunlengthandpluggingavoidance(includ-
ing distributors) can be achieved through regular
online cleaning. Remember that despite proper selec-
tion, trays and packing can still plug due to feed
quality or operating parameters, or due to unit upset.
Online cleaning allows you to take immediate action
tosolvetheproblemandavoidfurtherdrawbacks.
Q
In what circumstances does using a gas oil hydrotreater
to raise the quality of feed to the FCC unit make economic
sense?
A
Raul Arriaga, Global VGO & Optifuel Specialist,
FCC, Albemarle, raul.arriaga@albemarle.com Ken Bruno,
Global Applications Technology Manager, FCC, Albemarle,
ken.bruno@albemarle.com
In the majority of cases, using a gas oil hydrotreater to
improve the quality of feed to the FCC unit makes
economic sense and there are numerous benefts. There
are several types of reaction taking place in a gas oil
hydrotreater, some of the most important ones being:
hydrodesulphurisation (HDS), hydrodenitrogenation
(HDN), hydrodemetallisation (HDM) and polyaromat-
ics saturation (HDA), among others. In a gas oil
hydrotreater, the hydrogen content of the FCC unit
feed increases after hydrotreating, resulting in a more
crackable molecular distribution of paraffns, naph-
thenes and aromatics (PNAs). Aromatic saturation is
one of the most important types of reaction taking
placeinagasoilhydrotreater;whenaromaticsdecline,
the naphthenes and paraffns increase. At the same
operating conditions and catalyst activity in the FCC
unit, paraffns are relatively more crackable than naph-
thenes, which in turn are more crackable than
aromatics. Therefore, the overall FCC unit conversion
increases. Another beneft is a lower contribution from
feed-related coke to the FCC units delta coke, both
through a reduction in aromatics and a decrease in
feed metals. These positive feed quality effects are
manifested in a lower regenerator temperature, higher
catalyst circulation and consequently higher conver-
sion. Of course, a lower metals level also improves
catalyst activity and selectivity, including lower dry
gas. Last but not least, sulphur and nitrogen in the
feed decline, meaning that product contaminants and
emissionsdeclineaswell.
Even with all of the benefts explained above, there
aresomeimportantimplicationsfortheuseofagasoil
hydrotreater that need to be taken into account for
proper economic analysis. For example, if there are
capacity limitations at the fractionation or gas concen-
tration sections of the FCC unit, a revamp may be
needed.Or,sinceFCCunitdeltacokewoulddecline,if
the circulation rate is at maximum the refner may
need to increase catalyst additions or reformulate the
catalyst towards higher activity in order to restore the
lost delta coke. If these options are not enough to
satisfy the riser heat demand, the refnery may also
have to install a new FCC feed pre-heater. Another
caveat to consider is that gasoline olefnicity tends to
decline with hydrotreated feeds, meaning that octane
would decline. This is because with higher hydrogen
contentinthefeed,theratesofhydrogentransferreac-
tions and olefns saturation increase. In addition, LPG
olefnicity would decline, potentially resulting in lower
alkylate yields. If a refner is short on the pool octane,
a ZSM-5 additive could be used to improve not only
gasoline octanes but also alkylate production. In most
cases, it is worthwhile evaluating the feasibility of
installing a catalytic reformer along with a gas oil
hydrotreater, because the catalytic reformer would
produce some or all of the hydrogen needed for the
gas oil hydrotreater while it contributes with a high-
octane component for the gasoline pool. If a catalytic
reformer cannot be installed, hydrogen would need to
be produced, either on site or imported from an
outsidesupplier.
A typical range for the economic improvement from
the use of a gas oil hydrotreater is between 0.50 and
1.00 $/bbl after deducting operating costs, or $918
million annually for a 50 Mbpd FCC unit. Therefore,
assuming the refner has the budget to incur the capital
expenses required and the FCC unit has the available
capacity to process the extra amounts of LPG and
gasoline, a gas oil hydrotreater is usually a very posi-
tiveinvestment.
18 PTQ Q3 2011 www.eptq.com
www.ptqenquiry.com
for further information
Q&A copy 7.indd 3 12/6/11 21:01:28
Realize the value of BASF innovation.
www.catalysts.basf.com/innovation
Introducing Fortress

The First Catalyst Solution


from the Innovative Multi-Stage
Reaction Catalyst (MSRC) Platform

Fortress is designed for resid
cracking applications where
contaminant feed metal
passivation is crucial

Specialty alumina passivation
components are concentrated
on the outer stage of the catalyst
to trap nickel directly where it
deposits

Inner stage incorporates BASFs
award-winning Distributed Matrix
Structure (DMS) platform to
maximize gasoline and Light Cycle
Oil (LCO) yields from resid feeds
www.ptqenquiry.com
for further information
basf.indd 1 7/6/11 17:17:05
High pressure. Extreme temperatures. volatile products. lt's all part of the job in
hydrocarbon processing. But so is the goal of maximizing safety integrity. We make
the process more secure with our innovative valves and controls, which is why the
industry relies on us to keep their workers safe and their plants running smoothly.
/ea|n mo|e aoot o| o|ant oe|fo|mance so|t|ons at CWFC.com
Eng|nee||ng t|ansfo|mat|on.

9PZROHZ
HS^H`ZILLU
WHY[VM
[OPZQVI
A part we can do without.
www.ptqenquiry.com
for further information
c wright.indd 1 13/9/10 12:33:58
Automating procedural operations for
continuous processes
P
rocess plants are run accord-
ing to operational procedures.
These procedures consist of a
set of tasks that are executed in a
consistent manner to achieve a
specifc objective, such as starting
up, shutting down or transitioning
a unit as part of making a product.
The level of detail, purpose and
frequency of use of these proce-
dures varies by process, company
and site, but they are the basis for
how to operate plants.
This article reviews progress and
developments in the recently
formed ISA106 standards and prac-
tices group: Procedural Automation
for Continuous Process Operations.
It outlines current models and best
practices to follow in order to
improve a plants operational
procedures. Several examples high-
light different requirements and
how various tools can be applied to
implement operational procedures.
These best practices will enable
plant operations to execute proce-
dures effciently and consistently.
Challenges affecting
process operations
As an example of the issues facing
the process industries, let us focus
on distillation, as it is the most
widely used separation process. In
the manufacturing sector, distillation
uses 24% of the total energy
demand. Oil refneries, in particular,
rely heavily on distillation, which
consumes approximately 40% of all
the energy used in refneries.
Consider the distillation process
from a steady-state perspective, with
most attention focused on achieving
an optimal steady state to ensure
that the process remains within safe
Progress towards a standard for automating operational procedures in
refnery operations
MauriCe Wilkins and MarCus TennanT
Yokogawa Electric Corporation
operating limits. Those who run and
manage continuous process units
understand that procedural activities
are critical to safe and effcient oper-
ations. Often this procedural
information is not well documented,
not readily available or can reside in
an operators individual knowledge
acquired through years of experi-
ence. Procedural activities in a
distillation process are most preva-
lent and critical at certain times of
the process operation
startup
Safe and effcient startup of a distil-
lation unit is critical to its overall
operation. Startup frequency can
vary, from once every day to once
every fve years. If startups occur
often, they can occur on different
shifts, and the time and skill of the
operator on duty can determine the
effciency of the startup. If startups
occur at longer intervals, companies
run the risk of not having
experienced personnel available to
run and oversee the startup. There
have been cases where companies
have had to bring operators out of
retirement to restart a unit after a
shutdown.
shutdown
Orderly and safe shutdown of a
distillation unit is just as critical as
a startup. One key item in the shut-
down procedures is recognising
that a shutdown might not be
scheduled. System problems or
severe weather such as approaching
hurricanes can require a shutdown
of distillation operations at very
short notice. As with startups, on-
shift operating personnel might not
have the most experience in shut-
ting down a system.
Feedstock and product output
transitions
Many chemical processing units
regularly operate at more than one
www.eptq.com PTQ Q3 2011 21
19801984
19751979
19851989
19901994
19951999
20002004
20052010
0 500 1000 1500 2000
US$, millions
Figure 1 Five-year loss totals in the refning industry have continued an upward trend
over the last few years. Piping failures or leaks (caused by corrosion or incorrect
metallurgy) and startup and shutdown events continue to be signifcant causes
wilkins.indd 1 8/6/11 13:15:37
22 PTQ Q3 2011 www.eptq.com
Need for procedural best practices in continuous
process operations
A 2008 survey by the ARC Advisory Group indicates
that continuous process manufacturers are now seeing
effective and repeatable transition management, along
with the use of sequence-based operating procedures,
as a competitive advantage. Additionally, the safety
aspect of automating procedures should be assigned a
high tangible value. The cause of some recent indus-
trial accidents has been linked, in part, to a lack of
good emergency shutdown procedures during an
abnormal condition. This put too much pressure on
operators in a crisis, leading to improper procedural
operations that resulted in disastrous consequences.
Process industries have used semi-automated and
automated procedures in the process industries for
years, especially in batch processes. These procedures,
when implemented in recipes, use a modular approach
in accordance with the ISA-88 batch control standard.
Semi-automated and automated procedures are not
widely adopted in continuous processes due to a lack of
general industry expectations and standards. Increased
operational excellence has put more focus on business
drivers that require increased safety, improved through-
put, reduced cost and knowledge capture to retain years
of operational experience that would soon be lost due to
retirement.
Automated procedures for continuous processes in
the past have typically been implemented using ad hoc
designs and programming techniques, usually result-
ing in difcult-to-maintain code. While this can provide
short-term operational benets, the total cost of owner-
ship of these procedures is higher than necessary. This
is a result of the increased implementation costs of
changing and updating successive procedures over
time due to a lack of reusable software modules.
Implementing a modular approach could also provide
companies with the ability to standardise functions
across plants, sites and enterprise, achieving corporate-
wide repeatability and reproducibility. This would help
to reduce engineering labour costs, provide consistent
operation and lower the total cost of ownership (TCO).
The ultimate aim is to obtain the best practices of the
best operator on their best day, every day, to provide
consistent, optimal and safe operations.
Modular procedural automation
Yokogawa introduced modular procedural automation
(MPA) to address industry needs and to improve and
apply a uniform approach to operating procedures. It
is a consultative methodology, whose purpose is to
document and automate procedural operations in
continuous processes. A modular design approach
facilitates standardised implementation within sites
and across companies. This approach optimises user
acceptance and reduces deployment costs.
Using a modular approach provides a number of
benets, such as:
Hierarchy of procedures Breaks large and complex
procedures into smaller modules and organises them
into a hierarchy. This modularisation provides easier
optimum steady state. In reneries, for example,
many reners purchase crude oil of different types in
tankers. As they run out of one type of oil, they must
adjust operations to process correctly the next type of
oil. Similarly, in chemical plants, a single plant will
often make a variety of products in a campaign fash-
ion by changing operations to meet the new product
specication.
Transitions inherently increase the risk of disruptions
that can lead to incidents. According to a report by
Marsh Ltd, over the period 19752009, the ve-year
loss rate (adjusted for ination) in the renery industry
continues to rise (see Figure 1). Incidents that occur
during startup and shutdown continue to be a major
factor.
An additional study by J & H Marsh & McLennan
shows that the examination of major incidents by the
average loss per incident indicates that operational
error represents the largest average dollar loss (see
Figure 2).
With distillation operations consuming a large
amount of the energy requirements in a rening or
petrochemical application, efcient production is a key
to meeting product specications and producing the
best possible yields of valuable products. Performing
procedural operations in an inefcient or time-consum-
ing manner will have a signicant economic impact on
the complete operation.
best mass transfer
highest mechanical
strength
lowest pressure drop
highest capacity
P. O. Box 552, D - 56225 Ransbach-Baumbach,
Phone +49( 0) 2623/895-0, i nfo@vff.com, http://www.vff.com
For further information
www.vff-twin-pak.de
+++Tower Packings+++Catalyst Supports+++Internals+++Droplet Separators+++
VFF-TWIN-PAK

nothing is random about
this packing.
www.ptqenquiry.com
for further information
wilkins.indd 2 8/6/11 13:15:49
verifcation of documentation and
implementation of the discrete
pieces of procedural logic and
knowledge
Reuse of procedural logic Whether
implemented with manual,
prompted or automated procedures,
MPA can help reuse the same
procedures and logic in different
parts of the plant and with different
products. When procedural logic
(that is, portions or modules of
procedures) is reused, business
gains repeatability and reliability
and it lowers the cost of developing
duplicate procedures
Multi-site consistency Companies
with multiple production sites can
experience an effciency gain and
product improvement by standard-
ising procedural knowledge
between plants. Without modular
procedural automation techniques,
this level of standardisation would
be very cumbersome to maintain.
Modular procedures, both
prompted and automated, allow
easy sharing of procedural knowl-
edge between facilities.
A scalable approach to automa-
tion avoids the one size fts all
approach. Different types of proce-
dure implementations contain
scalability for manual, prompted
and automated control. The hierar-
chy of modular procedures is easily
adjusted to ft the application
requirements. Application require-
ments can vary the level of
complexity (for instance, normal
sequencing only or complex excep-
tion handling).
Manufacturing benefts
Skills retention
The loss of operational skills due to
an ageing workforce has already
begun to hit process manufacturers
around the world. One plant
reported problems with a turna-
round because operators lacked
suffcient experience on restarting
the plant. Retired operators with
experience were brought back to
start up the plant safely and eff-
ciently. MPA provides a systematic
approach to capturing valuable
skills before they are lost. The
procedures developed can also be
used to train new operators who do
not have the beneft of the years of
www.eptq.com PTQ Q3 2011 23
knowledge developed by those
retiring.
Knowledge management
Capturing operational knowledge is
important. MPA provides a meth-
odology for the captured knowledge
to be documented, distilled into
best practices, and implemented in
reusable and repeatable procedures.
Without this ability, captured
knowledge is not well managed
and simply becomes information on
a shelf that is not utilised.
Improved safety
Prompted and automated proce-
dures can reduce the risk of health,
safety and environmental incidents.
Available and repeatable procedures
can lower the chance of accidents
due to human error. As indicated
above, operating procedures provide
the biggest source of risk for process
safety management. MPA can ensure
that a process is shut down safely,
automatically or at the push of a
button. This can be invaluable
during an incident, when the opera-
tor may have many things going on
to distract them and could poten-
tially make mistakes.
Reduced variability
When procedural knowledge is
implemented as prompted standard
operating procedures (SOPs) and
automated procedures they are
available to all operational teams
and all operator skill levels. Their
use can drive down variability in
operation, resulting in more consist-
ently on-spec product. The lower
product and operational variability
enables further process improve-
ment and refnement of the
procedures.
Increased productivity
The combination of consistent and
repeatable procedures and reduced
variability can increase productivity
by eliminating lost time due to poor
operational decisions and actions.
The combination of these benefts
can result in improved metrics used
to measure operational excellence.
Improving plant operations
MPA is an approach to improving
plant operations; it is not a single
product or software package. It
begins with a multi-step consulta-
tive process supported by a design
methodology, control systems and
applications.
The goal of this approach is to
distil best operating practices and
fnd the right balance between
manual, prompted and automated
procedures, documenting and
implementing the procedures and
then executing continuous improve-
ment cycles on them. Automating
every procedure does not always
provide the best solution; neither
does manually executing every
procedure. What does provide the
best solution is to consciously
examine all procedural operations
within a plant, document them and
determine what type of implemen-
tation will provide the best
economic return while improving
safety, health and the environmen-
tal metrics for the facility.
The criteria for how to implement
a procedure can vary between
Operational error
Mechanical failure
Unknown
Process upset
Natural hazard
Design error
Sabotage/arson
0 25 50 75 100
US$, millions
Figure 2 Average loss per major incident by cause Source: J & H Marsh & McLennan Inc
wilkins.indd 3 8/6/11 13:15:59
had an impact on product quality
and production efciency. It also
caused the inefcient use of utilities
such as fuel gas, power and cool-
ing. Additionally, with junior or
less experienced operators, there
was a higher incident of operational
errors, resulting in abnormal condi-
tions and off-spec product.
Yokogawas engineers worked
with the operational staff at the
renery. They interviewed the board
operators from different shifts and
were able to uncover and document
their best practices. For instance,
when ramping up feed tempera-
tures, junior operators would
typically ramp feed temperatures at
a linear rate throughout the temper-
ature zones. The veteran operators
had the operational experience to
change the temperature ramp at
different rates, depending on the
temperature zone of the column. It
was also discovered that operators
typically had to make more than 100
adjustments to the process through
the DCS system during the switcho-
ver. This was in addition to
responding to false alarms that were
set for normal operating conditions.
Implementing MPA methodology
and automation enabled the rener
to make signicant improvements in
the operations switchover time to a
predictable four-and-a-half hours
regardless of which operator was on
shift. Crude switchover reduced the
operators workload signicantly,
with over 100 control system
24 PTQ Q3 2011 www.eptq.com
industries and companies.
Characteristics of procedures that it
may be wise to automate are those
that are repeatable, demand fast
reactions, are well understood and
can reduce variability. While
procedures that cannot be dened
in algorithms or sequences, do not
require fast responses or low varia-
bility, are well known and not time
consuming may be better candi-
dates to be implemented as manual
or prompted procedures.
MPA can be implemented as
either a top-down or bottom-up
process. With a top-down approach,
the focus is initially put on product
knowledge to decide how the equip-
ment capabilities should be used to
most efciently and consistently
produce the product. In a bottom-up
process, identication of equipment
capabilities available to production
and a denition of limits that dene
the safe and desirable operating
envelope for any product would be
a starting point to build product-
related procedures.
Modular procedural automation
supports any level of procedural
hierarchies. In some cases, it may
be appropriate to automate only
equipment modules, such as pump
sets, dryers and manifolds, and not
product-specic procedures. In
other cases, multi-level product-
oriented procedural hierarchies
may be required to automatically
produce products, as is often done
in batch control applications.
Examples of applications are:
SOPs dening the sequence of
activities to start, stop or transition
a process unit
Groups of SOPs that dene how
to bring a production line or plant
up or down
SOPs that dene normal operat-
ing conditions for a continuous
plant
Consolidation of operating
procedures into best practices (see
Figure 3)
Starting and stopping pump sets
(the collection of a pump and asso-
ciated valves and instrumentation)
Transition from one operating
state to another due to higher or
lower production rates, grade
changes or crude switching (see
below)
Recovery from process upsets
Specic unit operations such as
decoking.
Case study: oil renery crude switch
An oil renery in Japan in normal
operations underwent crude oil
feedstock switches two or three
times a week. The efciency of the
operation depended on the experi-
ence and skill of the board operator
running the distillation unit. With a
skilled operator, the time to reach
normal steady-state operations was
typically ve hours. A junior, less
experienced operator could take
more than eight hours to reach the
same normal steady-state opera-
tions. These long transition times
A B
C
D
E F
A B
C
D1
D2
E F
Procedure of Operator A
A
B1
B2
C
D
E F
Procedure of Operator B
Procedure of Operator C
A
C
D1
D2
E F
Best-practices procedure
B1
B2
Figure 3 Consolidating operating procedures into best practices
wilkins.indd 4 8/6/11 13:16:10
www.eptq.com PTQ Q3 2011 25
adjustments reduced to 10 and more than 2000 process
alarms reduced by confguring the system to be opera-
tionally aware of process conditions. Additionally, there
was increased process knowledge sharing, a signifcant
reduction in operator errors and reduced operator
training.
ISA106: a procedural standard for continuous
process operation
In view of the need for an industry standard, the
International Society of Automation approved the
formation of a new standards committee in April 2010,
and the committee adopted the title Procedural
Automation for Continuous Process Operations. The
purpose of the committee at its frst meeting in June
2010 was to develop a standard, recommended prac-
tices and technical reports for the lifecycle of automated
procedures for process operations in industries.
Topics being considered for the standard are:
Models and terminology
Modularisation of procedural steps to foster reuse
and lower total cost of ownership
State models for procedural logic
Process unit orientation with operational
perspective
Exception handling of abnormal process conditions
Implementation of startup, shutdown, abnormal
conditions, hold states and transition logic
Recommended target platform (ie, control system
versus safety system) for different types of procedures
Training and certifcation best practices.
So far, these topics have input and support from 39
manufacturers and suppliers, including many compa-
nies that run distillation operations.
The committee has met face to face three times and
held monthly teleconferences since June 2010 and now
consists of fve working groups preparing material for
the initial technical report. The groups are:
Knowledge management
Defnitions and terminology and models
References (other standards and industry
publications)
Examples and use cases
Work process.
The initial goal is to publish a technical report based
on the good practices used today, with a target release
of June 2011. Following that, the committee plans to
refne the material and publish a standard.
Maurice Wilkins is Vice President of Yokogawas Global Strategic
Technology Marketing Center (USMK) in Dallas, Texas. He was head of
the team that proposed the ISA106 standard to the ISA Standards and
Practices Board.
Email: maurice.wilkins@us.yokogawa.com
Marcus Tennant is a Principal Systems Architect with Yokogawa
Corporations Global Strategic Technology Marketing Group in
Carrollton, Texas. He holds a BS in chemical engineering from Michigan
State University and an MS in operations and technology management
from Illinois Institute of Technology.
Email: marcus.tennant@us.yokogawa.com
SAMSON AG MESS- UND REGELTECHNIK
Weismllerstrae 3
60314 Frankfurt am Main Germany
Phone: +49 69 4009-0 Fax: +49 69 4009-1507
E-mail: samson@samson.de
Internet: www.samson.de www.samsongroup.de
A
0
1
0
5
1
E
N
Know-how
for your success
Due to global compettion, compa-
nies can only be sucessful on the
marketplaceif they produce out-
standingquality cost-efciently.
High-quality control valves and
accessories with low cost of owner-
ship are what it takes for an eco-
nomic production.
With 47 largely independent sub-
sidiaries and over 100 engineering
and sales ofcesspread across the
world, SAMSON ensures the safe-
ty and environmental compatibility
of your plants on any continent.
www.ptqenquiry.com
for further information
wilkins.indd 5 8/6/11 13:16:22
On November 1 2010, INTERCAT Inc became part of the Johnson
Matthey group of companies. An important step in Johnson Mattheys
strategy in the refnery catalyst market.
Johnson Matthey
is proud to present
www.jmcatalysts.com/refneries
UK
Tel +44 (0)1642 553601
Fax +44 (0)1642 522542
Catalysts for hydrogen
production, the market leaders.
Speciality purifcation solutions,
absorbants and process
technology for sulphur, chloride,
and mercury removal.
A complete service package for
PURASPEC
JM
catalysts.
FCC additives and addition
systems.
Services to improve plant
performance and proftability.
Hydroprocessing catalysts in
conjunction with SINOPEC.
Specialist measurement and
diagnostic services, from
column scans to FCC studies.
USA
Tel +1 630 268 6300
Fax +1 630 268 9797
Activate your potential
www.ptqenquiry.com
for further information
j mathey.indd 1 7/2/11 20:57:50
Naphtha to diesel via novel routes
I
ncreasingly, refners are having
to look for alternative ways to
dispose of naphtha. Traditional
markets such as petrochemical
naphtha sales are suffering compe-
tition from Middle Eastern
gas-based steam crackers. Also,
given the considerable amount of
spare gasoline production capacity
in the world refning system, oppor-
tunities to export regular gasoline
to markets such as the US will
continue to decline and reduce the
opportunity to incorporate naphtha
into the gasoline pool.
Rebalancing of the refnery prod-
uct slate away from naphtha/
gasoline towards diesel is required
to meet the projected global diesel
demand. Dieselisation is a term
commonly applied in the oil refning
industry to the practice of shifting
yields towards diesel by a combina-
tion of crude selection, low-severity
fuid catalytic cracker (FCC) opera-
tion, and increased hydrocracking
and residue upgrading.
More radical options, which are
explored in greater detail in this
article, include the conversion of
naphtha to heavier distillates via
steam reforming and subsequent
middle distillate synthesis, or
polymerisation (oligomerisation) of
cracked naphtha material. Case
studies based on an average
European refnery confguration
and crude feed indicate the poten-
tial that could be achieved.
Outlook
Growth in demand for diesel is
outpacing that of gasoline, and
the global diesel balance appears
to be tightening. So, if refners
cannot export gasoline, how about
An unconventional, synthetic approach to converting LPG and naphtha
components could boost diesel production
DAvID GIbbOns
Foster Wheeler
naphtha? Naphtha is also under
pressure and, with signifcant
world-scale ethane-based steam
crackers continuing to come on-
stream, the market for export
naphtha is expected to reduce.
European naphtha-based steam
crackers have also become less
competitive.
1
All of these drivers come together
and give increased impetus to the
shift towards diesel production.
However, rather than looking at
ways to increase diesel only, this
article looks at ways to achieve both
an increase in diesel output and a
decrease in naphtha and gasoline
production.
Average European refnery
Before looking at some of the
options, let us consider an average
European refnery
2
based on
average crude and downstream
process unit capacities. This refn-
ery has both an FCC unit and a
hydrocracker, with a visbreaker for
residue upgrading. This is a simpli-
fcation, of course, since most
European refneries have only one
main vacuum gas oil (VGO)
upgrading unit, but there are a
handful of sites that do, in fact,
have such a confguration.
Assuming an average crude slate
of North Sea Brent and Middle East
Arab Light, and with the chosen
process units, there is a good match
between this average feedstock and
the average refnery, based on a
simple generic refnery model with
typical product cut points (see
Table 1).
With this confguration, an aver-
age 135 000 b/d refnery already
produces more diesel than gaso-
line, although the total of LPG +
naphtha + gasoline is about the
same as diesel (see Table 2).
Production of jet/kerosene is
minimised.
Using average 2010 crude and
product prices (based on a typical
value of $80/bbl crude), there is
around a $20/bbl differential in the
price of diesel above that of gaso-
line.
3
If the average refnery can
achieve an extra 10% of diesel
production, this could be worth
around $40 million annually.
However, this gives us only around
$200 million as a reasonable invest-
ment (assuming a generous
fve-year simple payback). Therefore,
the challenge is to fnd the appro-
priate simple, low-cost technologies
to achieve the desired increase in
the diesel-to-naphtha plus gasoline
ratio.
www.eptq.com PTQ Q3 2011 27
Capacities, kbpd OGJ average
2
Model
Crude 135.7 136.0
Vacuum 55.9 55.0
Visbreaker 14.1 19.0
FCC 21.2 19.3
Catalyticreformer 20.2 21.2
Hydrocracker 11.6 11.2
Alkylation 2.3 2.3
C
5
/C
6
isomerisation 5.3 5.3
Average European refnery confguration
Table 1
Product bpd ktpa
LPG 9280 289
Naphtha 9939 423
Gasoline 34005 1471
Diesel 56907 2733
Average European refnery confguration
Table 2
fw gibbons.indd 1 8/6/11 13:54:51
28 PTQ Q3 2011 www.eptq.com
investment options, but what about
some unconventional approaches?
First, refners should consider again
what it is they are looking for.
Ideally, they want to convert naph-
tha and gasoline boiling range
material into diesel boiling range
material, and could also consider
the use of LPG as an additional
component for conversion.
But what are they looking to
create? What are the important
qualities of diesel? We have already
mentioned boiling range, but one
key quality is cetane (number or
index), which is a measure of the
ignition delay in a diesel engine,
with higher values producing a
shorter delay with more effcient
combustion. This encourages refn-
ers to consider the chemical nature
of the many components that
comprise a typical diesel fuel.
Cetane is the reverse of octane, such
that low-octane normal paraffns
have the highest cetane, followed
by branched or iso-paraffns, then
olefns, naphthenes and fnally
aromatics, which are really the
worst cetane components to be
included.
It is interesting to note that the
density range of diesel is very
narrow, 820845 kg/m
3
or 0.82
0.845 SG. And, of course, refners
want minimal sulphur and prefer
not to have any aromatics.
Diesel quality: carbon number
Now that we are considering the
chemical nature of diesel, we
should look at what chemistry
might be involved. If we look at the
ideal boiling range of 180360C, it
is clear that we need to consider
molecules with between 1022
carbon atoms.
Figure 1 (derived from various
sources
4,5
) shows lines for n-paraffns
and 1-olefns, which are easy to
plot, but iso-paraffns and other
components fll a broad band in the
middle due to their many isomers.
So, we are looking for roughly 10 to
20 carbon atoms in our molecules.
Diesel quality: hydrocarbon species
The very narrow density range was
mentioned earlier, and Figure 2
(derived from various sources
6,9
)
shows the very small window of
Dieselisation: conventional
(easier) options
Some of the easier options that
refneries can consider are crude
selection, cut points, operating
mode and catalysts.
Crude selection can make a large
difference, but there are other
constraints on a refnery that can
limit the choice. Refneries typically
produce diesel material from the
jet/gas oil boiling range of 180
360C TBP cut point, but can
effectively go a little higher with
upgrading facilities, such as a mild
or even a full hydrocracker process-
ing VGO, or a visbreaker, for
example, processing atmospheric
or, more usually, vacuum residue.
With the average refnery confg-
uration as before, it might be
possible to achieve up to 10% more
diesel output by the appropriate
choice of crudes, but, again, the
confguration limits the ability to
process varying yields of interme-
diate products and their sulphur
contents, while maintaining crude
throughput.
A number of process units are
designed with some ability to
switch from gasoline to diesel, such
as the FCC unit, or from naphtha to
diesel, such as a hydrocracker.
However, there is a limit to what
refneries can achieve and how far
they can go with operational
changes. Catalyst additives or
complete unit catalyst inventory
changes are often required.
If we consider the FCC unit,
refners can drop conversion and
increase light cycle oil (LCO) yield,
but this is a poor-quality compo-
nent with many qualities that need
upgrading. A typical LCO cetane
number is around 20, compared to
a range of 4060 for straight-run
and hydrocracked distillates, which
is much closer to the fnished prod-
uct specifcation of 51 or above,
and sulphur content is usually far
above 10 ppm for the fnished
product. While the FCC unit
contributes around 20 vol% to the
gasoline pool, LCO is only 5% of
the diesel pool, and with opera-
tional changes a refner might
achieve an increase in overall
diesel production of perhaps only
another 5% overall.
Dieselisation: conventional options
bottoms-up approach
If an existing refnery has already
achieved as much as possible in
terms of crude selection and opera-
tion in diesel mode, traditionally
refners would take a bottoms-up
approach. For instance, they would
look at technologies to break or
crack larger, higher boiling point
components into smaller, lower
boiling point components.
If a refner already has a
visbreaker and a hydrocracker, they
might typically look at adding a
coker to give more conversion than
a visbreaker, or a residue desul-
phurisation (RDS) unit to produce
additional distillate.
While using a coker instead of a
visbreaker can give a large increase
in diesel output (perhaps 30%), it
also produces more LPG, naphtha
and gasoline components, which all
require additional treatment. A
vacuum RDS (VRDS) unit can also
give a sizeable increase in diesel
(perhaps 20%), but, again, produces
more of the lighter components.
Another option is to include a
solvent deasphalting (SDA) unit, but
this provides only a small increase
in overall diesel production (23%).
Gasifcation is rather an extreme
solution for residue upgrading
within a refnery, but it does intro-
duce the idea of using synthesis
gas, or syngas, which is covered
later in this article.
One fnal bottoms-upgrading
option to consider is residue FCC
(RFCC), but, on its own, that will
do the opposite of what we want
and produce overall more gasoline
and light olefns, and lead to a
reduction in diesel make.
As the conversion level increases
in these conventional bottoms
upgrading processes, so do capital
investment and operating costs.
Many smaller European refneries
would struggle to justify the cost of
these relatively small process units,
which would not have the beneft
of economy of scale that world-
scale units would have.
Dieselisation: unconventional
options frst requirements
The previous section gave a very
brief overview of conventional
fw gibbons.indd 2 8/6/11 13:55:03
/#0+PUKFG
4GHKPGT[UQNWVKQPUHQTVJGNQPITWP
We prov|de essent|a| components for the efc|ent operat|on of a broad range of app||-
cat|ons |n rener|es. As key components |n u|d cata|yt|c crack|ng |FOO un|ts, our ax|a|-
and centr|fuga| compressors are des|gned to ach|eve the h|ghest |eve| of ava||ab|||ty
atta|nab|e today. So that tomorrow, your p|ant |s runn|ng w|thout |nterrupt|on, and w|th
very ||tt|e t|me devoted to ma|ntenance. lt |s our goa| to de||ver products that perform to
the very h|ghest standards of re||ab|||ty - your standards. Our approach has made us
the wor|d market |eader |n FOO p|ants, and has earned us the trust of customers ||ke
you. Your standards, our goa|. F|nd out more at www.mand|ese|turbo.com
/CTKPG2QYGT2NCPVU6WTDQOCEJKPGT[2TKOG5GTX
)GCTV[RGEQORTGUUQTU
%GPVTKHWICNEQORTGUUQTU
#ZKCNEQORTGUUQTU
'ZRCPFGTU
*KIJRTGUUWTGEQORTGUUQTU
5VGCOVWTDKPGU
2TQEGUUICUUETGYEQORTGUUQTU
MAN Diesel & Turbo_Hydrocarbon Engineering_210 x 297mm_issue 09/2010 10.08.10 18:57
www.ptqenquiry.com
for further information
man.indd 1 8/6/11 12:48:01
acceptable EN590
7
diesel quality
that we are aiming for, compared
with the wide range of cetane and
density that is possible from the
different molecular species.
It is clear that there is not one
component that will work on its
own, but instead we need a blend
of different components; for
example, some of the simpler
(mono-branched) iso-paraffns and
alkyl-benzenes.
Another important consideration
is that, in the context of an existing
refnery, any new diesel compo-
nents that can be created will be
blended with a much larger exist-
ing diesel pool (containing a blend
of hydrotreated straight-run and
hydrocracked diesel, often with a
30 PTQ Q3 2011 www.eptq.com
proportion of hydrotreated and
hydrocracked kerosene), so a refner
can afford to be off-specifcation in
some properties.
Dieselisation: unconventional
options second requirements
So what are we starting from?
Ideally, refners want to take the LPG
and naphtha components (especially
if any are in surplus) and convert
them. They would be starting from
components with between three and
10 carbon atoms, and the sources are
those streams that go to the naphtha
pool, as well as streams that go to the
gasoline pool.
Refneries produce mostly
paraffnic naphtha from a crude
unit and hydrocracker, plus
aromatic and olefnic gasoline
components (from the catalytic
reformer and FCC unit respec-
tively). Using unconventional
routes, we want to make simpler
iso-paraffns (mono-branched) and
alkyl-aromatics or alkyl-cycloparaf-
fns and blend these new
components into an existing,
conventional diesel pool.
Unconventional options: how do we
get there? Top-down approach
In contrast to the bottoms-up
approach of the refner, we can
consider the top-down approach of
the chemist by synthesising desired
molecules from smaller molecules.
There are two routes and two differ-
ent chemistries to achieve this.
Oligomerisation is the joining
together of molecules, and this could
be thought of as polymer diesel, or
poly diesel, similar to poly gasoline
that has been used on a small scale
for gasoline production in some
refneries. (By January 2011 in
Europe, there was 47 000 b/d of
polymerisation capacity
2
compared
to gasoline demand of 1.9 million
b/d.
8
Polymerisation therefore repre-
sents around 2.5 vol% of the
European gasoline pool.)
There are technologies and ongo-
ing research to take a light olefn
(propylene or butene in olefnic
LPG) and react it with a heavier
olefn, such as C
5
/C
6
olefns in light
FCC naphtha. The product itself is
olefnic, and it would need saturat-
ing with hydrogen to make a
normal paraffn in order to avoid
diesel fuel stability problems
through oxidation of olefns and
resultant gum formation.
Another technology that extends
existing technology (for example,
cumene) is to take these olefn
components (from LPG and FCC
light naphtha) and react them with
aromatics (in reformate and FCC
heavy naphtha) using aromatics
alkylation. These components
would be hydrogenated to produce
alkyl-cycloparaffns. But, as the
cetane/density chart indicates (see
Figure 2), some of these compo-
nents are a little low in cetane.
For the ultimate in top-down
chemistry, we could go back to
basics and consider synthesis gas,
300
400
350
250
200
6 8 10 12 14 16 18 20 22 7 9 11 13 15 17 19 21 23 24
C


,
t
n
i
o
p

g
n
i
l
i
o
B
Number of carbon atoms
150
1-o|ens
Mono branched parafns
n-a|ky| benzenes n-parafns
Bi-cyc|ic
Figure 1 Diesel component properties
60
100
120
80
40
20
700 750 800 850 900 950
C
e
t
a
n
e

n
u
m
b
e
r
Density, kg/m
3
0
EN590
limits
1-olefins
n-paraffins
n-cetane
Bi-cyclic
Mono branched paraffins
i-cetane
n-alkyl benzenes
Figure 2 Hydrocarbons contribution to diesel quality
fw gibbons.indd 3 8/6/11 13:55:16
yield and high cetane, and compro-
mise on density, as they would be
blending these new components
into a conventional refnery diesel
pool, which is typically constrained
more by cetane, and where there
might be an opportunity to intro-
duce more high-density components
to the blend.
Assume LT-FT
Considering the average refnery
and taking paraffnic naphtha to
make syngas, a refner might
achieve yields based on a syngas
feed of 50% diesel and 20% kero-
sene, with a typical LT-FT plant
targeting maximum distillate. The
remaining yields (on syngas feed)
could typically be 25% naphtha and
5% C
s
and lighter, and since the
naphtha is highly (95%) paraffnic it
would make a suitable steam
reformer feed.
Assuming that some of the kero-
sene would be blended into the
diesel, and assuming up to a 60%
diesel yield from LT-FT, this would
give an overall increase in diesel of
8% for the average European
refnery.
Another point to consider is that
FT processes can create olefns as
well as paraffns, and with the opti-
mum scheme it may be possible to
supplement the requirement of the
heavier olefns for poly diesel or
alky diesel production.
www.eptq.com PTQ Q3 2011 31
which is available relatively cheaply
from naphtha steam reforming
rather than gasifcation, but down-
stream refners could employ
conventional Fischer-Tropsch (FT)
synthesis technologies. Such tech-
nology could be referred to as
naphtha to (other) liquids, or NTL.
Some theoretical yields
Create from two smaller molecules
Returning to the average European
refnery, a refner can take olefnic
FCC LPG and light naphtha and
create some poly diesel. The refner
could achieve 50% conversion of the
FCC light naphtha olefns, limited
by the availability of propylene, and
be able to produce an additional 4%
diesel production overall with, of
course, a corresponding decrease in
LPG and gasoline production.
Alternatively, the heavier FCC
olefns (C
7
+) could be reacted with
aromatics (alkylation). Again
assuming only 50% conversion, and
the fact that the availability of
propylene is limited, an additional
5% of diesel overall might be avail-
able via additional alky diesel.
Of course, both steps could be
taken, at which point the availabil-
ity of the heavier olefns would set
a limit, although this could be
overcome by also considering dehy-
drogenation of C
5
+ paraffns. Based
on the average European refnery
confguration, both steps might
give an increase of around
7% in diesel production overall.
Although RFCC on its own was
earlier dismissed as a residue
conversion option, the fact that it
creates high yields of light olefns,
as well as heavier olefns and
aromatics, could help create the
right balance of these components
to maximise poly diesel yields in a
refnery upgrading confguration.
Back to basics
The other unconventional route,
NTL, uses a conventional steam
reformer and a conventional FT
process, but put together in an
unconventional way. A steam
reformer designed for methane feed
normally gives too much hydrogen,
with a H
2
:CO ratio of 3:1 compared
to the ideal ratio of around 2:1 for
FT feed. But with higher hydrocar-
bons, and typically with a
straight-run naphtha feed, the
H
2
:CO ratio is closer to 2.1:1 and
therefore ideal for FT.
10
While FT is not considered in
detail in this article, we can say that
low-temperature FT (LT-FT) is
better suited to higher yields of
diesel.
11
However, in the same way
that poly diesel or alky diesel is not
ideal, FT diesel is also constrained
by what is commonly referred to as
the cetane/density/yield triangle.
While refners cannot maximise all
three, they could aim for a high
Kerosene
Diesel
Naphtha
PP/BB
splitters
Steam
reformer
Oligomer-
isation
Aromatics
alkylation
LT
FT
Hydrogenate
Hydrogenate
FT
refinery
Diesel
Kerosene
Diesel
PT liquids
Kerosene
Diesel
Naphtha
Naphtha
Sat. LPG
>50%
olefins
PCC
LPG
C
3
, C
4

Naphtha
PCC
Light naphtha
>50%
aromatics
>70%
aromatics
CCP
Naphtha
SP, HCP
Naphtha
Syngas
2:l
PCC
Heavy naphtha
Figure 3 From naphtha to diesel via oligomerisation, aromatics alkylation and low-temperature Fischer-Tropsch
fw gibbons.indd 4 8/6/11 13:55:28
RENTECH breaks new trails in the
boiler industry with its focus on custom
engineering and design.
Theres no on the shelf inventory at RENTECH because we design and build each and every
boiler to operate at peak effciency in its own unique conditions. As an industry leader, RENTECH
provides solutions to your most demanding specifcations for safe, reliable boilers. From design
and manufacture to installation and service, we are breaking new trails.
www.ptqenquiry.com
for further information
rentech.indd 1 8/6/11 11:47:01
Maximum potential:
combined oligomerisation,
aromatics alkylation and LT-FT
Figure 3 shows how to get from
naphtha to diesel with a combina-
tion of:
Oligomerisation
Aromatics alkylation
Steam reforming and LT-FT
where light and heavy olefns, and
light aromatics are converted into
higher carbon number paraffns and
other components of poly diesel,
alky diesel and FT diesel, and
where a refner might be able to
achieve greater than a 10% increase
in diesel production, with a corre-
sponding reduction in LPG,
naphtha and gasoline production.
Conclusions
The demand/supply outlook is
more positive for diesel than for
naphtha and gasoline, making
increased diesel production an
attractive opportunity. However,
there is a limit to what can be
achieved with an existing refnery
in terms of crude selection and
operating modes.
www.eptq.com PTQ Q3 2011 33
References
1 Davis A, Outlook 10: Mideast naphtha
capacitycastsshadowoverEurope,ICISNews,
24Dec2009.
2 2010WorldwideRefningSurvey,Oil & Gas
Journal,6Dec2010.
3 ICIS Pricing 6-month delay sample spot
pricedata,www.icispricing.com.
4 Yaws C L, Thermophysical properties of
chemicals and hydrocarbons, 2008, ISBN 978-
0-8155-1596-8.
5 Diesel Fuels Technical Review, Chevron
ProductsCompany,2007.
6 Compendium of Experimental Cetane
Number Data,NREL,Sept2004(Ref:NREL/SR-
540-36805).
7 European Norm (Standard) EN590:2004
Automotive fuels Diesel Requirements
andtestmethods,CEN(EuropeanCommittee
forStandardisation).
8 Data from IEA Oil Market Report, 12 Apr
2011,www.oilmarketreport.org.
9 De Klerk A, Fischer-Tropsch Refning, PhD
thesis, University of Pretoria, Feb 2008 (Ch 9,
Fig15).
10 Ibid.(Ch5,Sec2,2,1).
11 Ibid.(Ch5,Fig6).
David Gibbons isaPrincipalProcessConsultant
in the Business Solutions Group at Foster
Wheeler in Reading, UK, where he undertakes
refnery and petrochemical plant feasibility
andconfgurationsstudiesforclients.
A conventional bottoms-up
approach would look at adding a
coker or residue hydrocracker,
which can yield signifcantly more
diesel, but these are also high-cost
upgrades.
This article has discussed
some unconventional top-down
approaches to convert lighter LPG
and naphtha components via poly
diesel, alky diesel and FT diesel,
and there are some good synergies
between the processes that could be
exploited.
A number of supporting technol-
ogies would be required to produce
the fnished diesel components.
While the total cost could still be a
hurdle to a refner who might have
only $200 million to invest in an
average European operation, it is
clear that some existing technolo-
gies and ongoing developments
could be applied in an unconven-
tional way to address the issue of
dieselisation.
Thisarticleisbasedonapaperpresentedatthe
ERTC15thAnnualMeeting,Istanbul,29Nov
1Dec2010.
www.ptqenquiry.com
for further information
fw gibbons.indd 5 10/6/11 14:07:41
From Engineering Trough Fabrication
www.aistco.co 24hr Mk0h0Y $kVI0 1-800-839-6374
For top performing mass transfer and tower internal components let
AMISTCO Separation Products use your existing drawings or modify them
to improve your process. Our experienced in-house engineering and full
scale fabrication capabilities can streamline your project from inception
to installation.
Catalyst Bed Supports
Basket Strainers
Nozzles
Outlet/Inlet Baskets
Distributors
Hub and Header Laterals
0I$TkI8T0k$ & $FF0kT$
Manufactured to customer
specifications or engineered to
meet performance requirements
00AL$0k$
0il water separations haze removal
from fuels Removal of tower wet reflux
Caustic treater applications
$Tk0Tk0 FA0kIh0
woven, sheet metal, and knitted
structured packing. 8uilt to spec
or performance requirement.
Mesh numerous alloys & plastics
Vane lChevron} 0ouble Pocket Vane
Candle or Fiberbed lnsertion MistFix

MI$T LIMIhAT0k$ kAh00M FA0kIh0


Valve lnumerous options} 8affle
$ieve or perforated 0ual flow
8ubble cap trays Cartridge trays
TkAY$
wide variety of random packing types,
sizes and materials in stock! Ask us how
our patented $uper8lend 2-Pac can
increase your capacity and efficiency.
www.ptqenquiry.com
for further information
amistco.indd 1 7/2/11 10:28:14
Optimising hydrogen production and use
H
ydrogen plays a critical role
in the production of clean
fuels, and its use has
increased with the introduction of
low-sulphur gasoline and diesel
fuels. The reduction of benzene in
gasoline via benzene saturation will
also increase hydrogen consump-
tion, as will the trend toward diesel
cetane improvement and aromatics
reduction.
Changes in marine fuel oil speci-
fcations are also expected to
increase hydrogen demand. In 2008,
MARPOL Annex VI regulations
were passed, setting a framework
for regional and global specifca-
tions on marine fuel oil quality.
These regulations are expected to
further increase the demand for
hydrogen for desulphurising resid-
ual fuel oils and, through the
increase in distillate fuel demand,
to replace residual fuel oil in marine
fuels (see Figure 1).
The overall reduction in demand
for heavy fuel oils has encouraged
many refners to install bottoms
upgrading capacity such as delayed
coking units. The streams produced
by coking typically contain higher
contaminant levels (sulphur and
nitrogen) than the equivalent
straight-run streams. Hydrotreating
these coker products has also
increased hydrogen consumption.
Hydrocracking has become
increasingly important for convert-
ing heavier crude fractions into
high-quality clean fuels. Increased
reliance on hydrocracking for clean
fuels production has also led to a
rise in hydrogen consumption. A
hydrocracking unit is typically the
largest hydrogen consumer in the
refnery, and hydrogen can account
Knowledge of hydrogen producing and consuming process technologies, systems
analyses and process controls can be leveraged to optimise hydrogen use
Ronald long, KatHy Picioccio and alan ZagoRia
UOP LLC, A Honeywell Company
for more than 80% of the units
operating cost (see Figure 2).
The quality of crude oil is gradu-
ally declining. Globally, crude API
gravity is declining and the sulphur
content is gradually increasing (see
Figure 3). Both of these trends in
crude quality will contribute to
increased hydrogen consumption
during refning. The use of synthetic
crudes derived from oil sands and
other unconventional sources is
expected to increase to 2 million
b/d by 2020. These synthetic crudes
will require additional hydrogen to
be refned into usable products.
One of the major sources of
hydrogen and gasoline pool octane
is the catalytic reformer. The blend-
ing of ethanol has reduced the
octane requirements from other
refnery product streams to main-
tain the gasoline pool octane; often,
refners respond to this situation by
reducing the catalytic reformers
severity to produce a lower-octane
reformate product. However, a
lower catalytic reformer severity
typically produces less hydrogen.
Lower hydrogen production from a
lower severity operation is in oppo-
sition to the increased demand for
hydrogen in the rest of the refnery
and compels the refner to obtain
hydrogen from other sources. Some
refners have decided to operate
their catalytic reformer for hydro-
gen production and tolerate some
www.eptq.com PTQ Q3 2011 35
3.0
5.0
4.0
2.0
1.0
2005 2010 2015 2020 2025 2030

S
u
l
p
h
u
r
,

w
t
%

m
a
x
0.0
ECA
Global
Global delayed
Figure 1 Bunker fuel oil sulphur specifcations
Catalyst
1% Utilities
15%
Hydrogen
84%
Figure 2 Typical hydrocracker operating
costs
zagoria.indd 1 8/6/11 14:11:14
36 PTQ Q3 2011 www.eptq.com
target hydrogen partial pressures,
process changes in producers and
consumers, catalysts, operating
procedures, revamped and new
purifcation capacity, pressure
swing absorption (PSA) unit feed to
product bypass, feed to hydrogen
plant, compressor modifcations,
ability of LP models to accurately
represent hydrogen availability
constraints, and header pressure
control system improvements.
For the optimisation of hydrogen
use, the benefts are driven by iden-
tifying and alleviating critical
constraints in the refnery-wide
hydrogen network. Every refnery
is different and, from time to time,
the active constraints in a refnery
can change with different crudes or
operating objectives. A refnery
network may be constrained by
total moles of hydrogen available,
hydrogen purity, hydraulics, puri-
fer capacity, compression, H
2
S
scrubbing, fuel system constraints
or other refnery-specifc issues.
What is ultimately constrained is
a refnerys proftability. When
there is insuffcient quantity or
purity of hydrogen, charge rates,
the processing of more diffcult
feeds or product quality are limited
and refnery margins are reduced.
When suffcient hydrogen is availa-
ble, the effect of ineffciencies is
higher operating (hydrogen produc-
tion) costs.
Hydrogen and process performance
Hydrogen has a signifcant effect on
process performance and proftabil-
ity. Hydrogen partial pressure, a
variable completely under the
operators control, can be utilised
to increase catalyst life in
hydroprocessing units, increase
throughput, increase conversion,
improve product quality, or process
more proftable feeds. The potential
benefts of affecting refnery
proftability through hydrogen
management are much greater than
those from simply reducing hydro-
gen production or purchase costs.
Of course, the benefts must be
considered in concert with the capi-
tal and operating costs of increasing
the hydrogen partial pressure in
order to determine the most proft-
able targets.
octane giveaway in the gasoline
pool.
Many refners produce or
purchase hydrogen to have a suff-
cient supply available for their
refnery. The steam reforming (SR)
process is used to produce most of
the additional hydrogen required
by refners. The cost of the
hydrogen produced is directly
proportional to the feed costs. In
the US, most of the hydrogen is
produced in steam methane reform-
ing (SMR) units and the cost is
typically tied to the price of natural
gas.
The production of hydrogen by a
steam reformer requires signifcant
energy; one tonne of hydrogen
produced requires 3.5 to 4 tonnes of
hydrocarbons as feed and fuel.
Hydrogen production can account
for up to 20% of refnery energy
consumption. Additionally, the
production of hydrogen generates
signifcant amounts of carbon diox-
ide (CO
2
); the production of 1 tonne
of hydrogen generates 812 tonnes
of CO
2
. Future environmental legis-
lation may regulate the amount of
CO
2
that can be generated and may
increase the costs of hydrogen
production.
Availability of hydrogen is a
requirement for the production of
clean fuels, and demand for hydro-
gen is at an all time high.
Anticipated future trends and regu-
lations are expected to further
increase hydrogen consumption. At
the same time, the production of
additional hydrogen is expected to
become more expensive.
While it is well understood that
the ability of a refner to produce
clean fuels depends on having suff-
cient hydrogen, what many refners
recognise is that optimum use of
hydrogen will maximise refnery
profts.
Hydrogen network analysis and
improvements
Refnery hydrogen networks typi-
cally interconnect many producers,
consumers and purifcation units
with different pressures, purities
and operating objectives. The
network grows with each subse-
quent refnery project, modifed in
ways that minimise complexity and
the interruption of existing units
rather than for refnery-wide opti-
misation. Hydrogen production
costs and constraints on availability
are typically much greater than
when the network was frst envi-
sioned. All of these factors lead to
the conclusion that most operating
hydrogen networks are not opti-
mised for todays environment not
for the minimal cost of hydrogen
production, nor for maximised
refnery margins.
Optimising the hydrogen network
In many hydrogen optimisation
schemes, it often occurs that the
greater the number of degrees of
freedom, the larger the improve-
ment that is possible. The most
successful programmes for improv-
ing the hydrogen network draw the
largest possible envelope and take
advantage of all the knobs that
are available to turn, including
network connectivity, increased
hydrogen production capacity,
1.25
1.30
1.20
1.15
1.10
1.05
1990 1994 1998 2002 2006 2010
%

,
r
u
h
p
l
u
S

1.00
34
33
32
y
t
i
v
a
r
g

I
P
A
31
API gravity
Su|phur
Figure 3 Global crude quality
zagoria.indd 2 8/6/11 14:11:28
Establishing and faithfully maintaining the target
hydrogen recycle purity of key hydroprocessing units
is an important component of effective hydrogen
management. For refners who want to maximise the
effectiveness of their hydrogen network, further opti-
misation is possible when targets for recycle hydrogen
purity are modifed through major operating changes,
such as variation in charge rate, feed properties and
severity. The measured variable that represents hydro-
gen partial pressure is recycle purity. Make-up purity
and purge rate can both affect the hydrogen partial
pressure, but they do not determine it.
Essence of hydrogen network optimisation
The frst step in improving the hydrogen network is to
clarify the objective. The objective should always be
overall refnery proftability rather than hydrogen
production costs. Operating costs, capital costs and
refnery margins are all part of the picture.
At a high level, the process of hydrogen network
optimisation is:
Identify the constraint that is most limiting
proftability
Identify ways to alleviate the constraint and select
the most cost-effective approach. Consider all the
options listed above. Many improvements can be
implemented quickly without capital projects.
Repeat these steps until a constraint is reached that
can not be cost-effectively relaxed or alleviated
Utilising a broad range of tools makes comprehen-
sive optimisation possible.
Tools
Network analysis
Hydrogen network pinch analysis is a valuable analyti-
cal method to identify the theoretical minimum
hydrogen requirements for a given network through
unconstrained modifcation and connectivity (including
turning hydrogen recycle units into once-through and
cascading the purge to another unit). It provides a
target and a pinch purity that can guide the analyst in
proposing changes to the existing network. However,
pinch analysis is limited in a number of ways. Its key
limitations are that it is not concerned with the pres-
sures of sources and sinks, it treats all light hydrocarbon
components the same (in reality, methane has a much
greater tendency to build up in the recycle loop than
propane), it does not consider the signifcant effects on
compressors and hydraulics of changing a hydrogen
recycle hydroprocessing unit into a once-through unit,
nor does it consider sulphur levels.
To evaluate potential network improvements, a
detailed refnery-wide model of the hydrogen network
should be employed. One of the critical features of this
model (as opposed to a spreadsheet model) is that it
simulates the non-linear relationships between charge
rate, make-up rate, make-up purity, and recycle purity
(hydrogen partial pressure) and purge rate. With this
model, potential modifcations can be tested, such as
changing make-up source, trading off purity against
recovery in an existing purifer, adding new purifca-
www.eptq.com PTQ Q3 2011 37
ARCA
Flow Group
worldwide:
Competence in valves, pumps & cryogenics
Subsidiaries and partners in Switzerland, the Netherlands,
India, P.R. China, Korea, Japan and Mexico!
ECOTROL

control valve
Advantages,
that should not be kept quiet!
High reliability guaranteed by
precision manufacturing processes
and quality control
Emission control and leakage
conforming to the highest international
standards
Tubeless, integrated mounting of
positioners acc. to VDI 3847
Minimal working life costs
A range of awarded patents
Take advantage of the most technically
innovative control valve in a generation,
up to 400 mm!
ARCA Regler GmbH, P. O. Box 2120, D-47913 Tnisvorst
Phone +49-2156-7709-0, Fax -55, sale@arca-valve.com
www.arca-valve.com
The State
of the Art
solution!
Request
Technical
Information or
a Quotation:
T
h
e

Q
u
i
e
t

W
o
r
k
-
H
o
r
s
e
!
A
R
C
A
_
I
n
s
_
E
C
O
T
R
O
L
_
E
_
9
0
x
2
6
0
.
q
x
d
Armaturen AG
von Rohr
ARCA BV
von Rohr
www.ptqenquiry.com
for further information
zagoria.indd 3 10/6/11 14:08:35
estimates, we expected that an
internally generated LPG stream
would have to be burned to meet
the fuel balance. After energy opti-
misation of the preliminary design,
the fuel gas balance shifted to posi-
tive, and we were able to utilise
lower-value fuel gas as hydrogen
plant feed. As an additional beneft,
we could isolate hydrogen-rich fuel
gas streams and utilise them as
hydrogen plant feed. Not only is
fuel gas less valuable than LPG, but
the hydrogen content enables a
signifcant reduction in fring and
energy consumption in the hydro-
gen plant furnace compared to
other hydrocarbon feeds.
Optimising beyond the hydrogen
network
Broadening the optimisation enve-
lope even further, it is possible to
consider optimising more than just
the hydrogen system at one time.
There are often additional benefts
when several systems are optimised
together. For example, when design-
ing a refning complex with new
catalytic reforming and hydrocrack-
ing units, it is possible to consider
the hydrogen and LPG recovery
systems at the same time. As stan-
dalone designs, each unit would
have its own LPG recovery system
and produce moderate- purity
hydrogen to be purifed/recovered
(reformer net gas and hydrocracker
fash gas). The reformer net gas
contains the recoverable LPG and
hydrogen, while the recoverable
38 PTQ Q3 2011 www.eptq.com
tion, or changing the target
hydrogen partial pressure in a
hydrogen-consuming process unit.
The same model can be used by the
refner for operations and project
planning.
Process and purifcation models
Process models of catalytic reform-
ers enable the analyst to understand
the effects of operating, mechanical
or catalyst changes on hydrogen
yield and purity.
The hydrogen network model
predicts the effect of changing a
hydrogen consuming units target
recycle purity on the required
make-up and purge rates. A hydro-
processing model will predict the
effect of changing the hydrogen
partial pressure on process
performance (catalyst life, product
quality, conversion) and hydrogen
consumption.
Purifer (PSA and membrane)
models are used to predict the
impact of changing a purifers feed
composition and operating condi-
tions on hydrogen product purity
and recovery. These models are also
used to identify the potential for
debottlenecking existing purifers.
Refnery economics
The right question to ask is not
How much money am I spending
on hydrogen?, but Am I utilising
the right amount of hydrogen to
maximise my refnery margin?
With the addition of a refnery
economic model to the toolkit, the
analyst can consider how the addi-
tion of lower-value, more diffcult
feeds affect hydrogen consumption,
desired hydrogen partial pressures,
the hydrogen network and refnery
margins.
Design-phase full optimisation
We have had the opportunity to
optimise grassroots refneries before
individual process unit design bases
are set. When we optimise at this
stage of a project, all parameters are
available for optimisation and we
can be that much more effective in
minimising capital and optimising
refnery margins.
For example, in the design work
for a major new refnery for
Petrobras, we had full freedom to
route and recover streams, since
design pressures and make-up puri-
ties were not yet fxed. We set the
separator pressure of one hydrot-
reater such that the fash gas could
be sent to the suction of the make-
up compressor of another
hydrotreater and the revised make-
up hydrogen purity could be taken
into account in the design of the
consuming unit. The fash drum in
another unit was set at a pressure
such that its fash gas could easily
be purifed and recovered in an
existing PSA unit.
Given that no external fuel could
be purchased for this refnery, the
fuel balance was not only critical
from an economic standpoint but it
determined the feed selection for
the hydrogen plant. In our early
Capital project
66%
No/low cost
34%
Other
operations
15%
Catalyst
2%
Other
25%
H
2
partial
pressure
12%
Purification
38%
Control
improvements
3%
Change
flow piping
5%
Figure 4 Typical results: summary of benefts was $137 million/y in seven studies
zagoria.indd 4 8/6/11 14:11:53
hydrogen and LPG are found in the
purge and fash gas streams of the
hydrocracker.
LPG recovery is much more eff-
cient in the hydrocracker than in
the reformer because the gas stream
has a lower hydrogen concentra-
tion. Also, it is more cost effective
to process both hydrogen streams
in one PSA unit. It is possible to
integrate the two systems together
by designing a single PSA unit that
takes both hydrogen streams as its
feed and sending the PSA unit tail
gas to the LPG recovery system in
the hydrocracker. Since the PSA
unit concentrates LPG in the tail
gas, recovering the LPG from the
tail gas in the hydrocracker recov-
ery system requires less energy and
capital than if it was recovered
separately in the reformer. An addi-
tional beneft of the integration is
that a PSA unit product bypass can
now be integrated with the hydroc-
racker to enable optimisation of the
hydrocracker make-up purity a
degree of freedom that would not
otherwise exist.
Hydrogen network improvements
The potential for improving hydro-
gen effciency is estimated by
evaluating the hydrogen-containing
streams currently going to fuel,
fare and hydrogen plant feed. The
potential fnancial benefts will also
be a function of the value of hydro-
gen in a refnery. The potential
beneft of improvements in process
performance through hydrogen
optimisation can be estimated
roughly by evaluating current
constraints to process performance,
refnery drivers and refnery
economics. Our hydrogen manage-
ment studies generally identify $2
million to $20 million in annual
benefts.
In seven studies, UOP identifed
a total of $137 million in annual
benefts. A third of the opportuni-
ties identifed were no/low-cost
changes and the remainder required
capital projects (all with a simple
payback of less than two years).
While 38% of the opportunities for
improvement came from adding or
improving hydrogen purifcation
capacity, a much broader scope of
evaluation is required to achieve
www.eptq.com PTQ Q3 2011 39
these benefts. Operating changes
and better management of hydro-
gen partial pressure targets were
important, as well as many other
issues. Some of the improvements
found in these seven studies
include:
Operators adjusting the PSA
capacity factor to improve hydro-
gen recovery
Addition of a new purifcation
unit
Cascading the purge from an
isomerisation unit to the make-up
of another unit
Reducing the make-up purity of
a hydrocracking unit to reduce
hydrogen purifcation losses, while
still meeting the minimum hydro-
gen partial pressure target
Increasing the make-up purity to
a diesel hydrotreating unit to
improve process performance
Changing feed streams to an
existing membrane purifer to
obtain more effcient purifcation
Sending a hydrogen stream of
moderate purity to hydrogen plant
feed rather than fuel, reducing the
operating costs of the hydrogen
plant
Modifying a compressor to elimi-
nate hydrogen leaking through the
seals to fare. (Wasting hydrogen to
fare is much more costly than
wasting hydrogen to fuel.)
Replacing pressure safety valves
with pilot-operated valves so that
the reactor operating pressure can
be increased by 5%, enabling an
increase in hydrogen partial
pressure.
Options for hydrogen recovery and
purifcation
In the refning industry, high-purity
hydrogen can improve the perform-
ance of hydroprocessing units
(hydrotreaters and hydrocrackers)
by increasing the recycle gas purity
and the hydrogen partial pressure
in these units.
In existing hydroprocessing units,
the use of high-purity hydrogen to
increase the reactor sections hydro-
gen partial pressure can deliver the
following benefts:
Reduce the quantity of make-up
gas required
Enable the processing of more
feed
www.ptqenquiry.com
for further information
producing commercial quantities of
both green diesel and green jet.
;bhfZllmhebjnb]l?M\hfie^q
In addition to the Ecofning and
Renewable Fuel Jet Processes for
producing renewable middle distil-
late blendstocks, UOP is actively
developing biomass-to-liquids
(BTL) technology. One such route
to biofuels is based on the gasifca-
tion of biomass to produce a syngas
composed of carbon monoxide and
hydrogen. Materials such as wood,
wood waste, agricultural waste
(corn stover, straw), green waste
(lawn and tree clippings) and
energy crops can be fed to a gasifer
to produce syngas. The syngas
mixture is fed to a FT synthesis
reaction section, in which it is
converted to predominantly
paraffnic liquids and waxes. The
paraffnic liquids and wax are then
catalytically upgraded to produce
products including naphtha and
very high-quality jet fuel and diesel-
range fuel products or, alternatively,
chemicals feedstocks. A simplifed
block fow diagram of a FT synthe-
sis complex is shown in Figure 8.
In 2008, UOP and Rentech formed
an alliance to offer a single solution
to resource owners for producing
refned diesel and jet fuels from
biomass via the BTL route. Rentech
will provide technology for the gasi-
fcation and FT synthesis sections of
the complex, while UOP will
provide technology for synthesis gas
clean-up, hydrogen purifcation and
upgrading the FT liquids and waxes
to end products such as premium
fuels, chemicals and intermediates.
ppp'^imj'\hf <:M:ERLBL+)**,,
Fuel
storage
Gasifier Combustor
Heat recovery
options
Cyclone
separator
Biomass
feedstock
Biomass
Feed
Steam Air
Flue
gas
Sand
Sand
&
Char
Syngas
?b`nk^2K^gm^\a&LbeoZ@Zl@Zlb\ZmbhgIkh\^ll
Gasification
FT liquid
synthesis
Upgrading
complex
Syngas
purification
Hydrogen
purification
&EEDSTOCK
#HEMICALS
INTERMEDIATES
0REMIUM
FUELS
(

,&4,
7AX
?b`nk^1?bl\a^k&Mkhil\alrgma^lblZg]ni`kZ]bg`\hfie^q
ppp'imj^gjnbkr'\hf
_hk_nkma^kbg_hkfZmbhg
Our knockout drum needs
a mist eliminator!
Can we add one without
overhauling the vessel?
Read more on this topic at
www.amistco.com
ITS A MORE COMMON problem than
you might think. A vertical knockout
drum removes free liquid from a
certain gas stream. But at the time the
plant was built, a mist eliminator was
not considered necessary. Now mist
is carrying over and causing trouble
downstream. There is no manway, so
adding a conventional mist eliminator
would require cutting the vessel open.
nstead, if the fanged exit nozzle
is at least six inches in diameter, you
can simply hang a patented MistFix
mist eliminator inside, securing it
between the fanges.
Even when there is a manway,
MistFix avoids entry, welding, and
recertifcation in retrofts. For new
and old vessels, it drastically cuts
installation cost and downtime.
Prore: 281-331-595 Fax: 281-585-1Z80
amistco@amistco.com
21-r( Ere(dercv 3e(v|ce: 800-839-3Z1
V|slF|x |rse(l|or r|sl e||r|ralo(
uop.indd 6 7/2/11 13:14:41
zagoria.indd 5 10/6/11 12:05:07
produced is at just slightly below
the feed pressure and is typically
upgraded to 99.9+% purity, with
hydrogen recoveries of 6090+%.
The Polybed PSA System operates
as a batch process. Multiple adsorb-
ers operating in a staggered
sequence are used to produce
constant feed, product and tail gas
fows. The vast majority of Polybed
PSA Systems have been in hydro-
gen service. The economic
justifcation for a PSA unit will
depend on the hydrogen content of
the feed stream and the how the
refner values chemical hydrogen
versus hydrogen as a fuel. Generally
the following rules apply:
Hydrogen feed concentrations
>55% are easily economically
justifed
Between 40 and 50% hydrogen
can be economically upgraded
dependent on site-specifc
requirements
Below 40% hydrogen, economics
become more diffcult to justify.
The UOP Polysep Membrane
System separates a gas mixture by
the differences in permeation rates
of various gases through the poly-
meric membrane. The more
permeable gas (hydrogen) is
enriched on the permeate side of the
membrane, while the less permeable
gas enriches on the feed side of the
membrane. The membrane separa-
tion of these gases is a
pressure-driven process and requires
a high feed pressure. The hydrogen
product stream (permeate) is
produced at a lower pressure by
taking a pressure drop across the
membrane. The non-permeate
stream is available at essentially feed
pressure. The membrane process is
continuous and produces permeate
40 PTQ Q3 2011 www.eptq.com
Provide the ability to process
more diffcult feeds
Improve product quality, espe-
cially distillates
Increase catalyst life
Reduce the quantity of purge gas
required to maintain recycle gas
purity
Debottleneck existing make-up
gas compressors.
The design of a new hydro-
processing unit can beneft in the
following ways from a higher
hydrogen partial pressure through
the use of high-purity hydrogen as
the make-up gas:
Reduced capital cost (from lower
total plant pressure, smaller make-
up gas compressors, smaller recycle
gas compressor, smaller reactors
and less catalyst.)
Reduced power and fuel
requirements.
Improved distillate product quality
There are various technology
options for the production of high-
purity hydrogen by recovering
hydrogen from lower-purity
streams. The major technologies
used for hydrogen recovery and
purifcation are PSA and
membranes. A few hydrogen cold
boxes have been constructed, but
they are only warranted when
recovery of a valuable liquid prod-
uct is required. Selection of
technology will be guided by the
specifc application. Table 1 is a
guide to selecting between PSA and
membrane technologies for hydro-
gen recovery and purifcation.
The UOP Polybed PSA System is
a cyclical process in which the
impurities in a hydrogen-containing
stream are adsorbed at high pres-
sure and subsequently rejected at
low pressure. The hydrogen
and non-permeate streams at
constant fow, pressure and purity.
The membrane process is the
most economical process for high-
pressure purge gas upgrading. The
membrane system is normally
designed to produce hydrogen at
300600 psig, 9298 vol% purity
and 8595% hydrogen recovery.
The product delivery pressure is
chosen to allow the product to enter
one of the stages of the make-up
hydrogen compressors.
In addition to adding a new PSA
or membrane unit, there are often
opportunities to improve the
performance of existing units. PSA
and membrane units are often
revamped to increase hydrogen
production, recovery and/or purity.
These revamps can be as simple as
replacing adsorbents or as complex
as adding additional equipment.
Frequently, refners elect to perform
the revamps in phases, where each
phase adds additional capacity. The
following are examples of revamps
conducted by refners in North
America.
Case study 1: adsorbent reload
A major North American refner
started up a plant with two identi-
cal steam reformer Polybed
PSA-based hydrogen units, each
with a product capacity of 27.5
million scfd and a third Polybed
PSA unit to upgrade net gas from a
UOP CCR Platforming Process unit
with a product capacity of 28.3
million scfd. The product hydrogen
from the three PSA units was
combined and used as the make-up
hydrogen to a hydrocracker.
The refner wanted to process
more crude and, therefore, the
demand for hydrogen increased.
The CCR Platforming units net gas
purity was greater than 90% hydro-
gen and was deemed acceptable for
direct feed to the hydrocracker
when blended with high-purity
hydrogen from the PSA unit. Re-
routing the net gas from the PSA
unit to the hydrocracker reduced
hydrogen loss to the PSA units tail
gas but, more importantly, it freed
up this PSA unit for other uses.
First, the PSA units were modi-
fed in a number of stages. The
refner debottlenecked the two
Variable Polybed PSA Polysep Membrane
Product purity 9999.999 mol% Up to 98 mol%
Remove CO
2
, H
2
S, H
2
O
High product pressure
Economy of scale
Feed pressure <1000 psi Favoured @ >1000 psi
Feed H
2
Preferred >50 mol% Preferred >70 mol% Min. 15%
H
2
recovery 7090% 7097%
Ease of expansion Easy Very easy
Comparing purifcation technologies
Table 1
zagoria.indd 6 8/6/11 14:12:20
Find out more at www.fwc.com/sru
Foster Wheeler Acquires
Sulfur Recovery Technology
This addition to our technology portfolio enhances our ability to help you
reach peak operating and environmental performance.
www.ptqenquiry.com
for further information
fw.indd 1 8/6/11 11:50:40
steam reformers, which were then
producing over 20% more raw
hydrogen than was originally
intended. The CCR PSA unit (that
is, the PSA unit processing net gas
from the CCR Platforming unit)
was revamped by changing the
software and design conditions to
allow it to operate on SMR gas in
parallel with the original two SMR
PSA units. The adsorbent in the
CCR PSA unit was, however, far
from optimum for service on SMR
gas. The three PSA units could
easily handle the amount of fow.
Since capacity was not a problem, a
study was made with the objective
of increasing the amount of hydro-
gen recovered.
The adsorbent in the PSA unit
originally treating the CCR
Platforming net gas was replaced
with adsorbent optimised for SMR
gas. This was done in conjunction
with the frst set of vessel inspec-
tions, and the PSA units were
balanced and optimised for the
revised fow scheme. The hydrogen
recovery in this PSA unit increased
by over 6% and simultaneously
resulted in an improved CO specif-
cation on the product hydrogen.
The next vessel inspection was of
one of the SMR PSA units. For the
inspection, the adsorbent was again
vacuumed from the vessel through
the top fange (manway) and then
screened and replaced in its origi-
nal position. About 15% of the
adsorbent was lost during this
procedure due to screening losses
and interface losses between
adsorbent layers. This presented an
opportunity to replace the
existing adsorbent with higher-
performance adsorbent to provide
higher recovery and capacity.
After reloading with higher-
performance adsorbents, the
previously identical PSA1 and PSA2
units for their respective SMR units
were in operation side-by-side with
advanced and original adsorbents
in service. PSA1 demonstrated a
capacity increase of 10% over the
original adsorbent (still installed in
PSA2) as well as a 2% increase in
hydrogen recovery.
Figure 5 shows these improve-
ments as trends recorded by the
distributed control system. The new
adsorbents in PSA1 enabled it to
produce more hydrogen from the
same or less feed.
Case study 2: phased revamps
A large residuum desulphurisation
(RDS) facility in the Americas was
designed using hydrogen make-up
from a steam reformer hydrogen
plant with a product fow of 55
million scfd. The hydrogen plant
employed a large ten-bed PSA unit
that removed essentially all impuri-
ties, including nitrogen, from the
steam reformers effuent.
Phase 1
As designed, the feed gas to the
steam reformer was predominantly
natural gas, and supplemental feed
was derived from the high-pressure
vent and the low-pressure fash
gases of the RDS unit. The high-
pressure vent gas was scrubbed of
H
2
S and throttled down to the
steam reformers feed pressure, and
the low-pressure vent was
compressed to match the steam
reformers feed pressure. Figure 6
shows the overall fow scheme.
Various revamps have taken place
to meet the refnerys increasing
needs for hydrogen over the years
(see Table 2).

Phase 2: frst revamp of steam
reformer PSA
The frst plant expansion saw total
hydrogen production increased
30
50
45
40
35
25
20
15
10
5
1 5 9 13 17 21 1 5 9 11 3 7 11 15 19 23 3 7
M
M
S
C
F
D
Time of day, hours
0
Tail gas
Product
Feed
PSA1
PSA2
Figure 5 Comparison of two identical PSA units loaded with different adsorbents
Steam
reformer
PSA ARDS
Hydrogen to
other users
HP vent
90 hydrogen
2000 psig
55 MMscfd
LP ash
60 hydrogen
500 psig
Figure 6 Case study 2: original fow scheme
42 PTQ Q3 2011 www.eptq.com
zagoria.indd 7 13/6/11 13:11:20
w
w
w
.
p
t
q
e
n
q
u
i
r
y
.
c
o
m

f
o
r
f
u
r
t
h
e
r

i
n
f
o
r
m
a
t
i
o
n
koch.indd 1 11/3/11 15:00:25
Moscow Rening, Gas
and Petrochemicals Week
19 23 SEPTEMBER 2011 LOTTE HOTEL, MOSCOW
IGTC
3rd International Gas
Technology Conference
& Exhibition
19 & 20 SEPTEMBER 2011
RPTC
10th Russia & CIS
Petrochemicals
Technology Conference
20 & 21 SEPTEMBER 2011
RRTC
11th Russia & CIS
Rening Technology
Conference & Exhibition
22 & 23 SEPTEMBER 2011
During the week over the 3 events you will have the opportunity to:
listen to more than 65 selected speakers presenting original case studies on latest projects and
technology innovations as well as market overviews, industry outlooks and insight on how Russian
producers are meeting the current challenges;
meet over 500 delegates;
visit over 50 exhibition stands;
network at several Cocktail Receptions in the newest and most luxurious Moscow hotels Lotte.
IGTC was a highly informative
event, covering a wide range of
special and specic matters in
gas processing and gas
utilization. The practical
examples let us see the
common problems from a non-
conventional point of view.
Oleg Mishin, Head of
Construction Development,
GAZPROM
We look forward to welcoming you in September to the most productive and informative
meetings of the Russia & CIS Rening, Gas & Petrochemicals Producers and their partners.
www.europetro.com
Participation in the annual
RPTC conference is a great
opportunity to be updated on
the petrochemical trends
through direct contact with key
industry decision makers.
Alexander Samoilovich,
Deputy Director Rening &
Petrochemicals,
KAZMUNAYGAZ
RRTC is one of the best forums
on rening. Critical topics were
covered and the event has been
run at the highest efciency and
organisation level as usual. We
look forward to the next event
already
Anatoly Artyukh,
General Director,
NAFTAN
The most important technology events for the Russia & CIS Downstream industry
EARLY CONFIRMED SPONSORS SUPPORTED BY
The Russian Association for Reners
and Petrochemicals Producers
organised by
russia PTQ ad 16/5/11 12:01 PM Page 1
epc.indd 1 8/6/11 12:21:09
from 55 million scfd to 70 million
scfd. The initial capacity increase
was achieved through debottleneck-
ing of the steam reformer and SMR
PSA unit to increase hydrogen
output by 18% from 55 million scfd
to 65 million scfd. The SMR PSA
units debottlenecking was achieved
through a process redesign and
changes to the control system soft-
ware, with essentially no hardware
modifcations. Reducing the number
of pressure equalisations enabled
the unit to process a much higher
feed rate with a small decrease in
hydrogen recovery, while still
maintaining design product purity.
This increase in feed capacity more
than compensated for the small
decrease in hydrogen recovery; the
net result was an increase in hydro-
gen production by 18%.
The high-pressure vent stream
(over 2000 psig) was routed to a
membrane system. The hydrogen
product was delivered to the
suction of the hydrogen make-up
compressor. This change added
5 million scfd of hydrogen to the
refnerys hydrogen header.
Phase 3: a new PSA unit
Next, a UOP CCR Platforming unit
was installed and the net gas was
fed to a new 10-bed PSA unit. By
compressing the tail gas, it was
possible to maximise the hydrogen
recovery in the PSA unit while still
sending the tail gas to the refnery
fuel system. This new PSA unit,
processing net gas from the CCR
Platforming unit, added 50 million
scfd to the hydrogen balance. Five
years later, this unit was revamped
(see Phase 5).
Phase 4: second revamp of steam
reformer PSA unit
A second revamp took place to
further increase the capacity of the
steam reformer and its associated
PSA unit from 65 million scfd to 85
million scfd. This additional debot-
tlenecking required modifcations
to many of the control valves and
piping on the piping skid, but
maintained the existing adsorber
vessels and tail gas mixing tanks.
As fow rates had increased by over
50% since the original design, pres-
sure drop problems encountered in
the feed, product and tail gas
piping had to be overcome. This
was achieved by installing valves
with larger discharge coeffcients to
replace some of the existing valves,
and pressure drops through the
unit were reduced to acceptable
levels.
The new cycle was designed such
that any component could fail and
the unit would continue to operate
at design rates and maintain design
hydrogen purity. This further
improved on-stream factors and
unit reliability, as no single compo-
nent would cause a unit trip or
reduction in feed capacity.
Minor modifcations were made
to the skid instrumentation and
the entire control system software
was reprogrammed to implement
the new cycle. The revamp, design
and hardware was completed,
ready for installation, in less than
six months after the project was
authorised. All feld modifcations
were completed during a two-week
turnaround.
Phase 5: revamp of CCR PSA unit
Next, the PSA unit processing net
gas from the CCR Platforming unit
Table 2
Phase Steam reformer Membrane CCR Total
1 55 55
2 65 5 70
3 65 5 50 120
4 85 5 50 140
5 85 5 60 150
6 85 5 75 165
History of hydrogen requirement, million scfd
www.eptq.com PTQ Q3 2011 45
Steam
reformer
PSA ARDS
Membrane
Hydrogen to
other users
85 MMscfd
Catalytic
reformer
PSA
60 MMscfd
5 MMscfd
recovered
hydrogen
Figure 7 Casestudy2:revampedfowschemeafterphase5
zagoria.indd 8 8/6/11 14:12:43
revamp would fully utilise all the
tail gas compressors to their design
capacities.
Additionally, partial adsorbent
replacement with the current high
performance adsorbents would
allow hydrogen recoveries of both
units to improve, thereby further
increasing hydrogen production.
Implementing Phase 6 would bring
the total hydrogen availability for
this refnery to 165 million scfd,
three times the original capacity.
Case study 3: adsorbent and cycle
changes add 30% capacity
A Polybed PSA system was origi-
nally designed as a six-bed unit
processing 12 million scfd of SMR
feed and producing hydrogen with
10 ppmv CO. The plant needed
additional hydrogen and had avail-
able a refnery off-gas stream
containing ~76% hydrogen and
C
1
-C
6
hydrocarbons. Two choices
were considered. The frst was to
process the new feed in the SMR
and send the total effuent to the
PSA unit. The second was to send
the gas blended with the current
SMR gas directly to the PSA unit.
In the frst case, the hydrogen
would pass through the steam
reformer on a free ride and there
would be a need for additional
modifcations to the SMR to process
the gas. In the second case, the SMR
fow rate would stay constant and
the PSA unit would need an
adsorbent replacement for the
heavier hydrocarbons in the feed,
plus a new process design.
The refner chose the second
option, to replace the PSA adsorb-
ents and modify the PSA cycle. This
PSA unit revamp increased hydro-
gen production by ~30% from the
46 PTQ Q3 2011 www.eptq.com
was debottlenecked, as additional
hydrogen net gas feed was availa-
ble from the CCR Platforming unit.
By installing more tail gas compres-
sion and updating the cycle, the
PSA units hydrogen production
was increased to 60 million scfd
with the design hydrogen purity
maintained. Fabrication and instal-
lation of the new compressors
determined the projects overall
schedule, and changes to the PSA
unit were implemented well within
the time frame.
Phase 6: planned future expansion
Due to changing demands, the
refnery is still short of hydrogen,
and UOP was asked to evaluate
options to further increase the
CCR PSA units capacity. The
most recent PSA unit revamp
resulted in the ability to process
all of the available CCR
Platforming net gas and, at the
time, there was still some spare tail
gas compression capacity available.
The CCR PSA unit can be further
revamped to meet current demand
by fully utilising the existing
compression.
One approach being considered is
to make cycle changes similar to
those implemented in the steam
reformer PSA unit at this plant.
Future hydrogen production is
predicted to increase to 75 million
scfd. This revamp would reuse the
existing adsorber vessels and
adsorbents, but would require
changes to the existing valves and
piping skid. These changes would
allow the CCR PSA unit to produce
50% more hydrogen than the origi-
nal design and maintain the
hydrogen recovery already obtained
from the previous revamp. This
combination of a new cycle and
new adsorbents (see Table 3).
Hydrogen optimisation: sustaining
the benefts
It is one thing to optimise your
hydrogen network on paper. It is
quite another to actually realise the
benefts. Daily operating targets
must be optimised to refect the
day-to-day changes of the refnery.
Operations must know the critical
operating parameters of the
hydrogen network to monitor and
manage, and have that data readily
available. Ideally, one person is
responsible for the network as a
whole and can manage the network
to maximise the overall refnery
margin, to avoid each individual
operator making decisions based on
just their own unit.
It is common practice to produce
additional expensive hydrogen and
burn it, just as a safety margin in
case it is required in a hurry.
Operating procedures, control
improvements, automation includ-
ing multivariable control and better
operator communications all can
mitigate this ineffciency and waste.
To optimise the network continu-
ally, the refner must understand
the key constraints within the
network (purity, compression and
so on) and aim to meet those
constraints every day. Operations
must understand and monitor these
constraints and know what adjust-
ments they can make to increase
the hydrogen networks effciency
by pushing closer to a real network
constraint. For example, in a
cascaded system, one might regu-
larly reduce the DHT make-up
purity (bypass around a PSA) while
increasing the make-up and purge
rates and maintaining target recycle
purity, up until the make-up
compressor is at its maximum
capacity. Monitoring the compres-
sor spillback and adjusting regularly
will minimise hydrogen losses in
the PSA under all refnery operat-
ing conditions.
Every operator should be aware
of the value of hydrogen, the costs
of sending hydrogen to fuel and the
penalties for operating too
conservatively. Running a PSA unit
so that there is no detectable
Original design Revamp
PSA type 6 bed 6 bed
Feed SMR 12 MMscfd SMR & ROG 15.5 MMscfd
Product 8 MMscfd 10.5 MMscfd
10 ppm CO max 10 ppm CO max
Off-gas 5 psig 5 psig
Recovery 84.5% 87%
Adsorbent replacement & process cycle changes resulted in 30% greater capacity plus higher recovery
Case study 3
Table 3
zagoria.indd 9 8/6/11 14:13:00
impurity in the hydrogen product is safe, but it can
represent a 110% decrease in PSA recovery, thus wast-
ing hydrogen. Operating a hydroprocessing unit with
higher than target purity for recycle hydrogen is safe,
but it represents unnecessary losses of hydrogen to fuel,
either as purge or as an excessive feed rate to a purifer
with associated hydrogen losses to the tail gas.
When analysis of recycle gas purity is infrequent or
unreliable, the operator is almost forced to run conserv-
atively. In this case, refners should consider installing
one of the new inexpensive, very low maintenance,
direct hydrogen-reading analysers that are now on the
market.
Representation of the hydrogen network in the refn-
erys linear programming (LP) model is an often
overlooked opportunity to signifcantly enhance proft-
ability while evaluating the hydrogen network. This is
not signifcant if hydrogen does not constrain the refn-
ery, but if charge rate and severity targets are set in the
LP model or in the feld in response to hydrogen
constraints, it is critical that the LP model accurately
refects the actual constraints. While, typically, LP
models do refect the hydrogen yields in catalytic
reformers with feed properties and severity, they can
be modifed and maintained to refect accurately hydro-
gen compressor constraints and the impact of
hydroprocessing feed properties and severity on hydro-
gen consumption, partial pressure, purge rates and
make-up rates. Where even greater detail is warranted,
the LP model can refect the relationship between
hydrogen consumption and product properties in these
units.
Conclusions
Hydrogen is an increasingly important component of
refning, particularly in view of the increased demand
for clean fuels. There are opportunities to optimise the
use of hydrogen in a refnery to maximise profts:
Hydrogen network studies and fow scheme
optimisations
PSA and membrane technologies to recover and
purify hydrogen
Reforming process and catalyst technologies to
produce more hydrogen
Hydroprocessing and catalyst technologies to
consume less hydrogen.
Alan Zagoria is Engineering Fellow in the Optimization Services
Department at Honeywells UOP. He has spent the last 12 years assisting
customers in optimising their refnery hydrogen networks. He holds a BS
chemical engineering from Northwestern University.
Ron Long is Product Line Manager for Hydrogen with Honeywells UOP.
He has worked in a variety of felds including R&D, feld operating services,
operating technical services, Far East and Americas customer services,
engineering project manager and Americas customer sales. He holds a
BS in chemical engineering from the Illinois Institute of Technology in
Chicago.
Kathy Picioccio is a Senior Account Manager for Gas Purifcation at
Honeywells UOP, responsible for continuing support of its installed base
of PSA systems and Polysep membranes. She holds a BE in chemical
engineering and a Masters in electrical engineering/computer science
from Stevens Institute of Technology.
www.eptq.com
www.ptqenquiry.com
for further information
B
O
R
S
I
G
BORSIG GROUP
Egellsstrasse 21
D-13507 Berlin/Germany
Phone: +49 (30) 4301-01 Fax: +49 (30) 4301-2236
E-mail: info@borsig.de www.borsig.de
Leading Technology for
Innovative Solutions
RECIPROCATING COMPRESSORS
for Process Gases
The compressors are of modular design and comply with
the API 618 design standards. The series has been deve-
loped for the heaviest continuous operation.
Discharge pressure: 10 ... 1,000 bara
Capacity/flow: 100 ... 115,000 m/h
Power: 50 ... 16,000 kW
CENTRIFUGAL COMPRESSORS
for Process Gases
The BORSIG ZM T-Jet-series comprises multistage
integrally geared centrifugal compressors which comply
with the corresponding API design standards, such as
API 617.
Discharge pressure: up to 80 bar
Capacity/flow: 1,100 ... 40,000 m/h
Power: 100 ... 12,000 kW
Other BORSIGProducts:
Reciprocating Compressors for CNG Filling Stations,
Compressor Valves, Pressure Vessels, Heat
Exchangers, Membrane Technology, Boiler and Power
Plant Technology, Industrial and Power Plant Services
zagoria.indd 10 8/6/11 14:13:16
1
1
5
8
_
e
Build on our technologies.
Call us: +49 (0) 69 58 08-0
www.lurgi.com
A member of the Air Liquide Group
We help the environment and reneries grow together.
Reneries have to exploit oil resources more efciently than ever. At the same time, they
also have to protect the environment. These two aspects frequently represent considerable
obstacles that must be overcome. This is possible with the right combination of expertise
and matching processes.
We understand reneries in all their complexity. We know their mass ows, their energy
balances. We command technologies for optimal desulfurization and fuel upgrading
with a much deeper conversion than before.
As you can see, with Lurgi, you can combine what are increasingly perceived as two sides
of the same coin: efciency and environmental protection.
www.ptqenquiry.com
for further information
lurgi.indd 1 10/3/11 10:45:42
Improving residue hydrocracking
performance
M
any ebullated-bed residue
hydrocracking units are
conversion limited as a
result of fouling in downstream
equipment. Increasingly diffcult
feedstocks have forced operators of
ebullated-bed upgrading units to
impose limitations on reactor
temperatures and throughputs.
Heavy oil and bitumen upgrading is
often limited by poor conversion of
asphaltenes, which in turn leads to
the formation of sediment, coke and
downstream foulants. The most
likely cause is hydrogen-transfer
limitations at a higher residue
conversion, which cause aromatic
structures in the asphaltenes to
grow, aggregate and then fall out
of solution as the oil becomes
unstable.
HCat technology was developed
by Headwaters HTI group as a
molecular catalyst to limit sediment
formation and fouling in ebullated-
bed residue hydrocracking units.
The function of the catalyst in a resi-
due hydrocracking unit is to
facilitate hydrogen transfer into the
asphaltenes. This function gives
higher asphaltene conversion and
prevents oil instability when the
converted oil is cooled in down-
stream equipment. The ability to
hydrogenate and convert asphaltenes
provides the mechanism to increase
residue conversion without the asso-
ciated oil instability and subsequent
formation of sediment.
Neste Oil and HTI have been
working together to implement the
technology at Neste Oils Porvoo
refnery and have successfully
completed a 40-day commercial trial.
This demonstrated a signifcant
improvement in the ability of Neste
Trials with a catalyst system for enhanced transfer of hydrogen to asphaltenes show
reduced sediment formation and fouling in ebullated-bed residue hydrocrackers
Joni Kunnas Neste Oil
Lee sMiTh HTI
Oils ebullated-bed LC-Fining unit
to operate with a diffcult feedstock
at an elevated conversion for long
periods of time. Throughout the
trial, the combined feed to the unit
was mostly Russian Export Blend or
Urals-based vacuum residue.
The primary objective of the trial
was to demonstrate the ability to
control sediment in the vacuum
tower bottoms product stream,
while signifcantly increasing
560C+ residue conversion. A
second objective was to evaluate
the ability of HCat to reduce prod-
uct exchanger fouling rates, even at
elevated residue conversions levels.
The purpose of this article is to
provide an overview of the technol-
ogy and the Porvoo refnery, and to
review the results of the trial.
Technology overview
HCat residue hydrocracking tech-
nology is based on the in situ
formation of a molecularly
dispersed catalyst intimately mixed
throughout the heavy oil feedstock.
Molecular dispersion of the catalyst
in heavy oil is achieved by blending
www.eptq.com PTQ Q3 2011 49
Vac.
resid.
feed
Hydrogen recycle
and make up
Hydrogen and
liquid quench
HCAT
catalyst
precursor
Feed
conditioning
Separator
Fractionator
tower
Vacuum
tower
R-1 R-2
Hydrogen
recycle
H
2
S removal
C1/C2 to
fuel gas
C3/C4
Naphtha to
gasoline
processing
Diesel to
hydrotreating
AGO to
upgrading
LVGO to
upgrading
HVGO to
upgrading
Unconverted
resid. to fuel oil
Figure 1 Addition of HCat in an ebullated-bed residue hydrocracker
smith.indd 1 8/6/11 14:42:35
50 PTQ Q3 2011 www.eptq.com
typical supported ebullated-bed
catalyst. Therefore, the catalyst
complements the function of the
solid supported catalysts by facili-
tating hydrogen addition into heavy
components of the cracked residual
oil. It should be made clear that the
technology does not replace the
function of the supported catalysts.
It is, in fact, synergistic with the
supported catalysts and allows the
ebullated-bed reactor system to
operate more efciently.
The conversion of residue feed-
stocks is typically limited by the
poor conversion of asphaltenes,
which lead to coke and sediment
formation at reactor conditions
subsequent to downstream fouling.
The most likely cause is hydrogen-
transfer limitations at a high reactor
severity, which cause aromatic
structures in the asphaltenes to
grow, aggregate and become unsta-
ble. This leads to mesophase and
sediment formation, and fouling in
the downstream separators, distilla-
tion towers and product heat
exchangers. The function of HCat is
to facilitate hydrogen transfer into
these asphaltenes and increase
asphaltene conversion. This func-
tion inhibits asphaltene instability
and associated fouling.
The technology does not affect the
kinetics of residue conversion.
Residue conversion in an ebullated-
bed hydrocracker is dependent on
temperature, space velocity and
other conditions. HCat does not
increase or decrease this conversion.
However, its ability to hydrogenate
and convert asphaltenes gives a
rener the ability to raise reactor
temperature and increase residue
conversion without the associated
oil instability and the subsequent
formation of sediment and fouling
that typically limits an ebullated-bed
residue hydrocracker.
The technology offers signicant
economic benets in converting
low-value residues into high-value
products. Some of the potential
economic advantages can include:
Increased asphaltene conversion
improves product stability and
exibility to increase residue
conversion
n Residue conversion can be
increased
an oil-soluble catalyst precursor
with a residue feedstock using
HTIs proprietary mixing system
before the feed preheat system.
Figure 1 shows its addition in a
simplied ebullated-bed residue
hydrocracker.
As the feed is heated up to reac-
tor conditions, the precursor breaks
down and a molecularly dispersed
catalyst is formed upstream of the
ebullated-bed reactors. In addition,
the highly polar catalyst preferen-
tially associates with the asphaltene
molecules that typically constitute
the most polar fraction in heavy oil.
One of the primary functions of a
hydroprocessing catalyst is to disso-
ciate molecular hydrogen, which
can readily hydrogenate unsatu-
rated and cracked oil. Therefore,
once in the ebullated-bed reactors,
the catalyst facilitates hydrogen
transfer, particularly into large
asphaltenes that cannot readily
access the solid heterogeneous cata-
lyst (see Figure 2). This is because
the catalyst remains dispersed in
the heavy oil feedstock outside of
the supported heterogeneous
ebullated-bed catalyst.
A standard ebullated-bed residue
hydrocracker without HCat is
shown in Figure 3. The supported
catalyst is ebullated in the reactor
that denes the catalytically active
zone. Large asphaltenic oil mole-
cules responsible for sediment and
fouling behaviour have limited
access to the catalytically active
sites in the supported catalyst.
An ebullated-bed residue hydroc-
racker with HCat is shown in
Figure 4. The entire reactor volume
now has catalytic activity, as the
technology facilitates hydrogen
transfer into the large asphaltenic
oil molecules. This hydrogenation
function limits the formation of
sediment and the resulting fouling
behaviour of the oil, even as the
residue conversion is increased.
As a result, HCat is available to
catalyse benecial hydrogenation
reactions involving asphaltenes and
other large molecular components
of heavy oil feedstocks that are too
large to diffuse into the pores of
(
(
(
(
(
(
(
(
(#AT MOLECULE
Figure 2 The catalyst facilitates hydrogen
transfer, particularly into large asphaltenes
that cannot readily access the solid
heterogeneous catalyst
Figure 3 Ebullated-bed residue hydrocracker without HCat
H
2
H
Pore
Active catalyst
site
Products and
sediment
Thermal
reaction
zone
Catalytic
reaction
zone
Catalyst
addition
Hydrogen
Hot feed
Supported
catalyst
pellet
smith.indd 2 10/6/11 11:19:58
For more information about UOP, visit www.uop.com
2010 UOP. All Rights Reserved.
refning your proft
UOP helps you exceed your goals with innovative technology,
catalysts and optimization solutions specifcally designed to
meet your needs.
UOP hydroprocessing solutions and optimization services are designed to help
you maximize your return on investment and grow your business. As regional
market demands shift, we provide the process technologies, catalysts and
services that will meet your changing business needs. Our experts work closely
with you to meet your desired yields and product specifcations including
ultra-low sulfur diesel standards while improving your operational effciency.
Backed by over 50 years of hydroprocessing innovations, UOP offers the best and most advanced solutions
to keep your business one step ahead.
UOP_Refining Your Profit_Ad_A4_PRINT.indd 1 5/18/10 2:10:44 AM
www.ptqenquiry.com
for further information
uop.indd 1 11/6/10 14:23:56
n Improved reactor stability even
at higher conversion levels
Increase in residue feed rates
Signicant increase the units reli-
ability and on-stream factors
n On-stream time of vacuum
tower bottoms exchangers increased
signicantly
n Lower fouling rates in unit
separators, atmospheric and
vacuum towers
Ability to processes a heavier
crude slate in the renery or lower-
quality residue feeds
Unconverted residue products
have a higher value
n Lower density, lower sulphur,
lower viscosity
n Higher blending values into
typical fuel oil products
n Lower vacuum tower bottoms
product sediment.
Porvoo renery
Neste Oils reneries at Porvoo and
Naantali in Finland produce a range
of major petroleum products. Their
combined capacity accounts for
over 20% of the Nordic regions
overall rening capacity. The State
of Finland maintains a controlling
interest (50.1%) in Neste Oil.
The Porvoo renery, located at
Kilpilahti, 35 km east of Helsinki,
focuses on the production of high-
quality, low-emissions motor fuels.
In operation since the mid-1960s,
the renery currently has a capacity
of 205 000 b/d. Crude oil processed
in the renery is primarily Russian
52 PTQ Q3 2011 www.eptq.com
Export Blend or Urals supple-
mented with North Sea crude oil.
The residue hydrocracking unit at
the Porvoo renery was commis-
sioned in 2007 to convert heavy
vacuum residue feedstocks into
virtually sulphur-free transporta-
tion fuels. This project also gave the
renery the exibility to use up to
100% Russian crude. The residue
hydrocracking unit integrates a LC-
Fining ebullated-bed residue
hydrocracker and an IsoCracking
vacuum gas oil mild hydrocracker.
The units capacity is about 40 000
b/d and includes one of the largest
hydrogen plants in Europe.
HCat technology was applied to
the LC-Fining ebullated-bed unit.
The protability of the combined
unit had been limited by high foul-
ing rates at increased conversion of
the Urals vacuum residue feedstock
in the ebullated-bed unit. Conversion
of the residue feedstock was
expected to increase signicantly.
Trial objectives
The primary objective of the trial
was to demonstrate the ability to
control sediment in the vacuum
tower bottoms product stream at a
less than maximum target sediment
level (or baseline), while increasing
the 560C+ residue conversion by
8 wt%. A second objective was to
evaluate the ability to improve
product exchanger fouling rates at
both baseline operating conditions
and elevated conversions levels.
A third objective was to evaluate
the effect that HCat had on the
solid catalyst performance in the
LC-Fining reactors. Throughout the
trial, the feed to the unit was mostly
Urals-based vacuum residue, along
with some straight-run heavy
vacuum gas oil (HVGO) and recy-
cled product HVGO.
The following is a summary of
the objectives going into the trial:
Demonstrate HCats efcacy at
current conversion levels
Demonstrate the ability to raise
conversion by 8 wt%
Maintain IP-375 sediment at less
than maximum target
Monitor unit response to HCat
ramp-down
Solid catalyst response
Watch catalyst activity
Monitor spent catalyst metals
content.
Operating conditions in the LC-
Fining reactors were monitored and
adjusted throughout the test run.
The solid catalyst inventory and
quality were also monitored and
adjusted during the baseline and
test run periods to maintain a stable
solid catalyst activity. Neste Oil has
dened residue as material boil-
ing above 560C. Residue
conversion is dened as the net
weight per cent disappearance of
560C+ material compared to the
feed oil.
Trial of residue conversion and
sediment performance
The trial is summarised in Figure 5.
Baseline unit conditions, reactor
temperature and 560C+ conversion
were established seven days prior
to the introduction of HCat. Two
days after the initial introduction,
the unit encountered a hydrogen
outage that delayed the test by
about 12 days. Following the hydro-
gen outage, conditions in the
LC-Fining unit were returned to
baseline conditions and the trial
was restarted. The effects of the
outage were evident in the units
performance, but it was decided to
continue with testing as soon as
possible due to the limited time
available.
After holding baseline conditions
for seven days, reactor tempera-
tures were increased in stages of
H
2
H
Pore
Active catalyst
site
Products
with less
fouling
Catalytic
reaction
zone
Catalyst
addition
Hydrogen
Hot feed
Supported
catalyst
pellet
H
Figure 4 Ebullated-bed residue hydrocracker with HCat
smith.indd 3 10/6/11 11:20:08
BELCO is a leader in complete air pollution
control solutions, ready to serve your facility
anywhere in the world. Tousands of refnery,
industrial and municipal clients trust us for
proven technology leadership and customized,
turnkey services.
From consulting to design, fabrication,
installation, training and maintenance, the
world turns to Belco Technologies Corporation.
For more information and online access to
our extensive Technical Library of white papers
and brochures, visit www.belcotech.com.
MANY SOLUTIONS, ONE SOURCE. ITS A BELCO

WORLD.
Reducing refnery emissions has never been so easy.
) la|ia iea|. la|siaa. \| ))+. l:\ l|ea:. )-+-+)) lax. )-+-+ :-aail. ia|e|:lce|:c|.cea .|:lce|:c|.cea
Copyright 2008 DuPont. The DuPont Oval Logo, DuPont

, and BELCO

are registered trademarks or trademarks of E.I. du Pont de Nemours and Company or its affliates. All rights reserved.
SO
x
ACID MIST
NO
x
PARTICULATES
www.ptqenquiry.com
for further information
belco.indd 1 17/3/10 14:33:00
baseline conditions to maximise
residue conversion and the forma-
tion of vacuum tower bottoms
sediment increased accordingly. In
fact, the sediment increased above
the maximum level before HCat
was added to the LC-Fining
operation.
Following the hydrogen outage,
and before HCat was added to the
unit a second time, we do not
believe that the unit had stabilised
at equilibrium conditions and it can
be seen that formation of sediment
appeared to spike during the initial
reintroduction of HCat. However,
the sediment eventually stabilised at
a level well below the target level.
As reactor temperatures and the
resulting residue conversion were
increased, the sediment was gener-
ally maintained well below the
sediment baseline or target level.
One of the primary objectives of
the trial was to increase residue
conversion by 8 wt% above the
baseline conversion level, while
maintaining vacuum tower bottoms
sediment below the maximum target
level. This objective was not only
achieved, but exceeded. At the end
of the trial, the average temperatures
of all three LC-Fining reactors had
been increased by 9C compared to
baseline reactor temperatures. The
560C+ residue conversion increased
10 wt% compared to the baseline.
Sediment levels averaged about
0.10 wt% less than the maximum
sediment level or baseline sediment
level in the vacuum tower bottoms,
even when residue conversion had
increased 10 wt% above the baseline.
Furthermore, all three LC-Fining
reactors were stable at the increased
residue conversion level. However,
the test run at high conversion
conditions lasted a short period and
further testing was needed
to confrm optimal operating
conditions.
The advantage of adding HCat to
the process is summarised in
Figures 7 and 8. A longer-term view
of the trial can be found in Figure
7, where about 280 days of unit
operations are presented. The data
show that 560C+ residue conver-
sion was signifcantly higher during
the trial, including baseline residue
conversion before HCat was added
54 PTQ Q3 2011 www.eptq.com
about 2C over the succeeding
weeks of the test. Reactor condi-
tions and stability were closely
monitored as temperature and resi-
due conversion were increased.
Stable unit operations were required
before each increase in reactor
temperature.
After four weeks, the reactor
temperature was eventually
increased by 9C over baseline
conditions and held for 10 days.
This increase in reactor temperature
resulted in an increase in 560C+
residue conversion by about
10 wt%, exceeding the trials objec-
tive by 2 wt%. In addition, the
vacuum residue feed rate to the
unit was increased by 5 t/h above
the baseline feed rate during the
second week of operations. The
LC-Fining unit was shut down for
a scheduled turnaround at the
conclusion of the trial.
Naphtha, diesel and light vacuum
gas oil (LVGO) product yields all
increased as a result of increased
conversion. The HVGO yield
decreased with increasing conver-
sion, thus indicating conversion of
the HVGO fraction with increasing
residue conversion, which was
expected. The vacuum tower
bottoms product rate was reduced
by over 20 wt%.
The behaviour of the vacuum
tower bottoms sediment during the
trial is shown in Figure 6. The sedi-
ment is clearly suppressed, even
with an increase in residue
conversion. Before the trial began,
unit conditions were optimised at
8
12
10
6
4
2
Base|ine
-2
0 7 14 21 28 35 42 49 56 63 70 77 84
%
t
w

,
n
o
i
s
r
e
v
n
o
c

+
C

0
6
5

C


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

r
o
t
c
a
e
r

F
C
L

e
g
a
r
e
v
A
Run day
-4
560C+ conversion
HCat
Ave. reactor temp.
Figure 5 Trial results
5
15
10
-5
-10
-15
Base|ine
-20
0 7 14 21 28 35 42 49 56 63 70 77 84

s
t
c
u
d
o
r
p

B
T
V
%
t
w

,
t
n
e
m
i
d
e
s

r
o
t
c
a
e
r

F
C
L

e
g
a
r
e
v
A
C


,
e
r
u
t
a
r
e
p
m
e
t
Run day
-25
8
12
10
6
4
2
Base|ine
-2
-4
VTB sediment
HCat
Ave. reactor temp.
Figure 6 Behaviour of vacuum tower bottoms sediment during the trial
smith.indd 4 13/6/11 14:15:30
to the unit. However, even with a
higher residue conversion, the level
of sediment was reduced compared
to the historical data when HCat
was used.
Figure 8 shows the effect of HCat
on sediment as a function of 560C+
residue conversion. Without HCat,
as the residue conversion is
increased, sediment increases rather
quickly to exceed the maximum
sediment operating target. When
HCat is added to the LC-Fining
unit, sediment is reduced when the
unit is operated near baseline
conditions. Furthermore, as reactor
temperatures and residue conver-
sion are increased, the resulting
sediment is minimised.
Downstream equipment fouling
performance
The key effect of lower VTB sedi-
ment is lower fouling rates in
downstream separators, distillation
towers and exchangers. Fouling in
one or more of these item of equip-
ment typically limits the ability of
an ebullated-bed operation to
increase the units residue conver-
sion, because sediment and fouling
rates rise substantially as residue
conversion is increased. The result
is reduced run length and lower
on-stream time as residue conver-
sion is increased.
The downstream heat exchangers
were the items of equipment prima-
rily evaluated during the trial.
Other key downstream equipment
were also assessed but, because of
the short duration of the trial, it
was diffcult to come to any conclu-
sions with regard to fouling rates in
the separators and distillation
towers. The fouling rates of the
downstream heat exchangers, even
at the higher residue conversion
level, were signifcantly reduced
with the addition of HCat. Most of
the downstream exchangers were
changed out two to three times
during the 20-day January baseline
period, compared to one or two
times during the 40-day trial. The
cycle lengths of the vacuum tower
bottoms exchangers increased from
about eight days to more than 20
days on average.
The effects of HCat on fouling
rates on a key vacuum tower
www.eptq.com PTQ Q3 2011 55
exchanger, EA-73019, are shown in
Figure 10. The EA-73019 exchanger
transfers the heat from the vacuum
tower bottoms to the LC-Fining
units total feed stream. The EA-
73019 was switched over at about
eight-day increments in January
due to a high pressure drop across
the exchanger (600 kPa) before the
introduction of HCat. Following the
introduction, the exchangers P
increased from 150 kPa to only 300
kPa over 20 days before the
exchanger was switched to the
second exchanger bank. The P and
fouling factor increased as the resi-
due conversion was increased and
then stabilised before the trial
concluded. The exchanger was
changed out once throughout the
duration of the trial. Overall, the
fouling performance of the
bottoms exchanger, EA-73007, are
shown in Figure 9. The EA-73007
exchanger transfers heat from the
vacuum tower bottoms product
stream to the mild hydrocracker
stripper bottoms product stream.
The EA-73007 was switched over at
10-day increments during January
due to a high pressure drop across
the exchanger before the introduc-
tion of HCat. Following the
introduction, the exchangers P
was stable and actually declined
throughout the duration of the trial
due to a lower fow rate as a result
of increased residue conversion.
The fouling resistance factor also
declined. The exchanger was not
changed out throughout the dura-
tion of the trial.
The effects on the fouling rates in
another key vacuum tower bottoms
300
250
200
150
100
50
0
1
.
6
.
7
0
1
.
7
.
5
0
1
.
8
.
2
0
1
.
8
.
0
3
0
1
.
9
.
7
2
0
1
.
0
1
.
5
2
0
1
.
1
1
.
2
2
0
1
.
2
1
.
0
2
0
1
.
1
.
7
1

s
v

n
o
i
t
a
m
r
o
f

t
n
e
m
i
d
e
S

%

,
t
/
t

e
n
i
l
e
s
a
b

s
i
o
n
r
e
v
n
o
c

+
C

0
6
5


e
n
i
l
e
s
a
b

s
v
0
0
1
.
2
.
4
1
0
1
.
3
.
4
1
120
100
80
60
40
20
0
Sediment formed vs base|ine
HCat feed
560C+ conversion vs base|ine
%
,
t
/
t
Figure 7 560C+ residue conversion was signifcantly higher during the trial
0.10
0.15
0.05
Base|ine
-0.05
-0.10
-0.15
-4 -2 Base|ine 2 4 6 8 10 12
%
t
w

,
t
n
e
m
i
d
e
s

B
T
V

Resid. (560C+) conversion, wt%
-0.20
Base|ine (no HCat}
HCat
Figure 8 Effects on sediment as a function of 560C+ residue conversion
smith.indd 5 8/6/11 14:43:17
EA-73019 vacuum tower bottoms
exchangers was improved substan-
tially relative to the baseline.
Other key downstream exchang-
ers also showed similar performance
and much lower fouling rates after
the introduction of HCat. The
change in fouling behaviour exhib-
ited by the behaviour of key
downstream heat exchangers would
also indicate lower fouling rates in
the units separators and distillation
towers. An analysis of the medium-
pressure separator showed some
positive effects on its fouling
performance.
The lower fouling rates observed
in key downstream exchangers,
separators and distillation towers
give an ebullated-bed operator the
ability to increase residue conver-
asphaltenes. Increasing hydrogen
transfer leads to reduced aromatisa-
tion of the polynuclear aromatics,
which inhibits the molecular growth
of asphaltenes.
The primary function of the tech-
nology is to maintain the
concentration and reactivity of
hydrogen donors in the asphaltenes
during high-temperature hydrocon-
version. This function prevents the
growth of polynuclear aromatics
and makes them less likely to come
out of solution as either coke or
downstream asphaltenic foulants,
even at increased conversion.
This function is evident in many
aspects of the trial. When conver-
sion was increased by 10 wt%, the
sediment averaged 0.10 wt% less
than the maximum target level in
the vacuum tower bottoms product.
Fouling rates in downstream heat
exchangers, particularly the vacuum
tower bottoms heat exchangers,
including EA-73007 and EA-73019,
were reduced substantially.
Asphaltene conversion increased
signifcantly over the baseline by
about 1015%. This would indicate
that the asphaltene content in the
resulting vacuum tower bottoms
product is lower during HCat addi-
tion compared to the baseline, even
when residue conversion was
increased by 10 wt% over baseline
conditions. This increase in
asphaltene conversion and the
resultant lower asphaltene content
in the vacuum tower bottoms is a
primary reason why sediment
formation declined during the trial
and downstream fouling improved
greatly.
Other effects on the performance
of the LC-Fining unit were also eval-
uated and are summarised briefy. A
metals balance was performed to
ensure the catalyst was not laying
down on the supported catalyst or
somewhere else in the unit. The
average metals balance for the
period of the trial was acceptable. It
was determined that the catalyst
was not deposited either in the unit
or onto the LC-Fining catalyst.
The vanadium and nickel balances
were essentially the same before
and after the addition of HCat. This
is the expected result, since the
removal of vanadium and nickel is
56 PTQ Q3 2011 www.eptq.com
sion. HCat changes the relationship
of downstream fouling when resi-
due conversion is increased.
Asphaltene conversion
The key part of the residue feed to
the LC-Fining unit that causes sedi-
ment and fouling limitations is the
asphaltene component and its
behaviour during high-temperature
hydroconversion. It has been shown
that large polynuclear aromatics
that predominate in asphaltenes
limit the conversion of residue feed-
stocks due to the formation of coke
and asphaltenic sediments down-
stream. The key is that coke in a
reactor is triggered by a liquid-
liquid phase separation at thermal
cracking temperatures, and sedi-
ments are the result of insoluble
1.20
1.40
1.00
0.80
0.60
0.40
0.20
0
1
.
1
.
1
0
1
.
1
.
8
0
1
.
1
.
5
1
0
1
.
1
.
2
2
0
1
.
1
.
9
2
0
1
.
2
.
5
0
1
.
2
.
2
1
0
1
.
2
.
9
1
0
1
.
2
.
6
2
0
1
.
3
.
5
0
1
.
3
.
2
1
0
1
.
3
.
9
1
R

e
c
n
a
t
s
i
s
e
r

g
n
i
l
u
o
F
i
f
m

,
2
W
/
K
0.00
600
700
800
900
500
400
300
200
100


a
P
K

,
p
o
r
d

e
r
u
s
s
e
r
P




0
R

A AP
B AP
HCat
Figure 9 Effects on fouling rates in vacuum tower bottoms heat exchanger EA-73007
3.0
5.0
4.5
4.0
3.5
2.5
2.0
1.5
1.0
0.5
0.0
R

e
c
n
a
t
s
i
s
e
r

g
n
i
l
u
o
F
i
f
m

,
2
W
/
K
0
1
.
1
.
1
0
1
.
1
.
8
0
1
.
1
.
5
1
0
1
.
1
.
2
2
0
1
.
1
.
9
2
0
1
.
2
.
5
0
1
.
2
.
2
1
0
1
.
2
.
9
1
0
1
.
2
.
6
2
0
1
.
3
.
5
0
1
.
3
.
2
1
0
1
.
3
.
9
1
600
1000
900
800
700
500
400
300
200
100
0


a
P
K

,
p
o
r
d

e
r
u
s
s
e
r
P




R

AB AP
CD AP
HCat
Figure 10 Effects on fouling rates in vacuum tower bottoms heat exchanger EA-73019
smith.indd 6 8/6/11 14:43:26
primarily infuenced by the metals-
loading capacity of the supported
catalyst and the catalyst replace-
ment rate. HCat should have a
positive effect on sulphur removal
in the vacuum tower bottoms prod-
uct. However, data from the trial
did not conclusively indicate that
sulphur removal in the vacuum
tower bottoms was improved.
Sulphur removal rates increased
with rising reactor temperatures.
Similar performance was observed
with respect to MCR conversion.
Highlights
The highlights of the trial include
the following observations:
Reactor temperatures increased
by 9C over the baseline
Residue conversion increased by
approximately 10 wt%
Vacuum tower bottoms sediment
was well below target during the
trial, including a 10 wt% increase in
residue conversion
Fouling rates were substantially
reduced in key downstream prod-
uct heat exchangers
Vacuum residue feed rate
increased by 5 t/h above baseline
All LC-Fining reactors were stable
at increased residue conversion
No fouling was observed in the
LC-Fining feed preheat section
Asphaltene conversion increased
by approximately 10 wt%.
Conclusions
Neste Oil and HTI completed a 40-
day commercial trial at the Porvoo
refnery. The trial successfully
demonstrated a signifcant improve-
ment in the LC-Fining units
performance. The test was successful
in that the residue conversion
increased by about 10 wt%, with
substantially reduced fouling
observed in key downstream equip-
ment, including the vacuum tower
bottoms heat exchangers.
The technology was used in the
LC-Fining unit following a spring
2010 turnaround and has been used
continuously since the summer of
2010. During this longer-term test-
ing, the advantages observed
during the 40-day test were
confrmed. The vacuum tower
bottoms sediment was reduced well
below the maximum operating
target, even at increased residue
conversion and vacuum residue
feed rate. The downstream exchang-
ers fouling performance was very
similar to that observed during the
40-day test. With the continuous
use of HCat, Neste Oil was able to
increase the length of continuous
operation signifcantly at increased
residue conversion.
Neste Oil and HTI plan to
continue to optimise the residue
hydrocracker operations with HCat
to increase the units operability
and proftability.
HCAT is a trademark of HTI, a Headwaters
Company. LC-FINING and ISOCRACKING are
trademarks of Chevron Lummus Global LLC.
Joni Kunnas is the Development Manager for
the residue hydrocracking unit at Neste Oils
Porvoo refnery, Kilpilahti, Finland. He holds a
masters degree in chemical engineering from
Helsinki University of Technology.
Email: joni.kunnas@nesteoil.com
Lee Smith is Strategic Account Manager with
HTI, a Headwaters company. He holds BS and
MS degrees in chemical engineering from
Brigham Young University, Utah, and has an
MBA from the same university.
Email: lsmith@headwaters.com
www.eptq.com PTQ Q3 2011 57
What Goes Around...
Remove Solids and Hydrocarbons from
your solvent with MPRs High Capacity,
Easily Regenerated Media.
Available Worldwide
The MPR SSX and HCX processes use our unique and proprietary media, which
can easily be regenerated after exhaustion. They complement each other in way that
allows them to surpass the capacity of standard cartridges and activated carbon
lters, while bringing outstanding longevity.
SSX and HCX Solve The Following Common Problems:
Plugging of heat exchangers, piping, towers
and instruments
Poor hydrocarbon separation from
amine solutions
High Amine losses
Unstable Unit operations
Accelerated corrosion rates due
to erosion
Deactivation of Claus Catalyst in the SRU
SSX and HCX processes work by removing solids to
submicron size particles, while also removing hydrocarbons
(including diesel, gas oils, parafns, BTEX and various petroleum
intermediates) from amine solutions.
Eliminating cartridge lters and activated carbon means eliminating lter replacement,
disposal and labor
SSX and HCX processes are available exclusively from MPR Services, Inc..
www.ptqenquiry.com
for further information
smith.indd 7 8/6/11 14:43:38
www.ptqenquiry.com
for further information
mogas.indd 1 8/6/11 11:55:48
Industrial-scale production of
renewable diesel
A
renewable diesel has
emerged as a drop-in substi-
tute for todays mineral
diesel fuels. Its production is not
limited to seed oils, but is fexible
in that it can use a number of differ-
ent feedstocks with very little
change in operating conditions yet
still produce a renewable and clean
product. Renewable diesel is
produced by hydrotreating,
whereby the oxygenates in bio-
feedstocks are hydrogenated into
hydrocarbons. Hydrotreating takes
place in existing diesel hydrotreat-
ers in a co-processing scheme,
where the biofeed component is
deoxygenated and the fossil diesel
component is simultaneously desul-
phurised, or it can take place in a
standalone unit processing 100%
biofeed. In both cases, conversion
occurs over hydrotreating catalysts
and in conditions similar to those
used in hydrotreaters today. The
product can, in some cases, require
its cold fow properties to be
upgraded by, for instance, catalytic
dewaxing.
Haldor Topse has supplied cata-
lyst for a number of operating units
producing renewable diesel and has
also licensed process technology for
standalone units and revamps to
enable co-processing. An earlier
article in PTQ (Q2 2010, p101)
discusses the chemistry, catalyst
technology and challenges of
renewable diesel hydrotreating.
This article describes how innova-
tive technology solutions were
implemented at the Preem
Gothenburg refnery. Operating
data from the current cycle explain
how the revamp at Gothenburg has
made possible the large-scale
A revamp has made possible the large-scale production of clean diesel with a
signifcant percentage of renewable content
RAsmus EgEbERg and Kim KnudsEn Haldor Topse
stEfAn nystRm, EvA Lind gREnnfELt and KERstin EfRAimsson Preem
production of clean diesel with a
signifcant percentage of renewable
content.
Rebuilding for renewables
This case study shows how the
challenges of renewable diesel
hydrotreating can be overcome
through technology solutions and
the use of a specifcally designed
catalyst series. It also serves as an
example of how a non-edible and
low-value stream can be upgraded
into a high-value transportation
fuel by hydrotreating. Preem
contacted Topse in order to scope
out the possibilities of revamping
an existing mild hydrocracking unit
to a green hydrotreating unit, where
large quantities of raw tall diesel
(RTD) could be co-processed
together with light gas oil. The
capacity of the unit is approxi-
mately 10 000 b/d, and a target was
set of co-processing up to 30% RTD
in the feed. A development agree-
ment was signed, stipulating that
Topse would carry out the basic
design of the unit.
RTD is produced from tall oil by
Sunpine from Kraft paper mills in
the north of Sweden. Tall oil mainly
consists of resin acids and free fatty
acids, as well as a number of
contaminants in smaller concentra-
tions. Through an esterifcation
process, the majority of the free
fatty acids are converted to fatty
acid methyl esters (FAME), while
the resin acids are left almost
unconverted. RTD differs from
other feedstocks used for renewable
diesel production in that it is non-
edible, and thus this technology
does not affect the global food
shortage negatively or add to food
prices. The composition of RTD
varies, depending on the type of
wood used in the paper mill and
the severity of the esterifcation
process. Unlike other bio-oils, RTD
also contains several hundred ppm
sulphur. Thus, fexibility was
needed to ensure that all feed types
could be processed. This was veri-
fed in the pilot plant phase, where
several different compositions of
RTD were successfully converted,
confrming the performance of the
catalyst and the revamp design.
The design has enabled Preem to
co-process 30% RTD with light gas
oil to produce a renewable diesel
meeting all ULSD EN 590 diesel
specifcations. A level of 30% is a
relatively high fraction of biofeed in
an existing unit not built to handle
this type of feed, and unique chal-
lenges were encountered, in
particular related to hydrogen
consumption, exotherm, catalyst
selection and corrosion, both
upstream and downstream of the
reactor. The revamped unit was
started up in 2010.
www.eptq.com PTQ Q3 2011 59
the challenges of
renewable diesel
hydrotreating can be
overcome through
technology solutions
and the use of a
specifcally designed
catalyst series
egeberg.indd 1 8/6/11 14:49:35
60 PTQ Q3 2011 www.eptq.com
quench, which makes it possible to
control heat release from the
exothermic reactions, thereby
lengthening the lifetime of the
hydrotreating catalysts by a signif-
cant degree. With fresh RTD used
as a liquid quench between the frst
two beds, more hydrogen can be
used upstream of the reactor to
increase the hydrogen partial pres-
sure. This measure helps prevent
gum/coke formation and corrosion,
thereby ultimately delivering higher
unit reliability and lower invest-
ment cost.
Selection of catalysts
The implementation of new process
technology solutions goes hand in
hand with the selection of catalysts,
as catalyst activity and selectivity
determine the necessary reaction
conditions and the composition of
reactor effuent. The conversion of
high levels of RTD in the feed
constitutes a very fast reaction,
consuming substantially higher
amounts of hydrogen than in the
case of conventional hydrotreating,
and this requires specialised cata-
lysts for the conversion of
renewable material. The correct
balance between activity and stabil-
ity depends on the feed and
conditions and has important impli-
cations for the operation of the
unit.
For the Gothenburg unit, Preem
chose a catalyst loading consisting
of an extended grading system,
Topses TK-339 and TK-341 renew-
able fuels catalysts and a Topse
high-activity Brim NiMo catalyst.
The NiMo catalyst ensures that
suffcient HDS activity is available
to reach a 10 ppm sulphur target
throughout the cycle. Since the RTD
is split between the frst two beds,
the reactions in both of the top beds
are quite similar, and the catalyst
loading is also almost the same. The
renewable fuels catalysts are opti-
mised for high stability and do not
possess an equally high HDS activ-
ity as the latest generation of
Brim-type catalysts. However, if a
simple old-generation catalyst with
a low activity is used, the conver-
sion of RTD is merely shifted to the
bottom of the reactor, where the high-
activity catalysts are situated. Thus,
Handling of corrosive feed and
high exotherm
The introduction of RTD poses a
special challenge with regard to
procedures for material selection, as
this is a very unconventional feed
with a very high total acid number
(TAN). It contains many uncon-
verted free fatty acids, as well as
resin acids, which may increase the
corrosion rate in pipes, heat
exchangers and the fred heater
upstream of the hydrotreating reac-
tor, where these compounds are
converted. The corrosion issue has,
in other cases, been design-limiting
when deciding the amount of tall
oil or derived material that can be
co-processed. Therefore, a high
priority was to design the feed
system to overcome these
challenges.
A special RTD feed system was
designed for this purpose by Preem
and Topse. Injection of RTD and
mixing with the mineral feed is
carried out in multiple steps (see
Figure 1). Part of the RTD is intro-
duced at an injection point after the
fred heater and prior to entering
the reactor. In this way, all existing
process equipment upstream of this
injection point is left unaffected.
The remaining part of the RTD feed
is introduced between the frst two
beds of the reactor to control the
temperature profle, but also to
control the TAN and thereby mini-
mise corrosion. Since RTD is only
injected after the fred heater and as
a liquid quench to the second reac-
tor bed, the hardware has very
limited exposure to the highly
corrosive RTD. Thus, only minor
changes to the material selection
were necessary. These changes
meant that the unit is prepared for,
and has already experienced, opera-
tion with an even higher fraction of
RTD feed.
Following its revamp to renewa-
ble diesel production, the unit
operates at a lower average temper-
ature, but with signifcantly higher
hydrogen consumption. As a result
of exothermic hydrodeoxygenation
reactions, the heater duty and fuel
consumption of the unit are lower
compared with what is seen
in normal hydrodesulphurisation
mode. This means that the added
reaction heat is also utilised in fossil
light gas oil hydrodesulphurisation,
so this operation is less energy
consuming. However, the large
amount of heat released due to the
hydrogenation of RTD in the reac-
tor must be controlled by a proper
quench system. Although the main
purpose of splitting the RTD into
several streams and delaying the
mixing is to eliminate the risk of
corrosion, another equally impor-
tant purpose is to provide a liquid
Fired
heater
Hydrotreating
reactor with
four catalyst
beds
LGO
Feed
RTD
Make-up +
recycle H
2
To amine unit
Product
Hot
separator
Heat
exchanger
Figure 1 Revamped hydrotreater for RTD co-processing at Preem Gothenburg refnery
egeberg.indd 2 8/6/11 14:49:43
coking and delta P build-up in this
part of the reactor must be foreseen.
Tests in a semi-adiabatic pilot unit
demonstrated that the chosen cata-
lyst system was able to gradually
convert the RTD, indicated by a
linear temperature increase down
through the reactor bed. The tests
also showed that this confguration
was very stable and was operable
for extended periods without pres-
sure drop issues.
Implications of increased CH
4
, CO
and CO
2
formation
The reactions of RTD result in the
formation of water and oxygen-
containing gases such as CO and
CO
2
. Both CO and CO
2
pose prob-
lems that had to be solved through
process modifcations. The unit is
equipped with an amine wash for
the removal of H
2
S, and this opera-
tion will also remove a large
fraction of the CO
2
formed in decar-
boxylation reactions. However,
downstream of the reactor CO
2
can
form carbonic acid if liquid water is
present. This means that the risk of
carbonic corrosion in the air cooler
and cold separator is high and
proper measures had to be taken to
eliminate this risk.
The increased amount of methane
produced will build up in the recy-
cle gas loop if a sizeable purge is
not used. CO can be removed by a
chemical transformation but, in this
case, the increased purge rate was
suffcient to keep the CO at a toler-
able level. Furthermore, the choice
of catalysts ensured that CO inhibi-
tion and the associated activity
penalty did not become an issue.
However, in downstream units, CO
can cause problems, and the impli-
cations of this must be scrutinised.
The existing purge gas recovery
unit is a cryogenic unit that cannot
remove CO and, since the purge
gas is used in another refnery unit
where CO would be a catalyst
poison, a way of removing CO had
to be implemented.
To remove CO from the purge
gas stream, Topse designed a
methanator, where CO reacts with
hydrogen to form methane by the
following reaction:
CO + 3H
2
CH
4
+ H
2
O
www.eptq.com PTQ Q3 2011 61
This elimination of CO by means
of a nickel-based methanation cata-
lyst relies on experience in the
design of ammonia plants, where
methanation can be regarded as a
proven technology. Alternatively,
these components can be removed
by pressure swing absorption (PSA)
if the refner has spare capacity in a
PSA unit.
Operating experience with
renewable diesel at Gothenburg
General operating strategy
The reactor was successfully loaded,
the catalysts were sulphided and
the unit was brought on-stream in
May 2010. The unit operates in
several modes in order to process
different streams into various prod-
ucts. Generally, three modes of
operation are employed: a light cut
of a light gas oil (LLGO), a heavy
cut of a light gas oil (HLGO) and
renewable diesel. In the latter mode,
LLGO is co-processed with the
biofeed. As an additional challenge,
the RTD feed was not available for
processing at the start of the cycle,
so it was decided that rapeseed oil-
derived fatty acid methyl esters
(RME) should be processed instead.
A design check of the revamped
unit showed that there were no
major issues in using RME as feed-
stock, although a somewhat higher
CH
4
yield can be expected
compared with the RTD case. This
demonstrates the feed fexibility of
the hydrotreating approach to
making renewable fuels, which
reduces the commercial risk of the
operation, since it does not depend
on a single source of feedstock.
Apart from a few weeks of
planned shutdown, the unit has
been on-stream since the start-up
and been continuously operated in
all three modes. The units perform-
ance has been very satisfactory and
in good agreement with design
predictions. In the following
sections, key performance data are
shown to illustrate how the chal-
lenges usually faced in renewable
co-processing have been overcome.
Reactor exotherm and hydrogen
consumption
Figure 2 shows the reactor exotherm
for the different operating modes.
As can be seen, the exotherm is
much higher in renewable mode
due to the hydrodeoxygenation
reactions. Most of this exotherm is
related to reactions in the frst two
beds. In HLGO and LLGO modes,
the operation of the unit is quite
stable and the slightly lower
exotherm towards the end of the
period is related to slightly differ-
ent operating conditions and type
of crude processed. The measured
60
70
50
40
30
20
10
0 30 60 90 120 150 180 210

R
e
a
c
t
o
r

e
x
o
t
h
e
r
m
,

C
Run day
0
HLGO
LLGO
Renewable co-processing
Figure 2 Reactor exotherm versus run days
To remove CO from
the purge gas stream,
Topse designed a
methanator, where
CO reacts with
hydrogen to form
methane
egeberg.indd 3 13/6/11 14:16:32
both exotherm and hydrogen
consumption are directly related to
the amount of renewable material
in the feed. Figure 3 shows how the
unit has been co-processing a
widely varying amount of RTD and
RME, and how the exotherm and
hydrogen consumption have
responded to this. Clearly, a quite
excessive strain is put on the heat
management and hydrogen require-
ments when co-processing even
minor amounts of RTD or RME. At
the targeted 30 vol% co-processing
level, hydrogen consumption was
in line with pilot plant testing and
design predictions.
Although the target was origi-
nally to co-process no more than
30% of RTD, a small test was made
to evaluate the possible use of
higher amounts of RTD. Figure 4
shows that up to 55% of RTD has
been processed, with an associated
increase in hydrogen consumption.
Based on this trial, it was concluded
that the unit responded well and
can be expected to be able to handle
these higher amounts for longer
periods of time, but, of course, this
will require a higher availability of
hydrogen.
Catalyst activity and deactivation
Since catalyst HDS activity must be
suffcient to reach 10 ppm sulphur
in all modes throughout the cycle,
it is important to monitor constantly
feed, conditions and product prop-
erties. Topse aids in evaluating
this data and in following the
development in catalyst activity.
Based on this, recommendations are
made to ensure that, for instance,
maximum catalyst life or maximum
number of barrels processed is
reached. In this case, where differ-
ent feed modes are used and where
renewable co-processing increases
the risk of catalyst deactivation
when generic catalyst types are
used, it is particularly important to
monitor the catalyst activity closely.
Figure 5 shows how the weighted
average bed temperature (WABT)
has evolved in the cycle so far, and
shows that fairly equal tempera-
tures have been used in the three
modes of operation. The HLGO
mode requires slightly higher
temperatures than the LLGO mode.
62 PTQ Q3 2011 www.eptq.com
exotherm is in good agreement with
the design value.
It was more attractive to Preem to
produce a diesel product with a
high renewable content for shorter
periods than to run with a lower
content for longer periods. Figure 2
demonstrates that this strategy has
been implemented and that it is
possible to shift quite frequently
between different modes of
operation, despite the signifcant
difference between feeds.
Since all of the hydrogen-consum-
ing reactions are exothermic, the
exotherm is closely related to
hydrogen consumption. In the
production of renewable diesel,
60
70
50
40
30
20
10
0 5 10 15 20 25 30 35 40 45

T
o
t
a
l

T
,

C
160
180
200
140
120
100
80
60

H
y
d
r
o
g
e
n

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

N
m
3
/
m
3
Renewable in feed, %
0
Total T
H
2
consumption
40
Figure 3 Reactor exotherm and hydrogen consumption vs vol% of renewable component
in the feed
180
220
140
30% 35% 40% 45% 50% 55% 60%

H
y
d
r
o
g
e
n

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

N
m
3
/
m
3
Renewable in feed, %
100
Figure 4 Hydrogen consumption during trial with high amounts of RTD
360
370
350
340
330
320
310
0 30 60 90 120 150 180 210

W
A
B
T
,

C
Run day
300
HLGO
LLGO
Renewable co-processing
Figure 5 Weighted average bed temperature
egeberg.indd 4 10/6/11 10:55:50
Bryan Research & Engineering, Inc.
P.O. Box 4747 Bryan, Texas USA 77805
979-776-5220 www.bre.com sales@bre.com
Comparing Physical Solvents for Acid Gas Removal
PROCESS INSIGHT
Physical solvents such as DEPG, NMP, Methanol, and Propylene Carbonate
are often used to treat sour gas. These physical solvents differ from chemical
solvents such as ethanolamines and hot potassium carbonate in a number of
ways. The regeneration of chemical solvents is achieved by the application
of heat whereas physical solvents can often be stripped of impurities by
simply reducing the pressure. Physical solvents tend to be favored over
chemical solvents when the concentration of acid gases or other impurities
is very high and the operating pressure is high. Unlike chemical solvents,
physical solvents are non-corrosive, requiring only carbon steel construction.
A physical solvents capacity for absorbing acid gases increases signicantly
as the temperature decreases, resulting in reduced circulation rate and
associated operating costs.
DEPG (Dimethyl Ether of Polyethylene Glycol)
DEPG is a mixture of dimethyl ethers of polyethylene glycol.
Solvents containing DEPG are marketed by several companies including
Coastal Chemical Company (as Coastal AGR

), Dow (Selexol), and


UOP (Selexol). DEPG can be used for selective H
2
S removal and can be
congured to yield both a rich H
2
S feed to the Claus unit as well as bulk CO
2

removal. DEPG is suitable for operation at temperatures up to 347F (175C).
The minimum operating temperature is usually 0F (-18C).
MeOH (Methanol)
The most common Methanol processes for acid gas removal are
the Rectisol process (by Lurgi AG) and Ifpexol

process (by Prosernat). The


main application for the Rectisol process is purication of synthesis gases
derived from the gasication of heavy oil and coal rather than natural gas
treating applications. The two-stage Ifpexol process can be used for natural
gas applications. Methanol has a relatively high vapor pressure at normal
process conditions, so deep refrigeration or special recovery methods
are required to prevent high solvent losses. The process usually operates
between -40F and -80F (-40C and -62C).
NMP (N-Methyl-2-Pyrrolidone)
The Purisol Process uses NMP

and is marketed by Lurgi AG.


The ow schemes used for this solvent are similar to those for DEPG. The
process can be operated either at ambient temperature or with refrigeration
down to about 5F (-15C). The Purisol process is particularly well suited
to the purication of high-pressure, high CO
2
synthesis gas for gas turbine
integrated gasication combined cycle (IGCC) systems because of the high
selectivity for H
2
S.
PC (Propylene Carbonate)
The Fluor Solvent process uses JEFFSOL

PC and is by Fluor
Daniel, Inc. The light hydrocarbons in natural gas and hydrogen in synthesis
gas are less soluble in PC than in the other solvents. PC cannot be used for
selective H
2
S treating because it is unstable at the high temperature required
to completely strip H
2
S from the rich solvent. The FLUOR Solvent process
is generally limited to treating feed gases containing less than 20 ppmv;
however, improved stripping with medium pressure ash gas in a vacuum
stripper allows treatment to 4 ppmv for gases containing up to 200 ppmv H
2
S.
The operating temperature for PC is limited to a minimum of 0F (-18C) and
a maximum of 149F (65C).
Gas Solubilities in Physical Solvents
All of these physical solvents are more selective for acid gas than
for the main constituent of the gas. Relative solubilities of some selected
gases in solvents relative to carbon dioxide are presented in the following
table.
The solubility of hydrocarbons in physical solvents increases with
the molecular weight of the hydrocarbon. Since heavy hydrocarbons tend
to accumulate in the solvent, physical solvent processes are generally not
economical for the treatment of hydrocarbon streams that contain a substantial
amount of pentane-plus unless a stripping column with a reboiler is used.
Choosing the Best Alternative
A detailed analysis must be performed to determine the most
economical choice of solvent based on the product requirements. Feed gas
composition, minor components present, and limitations of the individual
physical solvent processes are all important factors in the selection process.
Engineers can easily investigate the available alternatives using a veried
process simulator such as ProMax

which has been veried with plant


operating data.
For additional information about this topic, view the technical
article A Comparison of Physical Solvents for Acid Gas Removal at
http://www.bre.com/tabid/147/Default.aspx. For more information about
ProMax, contact Bryan Research & Engineering or visit www.bre.com.
Typical Physical Solvent Process
Gas Component
DEPG
at 25C
PC
at 25C
NMP
at 25C
MeOH
at -25C
H
2
0.013 0.0078 0.0064 0.0054
Methane 0.066 0.038 0.072 0.051
Ethane 0.42 0.17 0.38 0.42
CO
2
1.0 1.0 1.0 1.0
Propane 1.01 0.51 1.07 2.35
n-Butane 2.37 1.75 3.48 -
COS 2.30 1.88 2.72 3.92
H
2
S 8.82 3.29 10.2 7.06
n-Hexane 11.0 13.5 42.7 -
Methyl Mercaptan 22.4 27.2 34.0 -
ProcessInsight 2008APR11.a4.rev2.indd 1 4/21/2008 9:00:47 AM
www.ptqenquiry.com
for further information
BRE 1 9/3/09 11:23:53
However, the normalised WABT in
the LLGO mode is quite similar to
that seen in the renewable mode,
which means that the catalyst is not
inhibited by the renewable compo-
nent or by any of its products.
When taking the variation in feed
properties and process conditions
into account, catalyst deactivation
is as guaranteed or lower, and the
64 PTQ Q3 2011 www.eptq.com
The product sulphur content is
shown in Figure 6. The target of 10
ppm sulphur is reached in all three
modes and does not show any
increasing trend with time, other
than can be explained by changes
in operating conditions. When
calculating the normalised WABT,
the HLGO is quite clearly 1015C
less active than the LLGO mode.
deactivation appears to proceed
gradually and is not accelerated by
running in renewable mode.

Pressure drop
With the introduction of fast and
very exothermic reactions in the top
two beds during renewable co-
processing, it is particularly
important to monitor the build-up
of pressure drop, as any gum or
coke formation can quickly lead to
an increased pressure drop, which
in turn will lead to very short cycle
lengths. As described above, this
was an important point in the tech-
nology design and catalyst selection.
It was also a parameter that was
monitored during pilot testing.
The data shown in Figure 7 illus-
trate that there is a great deal of
variation in the observed pressure
drop, mainly attributable to experi-
mental scatter. Most importantly,
there is no increasing trend with
time and no sign that any perma-
nent dP build-up is taking place.
Also, the pressure drops over all
four beds are comparable and not
higher over beds 1 and 2, where the
conversion of renewable material is
taking place.
To further elucidate the causes of
the trends in the observed dP,
Figure 8 shows the total pressure
drop along with the feed rate and
fraction of renewable content in the
feed. As can be seen, most of the
variation in pressure drop is related
to the changing feed rate. When
operating in renewable mode, a
much higher treat gas rate is used,
and the pressure drop thus also
increases. However, this is a purely
temporary phenomenon and, when
returning to LLGO or HLGO mode,
the pressure drop returns to a lower
value and no permanent increase is
observed.
A test run was performed to
validate the design and demonstrate
that all design predictions could be
met. This test run was performed
with 32 vol% RTD and showed that
all design guarantees were met,
including hydrogen consumption,
yields, RTD conversion and all guar-
anteed product properties such as
sulphur content. In short, all
concerned are satisfed with the
performance of the revamped unit,
15
10
5
0 20 40 60 80 100 120 140 160 180 200

P
r
o
d
u
c
t

s
u
l
p
h
u
r
,

w
t

p
p
m
Run day
0
Figure 6 Product sulphur content
2.5
2.0
1.5
1.0
0.5
0 30 60 90 120 150 180 210
R
e
a
c
t
o
r

p
r
e
s
s
u
r
e

d
r
o
p
,

b
a
r
Run day
0
dP bed 1 dP bed 1
dP bed 2
dP bed 3
dP bed 4
Figure 7 Pressure drop over four catalyst beds
60
70
50
40
30
20
10
0 30 60 70 120 150 180 210

F
e
e
d

r
a
t
e
,

m
3
/
h
R
e
n
e
w
a
b
l
e

i
n

f
e
e
d
,

v
o
l
%
Run day
0
6
7
5
4
3
2
1

T
o
t
a
l

p
r
e
s
s
u
r
e

d
r
o
p
,

b
a
r
0
Feed rate
Renewable in feed
Total dP
Figure 8 Total pressure drop, feed rate and amount of renewable component in feed
egeberg.indd 5 8/6/11 14:50:11
proving the successful implementa-
tion of a large-scale production of a
middle distillate fuel with a high
content of renewable diesel from a
non-edible waste material. The
production of a renewable diesel
does not increase deactivation or dP
build-up rates and is fully integrated
with the desulphurisation of conven-
tional petroleum diesel.
Conclusions
Renewable diesel from the hydrot-
reating of biofeedstocks is a drop-in
fuel that can replace mineral diesel
using existing infrastructure and
conventional engine technology. A
co-processing scheme reduces
energy consumption and increases
the energy effciency of the unit.
The process is versatile in terms of
feed type and thus offers great
potential for future operation on,
say, algae oils or other unconven-
tional second-generation feedstocks.
When compared with conven-
tional hydroprocessing of petroleum
diesel to make ULSD, the conversion
of fast-reaction oxygenates requires
quite different catalyst characteristics
and poses new challenges with
respect to equipment and process
design. If unaddressed, the conse-
quences of introducing new reaction
chemistry, such as high hydrogen
consumption and large exotherms
across the catalyst beds, will quickly
lead to catalyst deactivation and
fouling. An important tool to over-
come these issues is the development
of speciality catalysts for renewable
fuel operation to ensure low deacti-
vation rates and a high stability
towards fouling. These catalysts may
be combined with a high-activity
Brim catalyst to ensure that ULSD is
produced and also with the dewax-
ing catalysts TK-928 and TK-932,
which, through isomerisation, will
provide the necessary improvement
in cold fow properties to meet cloud
point specifcations.
BRIM is a mark of Haldor Topse.
Rasmus G Egeberg is General Manager,
R&D, Refnery Technologies with Haldor
Topse, Lyngby, Denmark, responsible for
the development of process technology for
hydrotreating and hydrocracking. He holds
an MSc in physics from the University of
Copenhagen and a PhD in surface physics/
chemistry from the Technical University of
Denmark. Email: rce@topsoe.dk
Kim Grn Knudsen is Vice President, R&D, with
Haldor Topse and Director of refnery catalyst
and technology research and development
projects and Director of new concepts and
discovery research and development projects.
He holds MSc and PhD degrees from the
Department of Chemical Engineering, Technical
University of Denmark. Email: kik@topsoe.dk
Stefan Nystrm has worked in petroleum
refning for since 1977 and is part of the Preem
refnery development team in Gothenburg,
Sweden, involved in the development of
renewable production. He holds a MSc degree
in chemical engineering from Chalmers
University of Technology, Sweden.
Email: stefan.nystrom@preem.se
Eva Lind Grennfelt works within the refnery
development department of Preem and was
previously a process engineer. She holds a
MSc in chemical engineering from Chalmers
University of Technology.
Email: eva.lind-grennfelt@preem.se
Kerstin Efraimsson has been with Preem since
1990, working mainly as a process engineer.
She holds a MSc in chemical engineering from
Chalmers University of Technology.
Email: kerstin.efraimsson@preem.se
www.eptq.com PTQ Q3 2011 65
www.ptqenquiry.com
for further information
HOERBIGER valves are
the Engineer's best friend.
Over 100 years of engineering
know-how and expertise have led to
our latest innovation: the CP valve.
T
h
e

n
e
w

C
P

v
a
l
v
e

f
o
r

s
m
a
l
l

p
o
c
k
e
t
,

h
i
g
h

s
p
e
e
d

c
o
m
p
r
e
s
s
i
o
n
.
www.hoerbiger.com
anzeige_ptq.indd 1 14.02.2011 15:06:23
egeberg.indd 6 8/6/11 14:50:21
GET A RETURN ON INVESTMENT
WITHIN 24 HOURS
Using ITW Online Cleaning on a proactive way will help improve
TURNAROUND OPERATIONS and OPERATIONAL EXCELLENCE
An entire Process Unit can be ITW Online Cleaned in as low as 24 hours on an oil-to-oil basis. Our cleaning technology will
clean all of the Units equipment quickly, safely and environmental friendly.
To recover a Units performance, ITW Online Cleaning can be applied in a Turnaround during shutdown operations, to
reduce/eliminate the need for mechanical cleaning, or during the run.
Furthermore, use of our cleaning technology eliminates the need to open or enter hazardous process equipment.
ITW Online Cleaning is the only technology which has an immediate Return On Investment (ROI).
By using ITW Online Cleaning during the run, to recover Units performance, at Unit start-up (just after 24h, ITW
Technology has been applied), all the benefits are immediately monetized in terms of:
- improved throughput
- improved energy efficiency
- improved product quality
- reduced emission taxes
Regular application of ITW Online Cleaning will target an increased run length under clean conditions, with a far higher
ROI.
For Turnaround applications, a further ROI increase can be added by applying ITW Improved Degassing/Decontamination
to achieve quick and effective safe entry conditions.
EVALUATE ITW ONLINE CLEANING TODAY TO IMPROVE YOUR PLANTS PROFITABILITY !
For more informations contact : ITW S.r.l.- C.da S.Cusumano - 96011 Augusta - Italy
Tel. +39 (0931) 766011 E-mail: info@itw.it - www.itw.it
J
Join ITW Team worldwide and send your Curriculum Vitae to : info@itw.it
ITW
Innovative
Technologies
Worldwide
Now hiring
Professionals
Worldwide
www.ptqenquiry.com
for further information
itw.indd 1 10/12/09 12:46:08
Improving solids handling in
a delayed coker
P
etroleum refneries are
increasingly adding cokers
and hydrocrackers to meet
refning challenges from heavier/
more sour feedstock and increasing
demand for higher distillate/
gasoline ratios with very low
sulphur levels. Uplift margins on
US Gulf Coast cokers have moved
to over $30 per barrel, so that cokers
are among the most important units
in a modern complex refnery.
Unfortunately, these new process
confgurations create very high
volumes of solids. A typical large-
scale delayed coker produces well
over 1 million t/y of solids, which
would fll 3035 railcars in a day.
That is a lot of solids handling,
which, for most refners, is not a
core operational competency.
Interruptions in coker units are
widespread due to unplanned
outages in the industry as a result
of breakdowns in the solids
handling systems. Shortening coke
drum reheat cycles to compensate
for downtime can lead to prema-
ture drum cracking, resulting in
signifcant downtime and lost
proftability.
Marsulex has developed a total
maintenance and reliability model
for delayed coking units (DCU) to
help refners plan a preventive
maintenance strategy to signif-
cantly reduce the total cost of
ownership for coke handling. This
model maximises a cokers profta-
bility by reducing equipment repair
and replacement costs, as well as
minimising unplanned outages.
Refners who choose to carry out
their own solids handling are
encouraged to adopt a similar total
maintenance cost approach.
A reliability model applied to a coke handling system demonstrates the
benefts of compliance with a preventive maintenance programme
RAndy Hull, Steve StewARt and JAmeS teetS
Marsulex
Since most refnery operations are
characterised as continuous proc-
esses handling liquid and gas
streams, handling solids in a batch
process operation is usually not an
area of excellence within refnery
operations. Effective performance
of the coke handling and batching
operation requires effective
performance in safety, operations
and reliability.
Operations know-how and expe-
rience constitute the second critical
area for the coke cutting business.
The coke drum heading and
unheading systems, the coke pit
and the coke conveyance areas are
highly automated with solids
handling technology that requires
well-trained operators.
The third area is reliability, which
is the focus of this article. Since the
coke cutting and movement busi-
ness is so mechanically intensive,
this is an area of high cost with
many opportunities for cost reduc-
tion. Reliability separates the
competition in DCU solids handling
and maintenance; however, with
the constant pressure to reduce
costs, eliminating reliability
programmes is a short-term fx with
long-term consequences. It is
Marsulexs philosophy and experi-
ence that reducing the reliability
investment namely, scaling back
preventive maintenance and
systems investment can be an
expensive decision in the long run.
The acquisition cost dollars saved
when bidding on a services contract
are insignifcant when compared to
the total lifetime cost from unrelia-
ble, poorly maintained equipment.
A world-class preventive mainte-
nance (PM) programme is key to
delivering a successful operation,
whether a refner chooses to self-
perform or elects to outsource these
services.
Reliability philosophy
Marsulexs philosophy of reliability
is best described by the Maintenance
Parabola shown in Figure 1.
The Maintenance Parabola is an
www.eptq.com PTQ Q3 2011 67
T
o
t
a
l

c
u
s
t
o
m
e
r

c
o
s
t
,

$
PM cost, $
A well managed PM
programme should
function just to the left of
the minimum maintenance
cost line
Figure 1 Maintenance Parabola: total maintenance vs PM costs
marsule hull.indd 1 8/6/11 14:53:56
68 PTQ Q3 2011 www.eptq.com
compliance levels, which were esti-
mated in the development of the
reliability model.
The values vary by equipment
type, but the trend is consistent:
increased PM compliance yields a
higher distribution of Type I fail-
ures. Conversely, decreasing PM
compliance, or an ineffective PM
programme, yields an increased
percentage of Type II through Type
IV failures.
The objective of an effective PM
programme is to reduce costly and
unplanned failures by increasing
the less expensive preventive and
planned maintenance activities.
Reliability study
A company operating a large coker
on the US Gulf Coast was chosen
for a detailed reliability assessment,
with the objective of building and
correlating the reliability model of
the coke handling system. The
purposes of the study were to
examine our reliability philosophy,
quantify site PM compliance, and
determine the relative benefts and
costs from the identifed PM
compliance levels. The assessment
consisted of a detailed review of
the preventive maintenance
programme, development of an
equipment repair table, a multi-year
review of actual maintenance repair
statistics, and development of a
Monte Carlo reliability model, relat-
ing PM compliance to maintenance
repair cost.
Preventive maintenance review
The PM review involved a detailed
examination to identify eight key
elements. A sample PM plan for the
coke crusher outlines this review
and is shown in Table 1. (Note:
operator-assigned PM is one of the
established but subtle concept. The
curve shows the relationship
between PM expenditure and total
maintenance cost. Starting at the left
of the curve and moving right, as
PM activity is increased maintenance
costs are reduced, signifcantly at
frst. As PM expenditure increases,
moving right on the curve, mainte-
nance cost reductions reach a point
of diminishing returns. To the right
of the curve, minimum total mainte-
nance costs begin to rise with
increasing PM expenditures.
Marsulexs philosophy is to stay
focused on delivering an effective
PM programme to model the above
relationship. The programme opti-
mises PM activities to deliver the
lowest total maintenance cost, which
includes total preventive and correc-
tive maintenance, as well as
consequential opportunity costs
related to lost production experi-
enced during unplanned outages.
Equipment failure background
Marsulex categorises repairs, or
failures, into four basic types:
Type I: planned or expected
failure
Type II: unplanned premature
failure
Type III: incident
Type IV: wreck.
Type I, a planned or expected
failure, is rarely complicated and
can be assigned a repair cost of 1X.
Type II, an unplanned premature
failure, has a cost that typically
averages 3X, or three times that of a
Type I failure. A Type II failure
occasionally has additional costs,
such as an operational outage. A
Type III failure, considered an inci-
dent, is a premature failure with a
typical cost of 6X. Type III failures
often involve an operations loss, or
a logistics cost to avoid an opera-
tions loss. Type IV, a wreck, is the
most severe failure. Type IV failures
average 9X or greater repair cost
versus the Type I planned/expected
failures and can include signifcant
premature failures, operations or
logistics costs and insurance claims.
Figure 2 shows a typical failure
distribution as a function of PM
15
25
20
10
5
65 75 85 95
T
y
p
e

I
I
-
I
V
,

%
0
60
100
80
40
20
T
y
p
e

I
,

%
0
PM, %
Type I expected
Type III incident
Type II expected
Type IV wreck
Figure 2 Failure types expected vs PM compliance levels
Crusher daily PM Task Time min Freq Period Number Total FTE Assigned
Check oil level in sight glass on guard (85 - 140 Super Red) Oil 10 2 D 1 20 0.042 Oper.
Inspect hopper for un-crushables Visual 10 2 D 1 20 0.042 Maint.
Verify all safety guards are in place Visual 15 2 D 1 30 0.063 Maint.
Inspect all mountings for tightness Visual 2 1 D 10 20 0.042 Oper.
Listen for unusual noises; eg, metal to metal, belts slapping Visual 10 2 D 1 20 0.042 Oper.
Inspect crusher for leaks Visual 5 2 D 1 10 0.021 Maint.
Check for excessive vibration Visual 5 2 D 1 10 0.021 Oper.
Coke crusher preventive maintenance
Table 1
marsule hull.indd 2 8/6/11 14:54:07
INTERCAT EUROPE B.V.
Stationsstraat 50
3451 BZ Vleuten
The Netherlands
Tel: +31 30 677 7010
Fax: +31 30 677 1340
www.intercatinc.com
Intercats CAT-AID line of additives offers cost effective opportunities for refners who
run heavy feeds or have problems maintaining catalyst activity and selectivities due to
metals contamination.
Our CAT-AID technologies help refners:
Improve FCC conversion
Increase ability to process heavier feeds
Reduce catalyst additions
Improve FCC coke/gas selectivities
Lower FCC SO
2
emissions
Increase propylene yield
CAT-AID allows for the immediate response to operating changes without requiring a
reformulation of the base catalyst. Couple with our state-of-the-art additive addition
systems refners can now optimize additive effciency and minimize costs.
www.ptqenquiry.com
for further information
intercat.indd 1 8/2/11 11:22:16
most effcient and effective ways to
perform PM activities.)
Repair expectation table
The repair expectation table (a
portion of which is shown in Table
2) uses the same detailed equipment
70 PTQ Q3 2011 www.eptq.com
list as the PM effort to set realistic
goals for repair frequencies and
costs.
The failures were divided into
classes based on mean time between
failures (MTBF). Generally, the
shorter-term MTBF repairs had
lower costs than the longer MTBF
repairs. All losses were evaluated
on a yearly equivalent basis, which
considered the time value of money.
For example, a repair cost of $5000
with a one-year MTBF frequency,
using a net present value (NPV)
method with a 6% interest rate,
would be equivalent to an expected
20-year MTBF repair cost of
$184 000 instead of the simple
yearly equivalent of $100 000. Using
the NPV method with a $100 000
repair cost and 20 years MTBF
frequency would produce a yearly
equivalent of $2718.
Detailed review of actual
repair cost
The third element of the study was
a multi-year review of maintenance
repair costs on the DCU used to
defne and correlate our model. The
key components of the review were
the repair cost, the number of fail-
ures per year, the cost per year
equivalent, and the failure type, I
through IV. Table 3 shows a sample
of the review.
Study feedback
The studys feedback and correla-
tions are summarised in Tables 46.
Table 4 is the PM analysis feedback
for a set of PM procedures. The
column on the left shows the type
of PM activities. Note that all of
these activities were relatively
simple tasks, such as visual checks,
changing oil, lubrication, sampling
oil and housekeeping. There were
156 activities in this PM stack. By
separating activity (time) and total
cost, this table shows how PM tasks
are generally inexpensive, but they
must be performed routinely, on
time and they must initiate
corrective action when problems
are identifed.
Table 5 shows the results of the
maintenance repairs analysis on the
unit against the ideal maintenance
levels required to deliver the lowest
total cost of ownership. The costs
were calculated considering the
time value of money to establish an
ideal maintenance base repair cost,
expressed as a dollar yearly
equivalent.
The costs are displayed as a
percentage of the total ideal
Equipment list Material cost, MTBF, # similar Repair time, Cost,
$/failure years comp. Mhrs $/y equiv.
Bridge crane
Hold cable $1000 1 2 24 $2000
Close cable $500 2 4 24 $971
Bearing $2100 4 2 4 $960
Coupling $12 500 5 2 4 $4435
Housing $30 000 20 1 2 $816
Component failure expectations
Table 2
Analysis sheet Repair expense Failures/ yr Yearly expense Failure type
Top head assy. 4.17 5.0 20.8 I
Bolts and nuts 3.47 0.3 1.0 I
Bolts and nuts 0.80 3.0 2.4 I
Gasket 1.33 0.5 0.7 I
Gasket surface 1.10 0.5 0.6 I
N2 skid:
N2 pumps 1.00 0.2 0.2 I
N2 skid controls 75.00 1.0 75.0 II
N2 bottles 75.00 1.0 75.0 II
North console 7.00 1.0 7.0 II
South console 0.50 3.0 1.5 I
Terminal load out:
Filters 2.50 4.0 10.0 I
Silo 2.00 1.0 2.0 I
Scale 3.00 0.5 1.5 I
Hydraulic pressure unit 4.00 1.0 4.0 I
Hydraulic pressure unit 8.33 1.0 8.3 II
HPU pump 1.00 0.5 0.5 I

Maintenance repair cost analysis sheet
Table 3
PM # of activities % of total activities % of total cost
Change flters 1 0.6 23.3
Clean 5 3.2 14.1
Housekeeping 1 0.6 11.1
Check visual 78 50.1 26.1
Lubricate 5 3.2 2.6
Change oil 27 17.4 3.5
Refurbish 10 6.4 2.7
Test 2 1.3 0.4
Grease 18 11.6 1.4
Calibrate 4 2.6 1.4
Torque 1 0.6 0.1
Inventory 1 0.6 8.9
Hydro-blast 1 0.6 3.3
Change bulbs 1 0.6 1.1
Sample oil 1 0.6 0.0
156 100.0 100.0
PM procedures
Table 4
marsule hull.indd 3 10/6/11 10:56:41
1MBUJOVNt1BMMBEJVNt3IPEJVNt3VUIFOJVNt3IFOJVNt(PMEt4JMWFS
Sabins precious metals recovery and rening
processes raise environmental protection to an advanced
state for safety and regulatory compliance. Your
advantage: assured peace of mind, and our promise
of maximum value returns from your precious metal-
bearing materials.
Learn more at sabinmetal.com
Think globally. Act responsibly.
Processing facilities and technical service offices in the Americas, Europe and Asia
Sabin-ThinkGlobally.indd 1 2/26/09 2:28 PM
sabin.indd 1 26/2/09 21:38:16
www.ptqenquiry.com
for further information
sabin.indd 1 8/6/09 14:22:15
severity from expected, operations
and logistics, and spare parts.
The frst two cost elements, PM
and training, increase as percentage
PM compliance increases. The third
cost element (repairs) remains
constant regardless of PM compli-
ance levels. The remaining cost
elements increase with decreasing
PM compliance. Note that, with a
decrease in PM investment from
9065% compliance, which would
save $250 000 in budget investment,
the total cost of maintenance actu-
ally rises by over $2 000 000. The
costs predicted by the model repli-
cated our actual experience over
several years on this large DCU.
The initiative to reduce budgeted
and planned costs such as PM can
be initiated by numerous events,
including budget cuts to reduce
operating expense and/or head-
count, short-term production
increase requirements that tend to
drive operators to defer PM, evalu-
ating service suppliers based solely
on lowest bid price and not lowest
total cost of ownership, and many
others. However, these planned
savings are often only an illusion.
Conclusions from modelling
exercise
The model assumptions and devel-
opment yielded predicted results
that closely matched actual experi-
ence and costs. This outcome
suggests that the model can provide
an effective planning and opera-
tions tool. Ancillary cost, MTBF,
severity, operations losses and
spare parts costs are much larger
than the base repair cost at low PM
compliance. Placing a low priority
on, or deliberately reducing, PM
expenditures can be an expensive
strategy, as costs rise quickly when
PM compliance falls below 90%.
Diligent compliance with an effec-
tive PM programme has an excellent
return. It is not just about doing the
correct volume of PM; it is also
about knowing the right PM activi-
ties to perform.
Additional benefts of
reliability study
The study helped focus the
reliability programmes continuous
improvement efforts and validated
72 PTQ Q3 2011 www.eptq.com
maintenance base cost. The largest
variance was in the miscellaneous
category, which tends to be a catch-
all for maintenance cost. The total
cost was 167.8% of the ideal base,
which is broken down by failure
type in Table 6.
There were 85.3% planned,
expected Type I repairs, 12.6%
unplanned Type II repair costs, and
2.1% Type III incidents, with no
Type IV wrecks. Referencing the
projected repair distributions shown
earlier, the model suggests 90% PM
compliance.
Development of Monte Carlo
PM model
A statistically based Monte Carlo
method analysis was undertaken to
expand and enhance the site study
and to examine all of the cost
components affected by PM compli-
ance levels.
Using data collected from the
study, specifcally the failure expec-
tation table and historical reliability
information, site-specifc failure cost
tables were developed. A Weibull
analysis was then used to simulate
failure frequencies at varying PM
compliance levels. The relationship
assumed that, as PM compliance
was reduced, the failure frequency
widened and the MTBF was
reduced or shifted left, as shown by
the failure distribution curves in
Figure 3.
The results of the model runs,
which simulated numerous 20-year
cycles, are shown in Figure 4. These
results were based on actual main-
tenance schedules, failure rates and
costs from the DCU reliability
study.
Explanation of model results
The chart shown in Figure 4 uses
the following cost categories to
calculate overall maintenance
expense: preventive maintenance,
training and safety, repairs,
frequency increase from expected,
60
100
80
40
20
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
P
r
o
b
a
b
i
l
i
t
y
,

%
MTBF, years
0
70% PM
80% PM
90% PM
Figure 3 Predictive DCU model assumed probability of failure vs PM investment
Equipment Maint, % ideal Maint % actual
Misc 0 42.10
Crane 27.24 25.64
Feeder 2.31 8.65
Crusher 3.83 0.55
Heads 9.61 28.02
Cutting 30.73 34.24
Transport 2.49 5.70
Conveying 9.67 5.34
Rail 12.63 15.50
Loading 1.50 2.10
Total 100.0 167.8
Ideal vs actual maintenance cost
comparison
Table 5
Failure type % task Count
I 85.3 81
II 12.6 12
III 2.1 2
IV 0.0 0
Total 100.0 95
Actual failure types observed
Table 6
marsule hull.indd 4 8/6/11 14:54:32
our reliability philosophy. It also
highlighted several areas for
improvement in the sites PM
programme, expanded the number
of right PM activities to execute,
and further refned the relationship
between PM compliance and the
total cost of repairs.
This exercise and the successful
development of a diagnostic main-
tenance planning model reinforced
the need for additional maintenance
planning and execution systems.
The benefts of computerised main-
tenance management systems,
maintenance planning and schedul-
ing activities, critical spare parts
identifcation and risk management
were realised, as they maximise
the effectiveness of maintenance
resources. Additionally, condition-
based equipment inspection, life
cycle analysis and root cause analy-
sis play key roles in reducing the
severity of equipment failures and
preventing recurrence.
Conclusions
While planned operating costs for
the delayed coking unit are fairly
transparent and measurable, the
ability to quantify the effectiveness
of the total maintenance investment
is much more diffcult. Total main-
tenance investment is generally
www.eptq.com PTQ Q3 2011 73
comparable to the investment in
operating labour and materials.
This tempts refners to scale back
maintenance investment for short-
term gain, whether maintenance is
performed in-house or by outsourc-
ing. The cost of this short-term
focus can be signifcant, with severe
effects on refnery proftability.
Refners are encouraged to resist
this short-term cost reduction
approach to maximise the long-
term performance of critical assets.
Randy Hull is Senior Vice-President, Sales,
Marketing and Development, for Marsulex Inc
in Houston, Texas. He holds a BS in mechanical
engineering and a MBA, and is a member of
AIChE, ASME and Tau Beta Pi.
Email: rhull@marsulex.com
Steve Stewart is Director of Operations,
Refnery Services, Southern US, for Marsulex in
Texas City, responsible for refnery operations
in delayed coking units, as well as terminal
operations. He holds a BS in industrial
engineering and is a member of Alpha Pi Mu
and Tau Beta Pi.
Email: sstewart@marsulex.com
James Teets is the Reliability Manager, Refnery
Services, Southern US, for Marsulex in Texas
City, responsible for reliability systems in
refneries delayed coking units. He holds a BS
in mechanical engineering from the University
of Cincinnati and is a member of Pi Tau Sigma
and Tau Beta Pi. Email: jteets@marsulex.com
1500
2500
2250
2000
1750
1250
1000
750
500
250
65 70 75 80 85 90 95 100
M
a
i
n
t
e
n
a
n
c
e

e
x
p
e
n
s
e
PM, %
0
3000
3750
3500
3250
2750
Spares
Oper.
Sever.
MTBF
Repairs
Train
PM
Total
Figure 4 Predictive working model of DCU total maintenance costs vs PM compliance
marsule hull.indd 5 9/6/11 11:36:36
JOHNSON SCREENS

:
TRUSTED NAME, TRUSTED QUALITY
From problem detection to recommissioning of the
reactor, Johnson Screens

feld service team comes


with a wealth of knowledge and experience applicable
to numerous applications:
Project engineering assistance for revamping or
upgrading
Technical assistance when damage (on reactor
internals) is found or assumed
lnternal expertise after catalyst unloading
On-site installation of Johnson Screens' products
Supervision of contractor's installation
Trouble doesn't usually give advance warning. The
moment something goes wrong, valuable productivity
is lost. Every minute costs more and more, so speed
is critical.
Johnson Screens' feld engineers can be on-site
within 48 hours of an emergency
Upon arrival, a damage assessment and
recommendation report will be generated within 12
hours of the inspection
Johnson Screens' worldwide manufacturing
capabilities and purchasing power mean that all
necessary materials can be delivered to the
production facilities with no wasted time
Trusted Experience in Inspection, Installation and Maintenance
www.johnsonscreens.com
A Weatherford Company
AUSTRALlA - ASlA PAClFlC
TEL.: +61 7 3867 5555
salesasiapacifc@johnsonscreens.com
EUROPE - MlDDLE EAST - AFRlCA
TEL.: +33 (0)5 4902 1600
saleseurope@johnsonscreens.com
NORTH, SOUTH & CENTRAL AMERlCA
TEL.: +1 651 636 3900
salesamerica@johnsonscreens.com
JAPAN
TEL.: +81 55 997 8511
sales.jpn@johnsonscreens.co.jp
www.ptqenquiry.com
for further information
j screens.indd 1 10/3/11 10:51:34
Refnery confgurations for maximising
middle distillates
R
efners globally continue to
face numerous challenges as
environmental laws become
increasingly stringent. Principal
among them in the near future will
be the International Maritime
Organisations (IMO) proposed
changes to bunker fuel oil sulphur
limits, from the current limit of over
3.5% down to 0.5% globally and
from 1% to 0.1% in Emission
Control Areas (ECA, see Figure 1).
Global demand for high-sulphur
residual fuel oil (HSFO) is steadily
declining too, by 35% since 1995.
Both of these changes will signif-
cantly impact a refners ability to
market any signifcant quantity of
HSFO at a price that will maintain
refnery proftability. Refneries
currently making a signifcant
amount of fuel oil and lacking the
complexity to upgrade the residual
oil to premium products (middle
distillates) will face two diffcult
options: either invest in commer-
cially proven and reliable solutions
to convert HSFO to more valuable
liquid products such as Euro V
diesel to greatly improve the refn-
erys proftability, or face a threat to
shut down the refnery as the opera-
tion becomes uneconomical to
continue.
Shift in product demand
The IMOs looming specifcation
changes (see Figure 1) are likely to
accelerate the decline in demand for
HSFO by the year 2020, if not earlier.
Worldwide, including emerging
markets such as China, India and
the Middle East, there is a shift in
product demand from gasoline to
diesel. Ethanol substitution in gaso-
line and improvements in engine
A comparison of confgurations for upgrading residual oil products for the
maximum production of diesel
ARun ARoRA and ujjAl MukheRjee
Chevron Lummus Global
technology are just two of the
reasons why the demand for diesel
continues to outpace that of gaso-
line. IMO regulations will indirectly
increase diesel demand further as
refners are forced to blend in addi-
tional low-sulphur diesel to meet
fuel oil sulphur specifcations.
Worldwide, production of mid-
distillates is projected to account for
55% of the rise in oil demand
expected over the next 20 years. The
shift to diesel puts emphasis on
bottom-of-the barrel processing.
Growing demand
Worldwide demand for refned
products is projected to increase
signifcantly in the next 20 years,
driven by population growth and
the transition of emerging markets
into the global economy, with the
majority of growth coming from
China in particular and Asia in
general. According to OPEC, global
demand for diesel fuels is expected
to grow by 10 million b/d by 2030,
driven by an increased share of
diesel-driven vehicles in Europe
and developing countries.
Current refning investment is
predominantly made in Asia, the
Middle East, Russia and Latin
America regions with growing
demand for refned products.
Tightening of product quality speci-
fcations will accelerate the
implementation of deep conversion
units in existing refneries, but often
these refneries are constrained by
plot space, hydrogen and other
infrastructural issues. Grassroots,
export-oriented refneries are all
geared towards high conversion to
mid-distillates.
For the strategically oriented
refner, stringent requirements for
high-quality products actually
present an opportunity to invest in
the right technologies to signif-
cantly improve refnery margins.
Based on increasing product
demand and the closure of multiple
non-performing refneries, refning
margins are expected to recover
by 2015.
www.eptq.com PTQ Q3 2011 75
3.0
5.0
4.5
4.0
3.5
2.5
2.0
1.5
1.0
0.5
2005 2007 2009 2011 2013 2015 2017 2023 2025 2019 2021
0.0
Max two-column width
Blue background around
brown.
Always translate into UK
English (except images).
Leading cap on rst word
only.
ECA
World
S
,

w
t
%
Figure 1 Requirements of IMO regulations Source: International Maritime Organisation
arora.indd 1 9/6/11 09:14:44
76 PTQ Q3 2011 www.eptq.com
current trends only such refneries
will survive in the future)
Ideally, should be part of a
conversion platform encompassing
complementary technologies.
Technologies on the cusp of
commercialisation were excluded,
because we did not want to prescribe
any solution without a reasonably
long operating history. For example,
there are several slurry-phase resi-
due conversion processes on the
verge of commercialisation, but
without a commercial operating
history there is no data on reliability
and on-stream factor a major
consideration in any residue upgrad-
ing process because of the diffcult
nature of the feedstock.
Major refnery processes included
in this evaluation were:
Delayed coking
LC-Fining (a high-conversion
residue hydrocracking process)
Residue desulphurisation (RDS)
Solvent deasphalting (SDA)
Combinations of the above, along
with secondary processes such as
hydrocracking, residue catalytic
cracking (RFCC) and gasifcation
(VR and coke), FCC feed/product
desulphurisation and various
gasoline-producing processes.
In the studies we conducted for
various clients, the residue conver-
sion technologies that rose to the
forefront were delayed coking, LC-
Fining and RDS. The screening
A wider and more intense
requirement for the deployment of
emissions reduction technologies
may also act as a catalyst for new
investments. Modern hydroprocess-
ing technology will eliminate the
need for expensive downstream
remediation technologies.
It is our view that refning should
be viewed as an ongoing business,
where long-term average margins
and product price differentials will
support the investments that are
needed.
Residue upgrading technologies
In view of the increasingly stricter
regulations expected in the near
future, along with the emerging
trends in product demand, CLG
evaluated multiple combinations of
residue conversion technologies,
keeping the intentions of a global
refner in mind. The conversion
technology:
Should be commercially proven
and reliable with a good on-stream
factor
Should maximise the most valua-
ble product (diesel) while retaining
the capability to address niche
product demands for the foreseea-
ble horizon
Should be fexible to handle more
diffcult feedstocks
Should be environmentally
compliant to meet future stringent
specifcations
Should have enough complexity
so that the refnery remains profta-
ble when margins remain depressed
for prolonged periods (based on
Coke
drum
Coke
drum
Gas oil
stripper
Reflux drum Fractionator
Residual oil
Heater
Wet gas
Unstablised
naphtha
Light gas oil
Heavy gas oil
Figure 2 Schematic of a delayed coking unit
www.ptqenquiry.com
for further information
embedded in the daily business
workfow of the end users via the
integration platform to ensure a
continuous improvement in terms of
the management of energy usage.
The end users review the results and
present them during meetings where
different business units are present.
K^lnemlZ\ab^o^]
The project yielded several benefts
for SIR. Some of them are described
below. More benefts are to be
expected as soon as further correc-
tive actions are undertaken by the
management of the refnery.
@Zbglbgmbf^Zg]^_\b^g\r
The integration of the EMS into the
daily workfow is the foundation of
a sustainable improvement through
increased awareness of energy
consumption at the refnery.
Running the application takes only
a few minutes every day, which
means a signifcant gain in time
and effciency: engineers can now
spend time on analysing data and
not waste time on collecting them.
The fow sheet-based reporting
tools continuously provide reliable
information on the use of energy at
the refnery in a transparent
manner. Previously, energy
balances were established on a
monthly basis not frequent
ppp'^imj'\hf IMJJ++)**-,
enough to take action in due time.
Since it is entirely integrated
through the integration platform,
the whole refnery can gain from
the information provided by the
EMS. For example, the refnery
production balances are established
partially based on results provided
by the EMS. It turned out that this
helped to calculate the balances
more quickly, with a 50% reduction
in time.
Ik^o^gmbhgh_ehll^l
Socit Multinationale de Bitumes
(SMB) is a company that is located
on the SIR site and strongly inte-
grated with SIR in terms of energy
consumption among other links. As
$l20 000
$l00 000
$80 000
$60 000
$40 000
$20 000
1anuary Pebruary March April May 1une
$0
?b`nk^,Ik^o^gm^]ehll^lbg_n^ebgohb\bg`%CZgnZkrCng^+)*)
ITW
Innovative
Technologies
Worldwide
ITW is a fast growing Company, marketing and implementing unique and patented Production Units Online
Cleaning, Tank Cleaning, Decontamination and Reclamation technologies, along with Specialty Chemicals.
Given the considerable success, ITW is expanding its markets and activities and is looking for experienced
professionals worldwide. The candidates should have minimum a 5 years experience in at least one of these
fields:
x refining/petrochemicals process specialty chemicals sales
x refining/petrochemicals process technology, operations, maintenance, turnaround
and should be willing to travel nationwide and worldwide, along with being prone to hard working and sales.
Having a technical degree and good market knowledge is also required.
The available positions will cover: technical sales, implementation of ITW technologies on the field,
sales and operations management.
Interesting compensation plans will be given along with serious career possibilities. Please contact:
ITW S.r.l. C.da S.Cusumano 96011 Augusta - Italy E-mail: info@itw.it www.itw.it
HIRING EXPERIENCED PROFESSIONALS WORLDWIDE
a consequence of this energy inte-
gration, SIR sells SMB fuel gas and
fuel oil to be used for treating crude
oil for the production of road bitu-
men and by-products. It turned out
that SMBs consumption of fuel has
been systematically underestimated
by about 215 tonnes of fuel oil
equivalent each month. As a result,
the average price for fuel per tonne
of crude was underestimated by
almost 20%. This underestimation
resulted in undercharging for fuel
over recent years. With the help of
the EMS, the true energy consump-
tion of SMB has been determined
with a higher precision, and losses
have been prevented by invoicing
the correct amount of energy
ppp'imj^gjnbkr'\hf
_hk_nkma^kbg_hkfZmbhg
belsim.indd 3 11/3/11 14:43:20
arora.indd 2 9/6/11 09:14:57
www.eptq.com PTQ Q3 2011 77
phase quickly ruled out several
technologies as being too expensive,
such as gasifcation, or not geared
towards maximising diesel, the
product of choice. A brief descrip-
tion of the primary upgrading
processes follows.
Delayed coking
Delayed coking is the most widely
used residue conversion technology
and is particularly valuable when a
long-term off-take arrangement for
coke exists. Almost every major
grassroots refnery in the world has
considered it as a primary residue
conversion process, with the excep-
tion of locations such as Scandinavia,
Western Europe and Eastern
Canada, where coking units are not
preferred. Fuel-grade coke is used in
infrastructure projects (cement,
power) and demand remains robust
in developing countries. However,
with even more large coking units
coming online, coke demand could
come under pressure.
Vacuum residue, normally
destined for fuel oil, is thermally
cracked to obtain nearly 70% of
distillate products. All distillate
products require further hydro-
processing to make fnished
products. Coker naphtha requires
special and more severe hydro-
processing compared to straight-run
naphtha. Light coker gas oil (LCGO)
that boils in the diesel boiling range
has a much higher nitrogen content
compared to straight-run diesel and
operating pressures required for
hydroprocessing are relatively
higher. Heavy coker gas oil (HCGO)
boils in the vacuum gas oil (VGO)
boiling range. HCGO has much
higher total aromatics, nitrogen,
polycyclic aromatics and asphaltenes,
and requires more severe operating
conditions compared to straight-run
VGO. HCGO is either sent to a FCC
feed pretreater (in a gasoline-
oriented refnery) or a hydrocracking
unit (in a diesel-oriented refnery).
The coke produced by a standalone
delayed coker (see Figure 2) is lower
value fuel-grade coke. If a hydro-
processing unit such as an LC-Fining
unit precedes the delayed coking
unit, the coke produced from the
delayed coking unit can be of supe-
rior anode-grade quality, suitable for
use in the aluminium industry.
Table 1 shows the main advantages
and disadvantages of the delayed
coking process.
LC-Fining
The LC-Fining process is a resid-
uum conversion process that
hydrocracks the most diffcult,
heavy, lower-value hydrocarbon
streams such as petroleum residua,
heavy oils from tar sands and shale
oils to lighter, more valuable prod-
ucts such as VGO, diesel and
naphtha. The process involves an
ebullated-bed reactor (see Figure 3)
that completely mixes oil and
hydrogen. Due to continuous addi-
tion and the withdrawal of small
quantities of catalyst, the run
lengths between shutdowns are
long. Unconverted oil from the LC-
Fining unit can be used as fuel oil
Table 1
Effluent
Thermowell
nozzle
Skin
TCs
Catalyst
withdrawal
line
Recycle
pump
Feed
Normal
bed level
Density
detectors
Density
detector
radiation
source
well
Catalyst
addition
line
Figure 3 Schematic of a LC-Fining reactor
with typical operating conditions
Table 2
Advantages Disadvantages
Lower on plot capital investment compared to hydrogen addition processes Coke handling, plot area limitations, and transportation and logistics
Can handle very poor-quality (high in contaminants) feeds Additional environmental health and safety (EHS) requirements
Widely used, with many references Hydrogen addition still required to upgrade products and the process
does not share the same process platform as other hydroprocessing units
Favoured in low crude oil price environment Loss of liquid yield compared to hydrogen addition processes
No residual liquid product to deal with Coke disposition is a major issue
Advantages and disadvantages of delayed coking
Start-up Client BPSD Unconverted oil
2011 Shell Canada 47 300 Stable HSSC
2010 GS Caltex 60 000 Stable FO
2007 Neste Oil 40 000 Stable FO
2003 Shell Canada 79 000 Stable HSSC
2000 Slovnaft 23 000 Stable LSFO
1998 AGIP Petroli 25 000 Stable LSFO
1988 Syncrude Canada 40 000 Coker feed
1984 BP-Amoco 75 000 Coker feed
Total 8 units 389 300
Installed LC-Fining units
Reactor temperature 410440C
Reactor pressure 110180 bar
Resid conversion 5580%
Hydrogen P. P. 75125 bar
Chem H
2
consumption 35300 Nm
3
/m
3
Desulphurisation 6085%
CCR reduction 4070%
Demetallisation 6588%
arora.indd 3 9/6/11 09:15:06
reduced if the feed quality
improves.
There are only two ebullated-bed
processes in the world that have
been proven by long commercial
history: LC-Fining and H-Oil. Table
2 is a list of operating LC-Fining
units, and Table 3 shows the main
advantages and disadvantages of
the process.
The unconverted oil from the LC-
Fining unit is normally used as fuel
oil. When combined with a delayed
coking unit downstream, the uncon-
verted oil is converted to distillates
and anode-grade coke, which
fetches a far higher price compared
to fuels-grade coke. While LC-
Fining can handle a relatively high
78 PTQ Q3 2011 www.eptq.com
or as feed to power plants or a
delayed coking unit. Maximum
conversion is dependent on feed-
stock. Operating unit conversion
ranges from 60% to over 80%.
The LC-Fining unit operates at
pressure levels similar to high-
pressure hydroprocessing and
therefore offers excellent opportuni-
ties for capital reduction by
permitting integration of either
hydrotreatment (Shell, Canada) or
complete hydrocracking (Neste,
Finland, see Figure 4).
The conversion of Conradson
carbon is economically important if
LC-Fining vacuum bottoms are fed
to a downstream coking unit. A
lower carbon-content resid product
to the coking unit means less coke
make and thus a higher yield
of liquid fractions that can
subsequently be converted to trans-
portation fuels.
The LC-Fining unit has inherent
fexibility to meet variations in
feed quality/throughput, product
quality and reaction operating
severities (temperature, space
velocity, conversion and so on).
This fexibility is a direct result of
the ebullated catalyst bed reactor
system. In an ebullated-bed unit, if
the metals or sulphur content of
the feed increases, the product
quality is maintained by increasing
catalyst consumption. Conversely,
the catalyst consumption is
LC-Fining reactors with
inter-stage stripper
To
fractionation
To
fractionation
LP amine
scrubber
LVGO
HVGO to FCC
Hydroprocessed
residue
External gas oils
for MHC/HDT
Vacuum
residue
and
diluent
Make-up
hydrogen
H
2
purification
HP amine
scrubber
HDT/MHC
reactor
Wash
tower
Hydro-
processed
distillates
and gas oils
Figure 4 LC-Fining process with integrated hydroprocessing
Table 3
Advantages Disadvantages
Higher liquid gain compared to delayed coking On plot investment is higher than delayed coking units
Can handle feeds higher in metals and other contaminants Residue stability may become a concern at high conversions (feed dependent)
compared to fxed-bed processes Middle Eastern feeds, for example, have no stability concerns even at high
conversions (proven commercially)
Long run lengths More complex process compared to delayed coking and requires better operator training
Can be integrated easily with other hydroprocessing units Not as much commercial experience as delayed coking units but adequate
Ebullated-bed technology is a mature technology and over Spent catalyst disposal (trucks, rail car) has to be considered. Spent catalyst normally
30 years operating experience has led to many technological sent to metals reclaimer
advances and made the process very reliable
Requires less plot space compared to delayed coking units Unconverted oil disposition can become an issue depending on sulphur/stability
specifcations
Advantages and disadvantages of LC-Fining
arora.indd 4 9/6/11 09:15:15
metals content in the feed, the high
level of nickel and vanadium in the
unconverted LC-Fining bottoms
could limit the production of anode-
grade coke in the downstream
delayed coking unit. LC-Fining by
itself produces signifcantly more
liquid yield compared to delayed
coking and improves the refners
volume gain.
The LC-Fining process is also easily
integrated with a solvent deasphalt-
ing unit either upstream (see Figure
5), downstream (see Figure 6) or as
an inter-stage process.
An upstream SDA signifcantly
reduces metals, CCR and
asphaltenes. Operating conditions
required in the LC-Fining unit
www.eptq.com PTQ Q3 2011 79
metals, nitrogen and sulphur from
petroleum residua in the presence of
hydrogen (see Figure 7). Conversion
results from the level of desulphuri-
sation required and is not by itself a
target. The process is normally used
to produce low-sulphur fuel oil or to
produce a feed stream that is suita-
ble for cracking in a residue FCC
(RFCC) unit.
The RDS process is by far the
most widely used residue upgrad-
ing process. Catalyst and process
innovations for RDS include the
upfow reactor (UFR) and onstream
become less severe and conversions
can be pushed much higher. The
yield slate shifts towards lighter
products and catalyst consumption
drops signifcantly. Without heavy
asphaltenes in the process, unit
operating factors improve as well.
The obvious disadvantage is the
loss of global conversion, as a
signifcant volume of residue is
removed as pitch and, without a
dedicated disposition of the large
volume of pitch (such as a gasifer),
the economics may not be favoura-
ble. The option becomes very
attractive in those situations where
an SDA is already in operation and
there is a need to upgrade the DAO
to diesel rather than routing to an
FCC unit for conversion to
gasoline.
The SDA process can also be
integrated downstream, where
deasphalting removes the heaviest
asphaltenic residue from the
unconverted oil. The DAO can be
recycled back to the LC-Fining
process, while the pitch can be
blended in with incremental VR to
an existing delayed coking unit
(BP, Texas City). Conversion is
boosted and the volume of pitch
to be handled is reduced
signifcantly.
Residue desulphurisation
RDS is a fxed-bed process that has
multiple beds of catalyst to remove
VDU
!TMOSPHERIC
RESIDUE
Solvent
deasphalting
LC-Fining
,OW3.
DISTILLATES
AND6'/
0ITCHPELLETS ,3&/
VDU
!TMOSPHERIC
RESIDUE
Solvent
deasphalting
LC-Fining
,OW3.
DISTILLATES
0REMIUM
&##2&##
FEED
0ITCHPELLETS
DELAYEDCOKING
Figure 5 SDA upstream of an LC-Fining
unit
Figure 6 SDA downstream of an LC-Fining
unit
Filter
Hot HP
separator
Cold HP
separator
LP
separator
Product
stripper
Make-up
hydrogen
Fresh feed
H
2
O
Sour
water
Recycle gas
scrubbing
H
2
S
To gas recovery
Unstabilised
naphtha
Product
Reactors
Figure 7 Schematic of RDS process
arora.indd 5 9/6/11 09:15:25
80 PTQ Q3 2011 www.eptq.com
CDU
VDU
DCU
CFB
NHT
HCR DHT
H
2
plant
Diesel
Fuel coke
Fuel oil
Jet A-1
Petro-chemical
Naphtha
LPG
LPG
Light naphtha
Heavy naphtha
Coker
Naphtha
HCGO
LCGO
UCO
Figure 9 A delayed coking-based refnery with no gasoline production
CFB: Circulating fuidised bed boiler
Atmospheric
tower
Vacuum
tower
HDT
FCC
Catalytic
reforming
Delayed
coking
Gasoline
Jet/diesel
Coke to
power plants
and
cement industry
Hydrocracking
Figure 8 Simplifed refnery with delayed coking as a residue conversion process
catalyst replacement (OCR). UFR is
typically used in revamp situations
or when concerns about metals
levels and catalyst pore mouth
plugging might shorten downfow
reactor run lengths due to excessive
pressure drop. OCR is used when
the metals level in the feed is
excessive.
RDS is a widely used technology,
especially in the Far East. It is the
only technology that can produce
<0.5 wt% sulphur fuel oil. The
technology is used in this context
in Japan, but the most prevalent
use of RDS is as a unit feeding a
RFCC unit for the production of
gasoline.
Upgrading confgurations
CLG explored several confgura-
tions, including some that are based
on the residue upgrading platforms
described here, along with other
major processes such as hydroc-
racker, hydrotreater and FCC
to maximise conversion to
mid-distillates.
Refnery with delayed coking as
primary upgrader
The confguration shown in Figure
8 is one of the most common refn-
ery confgurations and is a
benchmark against which other
confgurations have been evaluated.
The confguration is robust and,
depending on the crude slate, the
capacities of the hydrocracking and
FCC unit vary to obtain the right
balance between gasoline and diesel
production. In extreme situations,
where gasoline production is to be
avoided, the confguration will have
no catalytic reforming and no
FCC unit.
Figure 9 shows a refnery confg-
uration where there is virtually
no demand for gasoline and
the refner is only interested in
making middle distillates and
petrochemicals naphtha. Such a
arora.indd 6 9/6/11 09:15:35
Background
Refners operating FCCUs have adjusted their operating strate-
gies to maximize light cycle oil production to meet the increased
demand for automotive diesel. Catalyst formulations and reac-
tor conditions can alter yields, but the refnery cannot take full
advantage of the increased LCO recovery without an optimized
Main Fractionator operation and design. Sulzer offers several
methods of optimizing the separation between light cycle oil
and less valuable HCO or slurry with little or no investment.
The Driving Force
Product overlaps between light and heavy cycle oil or light cycle
oil and slurry can contribute to a signifcant portion of refnery
gas oil volumes. Shrinking these overlaps can be very prof-
itable for refnery operations and increase the overall refnery
diesel pool.

First, Optimize Your Base Case
Before considering design changes in the Main Fractionator, it
is important to begin with an optimized base case. LCO can
be downgraded to HCO or slurry product by the redistribution
of the pumparound and refux duties. The cooling duty to the
bottom pumparound is often provided by reboiling the Gas Con
Stripper or Debutanizer column. Sometimes the demands of
the gas plant columns necessitate high bottom pumparound
duties, resulting in low LCO product endpoints. In these cases,
Sulzer Chemtech
Tower Technical Bulletin
Maximizing Light Cycle Oil Recovery in the FCC Main Fractionator
the most benefcial investment may be a supplemental reboiler
to the Gas Con. This allows the Main Fractionator pumparound
duty to be reduced, which then allows more diesel-ranged ma-
terial to travel up the column to be pulled as LCO.
Many refneries also employ external jumpovers from the pump-
around directly to the LCO product stripper, in order to spike
the endpoint. This method can increase the overall distillate
product rate, but it provides relatively poor direct endpoint con-
trol.
Now, Consider Design Improvements
The balance of heat and pumparound duties in the column re-
quired to achieve maximum LCO endpoint may require higher
hydraulic capacity and / or improved stage fractionation eff-
ciency.
Increasing hydraulic capacity in the column is possible with the
installation of high performance mass transfer internals, such
as Sulzer VGPlusTM High Capacity Trays. With a combination
of fxed valves, high open area, and advanced downcomers,
these high capacity designs can accommodate high liquid and
vapor rates. The higher capacity allows more trays to be in-
stalled with reduced tray spacings to increase the number of
theoretical stages in the LCO-HCO fractionation section. This
improves separation effciency and reduces the losses of LCO
into HCO or slurry product.
Another very effective solution is to replace trays with Sulzer
MellapakTM or MellapakPlusTM Structured Packing. Struc-
tured packing can allow for increased stages of separation with-
in a fxed column height with the added side beneft of having a
much lower pressure drop. With this design, the refnery can
not only recover more LCO for the diesel pool, but potentially
debottleneck the unit by taking advantage of the lower DP to
free capacity in the Main Air Blower or Wet Gas Compressor.
The Sulzer Refnery Applications Group
Sulzer Chemtech has over 50 years of operating and design
experience in refnery applications. We understand your process
and your economic drivers. Sulzer has the know-how and the
technology to provide a scrubber internals design 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.sulzerchemtech.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.
www.ptqenquiry.com
for further information
sulzer.indd 1 9/6/11 16:27:55
confguration is likely to become
increasingly important in the next
decade.
Further optimisation of the delayed
coking-based refnery
The overall proftability and return
on investment in the confguration
depicted in Figure 10 improves
signifcantly with the addition of
coking unit, which also converts
part of the UCO to distillates to be
processed in downstream hydro-
processing units. This confguration
has no undesired or low-valued
products and is therefore truly
bottomless.
Furthermore, the confguration is
very amenable to phasing; the LC-
Fining unit can be built frst and
82 PTQ Q3 2011 www.eptq.com
the LC-Fining process to the
upgrading of residue.
The LC-Fining unit is the primary
residue conversion process, where
conversion is pushed to the maxi-
mum because unconverted oil
stability is not an issue. The uncon-
verted oil, low in sulphur and
metals, is converted to high-priced
anode-grade coke in the delayed
Vacuum
tower
HDT
Catalytic
reforming
Delayed
coking unit
LC-Fining
Hydrocracking
Gasoline
Jet fuel
Diesel
Anode-grade
coke
*ETKEROSENE
$IESEL
Atmospheric
tower
FCC
Vacuum
tower
Catalytic
reforming
Residual
desulphurisation
(RDS)
Hydrocracking
Gasoline
Diesel
Low sulphur
fuel oil
*ETKEROSENE
$IESEL
(#/
Atmospheric
tower
RFCC
,#/
Figure 10 Optimised residue conversion using LC-Fining and delayed coking
Figure 11 Refnery scheme with RDS as residue upgrader
arora.indd 7 10/6/11 14:09:17
will be proftable until such time as
there is a market for fuel oil. The
delayed coking unit can be phased
in after a few years.
The solution with RDS becomes
relevant when there is a high
premium for very low-sulphur fuel
oil and there is a fairly high
demand for gasoline, or alternately
a market exists for incremental
propylene from the RFCC. If
maximising diesel is the objective,
this confguration is not the opti-
mum one because it will either
make too much low-sulphur fuel oil
or too much gasoline, or signifcant
quantities of both, at the expense of
diesel. In a revamp, CLG has inte-
grated RDS and hydrocracking
technologies for ENIs Taranto
refnery in Italy to produce Euro V
diesel from residuum along with
low-sulphur fuel oil.
Recommended confguration
After detailed analysis, we came to
the conclusion that LC-Fining,
when combined with delayed
coking, provides the maximum
return and the highest NPV,
followed by the delayed coking
alone option. The recommended
technology platform is proven and
the solution is not dependent on
the unreliable future of fuel oil. The
solution is robust, because LC-
Fining and delayed coking can
handle very diffcult feeds.
Furthermore, with the proliferation
of delayed coking units worldwide,
the solution will provide a refner
with a competitive edge in terms of
higher volumetric gain and the
much higher priced anode-grade
product.
LC-FINING is a mark of Chevron Lummus
Global.
Further reading
1 Spieler S, Mukherjee U, Dahlberg A,
Upgrading residuum to fnished products
in integrated hydroprocessing platforms
solutions and challenges, 2006 NPRA, Utah.
2 Mukherjee U, LC-Fining: high conversion
residue hydrocracking, Middle East Petrotech,
2010.
3 Sieli G, Gupta N, Delayed coking and LC-
Fining technology a winning combination,
2008 ERTC Coking and Gasifcation conference,
Rome, Italy.
4 Purvin and Gertz, The global petroleum
market driving refning industry change,
Middle East Petrotech, 2010.
5 Klavers K, Global refning outlook
challenges and development, Middle East
Petrotech, 2010.
6 World Oil Outlook, 2010, OPEC.
Arun Arora is a Project Manager with Chevron
Lummus Global, Bloomfeld, New Jersey, USA.
His primary area of expertise is technology
management and process development in
distillate and residue hydrocracking and he
has led several high-pressure hydroprocessing
projects and refnery confguration studies.
Ujjal Mukherjee is Vice President, Technology,
with Chevron Lummus Global. With particular
expertise in technology development
for distillate and residue hydrocracking,
he has several patents in high-pressure
hydroprocessing and is the author of numerous
technical articles and papers relating to refning
technologies.
www.eptq.com PTQ Q3 2011 83
www.ptqenquiry.com
for further information
Environment. The daily availability
of these values has been acknowl-
edged by ISO 14001 auditors.
Lessons |earned
The implementation of an EMS
based on an advanced DVR tech-
nology has clearly demonstrated its
frst benefts. It is completely inte-
grated into the daily workfow at
the refnery of SIR. It is actively
used for providing continuously
reliable energy-related data.
Business decisions can now be
based on a trustworthy set of infor-
mation. SIR has acquired ownership
and therefore full control of the
application. Awareness in terms of
energy consumption has been
increased and losses can be
prevented without installing new
equipment or instrumentation.
The management of SIR has
noticed a huge time saving in the
collection of reliable information for
both energy balances and environ-
mental indicators as a result of the
implementation of this project. The
time saved is available for engineers
to analyse in greater depth the
ppp'^imj'\hf IMJJ++)**-.
refnerys current situation and to
propose operating improvements.
In other words, SIR has highlighted
a noticeable improvement in work-
ing productivity.
Actions taken by the management
of SIR are steps towards the
improved competitiveness of the
refnery. Moreover, the effects of
corrective actions can be detected
more easily and quickly as a result
of the removal of measurement
noise by the DVR technology.
Success factors
There were three important factors
in the success of this project. First,
there was a strong commitment to
the project by the management of
SIR and a positive attitude towards
the management of change. Second,
a dedicated, multi-disciplinary team
was established, formed by staff
from SIR and Belsim, working
together towards the defned objec-
tives. Third, the transfer of
knowledge from Belsim to SIR was
provided by training, on-site
missions and continuous support
during the project and following its
completion. These factors not only
guaranteed that the project was
successful for both parties, but also
ensured continuous improvement
after the project was completed.
kobert Chares ls MurkeLlng SoluLlon Lnglneer
uL elslm, Awuns, elglum, speclullslng ln energy
munugemenL und energy producLlon. He hus
u musLer's ln muLhemuLlcs Irom ChemnlLz
UnlverslLy oI 1echnology, Germuny, und u lhu
ln upplled muLhemuLlcs Irom UnlverslLy oI
Louvuln, elglum.
>fZbe3kh[^km'\aZk^l9[^elbf'\hf
Herv C|oson ls ulrecLor, Servlces und
1echnology, uL elslm. He hus u musLer's ln
chemlcul englneerlng Irom Lhe UnlverslLy oI
Llge, elglum, und ln munugemenL Irom HLC,
Llge. >fZbe3a^ko^'\ehlhg9[^elbf'\hf
Jean-C|aude Noisier ls lerIormunce und
QuullLy Munuger uL Slk ln Abld|un, lvory CousL.
He hus u musLer's ln englneerlng und u lhu
ln physlcs und muLhemuLlcs Irom Lhe lrench
leLroleum lnsLlLuLe, purL oI lurls UnlverslLy.
>fZbe3c^Zg\eZn]^'ghblb^k9lbk'\b
Hugues Stefanski ls un Advunced lrocess
Lnglneer uL elslm. He speclullses ln energy
munugemenL, perIormunce monlLorlng und
producLlon uccounLlng Ior downsLreum
uppllcuLlons und hus u musLer's ln chemlcul
englneerlng Irom Lhe UnlverslLy oI Llge,
elglum. >fZbe3an`n^l'lm^_Zgldb9[^elbf'\hf
www.imjenquiry.com
for further information
belsim.indd 5 10/3/11 13:15:32
arora.indd 8 10/6/11 14:09:26
For more information or to join our mail list, email or call now 979.845.7417 inquiry@turbo-lab.edu
Sponsored by:
The Turbomachinery Laboratory Texas A&M University
3254 TAMU College Station, TX 77843-3254
P: 979.845.7417 F: 979.845.1835
inquiry@turbo-lab.edu http://turbolab.tamu.edu
In-depth Short Courses
Solution-based Case Studies
Innovative Discussion groups
Hands-on Tutorials
Pioneering Lectures
Outstanding Exhibit Floor
Exhibits Open Free to the Public
Tuesday & Wednesday Thursday
2:30 p.m. - 7:00 p.m. 9:30 a.m. - 12:00 p.m.
Unparalleled Opportunities
George R. Brown Convention Center
.
Houston, Texas
Co-located with:
Meet your PE Requirements with the
Engineering Ethics Tutorial and Earn CEUs!
N
E
W
!
CEU Credits Available
turbo.indd 1 8/6/11 12:43:01
Bitumen processing: crude unit revamps
C
anada is now the largest
supplier of imported crude
oil to the US and currently
provides approximately 10% of
crude refned in the country. A
signifcant portion of this crude is
derived from heavy oil sands bitu-
men produced in Alberta and
primarily processed in the
Midwest.
Western Canadas plans for
growth in oil sands development,
coupled with numerous pipeline
expansion projects, will make bitu-
men more readily available to
refners not only in the Gulf Coast
and other regions in the US, but
also in other parts of the world.
Simultaneously, similar heavy
crudes are being produced in
Venezuela, the Middle East and
elsewhere. The fexibility to process
heavy crude can provide a signif-
cant competitive advantage by
reducing feedstock costs.
The ability to process heavier
crudes has been a refning trend
since the earliest facilities were
built. Refners worldwide have been
driven to upgrade their facilities
over time to handle heavier crudes
because of the following trends:
New sources of crudes are
predominantly heavier crudes with
increased sulphur content
Existing sources become heavier
as they pass their peak production
rates
Light-heavy crude pricing spreads
generally increase with crude price
Canadian bitumen represents one
of the largest potential future
sources of petroleum in North
America and is undoubtedly the
most secure supply available to US
refners
The ability to process bitumen offers a competitive advantage, but revamps of crude
and vacuum units to process bitumen require signifcant changes to equipment
Mike ArMsTrong, rob Henderson and Jon MoreTTA
Jacobs Consultancy
The next largest potential volume
of material is from heavy
Venezuelan crudes, which are simi-
lar to Canadian bitumen
With US refneries running at
lower utilisation levels, many refn-
eries have the cracking capacity to
increase heavy oil processing, if
primary distillation units are prop-
erly equipped to handle these feeds.
Modifying a refnery to process
heavy crudes typically involves
modifcation to primary crude and
vacuum distillation units, additional
gas oil and residue conversion
capacity, increased hydrotreating
and sulphur capability, and other
supporting changes to off-sites,
utilities and infrastructure.
While many refneries are already
geared up to process heavy crudes,
the inherent quality of bitumen
coupled with recent trends in
bitumen production and supply
present unique processing chal-
lenges. This article discusses these
trends and challenges, with specifc
focus on the revamp and reliability
of crude distillation units. While the
article focuses on processing
Canadian bitumen, many of the
challenges discussed are similar for
other heavy crudes.
bitumen production and
supply trends
Figure 1 depicts a potential growth
case for Western Canadian oil sands
developed by the Canadian
Association of Petroleum Producers
(CAPP). Canada currently produces
approximately 1.5 million b/d of oil
sands-derived crude oil. A signif-
cant portion of this material is
currently upgraded and exported to
the US as lighter synthetic crude oil.
www.eptq.com PTQ Q3 2011 85
3000
5000
4000
2000
1000
2005 2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
s
0
0
0
1

,
y
a
d

r
e
p

s
l
e
r
r
a
B
0
8itumen blend` ` 8itumen blend includes
some volumes of upgraded
heavy sour crude oil and
bitumen blended with diluent
or upgraded crude oil.
Upgraded light
Conventional heavy
Conventional light
Actual Porecast
1une 2009 forecast
Figure 1 Projected oil sands growth
Source: CAPP Crude Oil Forecast, Market, and Pipelines Report, Jun 2010
armstrong.indd 1 9/6/11 09:37:23
86 PTQ Q3 2011 www.eptq.com
Oil sands production could grow to as much as
4 million b/d by 2025. The majority of this new capac-
ity is expected to be supplied as bitumen blend instead
of upgraded light synthetic crude oil. Light diluent and
synthetic crude oil will be used as blendstock to trans-
port this bitumen to the US.
Bitumen is produced via two primary methods: ther-
mal in situ recovery using steam and mining, and
extraction of bitumen from oil sands. Current production
capacity is roughly split at 50% thermal recovery and
50% mining, and this trend is expected to continue. Most
of the mined bitumen currently produced is being
upgraded directly in Alberta. At the moment, bitumen
supplied to the US is primarily derived from thermal
recovery facilities. However, going forward, a signifcant
portion of new bitumen supply may be sourced from
mining production operations.
Due to these new production and supply trends, extra
diligence is required by refners to address issues such
as potential increases in chloride and solids levels in the
bitumen, as well as the choice, management and
processing of various diluents.
Bitumen sources, blends and properties
Sources
The term oil sands refers to heavy bitumen deposits co-
mingled with rocks and sand relatively near the surface.
At ambient conditions, the bitumen is effectively solid
and cannot be pumped or recovered using standard
crude oil well methods. For the purposes of this article,
oil sands and bitumen are used interchangeably and
refer solely to Alberta bitumen reserves.
As with any fossil fuel, bitumen properties vary from
site to site. Oil sands-produced bitumen properties are
further impacted by the production method. While
several theoretical and experimental methods of extrac-
tion are being studied and piloted, the primary
commercial methods of production are through mining
or thermal methods, such as steam-assisted gravity
drainage (SAGD).
Mined bitumen is produced in a similar fashion to
strip-mined coal. Mined sand interspersed with hydro-
carbon is mixed with water and then a light
hydrocarbon, to separate minerals and water-soluble
material from the hydrocarbon.
For deeper reserves, where mining is not currently
feasible, thermal production methods are used. The
most widely used thermal technology in Alberta is
SAGD, in which steam is injected into the reservoir via
one set of horizontally laid pipes. The steam heats the
bitumen, reducing its viscosity and allowing the oil to
settle or gravity drain to a low point in the reservoir.
The accumulated oil can then be piped from there to
the surface. SAGD-produced bitumen tends to have
lower concentrations of chlorides, clays and other solids
than mined bitumen.
Blends
Bitumen is never transported to the US as a discrete
product. It is always blended with some type of a cutter
stock for transportation purposes. One common blend
www.ptqenquiry.com
for further information
11IMJJ++)** ppp'^imj'\hf
The work was completed safely and on schedule,
proving that this unique method is a feasible option
for the replacement of column sections.
Ma^\aZee^g`^
Due to a change in duty, high levels of corrosion had
occurred on a 2m-high section of the main fractionator
column, just above the clad lower section of shell.
Non-destructive testing (NDT) during previous turna-
rounds had found certain corroded sections to be
thinner than required. These sections had been over-
laid as a temporary measure. As further corrosion was
expected, Petroplus and Foster Wheeler decided that
replacement of this section with a new band, complete
with cladding, was required.
?bg]bg`ma^kb`amlhenmbhg
There are several methods usually considered when a
section of a column has to be replaced. These include
replacement of the entire column top with new, or
replacement of the section through removal of the top
half to grade, insertion of a new band, and then rein-
statement of the old top half. The short time frame
meant that it was not deemed feasible to purchase an
entire new vessel upper-half due to the signifcant
engineering and fabrication time required. Even to just
replace the corroded band of column using the conven-
tional lift down top half approach would require the
use of a large crane. Since this work was added to the
turnaround scope at such short notice, the crane avail-
ability and plot space required for such a lift were not
guaranteed.
Added to this, a large amount of pipework is present
on the north side of the column, including a 42in main
overhead line, three 18in lines, four 8in lines and
several small bore lines. Each of these lines would
require cutting and bracing and rework should the top
half be removed. This could add signifcantly to the
planned turnaround duration. Replacement of the mid-
section in situ was, therefore, the most feasible option.
Ideally, the section would be replaced as a prefabri-
cated band, complete with internals that could be
slotted into place. For this to be possible, the existing
band would have to be removed while the entire tower
top remained above. The total weight of shell, internals
and piping above the section in question was an esti-
mated 200 tonnes. Supporting this would require a
substantial framework.
One option would be a framework linking the lower
half to the top. This was prevented by the metallurgy
of the column, the lower half being 1 Cr with a 304L
cladding. Although welding onto 1 Cr is possible, it
would require local post-weld heat treatment (PWHT).
It was felt that this was an option not worth pursuing
since it could have led to further problems and compli-
cations during the turnaround. The other option would
be to create a framework linking the upper half to
grade. A big advantage of this approach would be that
the framework could be erected pre-turnaround, with
the new section ready to be lifted and slotted into posi-
tion. However, this would require a huge amount of
ppp'imj^gjnbkr'\hf
_hk_nkma^kbg_hkfZmbhg
TruePeak Tuneable Diode
Laser (TDL) Analyser
Our TruePeak Process Laser Analyzer uses
a tunable diode laser and measures across
an infrared absorbance region. This means
you can make measurements in the most
demanding applications.
Rapid measurement (as fast as 1 second)
Process temperature up to 1500 Celsius
Optical measurement no sensor contact
with process
Aggressive applications high particulate,
corrosives, and more
Flexible installation options
TruePeak measurement
In situ analysis
Interference free
The Process TDL Solution
Features:
please visit us at www.yokogawa.com/eu
foster wheeler.indd 2 10/3/11 14:13:04 armstrong.indd 2 9/6/11 09:37:35
is called Dil-Bit, in which naphtha
is blended with the bitumen. The
concentration of naphtha varies,
depending on the bitumens prop-
erties, but typically the target blend
is 2035% naphtha. The source of
naphtha is equally variable, but
natural gas condensate is often
utilised.
Alternately, or in addition to
naphtha diluent, synthetic crude oil
(SCO) or partially upgraded bitu-
men is used as a transportation
blend stock. This blend is typically
referred to as Syn-Bit. As SCO type
and quality vary from one upgrader
to another, the properties of Syn-Bit
tend to be more variable than Dil-
Bit. When planning a refnery
upgrade to process bitumen, careful
attention should be given to the
type, quality and volume of diluent
and/or SCO blended with the
bitumen.
Properties
The following critical properties can
impact the design and operations of
a refnery crude unit processing
bitumen:
Viscosity Bitumen has signifcantly
higher viscosity than other crude
oils. This can result in a large loss
in crude unit heat transfer, as well
as desalting challenges. It is there-
fore critical to be able to accurately
model the viscosity of bitumen for
design purposes. Commercial simu-
lation tools cannot adequately
predict viscosities for these streams,
and it is advisable to use assay data
to develop viscosity curves outside
of the simulator being used.
Stability The blending of asphaltic
bitumen with other streams can
also lead to problems with stability.
An instable blend will lead to sepa-
ration and potentially condensation
of asphaltic material. This precipita-
tion can result in severe fouling of
heat exchangers, furnaces and other
equipment.
Stability problems may not be
apparent during initial blending or
even shipping, but when the mate-
rial is heated, separation often
occurs. Additionally, blending other
crudes with bitumen-based oils
may exacerbate instability issues.
Filterable solids Oil sands-produced
bitumens, as the name suggests, are
www.eptq.com PTQ Q3 2011 87
more prone to containing solids
than most other crudes. While the
mining process typically has more
diffculty in separating solids from
the bitumen product, SAGD and
other produced bitumen can also
have elevated levels of solids. Solids
content issues are also worse for
bitumens transported via pipeline,
since the material has a limited
settling time. This type of flterable
solid material can create multiple
problems, including: decreased oil
stability, water emulsions, erosion
issues, exchanger fouling, pump
damage and tank sludge build-up.
These solids silt out over time
within the feed tanks, creating
signifcant problems if the refner
does not take proper precautions.
Over time, large deposits can lead to
random excursions of solids materi-
als or water into the desalters,
causing operational upsets. Also,
refners incur high maintenance
expenses when removing and
disposing of these solids when the
tanks are periodically removed from
service for cleaning/inspection.
As mechanical fltration of crude
is not economically feasible, the
primary method of preventing these
solids from settling out in crude
tanks is by keeping the crude feed
system well mixed/agitated via
circulation pumps and tank mixers.
While this may seem counterpro-
ductive, as the solids will remain in
the oil sent for processing, the real-
ity is that as long as these solids
can be ratably fed to the CDUs
desalting system at manageable and
continuous amounts, properly
designed wash water and desalter
systems are able to remove the
material. Using desalting chemicals
such as de-emulsifers and foccu-
lants is also critical for achieving
effective removal of these solids
prior to them reaching the CDUs
preheat exchanger train hot end
and fouling exchangers, heaters and
towers.
Microfnes Of further concern in
mined bitumens is the presence of
so-called microfnes or ultrafnes.
These are typically clay particles
less than 5 microns in size, which
are so highly suspended in the bitu-
men that they do not settle out until
the constituent portions of the
bitumen have been coked or
hydrotreated, thus causing catalyst
fouling issues well downstream of
the CDU/VDU. These small fnes
cannot be fltered out via normal
methods and typically show up in
the latter stages of gas oil hydrot-
reating units catalyst beds as a
high pressure drop, much earlier in
the cycle run than would typically
be expected.
Maintaining a good wash zone
operation can help keep these fnes
in the resid and heaviest gas oil
feeds to the coker and/or FCC.
Fines in gas oil have become a
signifcant issue within the bitumen
upgrading industry and can
substantially increase operating
costs for heavy oil hydrotreaters
due to more frequent catalyst
changeouts. The only other solution
currently available for this issue is
the highly capital-intensive one of
increasing the catalyst cross-
sectional area by adding more
reactors in parallel.
Coking tendencies While there may
be laboratory tests for coking
tendencies, they are not commonly
discussed in the refning industry.
Coking tendency is generally deter-
mined by asphaltenes content and
oil stability, although microfne
content will also give rise to coking.
When comparing a bitumen opera-
tion to that of a typical crude unit,
the level of coking is one of the
more prominent differences.
Metallurgy requirements and
exchanger fouling can be dealt with
by specifc investments, but heater
tube coking issues may force either
a cut-point change that impacts
economics for the entire refnery or
investment in alternative reliability
provisions. Typically, crude and
vacuum tower cut-points must be
dramatically reduced versus opera-
tion with conventional crudes.
Naphthenic acid As with many
opportunity crudes, the acid content
of Canadian heavy oils tends to be
prohibitive to refneries not
equipped for this material.
Naphthenic acid is a generic name
referring to organic (carboxylic)
acids found in crude oils. For refn-
eries, the acids of concern are
primarily high molecular weight
acids concentrating in the 430750F
armstrong.indd 3 13/6/11 14:17:40
more often increases the tempera-
ture of the desalter
Increased water carryover Flash
drums provide a vent to offset water
carryover from the desalter, but high
levels of carryover can result in
signifcant temperature drops.
Jacobs Consultancy has devel-
oped an approach to revamp design
crude unit heat exchanger networks
to accommodate bitumen process-
ing. The general steps associated
with this approach are described as
follows.
Create simulation model
Model validation As part of any
revamp, a thorough survey of exist-
ing equipment and performance is
the frst step. This information can
be used to establish U-Values for
existing exchangers and to create an
accurate network model in the
chosen process simulator.
Bitumen characterisation Accurately
matching the transport and thermal
properties of Dil-Bit and Syn-Bit
blends is required. It is also
important to forecast the range,
88 PTQ Q3 2011 www.eptq.com
(220400C) boiling range. Different
crudes may have differing types of
naphthenic acids and the corrosive
potential can only be determined
through assessment and experience.
Chlorides Mined bitumen can
contain elevated levels of chlorides
and, in fact, chloride levels may
trend up as new mining operations
come on-stream. This requires extra
diligence in desalter design and
operation, metallurgy, and corro-
sion control and monitoring
practices, as well as maintenance
focus.
Optimising the heat
exchanger network
From a high-level perspective,
processing bitumen in an existing
crude unit usually comes down to
two primary issues: metallurgy and
heat exchange. In general, column
operation can usually be adjusted
to avoid replacement of the tower
itself (although, in some cases, this
is desirable for other reasons).
Heavier feeds such as bitumen
normally have less distillate content,
meaning furnace duties often
decrease. Some pump sizes may
need to be adjusted, but, in many
cases, larger impellers or motor
replacement is adequate. However,
switching to bitumen puts entirely
different demands on the heat
exchange network, and these gener-
ally need to be addressed by
redesign and the addition of serv-
ices and shells.
Bitumen processing results in:
Change in crude column heat
profles There will be much less
crude column distillate with more
vacuum gas oil and resid product
Reduced heat transfer coeffcients
Increased viscosity for bitumen and
resid streams
Increased exchanger fouling
Fouling factors for oil sands service
can be an order of magnitude
higher than for standard crudes
due to increased solids, asphaltenes
fouling, emulsions and poor
desalting
Increase in desalter temperature
Bitumen processing requires an
operation that at least maintains or
GE Energy
together
A shared vision of the futureit expands opportunities
and directs efforts toward a common goal.
As Consolidated and Masoneilan products become
part of the GE Energy portfolio, were bringing together
our expertise, innovation and a shared commitment
to our customers to develop advanced technologies
and solutions.
www.ptqenquiry.com
for further information
armstrong.indd 4 9/6/11 09:37:56
quality and properties that will
change as production in Alberta
expands.
Network adjustments The following
are typical network adjustments
required for bitumen design:
Establish new exchanger fouling
factors Fouling factors are depend-
ent on the stability and solids
content of the bitumen and can be
more than double those of the exist-
ing operation. High-temperature
exchangers will obviously see the
greatest increase in fouling
Establish the desired desalter
temperature Depending on the
quantity of bitumen and diluent
being processed, a large increase in
temperature may be required
Determine and model current
desalter water carryover Carryover
may more than double when
processing bitumen. Underestimating
water carryover is a common source
of optimisation error. Increased
water carryover to a fash drum is
an opportunity to increase low-level
heat utilisation. Furthermore, this
temperature in most cases sets the
pinch temperature. Due to the criti-
cality of this determination, more
sophisticated simulation modelling
techniques and laboratory testing
may be required for detailed engi-
neering calculations
www.eptq.com PTQ Q3 2011 89
Rigorously model exchanger services
Using the combination pinch curve
and simulation model, one can
readily identify services where
additional area is required and
services capable of meeting require-
ments. All such assumptions must
be checked through a rigorous
modelling of heat exchange
networks. A common source of
project creep is that process simula-
tion modelling of exchangers is
considered suffcient to defne
scope. Then, during detailed engi-
neering, it is determined that
additional area is required due to
initial inaccuracies related to viscos-
ity, heat capacity, fouling rates and
other properties, as discussed
previously.
Consider reliability asset
management strategy
During this phase of the design
work, the involvement of operations
and maintenance personnel is essen-
tial. There are many decisions to be
made related to modifcations or
new equipment, operability, unit
reliability and availability, and main-
tenance impacts. For example, there
may be trade-offs in adding new
surface area versus increased clean-
ing frequency. Another example
would be allowing higher than
Reduce furnace outlet tempera-
tures for heater tube coking
considerations Several revamps
have failed because refners
assumed that vacuum units can
continue to be operated as per
normal crudes. However, bitumen
coking tendencies are signifcantly
higher than for conventional crudes.
For high coil outlet temperature
designs, test run or lab testing
should be used to confrm the
proposed operation is sustainable.
Adjust process conditions
Crude and vacuum column cut-
points are generally determined by
economic and downstream process-
ing considerations. The degree of
fractionation is generally specifed
based on process guidelines and
rules of thumb. More critically for
revamps, the number of trays is
already fxed and, unless draws can
be moved or eliminated, scope for
increasing theoretical trays is
prescribed by the limits of packing
effciency and column space. Even
recognising that these parameters
are fxed, there are still degrees of
freedom that can be employed to
reduce network modifcation
requirements, such as changes to
column heat balances and equip-
ment operating pressure.
High temp.
exchangers
and piping
Column internals
and shell
Furnace tubes
and transfer
lines
Gasoil draws
Overhead
exchangers
and piping
Column top
section
Chloride attack
NAC / Erosion and sulphidation
HGO
LVGO
HVGO
Vac. resid.
Raw crude /
Dilbit
Figure 2 CDU: areas of concern for corrosion
armstrong.indd 5 9/6/11 09:38:05
90 PTQ Q3 2011 www.eptq.com
desired velocities in certain sections
of the plant, while enhancing
corrosion monitoring/management
and inspection programmes. There
are a variety of similar issues, such
as ensuring adequate plant fexibil-
ity for varying operations, providing
adequate process control and ensur-
ing adequate access for maintenance
when new equipment is added.
Corrosion considerations
While concerns with network opti-
misation are signifcant and major
changes may be required for a
desired capacity, project costs can
often be controlled by settling for a
lower capacity or reduced bitumen
concentration. Metallurgy issues are
not as easily avoided. Many articles
have been written summarising the
concerns of naphthenic acid corro-
sion in crude units, particularly
with heavy or opportunity crudes.
There are essentially four types of
corrosion that take place in a typi-
cal crude unit:
Naphthenic acid corrosion The
mechanism of naphthenic acid
corrosion (NAC) is less well under-
stood, but it is known that NAC
does not occur independently of
sulphur. It is suspected that naph-
thenic acid reacts with iron sulphide
and effectively pulls off the protec-
tive layer of scale, leading to pitting
and grooving. Velocity and wall
shear are also variables.
High-temperature sulphidation At
temperatures around 400F, hydro-
gen sulphide (H
2
S) begins to evolve
from oil. Along with an oxidising
agent (typically water) the H
2
S
reacts with the pipe walls to form
iron sulphide scale
Hydrochloric acid attack Chloride
salts, primarily magnesium and
calcium chloride, hydrolyse to form
hydrochloric acid. Acid attack then
occurs in the aqueous phase of the
column overhead. Similarly, the
acid reacts with ammonia to form
ammonium chloride salts, which
foul and corrode trays and exchang-
ers in the top section of the CDU.
Bitumen chloride levels may be
trending upward over time as more
mined bitumen enters the market
Sour water corrosion Sour water
corrosion is an issue in every crude
unit design except for ones
processing very low-sulphur crudes.
Critical to the mechanism is the
interaction of ammonia, H
2
S, chlo-
rides and their various salts. The
concentration of all these compo-
nents increases with bitumen
processing, raising the potential for
catastrophic failure in these systems.
We are aware of numerous facilities
experiencing issues in this regard.
Figure 2 shows the areas where
bitumen processing can increase
corrosion rates. Naphthenic acid
corrosion occurs in a specifc range
at higher temperatures. The overlap
between naphthenic corrosion and
sulphidation refects the fact that
the mechanisms work together. To
adequately determine the corrosion
rate for a given pipe or column
section, it is critical to know the
temperature, and H
2
S, sulphur and
naphthenic acid levels.
Naphthenic acid
Naphthenic acid is a generic term
for all organic acids in crude. More
specifcally, naphthenic acids refer
to any organic acids with a phenyl
or multi-phenyl group linked to a
carboxylic acid function, as shown
in Figure 3. Lighter molecular acids,
where they exist, are problematic in
that they create corrosion impacts
in lighter side cuts, such as heavy
naphtha and kerosene. In most
crudes, the concentration of these
components is low and only a select
few crudes have concentrations
high enough to create corrosivity
problems. Due to the relatively low
amount of kerosene and lighter
material in bitumen, these compo-
nents have a negligible impact on
corrosion and metallurgy selection
when considering the impact of
bitumen.
Heavier components boiling in
the 430750F (220400C) range are
typically the cause of bitumen
naphthenic acid corrosion. These
components are present in the
crude and heavier products, includ-
ing heavy gas oils, atmospheric
resid and vacuum resid.
NAC is often misunderstood
because of the interaction with
sulphidic corrosion. The following
equations represent the equilibrium
balance between the competing
NAC and sulphidic corrosion:
Fe + 2RCOOH Fe(RCOO)
2
+ H
2
Fe + H
2
S FeS + H
2
Fe(RCOO)
2
+ H
2
S FeS + 2RCOOH
An important difference between
the iron sulphide and iron naph-
thenic salt is that the sulphide
forms a stable flm on the pipe
surface, while the naphthenic salt is
solubulised in the hydrocarbon
stream. This equilibrium relation-
ship leads to conficting data about
when and where NAC occurs. If
there is a threshold concentration of
H
2
S, the pipe forms a protective
scale. As the acid concentration
increases, the equilibrium shifts
towards the naphthenic iron. In
addition, factors such as velocity
and turbulence can lead to removal
of the flm, increasing the amount
of acid metal loss at high H
2
S
concentrations.
In assessing the risk from NAC,
all of the following criteria must be
considered:
Partial pressure of naphthenic
acid component
Partial pressure of H
2
S
Level of turbulence (wall shear)
Temperature.
Figure 4 is a simplifed representa-
tion of the zones created by the
protective layer of sulphide scale.
Assuming constant levels of naph-
thenic acid concentration and
temperature suffcient for corrosion,
corrosion occurs at low levels of H
2
S
as there is insuffcient scale formed.
This theory of the interaction
between H
2
S and naphthenic acid
has some interesting and useful
corollaries. It is possible to reduce
NAC by artifcially increasing the
amount of H
2
S and/or iron sulphide
inside the pressure envelop. And,
as the H
2
S concentration increases,
system velocity becomes much
more important to the rate of
corrosion.
R
C
O
OH
C
O
OH
C
O
OH
(CH
2
)
n
Figure 3 Naphthenic acid component
structure Source: NACE Paper 04634
armstrong.indd 6 9/6/11 09:38:21
www.ptqenquiry.com
for further information
burkhardt.indd 1 20/9/10 10:10:44
This relationship has led to exper-
imentation with corrosion inhibitors,
sulphur components that keep the
sulphide layer intact. Such
inhibitors have been shown to be
effective to the extent that velocity
can be managed.
The steps to analyse the likeli-
hood of corrosion are:
Establish the concentration of
sulphur and naphthenic acid
throughout the system
n Use assay information to esti-
mate concentrations in columns and
sidecuts
n H
2
S concentrations and sulphur
concentrations should both be
estimated
n H
2
S should be determined from
evolution curves
Establish the temperatures associ-
ated with each concentration and
location
Identify areas where metallurgy
protection is required (upgrade to
Use risk assessment techniques to
identify critical equipment justify-
ing additional capital investment.
Sulphidation
The analyses for sulphidation and
NAC need to be carried out in
tandem. Sulphidation is less of a
concern, in that most refneries are
already designed for high levels of
sulphur and sulphidation. When
high levels of naphthenic acid
and sulphur are present, the
NAC mechanism typically over-
rides sulphidation concerns, while
changes to piping and exchanger
metallurgy to address sulphidation
are typically less expensive than
those to address NAC.
Unlike the other forms of fouling
and corrosion discussed here, there
is almost no mitigation strategy,
short of increasing the chrome
content of the metallurgy. Iron
sulphide forms a surface flm that
protects the metal from several
forms of corrosion and erosion. The
flm offers some resistance to heat
transfer, but it reduces metal loss
by protecting the surface under-
neath from NAC.
Hydrochloric acid attack
Refneries are well aware of the
dangers of chloride corrosion in the
overhead of atmospheric crude
towers. Chloride enters the column
in the form of salts that hydrolyse
at the temperatures found in the
crude furnace. The chlorides create
two distinct corrosion and fouling
problems acid attack and ammo-
nium chloride salt formation in
the overhead
Acid attack occurs as water
condensate forms on condensing
surfaces (see Figure 5). The frst
droplet of condensate preferentially
absorbs the hydrogen chloride gas,
forming a highly acidic solution
that attacks iron. This mechanism
can be mitigated by:
Continuous water wash
Neutralisation: control pH above 5
Corrosion inhibitors (flming
amines).
Even if ammonia is not used as a
neutralising agent, some ammonia
(ppm level) evolves during the
heating process. The ammonia and
chloride form a salt that lays down
92 PTQ Q3 2011 www.eptq.com
317SS is typically required)
n Areas of high corrosion rates
caused by a combination of high
sulphides, high napthenic acids and
high temperature
n Identify areas where high
velocities or turbulence may be of
concern
n Typical locations requiring
metallurgy upgrade for NAC are:
Transfer lines
Furnace tubes
Elbows and nozzle draws
Inlet horns
Tower lining and trays
Identifed piping and other
locations
Look at moderate areas where
corrosion can be mitigated by a
combination of monitoring, corro-
sion inhibitors and reduced velocity
In areas of low sulphur and low
velocity but high naphthenic acids,
consider the possibility of using an
inhibitor injection
H

g
n
i
s
a
e
r
c
n
I
2
n
o
i
t
a
r
t
n
e
c
n
o
c

S
Increasing turbulence/wall shear
Zone 3
(IGHCORROSION
4URBULENCESTRIPSIRONSULPHIDELAYER
)NCREASEDRATEOF.!#ONEXPOSEDMETAL
Zone 2
,OWCORROSION
3ULPHIDELAYERPASSIVATESMETALS
Zone 1
(IGHCORROSION
)NSUFFICIENTSULPHIDESTOPREVENT.!#
Figure 4 Relationship between NAC, H
2
S and turbulence Source: NACE Paper #02555
Cooling
Flow
H
2
S
HCI
H
2
O HCI HCI
HCI
Fe
HCI HCI
H
2
S
H
2
S
H
2
S H
2
S
High corrosion
High HCI concentration
High temperature
Low pH
Lower corrosion
H
2
S buffering
Lower temperature
Higher pH
Figure 5 Acid attack in crude unit overhead
armstrong.indd 7 9/6/11 09:38:32
on surfaces and trays. Although the
salt is water soluble, it generally
forms in areas with insuffcient
water distribution. In addition to
reduced heat transfer, under-
deposit corrosion occurs beneath
this salt layer. Methods to prevent
ammonium salt formation are:
Provide adequate water wash
with distribution to all exchangers
Maintain water pH below 7.0
Use inhibitors and amine
neutralisers.
Desalting
While steps can be taken to mitigate
both forms of chloride corrosion,
the best means of prevention is to
remove chloride salts upstream of
the crude furnace. Salt removal
tends to be more challenging for
bitumen feedstocks.
Among the factors that make
bitumen desalting more challenging
are:
Asphaltenes interact differently
with water compared with other
hydrocarbons, leading to an
increase in emulsions
High solids content
Saponifcation of naphthenic acids
High viscosity
Reduced gravity differential
between the crude and water.
Among the key considerations in
desalting are:
Temperature Higher temperature
reduces viscosity and increases the
gravity driving force for water-oil
separation
Mud washing High metals, solid
fnes and asphaltene precipitation
all inhibit the dehydration process.
Furthermore, these materials tend
to accumulate in the bottom of the
desalting vessel. While possibly
counterintuitive, it is more effective
to keep these materials well mixed
in the oil fraction so they can be
removed from the desalting process
Mixing valve/static mixer Shear at
the mixing valve is a common cause
of emulsion formation. Due to the
high viscosity of bitumen material,
it is critical to optimise the pressure
drop to prevent shear and minimise
emulsion formation, while main-
taining good mixing between water
and oil
pH control High pH leads to the
production of sodium naphthenates
that form emulsions and foul down-
stream exchangers. Typically, it is
suffcient to control the pH of the
make-up water, but impurities in
the feed may overwhelm the water
buffer chemistry, necessitating the
use of acid injection, into either the
make-up or circulating water
Flash bitumen after desalting
Bitumen production and upgrader
facilities have shown that asphaltic
bitumens retain free water even
after extensive water knockout
steps. If this water is not vapour-
ised, soluble salts in the emulsion
concentrate in the free water and
foul downstream exchangers.
Flashing the bitumen after the
desalter dissipates salts in the
hydrocarbon phase. In the hydro-
carbon phase, salts will not foul
exchangers, and chlorides may not
hydrolyse as readily. Recent studies
have indicated the hydrolysis of
chloride salts is more extensive for
the hydrated salt species (eg,
MgCl
2
_6(H
2
O))
Space velocity While the most
costly parameter to improve,
increasing separation time has a
marked infuence on desalter
performance. If possible, an addi-
tional desalting train should be
considered to process bitumen. The
cost of this train can be justifed in
terms of improved availability,
improved energy effciency and
reduced maintenance costs (less
exchanger cleaning and overhead
equipment replacement.) Space
velocity is determined based on oil
volume. Decreasing water addition
does not improve separation and
has been shown to increase water
carryover and emulsion formation
Rag layer control Withdrawal of a
rag layer should be considered a last
resort. A rag layer (an emulsion
between the liquid and hydrocarbon
phase) directly reduces the space
velocity of the desalter. The rag
layer often forms around solid parti-
cles. While withdrawal improves
performance, disposal of the rag
layer can be problematic. Generally,
high temperatures are required to
dehydrate the emulsion and most
refners do not have the facilities to
reprocess this material
Chemical addition and crude
pretreatment Emulsion breaking
chemicals and asphaltene stabilisers
can greatly improve separation eff-
ciency. For refners with suffcient
storage time, crude pretreatment
can help settle solids and water in
tankage prior to processing.
Capital costs: revamp or new build?
Revamp of existing units
Signifcant upgrades can be required
to existing crude and vacuum distil-
lation units to process bitumen. The
typical types of changes required for
revamping an existing heavy sour
crude unit to process oil sands-
derived bitumen include:
Addition of new exchangers and
more surface area to the CDU
preheat train
Lining of the main column,
usually with 317 SS
Retraying/packing of the main
column due to metallurgy concerns
and lower vapour/liquid traffc
Upgrading of most hot distillate
piping to SS
Lining of one or more of the side
strippers
Upgrading of most hot transfer
piping to SS
Improvement in the columns
overhead water wash system
Upgrading of the metallurgy of
hot pump impellers/casings
Desalter internals replacement/
upgrades
Fired heater retubing.
The cost of engineering and
executing this work can be highly
prohibitive in some cases, especially
as much of the work needs to be
done during turnarounds. Another
substantial cost of carrying out a
substantial CDU/VDU revamp is
the extended downtime required to
carry out this work. It is not
unheard of for the revamp work to
turn a normal 30-day turnaround
into a 60- to 90-day outage with the
signifcant attendant costs that a
long downtime entails. In some
cases, work scopes have been so
large that the work is executed over
more than one unit outage.
In addition to the actual costs
identifed here, there are other
concerns about the possible ramif-
cations for safety and reliability of
carrying out much of this work,
with some portions occurring while
the units are still operating. Typical
www.eptq.com PTQ Q3 2011 93
armstrong.indd 8 9/6/11 09:38:43
with upgraded synthetic crude.
Signifcant competitive advantage
opportunities exist for refners
having the capability to process
bitumen. Revamps of existing crude
and vacuum units to process bitu-
men are not trivial and entail
signifcant equipment changes,
especially related to heat exchange,
desalting and metallurgy upgrade.
In some instances, it may be prefer-
able to build new units rather than
revamp existing facilities, to avoid
excessive downtime and reap other
operating cost benefts.
Careful attention should be paid
to operations, reliability and main-
tenance when processing bitumen.
A thorough review of the refnery
asset management programme is
recommended.
References
1 Kremer L, Bieber S, Strategies for desalting
heavy Western Canadian feedstocks, NPRA
paper AM-08-36, 2008 NPRA National
Conference,911Mar2008,SanDiego,CA.
2 MesserB,TarletonB,BeatonM,PhillipsT,A
new theory for naphthenic acid corrosivity of
Athabascaoilsandscrudes,NACEpaper04634,
NACEInternational2008.
3 Kane R D, Cayard M S, A comprehensive
study on naphthenic acid corrosion, NACE
paper02555,NACEInternational2002.
4 GrayMR,EatonPE,LeT,Kineticsofhydrolysis
ofchloridesaltsinmodelcrudeoil,Petroleum
Science and Technology,26,19241933,2008.
Mike Armstrong is a Senior Consultant with
Jacobs Consultancy and has 22 years of
operations and process design experience in
the refning industry. He holds a bachelors
degree in chemical engineering from the
University of Illinois and a masters in
chemical engineering from the Colorado
SchoolofMines.
Rob Henderson is a Senior Consultant with
Jacobs Consultancy and has 19 years of
experienceinprocesstechnology,engineering,
commissioning, troubleshooting, operations,
projectdevelopmentandtechnicalsupervision.
He holds a BS in chemical engineering, a BA
in economics from Case Western Reserve
University and a MBA from Northwesterns
KelloggGraduateSchoolofManagement.
Jon Moretta is a Director with Jacobs
Consultancys Petroleum, Chemicals and
Energy Practice. He is an expert in heavy oil
upgrading technologies and been involved
in many recent oil sands bitumen projects.
He holds a bachelors degree in chemical
engineeringfromtheUniversityofMichigan.
94 PTQ Q3 2011 www.eptq.com
revamp costs vary widely and have
been found to run anywhere from
15% to almost 100% of an
equivalent-sized grassroots unit.
Grassroots considerations/costs
In view of the high costs of upgrad-
ing an existing CDU/VDU complex,
some refners have chosen to build
grassroots replacement units instead.
By building a new unit, diffcult-to-
overcome physical constraints can
be resolved (provided adequate plot
space exists) and additional operat-
ing cost benefts would also be
expected due to higher availability,
lower maintenance costs and more
energy-effcient operations for a
new, purpose-built unit, properly
sized and using the latest design
practices. The primary technical and
economic differences between a
typical heavy-sour CDU/VDU and
one confgured to process bitumen
are:
An increase in heat exchanger
surface area due to fouling/viscosity
issues
Use of higher metallurgy
throughout the units
An increase in column height to
accommodate space for a taller
wash oil section
A more robust desalter design
Changes in main column over-
head metallurgy and water wash
design may be desired.
Overall, these differences typically
lead to an approximate capital cost
increase of 1015% for a bitumen
processing design compared with a
typical heavy sour crude design.
Operations, reliability
and maintenance
It is important to note that revamp-
ing a refnery to process bitumen
warrants signifcant attention to
operations, reliability and mainte-
nance planning. There is a tendency
to treat a revamp as mainly an
engineering event, since operations
and maintenance already have
proven practices in place for the
unit. We fnd that the opposite
should be the case. Tasks such as
reliability modelling, maintainabil-
ity reviews and ready-for-operations
preparation of equipment strategies
should be key areas of focus in
these revamps.
Typical questions to resolve
include:
Is the equipment failure/fouling
rate going to increase?
Is the maintenance organisation
aware of the changes in practices/
frequencies they will have to adopt
(for example, increase the thickness
monitoring programme)?
Are there design alternatives that
will avoid the potential loss of
availability (for example, parallel
equipment) and how can they be
justifed?
Are their operating procedures
that we need to adopt to reduce the
failure risks (for example, increase/
reduce the settling time we provide
in the crude storage area)?
Should we use a risk-based
inspection approach to identify
corrosion loops and enable deci-
sions on monitoring such as the
inclusion of coupons
What equipment strategies need
to be developed to enable the
correct equipment care programmes
to be adopted from day one of start
up? truly ready for operations.
The list is certainly not exhaus-
tive, and it is advisable in any
situation to develop and implement
a strong asset management
programme. However, in the case
of this type of revamp, it is impor-
tant the refnery recognises that
future operations may present
issues and events which have not
been dealt with in the past, and to
look at the revamp with the same
diligence as they would the instal-
lation of a completely new process
unit in the refnery
Conclusions
Canadian oil sands represent a
signifcant source of secure energy
supply. Projected growth rates for
bitumen production are signifcant
and expected to more than double
over the next 15 years. In addition,
access to this bitumen will be more
readily available as pipelines are
expanded.
New bitumen supply will likely
include mined bitumen, which
presents potential added processing
challenges in the form of increased
chlorides and solids. In addition,
the volume of blended bitumen
import will increase compared
armstrong.indd 9 9/6/11 09:38:59
Unique Solution for Your
Petrochemical Application
Best for polyethylene, polypropylene,
ethylene glycol and ethylene oxide production
The ultimate contact-less labyrinth sealing
technology
No piston rings - no wear
Insensitive to particles and contaminated gases
Gas-tight design for zero emissions
Highly reliable components for highest
availability of your compressor
Count on our experience:
Thousands of compressors in operation,
hundreds of long-term customers
Your Beneft:
Lowest Life Cycle Costs
Like this image? Get more: www.recip.com/screensaver
Ad_Laby-A4.indd 3 07.01.10 09:43
www.ptqenquiry.com
for further information
burckhardt2.indd 1 9/3/10 14:25:09
Serving you along
the entire LNG Value Chain.
Rely on our expertise in process engineering as well as in the planning, project development and execution
of turnkey process plants. Proft from our competence along the entire LNG Value Chain:
Process plants Small and mid-scale LNG receiving terminals
BUhifU`[Ug`]eiYZUWh]cb LNG loading facilities for ship, truck and rail
BUhifU`[UgdfY!hfYUhaYbhB;@YlhfUWh]cb LCNG fuelling stations
B]hfc[YbfY^YWh]cb LNG fuelling systems
<Y`]iafYWcjYfmUbX`]eiYZUWh]cb Local storage and re-gasifcation
Coil-wound and plate-fn heat exchangers LNG tanker trailers for road and rail
Coldboxes Submerged combustion vaporizers
@B;ghcfU[YhUb_g @B;WUff]YfVc]`!cZZ[UgfY!`]eiYZUWh]cbgmghYag
Designing Processes - Constructing Plants.
Linde AG
Engineering Division
Dr.-Carl-von-Linde-Strae 6-14, 82049 Pullach, Germany
Tel. +49.89.7445-3784, Fax +49.89.7445-4928
E-Mail: natural-gas-plants@linde-le.com, www.linde.com
z4151.indd 1 07.09.2010 09:54:10
www.ptqenquiry.com
for further information
linde.indd 1 8/6/11 12:26:57
Improving protability despite lower
utilisation
O
ver the last few years,
the rening industry has
evolved from a golden era
of high margins and maximum util-
isation to tighter margins and
renery rationalisation. Reners
have had to quickly adapt to this
new reality or risk becoming
unprotable.
With reneries running at lower
throughputs than have been seen in
many years, reners are presented
with new challenges to retain a
protable operation. This article
focuses on ways that reners can
improve protability while operat-
ing in a low margin and low
utilisation environment.
For the last ve years, crude unit
throughputs have been declining,
with crude unit utilisation dropping
below 85% in the US in 2009 (see
Figure 1). Although crude unit utili-
sation recovered slightly in 2010
(85.3%), it continues to be well
below the average of the last 20
years. In addition, most forecasts
call for continued low utilisation
rates.
Some of the decrease in utilisation
is due to an addition in capacity
during a period of at or slightly
declining rened product demand.
However, over the last ve years,
low margins have forced reneries
to cut crude charge. After ve years
of relatively steady gross inputs, in
2008 they dropped nearly 4% and
dropped another 1% in 2009.
Although gross inputs rebounded
slightly in 2010, they are still lower
than they have been in nearly a
decade (see Figure 2).
In sharp contrast to recent history,
increased throughput does not
necessarily mean increased prot.
Alternative and novel strategies can maintain and improve protability despite
lower utilisation of individual process units
ERIC STREIT and ROBERT OHMES
KBC
Reners are forced to make deci-
sions that would have seemed
unlikely 10 years ago. Shutting
down units, extending turnarounds
and reducing unit capacities are all
strategies that are being done today
to try and compensate for lower
renery utilisation.
Another aspect of the market that
reners must deal with is that the
margins on all renery products are
eroding. During 2006 and 2007, the
price of ultra-low sulphur diesel
(ULSD) was much higher than that
of gasoline, which gave reners
a clear denition of direction:
www.eptq.com PTQ Q3 2011 97
90
100
95
85
80
90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12

%
January, year
75
CDU utilisation (RHS)
Yearly utilisation (RHS)
Figure 1 Crude unit utilisation history
15 500
16 500
16 000
15 000
14 500
14 000
90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12
d
/
b

s
0
0
0

January, year
13 500
6 mo. ro||ing average of gross inputs
6 mo. ro||ing average of crude oi| input
Figure 2 Renery input history
streit.indd 1 10/6/11 11:57:32
98 PTQ Q3 2011 www.eptq.com
were optimised for high rates now
need to be re-evaluated. Buying or
selling intermediates may have
been uneconomical in the past but,
with open capacity on some units,
may now be proftable.
The following sections outline
typical and novel example strate-
gies that refners can examine for
proft improvement opportunities.
Later, an industry-proven work
process is outlined that enables
refners to identify, capture and
sustain opportunities.

Fractionation
At higher unit throughputs, frac-
tionators throughout the refnery
are typically operated up to one or
more physical constraints. These
constraints may include fooding,
overhead condensing, furnace
outlet temperature or heat removal
limits, among others. At reduced
unit throughputs, many of these
former constraints may no longer
be an issue. There are a number of
ways that a towers operation could
be changed as a result of
the removal of these physical
constraints.
As tower fooding, reboiler duty
and condensing system constraints
are removed at a lower unit
throughput, increasing refux and
decreasing pumparound rates to
provide more internal refux in the
fractionators may be proftable.
These changes improve separation
in the tower and allow more valua-
ble products to be drawn to higher
rates or to better qualities.
The draw rate of individual prod-
ucts from a fractionator may have
been constrained at high fow rates
due to rundown hydraulics. As the
feed rate to the tower is reduced,
the unit may not need to run a
given product to its maximum
rundown rate. Instead, cutting back
some product draws that were
always maximised in the past may
make sense.
Each tower should be examined
on an individual basis to determine
if changing tower operational
targets makes sense at lower fow
rates. Each towers operation can
then be reviewed on a multi-system
and plant-wide basis to capture
proft improvement opportunities.
maximise distillate. Recently, ULSD
and unleaded regular gasoline
(ULR) have sold for similar prices
(see Figure 3). The driver to maxim-
ise ULSD over gasoline is no longer
present.
Historically, typical operating
plans maximised refnery and indi-
vidual unit throughputs. Even
conversion was frequently sacri-
fced to maximise feed rate. With
large margins on most products,
the planning challenge was to make
sure that as many units as possible
were fully utilised. The main opera-
tional challenge was to maintain
reliability while fnding ways to
relieve throughput constraints.
In the present and near future,
margins do not support maximum
throughputs. Economic drivers
have shifted from feed rate to
conversion (see Figure 4). Creating
volume in a refnery is always
desirable, but at lower unit feed
rates increased liquid volume yield
is more important than ever.
Therefore, refners must look
for and implement alternative oper-
ating strategies to maximise
proftability, including unique oper-
ational changes that may have been
sub-optimal in previous economic
conditions.
Maximising refnery proft at lower
throughputs
With the shift away from maximum
unit utilisation, many operational
opportunities may now be available
that have not been so in the recent
past. Constraints that previously
restricted unit throughputs have
been relieved. Energy inputs that
10
20
40
30
0
-10
-20
-30
-40
Jan
06
Ju|
06
Jan
07 07
Jan
08
Ju|
08
Jan
09
Ju|
09
Jan
10
Ju|
10
l
b
b
/
$

-50
3.5
ULSD
Un|. gaso|ine 87
Figure 3 Product differential history
30
20
10
0
10
20
2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020

R
e
a
l

c
r
a
c
k
s

v
s
.

W
T
I
,

$
/
b
b
l
30
3%S fuel oil to WTI
ULSD to WTI
LS No. 2 fuel oil to WTI
87 oct. reg. unl. gasoline to WTI
Forecast
Figure 4 Projected product cracks
streit.indd 2 9/6/11 09:49:12
Increasing hydrogen recycle
For many hydrotreaters, the gas-to-
oil ratio is set to a level to minimise
catalyst deactivation while maxim-
ising throughput. Frequently, this
has translated into maximising unit
throughput up to a minimum gas-
to-oil ratio (or minimum hydrogen
partial pressure). With lower unit
throughput, it makes sense to
revisit this strategy. Increasing the
gas-to-oil ratio on a hydrotreater or
hydrocracker will provide benefts
in terms of increased unit cycle
length and increased hydrogen
consumption along with improved
properties in hydrotreated diesel
(cetane) or FCC feed (H
2
content).
However, an increased gas-to-oil
ratio comes at a cost. The gas that
is being recycled has to be
compressed and heated back up to
reactor temperature, so raising the
gas-to-oil ratio requires more
energy input for the same unit
throughput. Each unit should be
evaluated to optimise the cost
versus benefts of increasing the
gas-to-oil ratio.
www.eptq.com PTQ Q3 2011 99
Rationalising units
In many refneries, there are multi-
ple units performing the same task.
Perhaps the most typical is to have
multiple distillate hydrotreaters
making similar ULSD blendstocks.
When refnery throughput is
reduced, often one of these units
can be shut down or assigned other
tasks and the remaining units can
pick up the load.
Frequently, the unit that is shut
down is the oldest unit in service,
which is a good choice, because
older units may run less effciently
than newer units and may suffer
more maintenance problems while
running. Older units, particularly in
hydrotreating service, frequently
operate at lower severities, making
them a better candidate to shut
down. However, the overall refn-
ery system should be evaluated
with each unit down to make sure
this is the correct choice.

Block operation
One key determiner in deciding
which unit to shut down is whether
or not the unit can be used in a
different service. For example, a
high-pressure diesel hydrotreater
can be used to hydrotreat FCC feed,
while a lower-pressure unit can
produce jet fuel.
When units are driven to maxi-
mum throughput, the downtime
involved in switching services may
preclude operating the unit in
different blocks. However, when
unit throughput is not the key
determiner to refnery proftability,
switching a unit from one service to
another may not be very costly,
particularly around slop or off-spec
product routing. As a result, block
operation may make sense where it
did not before.
Market solutions
In response to the new economic
drivers, there may be several ways
to increase proftability by exploiting
the market, including purchasing
different feedstocks or selling differ-
ent products. Examples include:
Produce solvent or lube stock
components
Adding value to the oil,
gas and petrochemical
industries?
Absolutely.
ABB provides technical consulting and engineering services to improve
performance in the areas of compliance, operations and engineering. We offer
expertise in the following areas: environmental services; industrial energy
effciency; inspection; integrity management; maintenance & reliability; pressure
relief; process safety; project services; technical software and technical training.
www.abb.com/consulting
ABB Consulting
Tel: +44 (0)1925 741111
E-mail: contact@gb.abb.com
PTQ Q3 2011 (125x178) May 2011.indd 1 26/05/2011 15:05:01
www.ptqenquiry.com
for further information
streit.indd 3 9/6/11 09:49:26
increased at high unit charge rates
to help maximise condensing in the
overhead or manage tower food-
ing. When the unit rate drops, the
overhead cooling or fooding
constraint may be relieved.
Lowering the tower pressure helps
in several ways:
Reduces the reboiler or feed
heater duty requirements
Improves component or cut
separation
Increases product recovery from
the bottoms stream in CDU, VDU
and conversion unit main
fractionators.

Turnarounds
When throughput was the main
factor in refnery economics, there
was usually a lot of pressure to
shorten turnarounds to get units
back on line as quickly as possible.
Now that throughput is not as valu-
able, a change in philosophy may be
in order. Extending turnarounds is
contrary to the nature of refnery
management but may be valuable if:
Costs can be signifcantly removed
by reducing overtime, the number
of contractors or manpower
Extending the turnaround allows
for additional time for equipment
repair, particularly for rush jobs
identifed during the shutdown
Additional turnaround time
allows for additional inspections to
avoid intermediate shutdowns in
the future
Unit modifcations can be
completed that improve the units
capability post-turnaround to
improve reliability or proftability.
Product blending/oil movements
Gasoline RVP is an area that can be
continuously optimised. Most refn-
ers purchase butane to minimise
their RVP giveaway. With a fully
loaded refnery, it can frequently be
diffcult to meet RVP specifcations
on some towers. As a result, some
towers may be run to lower RVP
targets to offset those towers that
produce a higher RVP product.
When the refnery load is decreased,
towers that had diffculty achieving
their target RVP may now be able
to reduce RVP below the target.
With more towers able to meet RVP
specifcations, a full optimisation of
100 PTQ Q3 2011 www.eptq.com
Increase reformer severity to
produce high-octane gasoline blend-
stock for producing boutique
gasoline products, gasoline
blendstocks or petrochemical
intermediates
Process purchased gas oils to fll
FCC capacity as opposed to process-
ing more crude
Convert discounted or low-
quality distressed purchased streams
into saleable products
Modify or renegotiate pay-as-
you-go and take-or-pay contracts
on purchased utilities, including
hydrogen.
Refners should look for opportu-
nities that ft within the facilitys
capabilities and with the strategic
goals.
Catalyst/chemicals
Another area of potential savings
that should be explored in the new
market is catalyst and chemical
purchases. Most process chemicals
injected in the refnery are dosed at
a fxed concentration to meet a treat
target (corrosion, contaminant
neutralisation and so on). At lower
unit charge rates, less process
chemical addition may be required.
Working with the chemical vendor
to ensure that the chemical addition
is varied with the unit feed rate
may be necessary.
In some cases, additional process
chemicals may be required at lower
fow rates. Lower velocities through
heat exchangers will tend to
increase fouling. If antifoulant is
injected to reduce the fouling rate
of these exchangers, it is possible
that increasing the antifoulant rate
at lower rates could be used to
offset the effect of lower velocities.
With reduced charge rates to
fxed-bed units, catalyst deactiva-
tion rates should be lower. One
obvious way to take advantage of
this lower deactivation rate is to
extend the time between unit
outages.
If the refnery plan calls for a unit
to operate for a specifed run length,
such as when the unit shutdown
must fange up with another unit
that is scheduled to shut down or
when the unit shutdown is driven
by inspection requirements, a fxed-
bed unit could be short-loaded.
This approach is a bit risky in that,
if margins go up and the plan calls
for an increase in unit feed rate,
the unit may not make it to the
scheduled shutdown. However,
particularly in the case of units
using noble metal catalysts, short-
loading can signifcantly reduce
costs.
Using regenerated catalyst may
be another way to take advantage
of a unit that is scheduled to run
reduced rates for an extended
period of time. Regenerated cata-
lyst has an activity lower than that
of new catalyst but can be much
less expensive. If the unit is
expected to run at lower than
historical rates but the cycle cannot
be extended, regenerated catalyst
is a good way to keep costs down
for the cycle.
For FCC units, reviewing catalyst
formulation may be warranted to
take advantage of lower feed rates.
If the unit is typically limited by
wet gas compressor capacity at high
rates, lower feed rates may result in
some available wet gas capacity.
The addition of additives such as
ZSM-5 could help to take advan-
tage of this new capacity by
increasing olefns production and
refnery gasoline yield.
Energy
One of the most obvious ways to
lower operating costs at lower feed
rates is to reduce energy consump-
tion. However, this is surprisingly
often overlooked.
Stripping steam to refnery frac-
tionators should normally be set on
ratio control to ensure that when
unit charge rates are reduced, steam
usage is reduced by a like amount.
However, in many instances, a
refnery that has been running at
maximum unit throughputs for an
extended period of time will have
the stripping steam set to a fxed
fow rate (usually due to a
perceived maximum). Then, when
unit charge rates are reduced, steam
rates are not reduced to the optimal
level. The same principle applies to
reboilers.
Another energy-saving measure
that may be less obvious is to
reduce fractionator tower pressures.
Pressures will frequently be
streit.indd 4 9/6/11 09:49:36
CHALLENGE
YOUR EXPECTATIONS
When a German oil renery experienced corrosion problems in its crude oil
distillation column, it turned to Sandvik for the solution. Since pitting and under
deposit corrosion were the reasons behind the short heat exchanger lifetime,
Sandvik suggested replacing the previously used carbon steel tubes with next
generation duplex stainless steel grade Sandvik SAF 2707 HD.
The new tube material was installed in 2006. Inspection after 4 years showed only
slight erosion corrosion. After cleaning, eddy current testing and hydro testing, the
bundle was reinstalled to run for another 4 years.
Reducing the number of shutdowns from 8 to 1 over a 4 year period has increased
plant safety and resulted in big savings in terms of material replacement, manpower
and lost production.
www. s mt . s andvi k. com/ oi l ref i ner y
Oil Refinery | Marcelo Senatore | Technical Marketing | Germany
If eight shutdowns could be
reduced to one, what would
that do for your productivity?
www.ptqenquiry.com
for further information
sandvik.indd 1 8/6/11 12:28:34
the refnery gasoline-producing
units may reveal a new optimum
operating point, minimise butane
import and reduce energy
consumption.
The tank farm is another area that
can be re-optimised at lower unit
charge rates. Reducing working
capital in intermediate storage can
allow the refnery to generate
immediate proft. Reducing work-
ing capital should be possible
because turnover time in tankage is
increased at lower fow rates. In
other words, less tank inventory is
required to maintain the same days
of inventory in tankage.
Another way to potentially take
advantage of lower refnery
throughput is to over-treat a prod-
uct and blend back to specifcation.
This technique can be used in distil-
late production, in particular,
although it may apply to gasoline
or fuel oil production. With spare
hydrotreater capacity, increasing
sulphur removal below the target
specifcation without increasing run
length may be possible. By over-
treating the distillate, not only is
the sulphur reduced but it may
allow the operator to increase
aromatic saturation and thus
improve cold properties and cetane.
By doing this, it could be possible
to blend raw or less severely
hydrotreated distillate directly into
the fnished product. This may help
improve overall energy effciency
and, in some cases, could reduce
the total amount of material proc-
essed in the refnerys hydrotreaters,
potentially allowing for a unit to be
shut down for extended periods of
time.
Recycle stream optimisation
When refnery throughput was the
driving force behind most economic
decisions, typically recycle streams
in the refnery were minimised. If a
unit was hydraulically limited,
recycling material back to that unit
would back out charge. If the unit
was constrained on furnace fring,
reactor or fractionator inlet temper-
atures would be reduced.
Minimising recycle was typically
the right economic answer.
With unit feed rates reduced,
recycling material to upstream units
102 PTQ Q3 2011 www.eptq.com
selectivity. Lower vapour rates can
reduce cyclone performance and
actually increase catalyst losses.
Recycling slurry is one possible
option to bring riser residence time
back up and improve cyclone
performance. As with other options,
individual unit modelling should
be used to determine the economic
incentive of recycling slurry.
Volume expansion
One of the best ways to make
money in a refnery is to maximise
volume expansion. The margin on
increasing the production of liquid
products is usually very large.
Running units at lower rates can
frequently make room for opera-
tional adjustments to maximise
volume expansion.
For hydrotreating units, a lower
unit charge rate means a lower
liquid hourly space velocity (LHSV),
which will automatically increase
hydrogen consumption. For high-
pressure ULSD units that are not
fully loaded, reactor temperatures
can frequently be increased higher
than in the past due to an unloaded
furnace. By running the reactor
outlet temperature up to the point
of maximum aromatic saturation,
volume expansion can be maxim-
ised. This is the philosophy used
for many FCC feed hydrotreaters
today.
A secondary effect of operating a
ULSD unit at higher reactor temper-
atures is that the lower sulphur
product produced can allow a
hydrotreater running at a lower
pressure (that cannot achieve the
same level of aromatic saturation as
a higher-pressure unit) to run to a
higher sulphur specifcation and
thus save energy and extend cycle
length. This same approach may
allow for bypassing and blending
of untreated or mildly treated mate-
rial while still meeting blended
product specifcations.
Another way to increase volume
expansion is to reduce volume loss.
The biggest volume loser in the
refnery is usually the naphtha
reformer. By lowering the reformer
feed rate and/or severity, volume
loss can be signifcantly reduced. In
some cases, routing the heaviest
reformer feeds to a jet or ULSD
is not as costly as it has been. Of
course, recycling still increases
energy usage and may increase
other operating costs, such as proc-
ess chemical addition. However, the
benefts may now outweigh the
costs. For instance, recycling some
streams back through crude units
or gas plants may allow for
increased product recovery or sepa-
ration, thereby justifying the
additional reprocessing expense.
Another place where recycle
might make sense is in hydrotreat-
ers. Recycling hydrotreater product
back to hydrotreater feed can help
to mitigate excessive exotherms in
those units. By recycling products,
a higher percentage of cracked
stocks may be able to be processed
in the unit, thereby increasing
liquid yield via hydrogen
consumption.
Slurry recycle in an FCC unit is
another option normally not consid-
ered at high unit feed rates. Low
FCC feed rates will increase riser
residence times and reduce yield
Benchmark and
set baseline
Identify portfolio
of opportunities
Evaluate options
Implement
Sustain
Figure 5 Proft improvement methodology
streit.indd 5 9/6/11 09:49:46
REFINED
PERFORMANCE
In todays economic environment optimising the performance of
your production facilities is critical to ensuring a safe and sustainable
future for your business.
DNVs unique performance simulation tools are designed to help you extract the power of refined performance from your
assets. Our global network of expertise allows us to deliver value adding services that help you maximise production, reduce
costs and enhance business performance.
&OUFSQSJTFSJTLNBOBHFNFOUt4BGFUZIFBMUIBOEFOWJSPONFOUBMSJTLNBOBHFNFOUt1SPDFTTTBGFUZNBOBHFNFOUt0QUJNJTJOHBTTFUEFTJHO
BOEPQFSBUJPOTt"TTFUSJTLNBOBHFNFOU
XXXEOWDPN
www.ptqenquiry.com
for further information
dnv.indd 1 8/6/11 12:30:12
hydrotreater can allow some of the
material to be transferred into the
distillate pool rather than reforming
it to high-octane gasoline blend-
stock. With the incorporation of
ethanol into the blend pool and the
continual pressure to lower aromat-
ics in the gasoline pool, reducing
the naphtha reformer feed rate is
typically a good economic decision.
Shift intermediate production
When fow rates to individual units
are lowered, re-routing streams to
different locations may be economi-
cal. Routing diesel to the FCC unit
was likely an unusual approach
when the FCC unit was running at
full rates. However, cracking diesel
in an FCC unit can be an economi-
cal operation when there is no other
feed available. The volume yield
provided by processing diesel in an
FCC unit is likely greater than can
be achieved in a diesel hydrotreater.
If the margin between gasoline and
product diesel is low enough, it
may be proftable to route the
diesel, or a portion of it, to the FCC
unit.
In fully loaded steam-methane
reformers, running hydrogen-rich
streams to the unit is usually not a
proftable operation. However, if
the unit is running low rates, feed-
ing hydrogen-rich streams to the
unit will likely make sense, as the
hydrogen can be recovered into the
high-purity hydrogen product
rather than going to fuel. In some
cases, it may be proftable to shut
down hydrogen purifcation units
that are energy intensive, such as
membrane or cryogenic units, and
capture the economic incentives for
improvement in a given area. For
example, if future margins on jet
fuel are expected to drop, opportu-
nities to increase jet fuel production
are likely to be less valuable than
opportunities for other products.
Taking future operation into
account can infuence where proc-
ess improvement development
should be focused.
Step 2: identify opportunities
Once the benchmarking stage is
complete, the next phase involves
identifying potential opportunities.
Several options exist and all should
be considered for utilisation:
Complete gap analysis on bench-
marking results from Step 1, as well
as other metric reviews
Review and collate existing proft
improvement idea and initiative
lists from multiple sources (opera-
tions, engineering, planning,
projects, central headquarters and
so on)
Unit and system brainstorming
sessions utilising a cross-functional
team.
The challenge most refners have
is in prioritising and ranking the
opportunities. One option is to use a
prioritisation matrix (see Figure 6).
Although depicted simplistically,
the ranking matrix should be cali-
brated by a cross-functional team to
set the basis for the axis and defne
the criteria for each trench. In the
referenced example, economics and
probability of success are the crite-
ria, and opportunities scored with
the highest value on each axis
should be pursued as the higher
priority. This same team should
apply the calibrated matrix to the
potential opportunities to outline a
work list for the evaluation phase.
In addition, this same matrix can
be utilised as part of the continuous
improvement phase to screen future
potential opportunities. Without a
sharp focus on the most valuable or
easiest-to-capture opportunities, the
organisation can waste valuable
time and resources on low-priority
initiatives.
Step 3: evaluate options
Once the opportunities are identi-
fed and prioritised, there will likely
104 PTQ Q3 2011 www.eptq.com
route the feed gases to the steam-
methane reformer.
Methodology
For most refners, the challenge is
not in generating ideas for proft
improvement, but in employing a
methodology to screen, evaluate,
implement, and sustain proft
improvement opportunities and
strategies (see Figure 5). The follow-
ing outlines an industry-proven
methodology, along with key issues
that should be addressed to success-
fully deploy the outlined approach.
Step 1: benchmark and set baseline
Benchmarking a refnerys opera-
tion is an important step prior to
any evaluation of opportunities. A
refnery can be benchmarked
against its peers or against best
practice operation. The latter
approach is preferred, as this ties in
closer to best achievable operation
or links with physical constraints
that must be managed to increase
proftability. Identifying areas
where gaps are the largest and
provide the most economic incen-
tive provides guidance on which
areas to focus on when developing
ideas for improvement. For exam-
ple, a better unit KPI to track for a
diesel hydrotreater may now be
hydrogen consumption rather than
unit throughput.
Determining how a refnery is
operating and how it is expected to
operate in the near future provides
a baseline against which any oppor-
tunities can be evaluated. It is
important to account for future
planned operation to properly
5
4
3
2
1 2 3 4 5

I
n
c
r
e
a
s
i
n
g

e
c
o
n
o
m
i
c
s
Increasing probability of success
1
Figure 6 Decision matrix
streit.indd 6 9/6/11 09:49:55
be several options to pursue the
opportunity. Evaluation of these
can be accomplished in a number
of different ways.
Relatively simple opportunities
can be feld tested as part of an
evaluation phase. A classic example
is FCC riser injection steam, where
test runs and rigorous data analysis
are required to fnd the optimal
operating range, although vendor
information and experience can
guide the optimisation.
For more complex or multi-
system opportunities, models and
simulations are required. Several
options exist, each with their own
inherent capabilities and limita-
tions. Most refners apply the use of
a linear program (LP) to not only
complete planning and scheduling
activities, but also to examine opti-
misation strategies. While an LP is
effective for evaluating many
opportunities, the tool does have
limitations. For instance, sub-
models and associated shift vectors
may have been calibrated at a
higher throughput than those
currently observed. An LP, by def-
nition, is linear in nature, so these
sub-models need to be validated
against current plant operation and
updated, as required. Also, an LP
may not be able to capture all of
the non-linear aspects required to
analyse unit and site-wide
opportunities.
Another popular toolset is proc-
ess simulation. Individual unit
process simulations are particularly
useful for examining the value of
changing operating conditions in
separation vessels and fractionators,
and completing detailed equipment
reviews as part of process integrity
checks. However, these simulations
are often complex and require
profcient users to be effective. In
addition, as with the LP, the models
should be validated against recent
plant data before being used on
opportunity evaluations.
Kinetic reactor models are critical
for effective optimisation studies.
Often, proftability is improved by
changing stream routing in the
facility or changing the qualities of
feeds to various units. Therefore,
kinetic models can not only esti-
mate the impact to product yields
and qualities, but also highlight
the changes on unit operating
requirements.
For a high degree of accuracy in
economic and technical evaluations,
the kinetic reactor models are
embedded into the process simula-
tion. This allows the optimisation
team to examine performance on a
holistic basis and even identify
additional opportunities, as well as
provide guidance on the implemen-
tation phase.
Some novel evaluation tools are
available. Since fnal product blend-
ing is critical to meeting production
specifcations with minimal givea-
way, non-linear blending models
are often used as part of the evalu-
ation phase. These tools not only
help with giveaway minimisation
opportunities but also validate that
upstream process changes can
produce on-specifcation products.
Computational fuid dynamic
(CFD) models can help to evaluate
www.eptq.com PTQ Q3 2011 105
To keep pace with the demanding quality requirements of
modern fuels, advanced, precise and easy to use analytical
technology is required. With a complete range of XRF and
ICP spectrometers, SPECTROs unique solutions for at-line
and laboratory elemental analysis are capable of meeting
the most demanding product specification testing
requirements.
Measure your fuels
- at sub 10-15 ppm levels of sulfur to ensure federal
government agency compliance
- at trace ppm levels for metal elements such as Cu, Ca, Mg,
Na, K, and P, to ensure low engine emissions as well as
trouble-free motoring
- with reliability, speed and accuracy for any particular application
- manually or fully automatically
Discover more exciting details,
visit SPECTROs e-Learning center
or contact us for additional
information about the SPECTRO
solutions for fuels analysis at
www.spectro.com/fuels
spectro.info@ametek.com and
Tel +49.2821.892-2102.
Elemental Analysis
of Fuels
Determination of Sulfur and other elements
at-line and in the laboratory
www.ptqenquiry.com
for further information
streit.indd 7 9/6/11 09:50:11
options. They identify specifc areas
of stagnant or low fows when fow
rates are lowered. If the fow to a
heat exchanger, for example, falls
too low, local fow rates in the head
of the exchanger may be low
enough to settle solids that would
typically be washed away at higher
rates. Problems resulting from this
can include an increase in pressure
drop, under-deposit corrosion or
equipment pluggage. Using a CFD
model can help determine potential
areas of low fow or fow maldistri-
bution, allowing the refner to
address the problem proactively.
Step 4: implement
The frst step in implementing a
process change is usually to feld
test the change. Management of
change (MOC) forms should be
flled out and proper procedures
followed to ensure that the change
does not cause a secondary nega-
tive effect that has not been
considered.
When the operational change is
made, a test run should be
performed to verify that the
response matches what was
expected. To prepare for a test run,
any suspect meters should be cali-
brated and sample points should be
verifed operational and be able to
provide a representative sample.
Documentation of the test run is
necessary to make sure all inter-
ested parties are aware of the result
of the change.
Good data quality is integral to
successfully evaluating any opera-
tional move made in a refnery.
There are several components to
good-quality data including, but
not limited to:
Closed loop sample points
Proper sampling and testing
procedures
Accurate and reliable metering
Accurate unit mass balances.
Although these items are impor-
tant for day-to-day unit operation,
they are essential for evaluating a
test run. Without good metering
and sampling, an accurate unit
mass balance cannot be calculated.
Without a valid mass balance,
determining the actual effect of an
operational move is much more
diffcult.
To perform an accurate test run,
the unit or units affected should be
kept as stable as possible and single
steps in operating parameters
should be made. Trying to evaluate
the effect of a change in stripping
steam on a crude tower, for exam-
ple, while changing crude slates to
the unit will be challenging. Instead,
the steam rate should be changed
to the new level as quickly as is
safely possible without upsetting
the unit. The unit should then be
given time to come back to a steady
state, and all samples and meter
readings recorded.
A successful test run provides
enough data to properly evaluate
the effect of the process move being
examined. With this information,
the true economics of a process
change can be determined. In addi-
tion, LP and process models can be
validated so that they can be used
to help make other decisions and
select strategies in the future.
Step 5: sustain/continuous
improvement
Sustaining proftable operation is
often one of the most overlooked
steps in the process. Once the idea
has been identifed, proven to be
proftable and implemented, it is
human nature to move on to the
next proft improvement item.
However, without enacting some
plan to sustain the existing
improvement, performance will
tend to back-slide to the more
comfortable and less proftable
operation.
Several options exist to ensure
that the desired operational change
is sustained:
Operating procedures updated to
incorporate any operational change
Operating personnel educated on
why the operational change is
necessary and proftable
Key performance indicators
(KPIs) and metrics developed to
measure the effect of the change
Unit-monitoring tools updated to
include the newly developed KPIs.
An acceptable range for this KPI
should be included in these unit-
monitoring tools
Decision-making tools and work
processes established, to track
opportunities against current pric-
ing and strategies put in place to
change philosophy as economics
dictate.
Ultimately, the entire fve-step
process is perpetually repeated as
part of continuous improvement
activities, such that new opportuni-
ties are incorporated into the
process and the facility maintains
focus on pushing for fundamental
improvement as part of normal
business practices.
Table 1
Table 2
Process variable % Change from base
Sweet naphtha to blending +0.5
of gasoline
Reformer feed -1.7
C
3
/C
4
product from reformer -1.9
Reformate production -1.6
Volume loss at reformer -1.8
Butane to blending +0.9
% Regular product +2.1
% Premium product -5.2
Total gasoline production +0.3
Key parameters for sweet
naphtha blending
Process variable Delta/% change
from base
DHT 1 WABT +46F
DHT 2 WABT -5F
DHT 1 H
2
consumption +16.6
DHT 2 H
2
consumption -3.0
Total H
2
consumption +13.5
DHT 1 cetane +3.1
DHT 2 cetane -0.2
ULSD2 cetane +1.5
DHT 1 sulphur -7.9 wppm
DHT 2 sulphur +6.2 wppm
ULSD2 sulphur
Total ULSD2 production +0.4
Key parameters for increasing
HTR severity
106 PTQ Q3 2011 www.eptq.com
Good data quality
is integral to
successfully
evaluating any
operational move
made in a refnery
streit.indd 8 13/6/11 14:18:29
Application of process
The following outlines examples of how the work proc-
esses and market information discussed above can be
applied in the industry.
Blending sweet naphtha
Hydrotreated crude naphtha is traditionally not a good
gasoline blendstock. It has a very low octane and there-
fore blending it to gasoline requires other high-octane
blendstocks. However, the other blending properties,
such as low RVP, moderate 50% point, low aromatics
and low olefns, are generally positive (see Table 1).
With the addition of ethanol to the gasoline blending
pool, along with a reduction in higher-octane gasoline
demand, the penalty associated with the low octane of
this stream is lower than it has been in the past.
As an example, routing sweet naphtha away from
the naphtha reformer and directly to blending was
analysed using KBCs Petro-SIM full refnery fow
sheet model and determined to be proftable. Reducing
the volume loss from the reformer was valuable
enough to offset the reduction in premium gasoline
product. The lower RVP of the sweet naphtha to blend-
ing allowed more butane to be blended to gasoline.
Thus, total gasoline production was increased due to a
reduction in volume loss across the reformer and
increased butane blending.
In this example, there were several items that contrib-
uted to the proftability of this move:
A high-purity hydrogen price was low and availabil-
ity was high, reducing the economic value of the
lower-purity hydrogen produced by the reformer
The reformer makes chemical feedstock, so a rela-
tively high-severity operation was required
Increasing hydrogen purity on the reformer is not
possible because of compressor issues
High-octane feedstocks are readily available.
Blending sweet naphtha in the past was not explored
because the value of gasoline octane was much higher.
With octane value lower under current economics,
avoiding volume loss on the reformer makes this
option feasible.
Increasing severity in a high-pressure ULSD
hydrotreater
Increasing the WABT on a high-pressure hydrotreater
allows a lower-pressure hydrotreater to operate at
lower WABT while still producing the same overall
product sulphur. The result is an increase in total
hydrogen consumption, boosting the liquid volume
yield.
Although this move shortens the cycle time of the
higher-pressure unit, the reduction in severity in the
low-pressure unit somewhat offsets that cost. In addi-
tion, due to lower refnery throughputs, the high-
pressure unit had suffcient catalyst activity to still
meet overall cycle length targets.
Another beneft of this move is that the overall cetane
is increased. This shift makes the processing of addi-
tional low-cetane feedstocks, such as LCO, possible.
Processing additional cracked stocks reduces cycle
www.eptq.com PTQ Q3 2011 107
length but provides signifcant proft due to further
increased hydrogen consumption. Table 2 summarises
the Petro-SIM simulation results for this strategy.
Conclusion
Given projected margins and refned product demands,
refners are driven to examine alternate and novel
approaches for maintaining and improving proftabil-
ity. Lower utilisation of individual process units and
entire refneries provides a unique opportunity to turn
an apparently negative situation into a positive
outcome. Applying a proven work process for bench-
marking, identifying, evaluating, implementing and
sustaining opportunities will facilitate successful trans-
formation of the facilitys proftability.
Eric Streit is a Senior Staff Consultant with KBC Advanced Technologies,
Houston, Texas. His primary responsibilities are consulting on initiatives
for general refnery proft improvement and hydroprocessing, and
reforming technical consulting for domestic and international clients.
He holds a BS in chemical engineering from Texas Tech University.
Email: estreit@kbcat.com
Robert Ohmes is a Senior Staff Consultant with KBC Advanced
Technologies, Houston, Texas. His primary responsibilities are centred
on initiatives for proft improvement and consulting on hydroprocessing
units for domestic and international clients. He holds a BS in chemical
engineering from Kansas State University (BSChE) and a MBA from
Tulane University, and is a licensed professional engineer in Louisiana.
Email: rohmes@kbcat.com
Drying Technology
www.newton-s.com
u
n
d
e
r
C
O
N
T
R
OL ...
W
a
t
e
r
s
p
e
c
i
f
i
c
a
t
i
o
n
K
e
e
p
Y
o
u
r
H
Y
D
R
O
C
A
R
B
O
N
www.ptqenquiry.com
for further information
streit.indd 9 9/6/11 09:50:30
C O M P R E S S O R S Q T U R B I N E S Q G L O B A L S E R V I C E
EBARA CORPORATION
www.elliott-turbo.com
Q Customer:
Ethylene plant, Saudi Arabia.
Q Challenge:
Overhaul and repair four critical steam turbines in 12 days.
Q Result:
Elliott coordinated six months of pre-planning
and completed the project four days early.
They turned to Elliott
for planning and execution.
The customer turned to Elliott because they understood that planning is everything
and that few companies can match Elliott in turnaround planning and execution.
Who will you turn to?
The world turns to Elliott.
www.ptqenquiry.com
for further information
elliott.indd 1 13/6/11 14:19:37
Multi-stage reaction catalysts

B
ASFs Multi-Stage Reaction
Catalyst (MSRC) platform
takes advantage of staged
reactions with different catalytic
attributes in much the same way
that staged hydrotreating catalyst
loading permits different reaction
zones in a fxed-bed reactor vessel.
The concept of staged reactions is
not new to the refning industry,
but its application to a circulating
system such as FCC is an advance-
ment in catalyst technology.
The manufacturing platform can
utilise any of the existing catalyst
technologies including BASFs
Distributed Matrix Structures
(DMS)
1,2
and Proximal Stable Matrix
and Zeolite (Prox-SMZ)
3,4
to create
the stages. The location of the vari-
ous stages can be specifcally
engineered to achieve maximum
performance. There are multiple
possibilities for combining different
catalyst technologies in the inner
and outer stages, depending on
specifc objectives such as the
processing of heavier feedstocks or
to maximise specifc product yields.
This manufacturing process is
based on in situ technology and
involves several key sequential
manufacturing steps. One of the
key success factors in the develop-
ment of MSRC technology is the
binding process, unique to the in
situ manufacturing process, where
Y zeolite grows across the bound-
ary between catalyst stages, acting
as a binder and giving the catalyst
particle high attrition resistance.
The frst product utilising MSRC
manufacturing is Fortress. Fortress
is designed for resid feed applica-
tions, where contaminant metals
passivation is critical. MSRC
Recently trialled manufacturing technology combines two or more FCC catalyst
functionalities within the same catalyst particle
Joesph MCLean, BRaM hoFFeR, GaRy sMith, DaviD stoCkweLL and aLexis shaCkLeFoRD
BASF
manufacturing technology was
scaled up and demonstrated in 2010
with Fortress, and two refnery
trials were initiated.
Halo, the second catalyst offering
from BASF under the MSRC manu-
facturing platform will be ready for
commercial trials in the third quar-
ter of 2011. Halo is engineered to
provide attrition resistance compa-
rable to Petromax with yields, coke
and bottoms selectivity equivalent
to NaphthaMax III. As with
Fortress, a two-stage system is used,
but in this case it delivers an outer
stage with highly active zeolite. The
staged approach allows for a reduc-
tion in the level of Y zeolite in the
particle, as well as a reduction in
the diffusion path length. The inner
stage is composed of activity modi-
fying material, which also serves as
an anchor for the outer stage. Due
to the reduced level of zeolite in the
catalyst particle, there is an addi-
tional beneft in the form of a
reduction of RE level with no
performance defcit.
Fortress
Fortress is designed for resid feed
applications, where contaminant
feed metal passivation is crucial.
Feed metals, in particular nickel,
catalyse dehydrogenation reactions,
producing the undesirable products
of hydrogen and coke. In resid FCC
catalysts such as Flex-Tec, a special-
ity alumina is integrated in the
catalyst formulation to trap the
nickel and form nickel aluminate,
which is less deleterious to dehy-
drogenation reactions in the FCC
riser. By examining spent FCC cata-
lysts from refneries with electron
microscopy, it was generally
observed that, while vanadium is
distributed homogenously through
the particles, nickel mainly deposits
and accumulates on the outer
surface of the catalyst (see Figure
1). It would thus be an advantage
www.eptq.com PTQ Q3 2011 109
0.4
0.6
0.2
50 25 0 25 50
V

o
r

N
i

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

w
t
%
Distance from centre, microns
0
V
Ni
Figure 1 Vanadium is distributed homogeneously, while nickel mainly accumulates on the
outer surface of the catalyst
mclean basf.indd 1 9/6/11 10:10:04
110 PTQ Q3 2011 www.eptq.com
sphere is then added to a second
slurry, which has a Flex-Tec-like
formulation, including elevated
concentrations of the nickel-passivat-
ing speciality alumina, and
spray-dried again to yield a two-
stage microsphere with the required
fnal FCC particle size.
The FCC catalyst is then manu-
factured by growing zeolite from
the kaolin nutrients in the multi-
stage microsphere. The zeolite
grows in both stages and across the
interface, and acts as a binder to
hold the stages together, preventing
selective attrition of the outer stage.
Zeolite growth and catalyst activity
are in line with current DMS prod-
ucts, as are all physical properties.
Development of Fortress
The basis for the development of
the Fortress catalyst was the obser-
vation that nickel accumulates
mainly in the outer 510 m of the
equilibrium FCC catalyst
5
(see
Figure 1). Geometric calculations
show that the weight ratios of the
two stages for a ~8 m outer-stage
thickness and a normal FCC parti-
cle size distribution are about 1:1.
The addition of a solid inner-stage
particle to the recipe makes the
production of MSRC more complex
than for existing FCC catalysts.
to concentrate the nickel-trapping
alumina at the outer layer of the
catalyst to make it more effective.
With current catalyst technology,
the speciality alumina is uniformly
distributed through the catalyst
microsphere. This makes a large
portion of it, located in the interior
of the particle, unavailable to react
with the nickel and is, in essence,
wasted.
In the MSRC approach, the inner
stage of the catalyst has the DMS
structure to allow enhanced diffu-
sion of heavy molecules and
selective pre-cracking on the
exposed zeolite surface, maximising
gasoline yields.
1
The outer stage is
also based on DMS technology, but
is enriched with speciality alumina
to trap the nickel directly where it
enters and deposits on the catalyst
(see Figure 2). The technology is
analogous to that employed with
Flex-Tec
2
and Stamina
4
, but the
improved spatial distribution of the
trapping alumina offers more eff-
cient material utilisation and better
performance potential.
The manufacturing process for this
two-stage reaction catalyst is based
on in situ manufacturing technology
and involves multiple consecutive
steps (see Figure 3). First, a precur-
sor microsphere with a reduced
particle size is formed. This micro-
Alumina
Flex-Tec Fortress (Resid. MSRC)
Clay slurry
H
2
O
Dryer
Calciner
Calciner
Rare-earth
exchange
Multistage reaction
catalyst
Base
exchange
Na-Y
Calciner
Alumina-free
inner-stage MS
(~65 m)
Step 1: Inner-stage microsphere
Step 3: Zeolite crystallisation and after-treatment
Clay slurry
+ alumina
+ inner-stage MS
H
2
O
Dryer
Two-stage MS
(~80 m)
Step 2: Outer-stage microsphere
Microsphere
caustic
silicate
Figure 2 Simplistic representation of alumina distribution in current resid catalyst
Flex-Tec (left) vs MSRC resid catalyst technology (right)
Figure 3 Manufacturing sequence of MSRC technology
mclean basf.indd 2 9/6/11 10:10:13
www.ptqenquiry.com
for further information
enersul.indd 1 9/12/10 12:23:52
Scanning electron microscopy of the
early samples demonstrates the
multi-stage confguration of the
particles (see Figure 4). By modify-
ing the microsphere formulation,
the density of inner- and outer-
stage components was optimised to
allow maximum diffusion rates
without impacting the staged
confguration. After demonstration
of manufacture on a pilot scale and
encouraging test data, the project
was transferred to a commercial
manufacturing scale. The plant
production trials required the capa-
bilities of multiple manufacturing
plants to make frst the inner-stage
microspheres, followed by the
construction of the outer-stage
material, and fnally zeolite crystal-
lisation and fnishing to make the
catalyst (see Figure 2).
The particle size distribution of
the resulting microspheres is shown
in Figure 6. The graph shows that
the inner-stage particle has an aver-
age particle size (d50) of 64 m and
a fnal particle (inner + outer stage)
of 80 m, resulting in an average
outer-stage thickness of 8 m. The
thickness is fairly constant through-
out the particle size distribution.
112 PTQ Q3 2011 www.eptq.com
This is important, since the expected
nickel penetration depth would be
fairly constant across the entire
particle size range, so that both
larger and smaller particle sizes
should have a comparable outer-
stage thickness.
The attrition resistance of the
commercially manufactured catalyst
was measured with two standard
attrition-testing techniques: Roller
and Airjet (see Table 1). The table
shows that attrition properties are
very similar to those of Flex-Tec,
indicating that the two stages are
well connected and form a mechan-
ically stable particle. Since the
zeolite grows between and
chemically bonds to the matrix
support materials in both stages,
attrition resistance and other physi-
cal properties such as density were
not any different from existing DMS
products such as Flex-Tec. This was
also validated in the commercial
performance, which is described
later.
The product from the frst manu-
facturing campaign was evaluated
in the lab using an advanced crack-
ing evaluation (ACE) unit. Testing
was done with a commercial resid
feed at 960F and variable catalyst-
to-oil ratios to generate yield
response curves. The performance of
Fortress was compared to the Flex-
Tec catalyst that was being supplied
to the trial refnerys FCC unit. Prior
to ACE testing, the samples were
deactivated and metallated using a
cyclic metallation and deactivation
unit, with intermediate steaming to
target approximately 2800 ppm Ni
and 1100 ppm V, as well as commer-
cially representative equilibrium
catalyst activity. Table 2 shows the
properties of the pseudo-equilibrium
catalysts. The chemical and physical
properties of Fortress are very simi-
lar to those of Flex-Tec; the
difference is in the distribution of
alumina concentrated in the outer
stage with Fortress, but uniform
with Flex-Tec.
The proper laboratory prepara-
tion of a pseudo-equilibrium
catalyst for a resid cracking
performance evaluation is not triv-
ial. This is because conventional
(Mitchell) pore volume impregna-
tion methods result in nickel being
distributed uniformly throughout
the catalyst microsphere, thus
10.0m
10.0m
10.0m
10.0m
Figure 4 SEM pictures of early MSRC lab samples show multistage concept
0.50
0.45
0.40
0.35
0.30
0.25
65.0 70.0 75.0 80.0
H
y
d
r
o
g
e
n
,

w
t
%
Conversion, wt%
0.20
Fortress
Flex-Tec
Figure 5 Hydrogen yield in ACE testing of Fortress compared to Flex-Tec. The samples
were deactivated with CMDU to 3000 ppm Ni and 1000 ppm V
Flex-Tec Fortress
Roller, wt% loss/h 18 16
Airjet rate,* wt% loss/h 2.7 2.9
* Losses in frst hour of operation are disregarded
Attrition properties
Table 1
mclean basf.indd 3 10/6/11 14:10:10
properties, activity and selectivities.
The activity and selectivities are
measured on an ACE unit. This is a
standardised testing method using
a fxed fuid bed, which simulates
SCT (short contact time) riser
performance. It runs all the E-Cats
under the same conditions (feed,
temperature and catalyst-to-oil
ratio), providing a good basis for
comparison. Tables 4a and 4b show
typical E-Cat values for before and
after the purchased E-Cat was
removed.
With removal of the purchased E-
Cat, metals levels increased due to
lower catalyst addition rates. Nickel
content alone increased from ~7500
ppm to ~9500 ppm. Despite this
penetrating much further into the
microsphere than is observed with
an FCC units equilibrium catalyst
(E-Cat). Since hydrogen production
from impregnated metals is a linear
process, Fortress and Flex-Tec
appear to have the same Ni toler-
ance by pore volume impregnation.
Resid FCC units, on the other hand,
result in Ni deposition mostly on
the alumina-enriched outer stage,
so the correct answer is for MSRCs
to outperform catalyst technologies
employing uniformly dispersed
active ingredients. This correct
answer cannot be obtained by pore
volume impregnation.
The ACE results are shown in
Table 3. Fortress catalyst hydrogen
and coke yields were reduced, with
about 15% at a metals content simi-
lar to Flex-Tec, and a corresponding
increase in gasoline yield. Clean
testing in the absence of metals
showed no differences in coke and
bottoms yields, demonstrating that
no diffusion limitations occur in
laboratory testing and that the effect
observed in metals testing is a
passivation effect. Since H
2
produc-
tion is a direct measure of metals
dehydrogenation activity, it is clear
that Fortress is achieving improved
passivation. This is in line with the
laboratory prototypes shown in
Figure 5.

Commercial performance
Fortress was employed in the frst
of two commercial trials in the
second half of 2010. The trial that is
furthest along is a short contact
time resid unit located in the US.
The unit operates in deep partial
burn and low reactor severity, with
the primary constraints of wet gas
compressor and air blower limited.
The refners economic objects are
to maximise residue rates, while
minimising bottoms yields. With
the metals tolerance of Fortress, the
expected reduction in hydrogen
yield can be reduced to alleviate the
wet gas compressor limit, and the
improvement in coke selectivity can
allow more operating fexibility
within these constraints.
The unit was using Flex-Tec with
purchased E-Cat for metals fush-
ing. Flex-Tec replaced a competitive
catalyst early in 2010, following a
www.eptq.com PTQ Q3 2011 113
catalyst evaluation. For the cleanest
comparison, the purchased E-Cat
was removed prior to the start of
the trial. As such, three periods of
operation were compared: Flex-Tec
with E-Cat, Flex-Tec without E-Cat
and Fortress without E-Cat.
The trial was supported by analy-
sis of the refnerys E-Cat and fnes
data. The refnery sent twice weekly
E-Cat samples and weekly fnes
samples. These are systematically
analysed for chemical and physical
60
100
80
40
20
0 30 60 90 120 150
F
i
n
e
r

t
h
a
n
,

%
Particle size, micron
0
Inner + outer stages
Inner stage
Figure 6 Particle size distribution of inner stage and fnal plant produced microspheres
Flex-Tec Fortress
Fresh
TSA, m
2
/g 300 301
MSA, m
2
/g 72 69
REO, wt% 3.1 3.7
APS, m 76 76
040 m, wt% 11 10
080 m, wt% 54 55
Deactivated (CMDU)
Ni, ppm 2797 2718
V, ppm 1133 1050
Steamed TSA, m
2
/g 115 123
Steamed MSA, m
2
/g 36 38
ZSA/MSA 2.2 2.2
Properties of catalysts for ACE testing
Table 2
Flex-Tec Fortress
Hydrogen 0.53 0.44
Total C
4
- 15.9 15.5
Gasoline 49.4 50.4
LCO 20.1 20.1
HCO 9.9 9.9
Coke 4.7 4.1
Product yields wt% at 70% conversion
Table 3
FACT 71 wt%
Nickel 7400 ppm
Vanadium 2600 ppm
Delta iron 0.23 wt%
Antimony 2000 ppm
Lead 100 ppm
Carbon on catalyst 0.20.5 wt%
Equivalent nickel 5600 ppm
With purchased E-Cat
FACT 71 wt%
Nickel 8900 ppm
Vanadium 2700 ppm
Delta Iron 0.30 wt%
Antimony 2300 ppm
Lead 100 ppm
Carbon on catalyst 0.20.6 wt%
Equivalent nickel 6800 ppm
Without purchased E-Cat
Table 4a Table 4b
mclean basf.indd 4 10/6/11 14:10:22
114 PTQ Q3 2011 www.eptq.com
rise in metals levels, the hydrogen
make of the ACE unit did not
increase (see Figure 7). Equivalent
nickel in Figure 7, calculated by Ni
vation effect. Changing from
Flex-Tec without purchased E-Cat
to Fortress without purchased E-
Cat, the hydrogen selectivity of the
ACE unit improved by 12%. The
cross plot of the data (see Figure 8)
shows how the average hydrogen
yield decreased from 0.410.36 wt%
with Fortress, demonstrating
improved metals passivation.
The improved metals tolerance of
Fortress can also be seen in the
improvement in coke selectivity.
Figure 9 shows the ACE coke factor
(coke yield adjusted for activity) as
it corresponds to the equivalent
nickel. As with the ACE hydrogen
yield, the removal of purchased E-
Cat did not increase the coke factor
despite the increase in equivalent
nickel of over 1000 ppm. With
Fortress, the coke factor has
decreased signifcantly. Comparing
the ACE coke yield from the transi-
tion of Fortress to Flex-Tec (see
Figure 10), the coke yield decreased
by 17%. This decrease in coke of
~0.7 wt% for the same conversion
is unexpected. A general rule of
thumb is a 7:1 ratio of hydrogen
reduction to coke reduction. With a
hydrogen yield reduction of
0.05 wt%, this corresponds to an
expected 0.35 wt% reduction in
coke. A reduction in coke of
~0.7 wt% is twice what can be
explained from the hydrogen
reduction.
Since less mass is being converted
into hydrogen and coke, the delta
mass is showing up as 0.7 wt%
more LPG and gasoline yield (see
Figure 11). A commercial unit can
further realise additional beneft
from the improved coke selectivity
by being able to lower the regenera-
tor temperature, thus increasing the
catalyst-to-oil ratio, or by lowering
the CO
2
/CO ratio, resulting in an
increased liquid yield.
As should be expected from the
fresh catalysts properties, there
were no changes observed in
attrition or unit losses for the tran-
sition from Flex-Tec to Fortress. The
particle size distribution of the fnes
from the ESP shows no change
across the transition (see Figure 12).
A post-audit study of the trial
was completed on the ACE
unit using the refners feed and
+ V/4 + Fe/10 + 5 Cu - 4/3 Sb,
represents the equivalent dehydro-
genation activity of the metals on
the E-Cat and the antimony passi-
0.50
0.70
0.60
0.40
0.30
-200 -150 -100 -50 0 50 100
H

E
C
A
2
d
l
e
i
y

Day
0.20
6000
8000
7000
5000
4000
5500
7500
6500
4500
3500
m
p
p

,
l
e
k
c
i
n

t
n
e
l
a
v
i
u
q
E
3000
F|ex-Tec with E-Cat F|ex-Tec Fortress without E-Cat
E-Cat equiva|ent nicke|
ACE H
2
yie|d
Figure 7 E-Cat lab data during the trial period, ACE hydrogen yield and E-Cat equivalent
nickel
0.50
0.55
0.45
0.40
0.35
0.30
0.25
5000 5500 6000 6500 7000 7500 8000
H

E
C
A
2
%
t
w

,
d
l
e
i
y

Equivalent nickel, ppm
0.20
Fortress without E-Cat
F|ex-Tec with E-Cat
F|ex-Tec without E-Cat
Figure 8 E-Cat lab data during the trial period, ACE hydrogen yield vs E-Cat equivalent
nickel
1.9
2.5
2.2
1.6
1.3
200 150 100 50 0 50 100
A
C
E

c
o
k
e

f
a
c
t
o
r
Day
1.0
6000
8000
7000
5000
4000
5500
7500
6500
4500
3500
E
q
u
i
v
a
l
e
n
t

n
i
c
k
e
l
,

p
p
m
3000
Equivalent nickel
ACE coke factor
Figure 9 E-Cat lab data during the trial period, ACE coke factor and equivalent nickel
mclean basf.indd 5 9/6/11 10:10:37
Moving science forward
Thermo Scientifc Prima PRO
Named 2010 Innovative Product of the Year
by the International Society of Automation
(ISA) Analysis Division for superior
performance and ease-of-maintenance.
Introducing Thermo Scientific Prima PRO, the online gas
analyzer that uses high-speed mass spectrometry to increase
yield, conserve energy, and enhance process safety.
In fact, a single Prima PRO can do the work of ten gas
chromatographs to dramatically lower your cost of ownership
and raise the bar on product quality.
This powerful yet easy-to-use, simple-to-maintain system
offers automatic tuning and calibration and produces higher
resolution for a number of process variables. So you have a
better understanding of process dynamics and can implement
more effective operating procedures. Plus, Prima PRO is built to
deliver 99.9% uptimemaking it a safe, reliable choice.
To learn more, visit www.thermoscientific.com/primapro,
email us at sales.process.us@thermofisher.com or call
1 (713) 2720404 or 1 (800) 4377979.
Imagine one gas analyzer that does the work of ten.


2
0
1
1

T
h
e
r
m
o

F
i
s
h
e
r

S
c
i
e
n
t
i
f
c

I
n
c
.

A
l
l

r
i
g
h
t
s

r
e
s
e
r
v
e
d
.

C
o
p
y
r
i
g
h
t
s

i
n

a
n
d

t
o

t
h
e

M
a
t
c
h
e
s

a
n
d

B
l
o
w
t
o
r
c
h

p
h
o
t
o
g
r
a
p
h
s

a
r
e

o
w
n
e
d

b
y

a

t
h
i
r
d

p
a
r
t
y

a
n
d

l
i
c
e
n
s
e
d

f
o
r

l
i
m
i
t
e
d

u
s
e

o
n
l
y

t
o

T
h
e
r
m
o

F
i
s
h
e
r

S
c
i
e
n
t
i
f
c

b
y

m
a
x
x

i
m
a
g
e
s

a
n
d

S
u
p
e
r
S
t
o
c
k
.

www.ptqenquiry.com
for further information
thermo.indd 1 8/6/11 12:35:09
mimicking the refners operating
conditions, comparing the fully
changed-out E-Cat samples of Flex-
Tec and Fortress. The ACE
comparison shows a decrease in
hydrogen of 15%, dropping from
0.650.55 wt%. The coke yield
decreased by 0.5 wt% for the same
conversion. This is in alignment
with both expectations and the data
already discussed from the routine
E-Cat monitoring. Some of the
results are shown in Figure 13.
A second trial of Fortress took
place at another US refnery. The
beneft of the improved nickel
passivation in this unit is expected
to be slightly less than the prior
unit because the nickel content is
lower at 3000 ppm. Like the frst
refnery, this is a comparison with
Flex-Tec, and purchased E-Cat was
removed from the inventory. The
typical E-Cat properties before and
after the removal of purchased E-
Cat are shown in Tables 5a and 5b.
Another feature of this unit is a
very high level of contaminant iron
(~1 wt%). At this time, the unit is
operating well with Fortress, but it
is too early to tell the results of the
trial.
Conclusions
The MSRC manufacturing concept
represents a breakthrough develop-
ment in FCC catalyst technology.
Through this novel multi-stage
approach to catalyst manufacturing,
BASF is able to leverage the success
of its proven FCC catalyst technolo-
gies and enable the development of
next-generation catalysts that
address the challenges and objec-
tives of refners in todays rapidly
evolving refning landscape.
The frst product based on BASFs
FCC catalyst concept MSRC,
Fortress, has been developed and
scaled up to commercial production
and is demonstrating improved
performance in metals passivation
in early refnery trials. Manufacturing
trials for the second product,
Halo, are currently under way.
Commercialisation of Halo is
currently targeted for Q3 2011.
BASF intends to expand the
5.0
6.0
5.5
4.5
4.0
3.5
3.0
2.5
66 68 70 72 74 76
%
t
w

,
d
l
e
i
y

e
k
o
c

E
C
A
ACE conversion, wt%
2.0
Fortress without E-Cat
F|ex-Tec with E-Cat
F|ex-Tec without E-Cat
64.0
66.0
65.5
65.0
64.5
63.5
63.0
62.5
62.0
61.5
67 68 69 70 71 72 73 74
%
t
w

,
e
n
i
l
o
s
a
g

+

G
P
L

E
C
A
ACE conversion, wt%
61.0
Fortress without E-Cat
F|ex-Tec without E-Cat
12
16
14
10
8
6
4
2
0.1 1 10 100 1000
F
r
a
c
t
i
o
n
a
l
,

%
Particle size, microns
0
Day 68
Day 103
Day 3
Day 72
Day 40
Figure 10 E-Cat lab data during the trial period, ACE coke yield vs ACE conversion
Figure 11 ECat laboratory data during the trial period, LPG and gasoline vs conversion
Figure 12 ESP fnes PSD shows no changes
FACT 71 wt%
Copper 70 ppm
Nickel 2700 ppm
Vanadium 2200 ppm
Delta iron 0.9 wt%
Antimony 600 ppm
Lead 200 ppm
Equivalent nickel 3400 ppm
With Purchased E-Cat
FACT 69 wt%
Copper 70 ppm
Nickel 3500 ppm
Vanadium 3200 ppm
Delta iron 1.1 wt%
Antimony 900 ppm
Lead 200 ppm
Equivalent nickel 4200 ppm
Without Puchased E-Cat
Table 5a and 5b
116 PTQ Q3 2011 www.eptq.com
mclean basf.indd 6 10/6/11 14:10:40
number of Fortress applications,
commercially launch Halo, and
introduce other new technologies
based on the MSRC concept.
Fortress, HALO, Flex-Tec, Petromax and Stamina
are marks of BASF Corporation.
References
1 McLean J B, Stockwell D M, NaphthaMax
breakthrough FCC catalyst technology for short
contact time applications, NPRA AM-01-58.
2 Weber W A, Harris D H, Distributed Matrix
Structures a technology platform for
advanced FCC catalytic solutions, NPRA AM-
03-38.
3 McLean J B, The role of FCC catalyst
technology in maximizing diesel production,
NPRA AM-09-34.
4 Kraus M, Kiser N, Fu Q, Fu J, Thornton O, Finch
J, Stamina new FCC catalyst for maximum
0.70
0.80
0.75
0.65
0.60
0.55
0.50
0.45
60 65 70 75 80
H
y
d
r
o
g
e
n
,

w
t
%
Conversion, wt%
0.40
Fortress
Flex-Tec
60
64
66
62
58
56
54
52
60 65 70 75 80
L
P
G

+

g
a
s
o
i
n
e
,

w
t
%
Conversion, wt%
50
Fortress
Flex-Tec
16
18
14
12
10
8
60 65 70 75 80
B
o
t
t
o
m
s
,

w
t
%
Conversion, wt%
6
Fortress
Flex-Tec
11
13
14
12
10
9
8
7
6
60 65 70 75 80
C
o
k
e
,

w
t
%
Conversion, wt%
5
Fortress
Flex-Tec
16
18
14
12
10
8
6 8 10 12 14
B
o
t
t
o
m
s
,

w
t
%
Coke, wt%
6
Fortress
Flex-Tec
3.2
3.6
3.8
3.4
3.0
2.8
2.6
2.4
2.2
60 65 70 75 80
D
r
y

g
a
s
,

w
t
%
Conversion, wt%
2.0
Fortress
Flex-Tec
Figure 13 ACE post-audit testing using the refnerys feed and mimicking the refners operating conditions
distillate yield demonstrated in Big Wests Salt
Lake City Refnery; NPRA AM-10-171.
5 Kugler E L, Leta D P, Journal of Catalysis 109,
1988, 387.
Joesph McLean is BASFs Global Technology
Manager and has led the market introduction
of DMS and Prox-SMZ platforms. With over
35 years experience, he holds a bachelors
degree in chemical engineering from Princeton
University, a masters in chemical engineering
from the University of California, and several
patents. Email: joe.mclean@basf.com
Bram Hoffer is a Senior Research Engineer
with BASF and led the Fortress R&D
programme from lab phase up to commercial
trials in the FCC manufacturing plants. With
over seven years experience in industrial
catalyst development, he holds a MS and a
PhD in chemical engineering from Delft
University of Technology, The Netherlands, and
six US patents. Email: Bram.hoffer@basf.com
Gary Smith is a Research Chemist with BASF.
He has over 30 years experience in FCC
catalyst R&D, has worked on the development
of a number of catalyst technologies, and holds
four patents. Email: Gary.m.smith@basf.com
David Stockwell is a Senior Research Associate
with BASF, has over 25 years experience
developing products and solutions for FCC
applications, and was the principal inventor
of the DMS technology platform. He holds a
BS degree from the University of Rochester,
US, a MS and a PhD from the University of
Connecticut, as well as 17 US patents.
Email: David.stockwell@basf.com
Alexis Shackleford is a Technical Support
Engineer with BASF and manages the technical
services programme for refners. She holds a
bachelors degree in chemical engineering from
Michigan State University.
Email: Alexis.shackleford@basf.com
www.eptq.com PTQ Q3 2011 117
mclean basf.indd 7 10/6/11 11:52:34
www.ptqenquiry.com
for further information
paharpur.indd 1 11/3/10 14:52:56
Preventing environmental excursions on
the FCC unit
S
ignifcant attention has been
given to the reactor, riser and
product recovery sections of
the FCC unit. Attention has also
been given to the regenerator when
necessary due to high catalyst
losses, when afterburning arises
and when environmental regula-
tions require. The FCC world is
presently undergoing a paradigm
shift, in which the regenerator now
needs to be as well tuned as the
reactor riser vessel.
Optimisation of the regenerator is
required to minimise emissions
to comply with environmental
regulatory permits. Fortunately,
optimisation of the regenerator is
often less complicated than that of
the other sections of the FCC unit.
Close attention to regenerator oper-
ations will enable the FCC process
engineer to effciently and effec-
tively avoid unexpected emissions
excursions and the possibility of
large penalties being imposed by
regulatory environmental bodies.
This article provides a list of the
most common causes of unexpected
SO
x
, NO
x
and CO emission exceed-
ances from FCC units today. Simple
and effective techniques for diagnos-
ing and troubleshooting these
problems are embedded within each
section. There have been many arti-
cles published over the decades
relating to particulate emissions,
hence troubleshooting catalyst losses
will not be addressed within this
article.
SO
x
emissions control
SO
x
emissions from the FCC regen-
erator have long been monitored
and regulated by environmental
regulatory bodies. Furthermore, the
Guidelines to enable the FCC engineer to identify and eliminate the root cause of
unexpected fue gas emissions
Ray FletCheR and MaRtin evanS
Intercat
technology for controlling SO
x
emis-
sions from the regenerator has been
well defned and developed. SO
x

additives are very effective at
controlling SO
x
emissions (see
Figure 1). The guidelines provided
here will enable the process engi-
neer to swiftly troubleshoot the
most likely causes for unanticipated
shifts in SO
x
emissions.
Change in feedstock sulphur content
and source
SO
x
emissions from the FCC unit
will frequently, but not always,
correlate directly with feedstock
sulphur concentration (see Figure
2). A more accurate correlation for
predicting SO
x
emissions has been
developed using the Gulf correla-
tion, which is based on slurry
sulphur concentration. Most FCC
engineers track feed sulphur to
predict required rates of addition of
SO
x
additive. However, it is well
known that the amount of feed
sulphur going to coke can vary
with different feedstock types. The
Gulf correlation
1
uses slurry sulphur
concentration to predict coke
sulphur levels. This has been found
to be a better ft for most FCC units.
The Gulf correlation was developed
by the Gulf Research &
Development Company and relates
sulphur in coke to sulphur in the
slurry oil. This is the preferred
correlation used by Intercat to eval-
uate SO
x
additive effciency. This
correlation is defned as:
Coke sulphur = UF * (Sulphur in slurry)
1.265

where UF is the unit factor.
The dilemma faced by FCC engi-
neers is that feed sulphur content
can be easily predicted (via the
projected feed slate), whereas slurry
sulphur content is not easy to
predict.
However, most FCC units process
a limited number of feedstocks. The
prudent engineer will evaluate
which feedstocks have the largest
www.eptq.com PTQ Q3 2011 119
1500
2500
2000
1000
500
0 0.3 0.6 0.9 1.2 1.5 1.8 2.1

S
O
x

e
m
i
s
s
i
o
n
s
,

p
p
m
Feed sulphur, wt%
0
Base
SSG
Figure 1 Additives are highly effective
fletcher.indd 1 9/6/11 10:36:51
120 PTQ Q3 2011 www.eptq.com
changes in SO
x
emissions have been
related to equilibrium iron concen-
trations in many cases. The half-life
for this iron activity is very short
(typically in the range of two to
three days), but the magnitude of
the effect may be very large.
Temporary increases in SO
x
addi-
tive injections will, in most cases,
be suffcient during these step
change events. Identifcation of the
feed source is recommended, with
consideration given to controlling
the rate of injection in the future.
Additive loader failure
An additive loader failure can, in
some cases, result in a SO
x
excur-
sion. It is strongly recommended
that the FCC operator uses an addi-
tive loader supplier with a proven
record of high reliability.
3
Furthermore, the additive loader
should be capable of manual opera-
tion in the event of a power outage.
It is recommended that the additive
loader control system be connected
to the distributed control system on
the FCC unit. A shutdown alarm
will then sound in the control room,
enabling the operators to swiftly
reinitiate injections should a prob-
lem occur.
Flue gas analyser failure
Regular preventive maintenance of
the fue gas analysers is strongly
recommended, with close attention
being applied to the fue gas sample
conditioning system. Flue gas
conditioning systems seem to be
the most common cause of poor
analyser readings. This critical
system may often be overlooked by
instrumentation groups. Two criti-
cal elements of a successful
conditioning system are the removal
of steam condensate and catalyst
fnes. Unexplained step changes in
SO
x
concentrations should be
followed by a spot check of the
analyser with standardised calibra-
tion gases.
Shift to diesel mode operations
The typical shift from gasoline to
diesel mode operations may include
a substantial drop in riser outlet
temperature. This reduction will also
result in a drop in the reactor strip-
per zone temperature. In certain
impact on SO
x
emissions. Regression
techniques are useful for identify-
ing such feeds. This enables feed
forward recommendations to opera-
tions prior to feedstock sulphur
content increases.
Additionally, units operating with
FCC feed hydrotreaters may observe
a gradual increase in SO
x
emissions
as the catalysts age increases, even
though the total sulphur content
may remain constant. This change in
sulphur distribution within the FCC
feedstock is a result of catalyst deac-
tivation leading to a different
distribution of sulphur between coke
and heavy oils. Even the Gulf corre-
lation can fail to predict this effect
adequately. It is important to antici-
pate this trend. These increased
emissions are easily controlled by
adjusting the SO
x
additive injection
rate.
Equilibrium catalyst iron step change
A step change increase in feedstock
iron may have an immediate and
negative impact on SO
x
emissions.
2
There is a direct relationship
between SO
x
emissions and contam-
inant iron concentration on the
equilibrium catalyst (see Figure 3).
Unexpected and unexplained step
800
900
700
600
500
400
300
0.0 0.5 1.0 1.5 2.0 2.5

S
O
x

e
m
i
s
s
i
o
n
s
,

p
p
m
Feed sulphur, wt%
200
Figure 2 Effect of feed sulphur on SO
x
Figure 3 Effect of equilibrium iron on SO
x
200
300
250
150
100
27/5/09 26/6/09 26/7/09 25/8/09

S
O
x
,

p
p
m

E
c
a
t
F
e
,

w
t
%
50
1.05
1.15
1.10
1.00
0.95
0.90
1.30
1.25
1.20
SOx
EcatFe
There is a direct
relationship between
SO
x
emissions and
contaminant iron
concentration on the
equilibrium catalyst
fletcher.indd 2 9/6/11 10:37:06
www.ptqenquiry.com
for further information
Its been called the box
that changed the world
and yet the industry that
transformed the face of
global trade and transport
in the space of just 50 years
now stands on the brink
of another era-defining
moment: whether to
continue with entrenched
habits, or rediscover its
desire and capacit for
innovation and change.
Ready, set, change!
Our industry and our
customers beneft every
day from the time, efort,
money and sheer belief in
the future that went
into building the global
container transport system
as we know it today. When
containers came along, we
had to change everything
and we believe the time has
come to make that change
again.
Maersk Line has identifed
three key challenges that
we believe the whole
container shipping industry
must now address to ensure
that it has a prosperous
future:
on-time delivery,
ease of business and
environmental performance.
We want to start the debate
that will redefne our future
and help us add tremendous
new value to our customers
supply chains worldwide.
We have to become an
integrated part of our
customers promise to
their customers to act as
responsibly as possible in
the marketplace. But why
wait to be asked? Why not
lead the way?
Join the debate on
Changing the way we think about shipping.com
maersk.indd 1 8/6/11 12:37:59
instances, this may lead to insuff-
cient SO
x
additive regeneration,
resulting in an increase in SO
x
emis-
sions (see Figure 4). Applying the
appropriate SO
x
additive technology
can help prevent such excursions.
Changes in CO promoter usage
Platinum-based CO promoters are
also effective at catalysing the
oxidation of SO
2
to SO
3
in the
regenerator. A refnery that is
controlling SO
2
emissions and also
utilising platinum-based CO
promoters may observe an increase
in SO
2
emissions at constant operat-
ing conditions if the injection rate
of the CO promoter is reduced or
stopped.
CO promoter poisoning
CO promoters are poisoned by
122 PTQ Q3 2011 www.eptq.com
antimony and lead. Once poisoned,
the promoters ability to oxidise CO
to CO
2
and SO
2
to SO
3
is reduced.
Reduced capacity to oxidise SO
2
to
SO
3
may result in increased SO
x

emissions at constant conditions.
Regenerator maldistribution
Changes in catalyst and air distri-
bution within the regenerator
combustion zone regularly affect
combustion effciency. On-site test-
ing using a reaction mix sampling
device (RMS) with a portable gas
analyser has revealed that some
apparently stable regenerators can
be operating simultaneously with
oxygen-starved and oxygen-rich
zones. Such maldistribution may
result in increased SO
x
emissions.
The effect will be most strongly
observed when the maldistribution,
as measured by afterburn,
is unstable and shifts from quad-
rant to quadrant within the
regenerator.
An example of maldistribution
resulting from a mechanical change
made to a regenerator has been
observed in one Gulf Coast refner.
This refner was able to achieve
25 ppm SO
2
emissions while using
Super SOxGetter prior to the modi-
fcation. After the revamp, a
degradation in combustion eff-
ciency was observed, measured by
afterburning. This refner was
subsequently required to increase
the additive injection rate by 45%
while achieving a SO
2
emissions
level of only 60 ppm. This example
demonstrates the effect maldistribu-
tion within the regenerator
combustion zone can have on SO
2
emissions.
Increased catalyst losses
Increased catalyst losses from the
regenerator will have a two-fold
negative effect on SO
x
emissions.
First, the SO
x
additive will be lost
with the catalyst, resulting in
reduced effciency. Second, the cata-
lyst average particle size will
increase within the circulating inven-
tory, resulting in poor mixing within
the regenerator bed. Reductions in
mixing effciency can result in
increased SO
x
emissions, as
mentioned in the previous section.
Misapplication of SO
x
additive
technologies
Additional care in the selection of
SO
x
additives is required for units
equipped with CO boilers, which
occasionally shift from partial- to
full-burn operations. Magnesium
oxide for SO
3
absorption is typically
the factor limiting SO
x
additive
performance in full-combustion
operations. Oxygen availability is
the most typical limiting factor in
partial-burn operations. Highly
specialised oxidation packages are,
therefore, used in SO
x
technologies
intended to be used in partial-burn
operations. A standard SO
x
additive
designed for full-combustion opera-
tions will demonstrate a lower
pickup rate in partial-burn operation
when compared to a full-combustion
operation. It is recommended that
600
1000
900
800
700
500
400
300
200
100
29/9/08 28/11/08 27/1/09 28/3/09
s
n
o
i
s
s
i
m
e

x
O
S
0
Riser
SOx
985
1005
1000
995
990
980
975
970
965
960
e
r
u
t
a
r
e
p
m
e
t

t
e
l
t
u
o

r
e
s
i
R
955
Figure 4 Effect of riser temperature on SO
x
320
370
270
220
170
120
70
6.2 6.4 6.6 6.8 7.0 7.2 7.4

S
O
x
,

p
p
m
Cat-to-oil, wt/wt
Figure 5 Effect of cat-to-oil on SO
x
emissions
fletcher.indd 3 9/6/11 10:37:17
FCC units today are not yet regu-
lated with respect to NO
x
emissions.
A FCC process engineer is well
advised to monitor and understand
the units response to operating
changes and how they relate to NO
x

emissions. NO
x
emissions can also
provide an excellent, albeit second-
ary, indication of the degree of
maldistribution present within the
regenerator. NO
x
emissions tend to
rise with increasing maldistribution.
Furthermore, a solid baseline will
prove invaluable when the regula-
tory agencies begin to discuss NO
x
emissions.
Flue gas excess oxygen
concentration
The single most important variable
with respect to controlling NO
x
emissions is fue gas excess oxygen
(see Figure 7). As the oxygen
www.eptq.com PTQ Q3 2011 123
the refner injects appropriate SO
x

additive technologies when switch-
ing between full-combustion and
partial-burn operations.
Reduced catalyst circulation rate
SO
x
reduction additives are tri-
functional catalysts. The frst two
steps in the catalytic process include
oxidation of SO
2
to SO
3
, followed
by absorption of the SO
3
by the
additive as magnesium sulphate.
The additive is then carried into a
reducing atmosphere within the
riser and reactor vessel, in which
the sulphate is converted to H
2
S.
The magnesium sulphate is
converted back to magnesium oxide
and is available for subsequent
absorption reactions in the regener-
ator. Any reduction in the catalyst
circulation rate will increase the
residence time of the additive in the
regenerator, thus directionally
reducing the amount of regenerated
additive available for further reac-
tion (see Figure 5). Conversely,
increasing the catalyst circulation
rate increases the rate of regenera-
tion, thus improving additive
effciency.
Reduced reactor stripper effciency
A reduction in stripper effciency
due to insuffcient steam or internal
damage results in an increase in the
heavy oil carried into the regenera-
tor vessel. This results in increased
sulphur loading to the regenerator,
since sulphur tends to distribute
itself into the heaviest oil cuts. An
increased hydrogen-on-coke ratio
can, therefore, be associated with
increased SO
x
emissions. Regular
monitoring of stripper effciency
and stripping steam rates allows
early detection of any changes in
this area.
Optimisation guidelines for
maximum SO
x
additive effciency
SO
2
and SO
3
gases are in equilib-
rium within the regenerator
atmosphere in most full-combustion
operations. The relative composi-
tions are typically at an
approximately 90:10 SO
2
-to-SO
3
ratio. SO
x
additives function by frst
shifting SO
2
to SO
3
. The additive
then quickly absorbs the SO
3
to
form a metal sulphate followed by
H
2
S release in the reducing atmos-
phere of the riser and reactor
stripper. Therefore, optimising the
FCC regenerator conditions for
maximum additive effectiveness is
a function of shifting the SO
x
equi-
librium to SO
3
. This equilibrium is
shifted in the direction of SO
3
under
the following conditions:
Increased fue gas excess oxygen
concentration (up to approximately
2 wt%; see Figure 6)
Lower regenerator temperatures
Increased regenerator pressure
Increased catalyst circulation rate
(accelerates additive regeneration)
Use of platinum-based combus-
tion promoters.
NO
x
emissions control
NO
x
emissions from the FCC unit
have only recently begun to be scru-
tinised by regulatory bodies. Many
60
70
50
40
30
20
0.8 1.2 1.6 2.0 2.4 2.8
S
O
x
,

p
p
m
Excess oxygen, %
10
Figure 6 Effect of excess oxygen on SO
x
140
160
120
100
80
60
1.0 1.5 2.0 2.5
N
O
x
,

p
p
m
Excess oxygen, %
40
Figure 7 Effect of excess oxygen on NO
x
fletcher.indd 4 9/6/11 10:37:32
Selas Fluid has supplied innovative heater and furnace technologies to reneries and
petrochemical plants in the Middle East and worldwide for more than six decades.
As an approved supplier of renery heaters, petrochemical furnaces and incineration technologies to the major oil and
gas companies in the Middle East, Selas Fluid is prepared to surpass your expectations. We have built a history of proven
results, with thousands of installations, by providing continuous innovation, quality products and reliable customer
support.
Selas Fluid is a full-service partner to the rening, petrochemical, and chemical industries worldwide.
Equipment supply
Process design and engineering
Modular construction
Erection services

Revamp, retrot and upgrade


Consultation and feasibility studies
Start-up and operator training
Spare parts and technical services

Selas Fluid
Subsidiary of The Linde Group
Results
Seven (7) ethane cracking furnaces supplied by Selas Fluid to Saudi Ethylene and Polyethylene Company (SEPC) on an
EPC basis. Completed ahead of schedule and contributed over 1 million accident-free man hours to the entire project.
Headquarters:
Five Sentry Parkway East
Blue Bell, PA 19422 USA
Telephone: 610-834-0300
Texas Ofce:
16225 Park Ten Place
Suite 250
Houston, TX 77084 USA
Telephone: 281-717-9090
Al-Khobar:
Linde Arabian Contracting
Telephone: +966-3-887-0133
www.selasuid.com
www.ptqenquiry.com
for further information
selas.indd 1 8/6/11 12:39:31
concentration increases, the NO
x
concentration also increases. It is,
therefore, recommended that the
FCC unit operates at minimum
excess oxygen concentrations,
consistent with stable combustion
patterns within the regenerator.
Regenerator maldistribution
Changes in the catalyst and air
distribution within the regenerator
combustion zone can have a greater
effect on NO
x
emissions than on
SO
x
emissions. Segments of the
dense bed operating in partial burn
will produce signifcantly higher
concentrations of ammonia and
hydrogen cyanide gases. These
gases will then be oxidised prima-
rily to NO
x
in the fue gas line or
the CO boiler in partial-combustion
operations. Increases in NO
x
emis-
sions while operating at constant
conditions can be a strong indica-
tion of changes in the catalyst and
air distribution. This is especially
true if there is a simultaneous
increase in NO
x
and particulate
emissions in a normally well-
behaved unit.
One European refner experienced
a step change increase in NO
x
emis-
sions after suffering a mechanical
failure in the regenerator. Figures 8
and 9 demonstrate a technique that
is frequently effective in identifying
which quadrant of the regenerator
is experiencing maldistribution. The
technique is to plot NO
x
emissions
versus differential cyclone outlet
temperatures between each cyclone
set. An examination of these plots
provides the process engineer with
an indication of which cyclone set
is affecting NO
x
emissions the most.
Multivariable linear regressions of
NO
x
emissions and the primary
independent variables is a second
useful technique available to the
process engineer. In this example,
the maldistribution appears in the
quadrant of the regenerator contain-
ing the second set of cyclones. This
observation was also confrmed
through multivariable linear
regressions.
Regression analysis is a valuable
tool to investigate the impact of
maldistribution, as evidenced by
after-burn, on NO
x
emissions. The
process engineer is advised to
www.eptq.com PTQ Q3 2011 125
promoters is highly effective at
oxidising nitrogen compounds
found in the coke on spent catalyst
to form NO
x
. Replacement of a
platinum-based CO promoter to a
non-platinum-based system will in
many units result in a 5070%
reduction in NO
x
emissions (see
Figure 10). However, the baseline
NO
x
emissions level will have a
strong effect on the actual reduc-
tions observed. A unit with very
good regenerator distribution and
NO
x
emissions of 2025 ppm will
experience little impact following a
change to a non-platinum promoter.
Conversely, a unit with less ideal
mixing and NO
x
emissions of
150200 ppm will likely observe
reductions of 5070%.
It is worth noting that (more
than) one refner, after switching
from a platinum-based to a non-
consider employing different time-
frames with this technique.
Regression of daily averaged oper-
ating data against NO
x
emissions is
valuable if daily feed nitrogen anal-
yses are available. Hourly averaged
operating data regressed against
NO
x
emissions is valuable for eval-
uating operating effects. Utilising
timeframes shorter than hourly
averages will frequently lead to
inconclusive or inconsistent results.
Correlations developed on operat-
ing periods without NO
x
additives
or non-platinum CO promoters
cannot be used with periods of
operations incorporating such addi-
tives. New correlations must be
derived for each period.
Use of platinum-based CO
promoters
The platinum found in many CO
300
500
400
200
100
5 3 1 1 3 5 7

N
O
x
,

p
p
m
Differential temp, Set 1 Set 2
0
Normal
Maldistribution
Figure 8 Effect of maldistribution on NO
x
300
500
400
200
100
-2 -1 0 1 2 3
m
p
p

,
x
O
N

Differential temp, Set 2 Set 3
0
Norma|
Ma|distribution
Figure 9 Effect of maldistribution on NO
x
fletcher.indd 5 9/6/11 10:37:43
platinum-based promoter, found,
much to their surprise, an old bin
of the platinum-based promoter
hidden within their storehouse.
Without realising that this additive
contained platinum, this bin was
inadvertently delivered for injection
into the unit, resulting in a large
increase in NO
x
emissions.
Verifcation that no platinum-based
promoter remains in storage is
recommended after switching to
a non-platinum-based promoter
system.
Change in feedstock nitrogen
concentration
The source of nitrogen for NO
x
formation within the FCC unit is
nitrogen in the feedstock. NO
x

emissions may increase as the feed-
stocks nitrogen content increases.
Many units have observed that the
NO
x
emissions should be verifed
by the FCC process engineer.
Changes in NO
x
additive addition
rates
NO
x
additives have an optimum
addition rate. Overdosing or under-
dosing NO
x
additive additions can
result in increased NO
x
emissions.
It is recommended that the process
engineer monitors this concentra-
tion and ensures that it does not
move outside the effective range of
the unit.
Non-platinum promoters have
been proven to reduce NO
x
emis-
sions in most regenerators.
However, it is not so commonly
understood that over-promotion,
even with non-platinum CO
promoters, can actually increase
emissions. Careful monitoring of
injections is therefore strongly
recommended, especially when
switching from one non-platinum
promoter to another.
Operating in deeper partial burn
In partial-burn regenerators, an
increase in the CO-to-CO
2
ratio will
result in additional nitrogen in the
spent coke being converted to
ammonia and hydrogen cyanide
gases. These gases will then be
oxidised primarily to NO
x
in the
CO boiler (see Figure 11).
Unfortunately, current NO
x
addi-
tive technologies can do little to
mitigate this effect.
Increased catalyst losses
An increased catalyst losses rate
will have a negative impact on NO
x

additives. This impact has been
discussed in the section related to
SO
x
emissions.
Flue gas analyser failure
Regular preventative maintenance
with special attention paid to the
sample preparation system is
crucial.
Optimisation guidelines for reduced
NO
x
emissions
Operating the full-combustion
regenerator with the lowest stable
excess oxygen concentration in the
fue gas directly reduces NO
x
emis-
sions. Operating the partial-burn
regenerator with the lowest stable
126 PTQ Q3 2011 www.eptq.com
relationship between feed nitrogen
concentration and NO
x
emissions is
not strong. However, it is recom-
mended that the FCC engineer
determines the relationship between
feedstock nitrogen and NO
x
emis-
sions for the unit. Operations may
then be alerted whenever large
changes are expected. This relation-
ship may be determined using
multivariable linear regression. A
warning to operations will then be
possible prior to a change in feed
slate.
Use of antimony for nickel
passivation
While not well understood, the use
of antimony for passivation of
nickel on equilibrium catalyst has
been shown to signifcantly increase
NO
x
emissions in many, but not all,
units. The impact of antimony on
300
250
200
150
100
50

N
O
x
,

p
p
m
0
CDP-850
CDP-NP
Figure 10 Effect of non-platinum promoter on NO
x
250
200
150
100
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

N
O
x

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

p
p
m
CO concentration, mol%
50
Figure 11 Effect of partial burn on NO
x
fletcher.indd 6 9/6/11 10:37:53
C
M
Y
CM
MY
CY
CMY
K
KTI-SCRad-PTQ2011-cmyk.pdf 5/31/11 9:38:09 AM
www.ptqenquiry.com
for further information
kti.indd 1 8/6/11 12:40:54
CO-to-CO
2
ratio also directly
reduces NO
x
emissions.
Additionally, avoiding the use of
platinum-based CO promoters in
favour of non-platinum promoters
reduces NO
x
emissions, in most
units, by 5070%. Carefully monitor
and control antimony injection for
the passivation of equilibrium
nickel. In order to minimise anti-
mony injections, consider applying
catalyst technologies specically
designed for nickel passivation.
CO emissions and afterburn control
A well-operated FCC unit operating
in partial burn will have sufcient
capacity to handle normal varia-
tions in CO concentration. The
guidelines provided will, therefore,
focus on CO emissions from the
full-combustion regenerator.
FCC unit startups
The most commonly observed
instance of CO emissions on the
FCC unit is during startups.
Occasionally, units get behind in
coke combustion, leading to CO
emissions to the atmosphere. Many
reners practise the addition of CO
promoter during startups as a
mandatory precautionary procedure
to prevent CO excursions.
Regenerator maldistribution
Poor regenerator mixing has led to
some regenerators emitting CO to
the atmosphere. In some of these
cases, injection of a CO promoter
has been sufcient to control CO
emissions.
Figure 12 shows the conguration
of a North American renery,
whose FCC unit suffered from
maldistribution of spent catalyst.
4
While considering a mechanical
modication to the spent catalyst
distributor, the rener initiated a
combustion gas analysis using a
RMS in conjunction with a portable
gas analyser. Table 1 shows that
CO-enriched gases were emitted
from the dense bed in the northern
half of the regenerator, while the
gases emitted from the southern
half contained excess oxygen. The
gas analyser CO reading on the
northern half of the regenerator
was off scale (over 10 000 ppm).
These results clearly demonstrate
the effects of maldistribution on CO
production in a full-combustion
regenerator.
The same rener cited in the
maldistribution case for SO
x
emis-
sions also observed an increase in
CO emissions after having made
mechanical changes to the regenera-
tor. After startup, it was
immediately observed that the
historical addition rate of CO
promoter had to be increased in
order to control CO emissions and
after burn. This rener observed
step changes in the concentrations
of CO, SO
x
and NO
x
emissions with
corresponding increases in required
additive injection rates after their
turnaround.
A second rener recently
conducted a CO emissions survey
on their ue gas stack using a port-
able gas analyser. It was observed
over four days of testing that the
CO concentration in the ue gas
was typically between 0 and 15
ppm. However, two unexplained
spikes were observed, of 405 ppm
lasting for ve minutes and 80 ppm
lasting for 20 minutes. It is interest-
ing to note that the NO
x
emissions
were observed to increase 43%
128 PTQ Q3 2011 www.eptq.com
Table 1
O
2
, vol% CO, ppm SO
2
, ppm NO
x
, ppm
Position 1 0.1 >10 000 482 17
Position 2 1.4 4276 127 51
Position 3 5.1 23 4 86
Regenerator gas concentrations
9
1
/
3
"
C
L
4' 11
7
/
16
"
8
7
/
8
"
Cata|yst
withdrawa|
Modied dip|eg
with offset
Modied dip|eg
with offset
5' 5
1
/
16
"
Regenerator
CP
CD
CMII
CMI
CR
CPII
CG
COI
P1
CA
CJI
CN
CE
CC
COII
CL
CB
20
0
5
22
62
125
135
140
150
No. 2
172
180
190
222
245
260
330
352
322
No. 1
No. 3
CF
N
Figure 12 Regenerator conguration
fletcher.indd 7 9/6/11 10:38:11
during the CO spike reaching 405
ppm. These observations taken
from a well-behaved regenerator
clearly indicate that transient condi-
tions can result in unexpectedly
high emissions levels.
Low feed concarbon concentration
A reduction in feed concarbon
concentration or processing a deeply
hydrotreated feedstock while oper-
ating the regenerator dense bed at
low temperatures can lead to CO
breakthrough from the regenerator.
This is the result of insufcient resi-
dence time for combustion gas at the
low temperatures within the dense
bed to achieve complete combustion.
This situation can be corrected by
increasing the dense bed tempera-
ture, increasing the bed depth and/or
through the addition of CO
promoter.
Conclusions
Careful attention to the FCC regen-
erator operation enables the process
engineer to maximise regenerator
combustion efciency and minimise
the probability of unexpected ue
gas emissions. Furthermore, the time
required to achieve an optimal oper-
ation is minimal when carried out as
a part of normal unit monitoring.
In the event of an unexpected
increase in ue gas emissions, care-
ful analysis using the guidelines
provided should enable the FCC
engineer to swiftly identify and
eliminate the root cause of these
emissions.
References
1 Huling G P, McKinney J D, Sulphur distribution
in high conversion riser catalytic cracking,
Gulf Research & Development Company; and
Readal T C, Gulf Oil Company Asia, Pittsburgh,
Pennsylvania.
2 Diddams P, Reagan B, Evans M, Aru G,
Feed iron effects on FCC SO
x
emissions, EPA
Consenters Team Meeting, Phoenix, Mar 2010.
3 Ferguson P V, Evans M E, State-of-the-art
catalyst addition systems provide accuracy,
reliability and operational stability for the
FCCU.
4 Fletcher R P, Evans M E, Optimizing &
troubleshooting the FCC regenerator for
reduced emissions, NPRA Annual Meeting, AM-
10-173.
Ray Fletcher is an Engineering Fellow with
Intercat, Manasquan, New Jersey, USA. With
a background as a process engineer working
on renery process units, and as a technical
service manager and FCC development
manager, he holds a bachelors degree in
chemical engineering from the University of
Washington.
Martin Evans is Vice President of Engineering
with Intercat, responsible for the companys
technical assistance on the use of FCC catalyst
additives and for the design and development
of catalyst addition technology. A Chartered
Chemical Engineer in the UK, he has a BSc in
chemical engineering from the University of
Wales and holds 19 patents for FCC catalyst
addition systems.
The time required to
achieve an optimal
operation is minimal
when carried out as
a part of normal unit
monitoring
www.eptq.com PTQ Q3 2011 129
www.ptqenquiry.com
for further information
3225 Gallows Road, Fairfax, Virginia 22037-0001, USA
+1-703-846-2568 fax +1-703-846-3872
www.exxonmobil.com/tsl tsl.email@exxonmobil.com
Advanced Solutions
for the Worlds
Toughest Energy
Challenges
TECHNOLOGI ES:
ExxonMobil Global Leader
in Fuels and Lubes
Process Technologies
ExxonMobil Technologies are applied across
our corporation and by licensees worldwide in
a growing list of process industry applications.
Take advantage of our vast experience and
ongoing commitment to continuously improve
our industry leading technologies.
fletcher.indd 8 9/6/11 10:38:25
11-13 October 2011,
GDANSK/SOPOT, POLAND
www.wraconferences.com/cee
YOUR INVITATION TO THE
14TH ANNUAL MEETING:
Where key leaders meet and shape the future
of the industry
KEY SPEAKERS:
Grupa Lotos, Vice-President Management Board, COO,
Marek Pawel Sokolowski
PKN Orlen S.A., Vice President and CFO, Slawomir Jedrzejczyk
NIS, CEO and JSC Gazprom Neft Deputy General Director
For Overseas Assets Management, Kirill Kravchenko
Europia, Deputy Secretary General, Chris Beddoes
MOL Group, Director, Downstream Development, Artur Thernesz
CESKA Ranerska, Chairman and CEO, Ivan Soucek
OMV/Petrom, Operations Manager, Petrobrazi Renery,
Bertram Muchan
CENTRAL AND EASTERN
EUROPEAN REFINING
AND PETROCHEMICALS
14
th
Annual Meeting
CEE
REFINING AND
PETROCHEMICALS
11-12 October
DOWNSTREAM
FINANCING
CHALLENGES
13 October AM
GDANSK
REFINERY VISIT
13 October PM
2011 Sponsors and Exhibitors: Ofcial Partner:
For further information please contact Jivko Gadjourov,
T: +44(0)207 067 1800 or E-mail:marketing@theenergyexchange.co.uk
wra.indd 1 8/6/11 12:44:09
Catalytic oxidation of spent caustic
C
austic has been a mainstay of
the oil refning industry since
its earliest days because of its
effectiveness in scrubbing organic,
inorganic and naphthenic acidic
components, especially mercaptans,
H
2
S and phenols, from crude oil or
its fractions.
1-3
Once it has been used
in the extraction of these acid
components, the resulting dilute
caustic stream becomes spent caus-
tic. The generation of spent caustic
is continually increasing as the
processing of sour crude oil in refn-
eries worldwide multiplies, and
levels of generation are expected to
rise signifcantly in the future. Fresh
caustics effectiveness and low cost
are the reasons for its widespread
use. But ineffcient use of alkali
resources, followed by improper
handling and inadequate disposal,
has a growing effect on refnery
margins.
Refnery spent caustic is recog-
nised as a hazardous stream in
view of a range of characteristics: it
is corrosive (high pH); it contains
toxic compounds such as H
2
S or
phenol; and it imposes high levels
of biological and chemical oxygen
demand (BOD and COD) on natu-
ral water supplies.
4,5
It also contains
sulphides and mercaptans, which
give rise to fouling or metallurgical
damage in refnery equipment.
Furthermore, there is no available
route for converting spent caustic
into a valuable product. Hence, it
has the least potential for reuse
within the refnery. Even worse, it
is discarded at high cost.
Usually, a caustic stream extracts
H
2
S, mercaptans and sulphides, and
generates sodium salts, including
sodium sulphide and sodium
A process to oxidise spent caustic in an existing treatment plant by a catalytic
route in the presence of air enhances the quality of wastewater releases
ViVek RAthoRe, ShAlini GuptA, t S thoRAt, p V C RAo, n V ChoudARy and G Biju
Bharat Petroleum Corporation Limited, India
bisulphide. The disposal of these
compounds in the open environ-
ment releases H
2
S, which has a
strongly objectionable odour, so
environmental agencies around the
world have tightened regulations
aimed at controlling its disposal.
Adequate effuent treatment proce-
dures are therefore required for its
safe disposal, but commercially
available disposal methods (for
instance, oxidative detoxifcation
using peroxide or ozonolysis) are
costly and not environmentally
benefcial. Consequently, there is a
pressing need to develop an
alternate, low-cost and environmen-
tal-friendly process for converting
sulphides and mercaptans into
water-soluble sulphate salts.
Oxidative detoxifcation of spent
caustic and demercaptisation can be
performed in the presence of molec-
ular oxygen, but the rate of oxidation
is found to be very low in the
absence of a catalyst. However, it is
also known that mercaptans and
sulphides can be oxidised with air/
oxygen in the presence of metal
phthalocyanines used as catalysts.
Phthalocyanines of metals, including
cobalt, iron, manganese, molybde-
num and vanadium, catalyse the
oxidation reactions in an alkaline
medium.
6
Currently, metal phthalo-
cyanine derivatives are used as
homogeneous catalysts for liquid-
liquid sweetening and alkali
regeneration in the extraction of
mercaptans from light petroleum
distillates, including LPG, pentanes,
light straight-run naphtha (LSRN)
and light thermally cracked naphtha.
They are also used as heterogeneous
catalysts for the sweetening of other
petroleum products such as heavy
naphtha, FCC gasoline, ATF and
kerosene. These catalysts are
prepared by adsorptive impregna-
tion of phthalocyanine solutions on
www.eptq.com PTQ Q3 2011 131
Gasoline caustic wash /
MEROX
(extraction/sweetening)
LPG caustic wash /
MEROX
(extraction)
Sweet gasoline
Fresh
caustic
solution
Sour gasoline
Spent caustic
(sulphide)
Spent caustic
(sulphide)
Spent caustic tank
Spent
caustic
Fresh
caustic
solution
Sweet LPG
Sour LPG
Draining of
bottom
gasoline tank
Figure 1 Generation of refnery spent caustic streams
rathore bpcl.indd 1 9/6/11 10:43:16
132 PTQ Q3 2011 www.eptq.com
loaded into a reactor column system
operated at 5090C and atmos-
pheric pressure. The spent caustic
was pumped into the system by a
metering pump. The spent caustic
stream was characterised for
sulphide and total mercaptans
content by the ASTM method
before being pumped into the reac-
tor system.
Metal phthalocyanine catalysts
It is well known that the phthalocy-
anines of metals, including cobalt,
nickel, iron, copper, molybdenum
and vanadium, catalyse the oxida-
tion of sulphides and mercaptans in
an alkaline medium. However,
metal phthalocyanines are not solu-
ble in an aqueous medium, so their
respective N-substituted derivates
were prepared, as explained in the
available literature.
7,8
To compare
and understand the effectiveness of
various metal phthalocyanine
derivatives as catalysts, experiments
were performed at the same cata-
lyst dosing (200 wppm) under
similar reaction conditions of
temperature (60C) and air (50%
excess).
It was found that cobalt phthalo-
cyanine derivative works as an
effective catalyst for both homoge-
neous and fxed-bed oxidation of
sulphidic spent caustic solution. A
level of sulphide conversion of
more than 30% was achieved in the
presence of a cobalt phthalocyanine
derivative (see Figure 2). The deriv-
ative was therefore considered as
an active catalyst for a study to
optimise reaction parameters.
Effect of catalyst loading
The effect of catalyst loading
(01000 wppm) was studied at a
constant temperature of 60C and in
porous supports (activated
carbon, alumina, bauxites, silica gel,
aluminium and magnesium
oxides).
8,9
The use of homogeneous
catalysts in sulphide oxidation proc-
esses is economically feasible due
to their low cost and ease of
availability.
In the present work, a novel proc-
ess has been developed, based on
various laboratory trials, for the
oxidation of sulphidic streams in
the presence of metal phthalocya-
nine compounds, which act as a
catalyst in the presence of air.
Additionally, the effect of various
process parameters, including
temperature, pressure, catalyst
loading, air fow rates and reaction
times, has been investigated for
process optimisation and more eff-
cient operation. The performance of
a refnery feld trial for the new
process is also discussed.
Spent caustic stream
The studies involved a spent caustic
stream from BPCLs Kochi refnery
in India. Physical chemical proper-
ties and other characteristics were
measured in the laboratory accord-
ing to ASTM methods. The
composition of spent caustic
streams is highly variable and can
be categorised into three groups,
depending on origin and composi-
tion (see Table 1 and Figure 1).
Typical spent caustic effuent
contains about 412% of sodium
hydroxide (NaOH) by weight,
although this could be as low as 1%
or 2%. The composition in a typical
refnery spent caustic stream is
based on sulphides (0.54% as
sulphur) and mercaptides (0.14%),
but naphthenic spent caustic
contains mostly naphthenic acids (as
high as 15%, mostly in diesel) and
minimal sulphide. This stream has a
tendency to foam vigorously and
creates additional problems if not
treated properly. Finally, a cresylic
spent caustic stream is rich in
phenols, cresols and other organic
acids. Phenol content can be as high
as 2000 ppm. The spent caustic
stream contains sulphides, mercap-
tides and other contaminants (see
Table 2, which characterises a typi-
cal spent caustic stream).
Laboratory studies
During the laboratory studies, a
calculated amount of catalyst was
Type Contaminants Source
Sulphidic Sulphidesandmercaptans EthelynefromLPGscrubbing
Cresylic Phenolsandsulphur Gasolinescrubbing
Napthenic Naptheniccompoundsandsulphur Keroseneanddieselscrubbing
Type and generation route for refnery spent caustic stream
Table 1
Contaminant Content
Sulphide,% 0.10.4
Mercaptide,% 0.10.4
Phenols,ppm 02000
TOC,ppm 600020000
COD,ppm 2000060000
BOD,ppm 5000015000
Characterisation of a typical spent
caustic stream
Table 2
30
35
25
20
15
10
5
0 ZnPc/D FePc/D CoPc/D AIPc/D
%

,
n
o
i
s
r
e
v
n
o
C
Metal phthalocyanine derivatives
0
Figure 2 Comparativeperformancesofsupportedmetalphthalocyaninesintheoxidation
ofsulphidicspentcaustic
rathore bpcl.indd 2 9/6/11 10:43:30
a continuous supply of air (50%
excess of the stoichiometric require-
ment). Two different concentrations
of sulphidic spent streams, named
sulphidic spent streams (SSCS) 1
and 2, in which the sulphide concen-
trations were 13 000 ppm and
21 000 ppm, respectively, were used
in the study. The results (see Figure
3) showed that at 400 wppm cata-
lyst loading, the sulphide conversion
level was 53% and 51% for SSCS 1
and 2, respectively. Thereafter, the
conversion level did not improve
signifcantly with the addition of
further catalyst to either of the
sulphidic streams. Hence, the effect
of catalyst charging on sulphide
conversion was insignifcant beyond
the 400500 wppm range. This was
taken as the optimum and preferred
catalyst dosing range for both of the
sulphide streams.
Effect of temperature and pressure
To investigate the effect of tempera-
ture on reaction, experiments were
performed with a continuous supply
of air (50% excess of stoichiometry
requirement) at 400 ppmw catalyst
loading of cobalt phthalocyanine
derivative. The temperature was
varied from 3090C for both SSCS 1
and 2. It was found that the sulphide
conversion level rises from 1374%
(approximate) by increasing the
reaction temperature from 3080C
for both the sulphidic streams (see
Figure 4). Reaction dependence on
temperature was high initially and
the rate of oxidation also increased
with temperature. It was also noted
that high temperature favours the
conversion of sulphidic compounds
into benign sulphates. The preferred
reaction temperature range for
oxidation is 7090C, depending on
the characteristics of sulphides and
other contaminants in the spent
caustic stream.
Oxidation of a spent caustic
stream containing sulphide,
mercaptans and other contaminants
such as phenols also proceeds
within a preferred pressure range
of 115 kg/cm
2
.
9-11
Effect of excess air supply and
reaction time
The oxidation of sulphidic spent
caustic in the presence of cobalt
www.eptq.com PTQ Q3 2011 133
phthalocyanine derivative (as a
catalyst) is affected by air, oxygen
or any other oxygen-containing gas,
air being preferred as an economi-
cal supply of oxidant. The chemistry
of oxidation involves free radical
formation, with the oxygen and/or
water-derived radicals attacking the
organic compounds and encourag-
ing the formation of organic
radicals. Experiments were
performed at a constant 80C
temperature with a 400 ppmw cata-
lyst loading of cobalt phthalocyanine
derivative in the presence of a
continuous supply of air ranging
from stoichiometry to 100% excess.
The results showed that 30%
(approximate) conversion was
achieved with a stoichiometric
supply of air, and increased to 75%
at an air supply ratio of 1.5 of stoi-
chiometry for both SSCS 1 and 2,
(see Figure 5). Hence, the reaction
rate was found to be much higher
with 50% excess air supply
compared to a stoichiometric level.
Beyond 50% excess air, there is no
signifcant effect on the conversion
of sulphides.
It was also observed that sulphide
conversion varies with time, and
depends on sulphide content and
the concentration of other contami-
nants present in the spent caustic
streams, with a constant supply of
air during the progress of the oxida-
tion reaction.
11-14
Refnery feld trial
A feld trial of the process was
carried out at the sweetening unit
of BPCLs Kochi refnery. Some
modifcations were also made to
the existing unit for handling spent
sulphidic streams. Normally, the
refnery generates 58 m
3
/d of
spent caustic from various units,
60
50
40
30
20
10
0 200 400 600 800 1000 1200
C
o
n
v
e
r
s
i
o
n
,

%
Catalyst, ppm wt
0
SSCS-2
SSCS-1
60
50
40
90
80
70
30
20
10
20 30 40 50 60 70 80 90 100
C
o
n
v
e
r
s
i
o
n
,

%
Temperature, C
0
SSCS-2
SSCS-1
Figure 3 Effect of catalyst loading at constant temperature (60C) and continuous supply
of air (50% excess)
Figure 4 Effect of temperature in the presence of catalyst (400 ppmw) and a continuous
supply of air (50% excess of stoichiometry requirement)
rathore bpcl.indd 3 9/6/11 10:43:43
level of sulphide releases below 100
ppm for meeting the environmental
regulations.
Conclusions
The prospect of oxidising a spent
caustic stream in an existing sweet-
ening plant by a catalytic route in
the presence of air opens a new
window for refners to bring fexi-
bility to their existing unit
operations and processes to treat all
spent caustics produced at the
refnery. The following conclusions
can be drawn from the present
study:
Oxidation of spent caustic
containing sulphides, mercaptans
and phenols can be performed in
the presence of metal phthalocya-
nine catalysts in the presence of air.
A water-soluble derivative of cobalt
phthalocyanine was found to be the
most effective, compared to the
phthalocyanines of Fe, Zn and Al
Sulphidic and hazardous organic
compounds in wastewater, such as
mercaptans, and naphthenic and
cresylic compounds, are destroyed
or converted to benign molecules.
Odour problems due to the libera-
tion of H
2
S from sulphides
following inadequate disposal are
eliminated. Additionally, the pH of
the effuent is close to neutral (~8).
Therefore, reactor effuent can be
directly discharged to on-site
biological facilities for fnal polish-
ing. This eliminates stockpiles of
spent caustic, reduces environmen-
tal and health hazards, and
improves environmental conditions
in the refnery
The process treats sulphidic spent
caustic in a safe, economically
viable and environmental-friendly
manner, even though operating
conditions for the process are less
stringent (temperatures <100C)
compared to commercially available
methods such as wet air oxidation,
peroxide treatment and ozonolysis
In addition, the performance
of the process is found to be
excellent, with lower catalyst
consumption and minimal energy
utilisation than commercially availa-
ble methods/processes
Initial feld trials are promising
and showed more than 85% conver-
sion for sulphidic compounds
134 PTQ Q3 2011 www.eptq.com
where the sulphide concentration
varies from 30 00040 000 ppm and
other contaminants include
mercaptans and phenols. Water was
used as the diluent for the concen-
trated sulphidic spent caustic
stream. One batch of spent caustic
(78 m
3
/h) with a sulphide concen-
tration ranging from 85009000
ppm (following dilution with water)
was used for the trial run. For ease
of operation, the entire batch was
treated in recirculation mode, until
sulphide concentrations reduced to
an optimum level. The fow
diagram for the process is shown in
Figure 6.
Catalyst concentration and air
fow rate were maintained at a
level determined by the reaction
stoichiometry of sulphide oxidation.
Other parameters of the process
such as pressure, temperature and
caustic circulation rate were main-
tained at 23 kg/cm
2
(g), 5070C
and 0.5 m
3
/hr, respectively. It was
found that the sulphide concentra-
tion fell from an initial level of 8500
ppm to 1200 ppm (85% conversion)
in less than 10 hours. Furthermore,
it was observed that the catalyst
and energy requirements were
minimal compared to a commer-
cially available sulphide oxidation
process such as wet oxidation,
peroxide treatment or ozonolysis.
Based on these plant trials, new sets
of experiments have been designed
for further improvement in the
sulphide conversion, to reduce the
DM/BFW
water
Catalyst
2-3 kg/cm
2
(G)
50-70C
To heater
Air
Spent
caustic
Caustic
heater
Distributor
Spent
caustic
storage
Anthracite coal mesh
Treated caustic
to ETP/biological
treatment plant
Figure 6 Process fow diagram for the for treatment of refnery spent caustic
60
50
40
90
80
70
30
20
10
1.0 1.2 1.4 1.6 1.8 2.0
C
o
n
v
e
r
s
i
o
n
,

%
Air supply ratio, actual/stoichiometry
0
SSCS-2
SSCS-1
Figure 5 Effect of air fow rate in presence of catalyst (400 ppmw) at a constant
temperature (80C)
rathore bpcl.indd 4 9/6/11 10:43:58
present in the spent caustic.
Experiments and feld trials are
planned for further improvement to
the level of sulphide conversion,
with the aim of reducing sulphide
levels below 100 ppm to meet envi-
ronmental regulations.
Acknowledgement
The authors would like to acknowledge BPCL
management for their constant support,
encouragement and permission to publish this
work.
References
1 Walter D M, US patent 4 088 569, 1978.
2 Wallace T, Schriescheim A, Hurwitz H,
Glaster M, Ind. Eng. Chem. Process Des. Dev. 3,
1964, 237.
3 Holbrook D L, Meyers R A (Ed), Handbook
of Petroleum Refning Processes, 2nd ed, 11.3,
McGraw Hill, New York, 1997.
4 Basu B, Satapathy S, Bhatnagar A K, Catal.
Rev.-Sci. Eng. 35, 1993, 571.
5 Wallace T J, Schriescheim A, Hurwitz H,
Glaser M B, Ind. Eng. Chem. Prod. Res. Dev. 3,
1964, 237.
6 Zwart J, Vander Weide H C, Broker N,
Rummens C, Schuit G C A, German A L, J. Mol.
Catal. 3, 1977, 151.
7 Sain B, Puri S N, Das G, Balodi B P, Kumar S,
Kumar A, Kapoor V K, Bhatia V K, Prasada Rao T
S R, Rai G P, US 6740619 B1, 2004.
8 Sain B, Puri S N, Das G, Balodi B P, Kumar S,
Kumar A, Kapoor V K, Bhatia V K, Prasada Rao T
S R, Rai G P, US 6565740 B2, 2003.
9 Brouwer W M, Piet P, German A L, J. Mol.
Catal. 29, 1985, 335.
10 Ansquer P, Orgebin J M, US patent 5 382
354, 1995.
11 Christian M, Philippe C, FR patent 2 688
223, 1993.
12 Di Cosimo J I, Dez V K, Xu M, Iglesia E,
Apestegua C R, J. Catal 178, 1998, 499 and
references within.
13 Yamaguchi K, Ebitani K, Yoshida T, Yoshida H,
Kaneda K, J. Am. Chem. Soc. 121, 1999, 4526.
14 Alcaraz J J, Arena B J, Gillespie R D, Holmgren
J S, Catal. Today 43, 1998, 89.
Vivek Rathore is a Senior Research Engineer
(R&D) with Bharat Petroleum Corporation Ltd,
India, specialising in resid upgrading technology,
LPG sweetening catalysts, biofuels, glycerol
valorisation and reaction modelling. He holds
a masters degree in chemical engineering from
Indian Institute of Science, Bangalore, India.
Email: vivekrathore@bharatpetroleum.in
Shalini Gupta is a Senior Research Scientist
with Bharat Petroleum, specialising in analytical
chemistry, new product development and
biolubricants. She holds a masters degree in
chemistry.
Email: guptashalini@bharatpetroleum.in
www.eptq.com PTQ Q3 2011 135
N V Choudary is a Chief Manager (R&D) with
Bharat Petroleum, with 26 years experience
of research in petroleum refning and
petrochemicals. He holds MSc and PhD degrees
in chemistry.
Email: choudarynv@bharatpetroleum.in
G Biju is a Deputy Manager, Energy &
Environment, at BPCL Kochi Refnery and a
chemical engineer with 12 years refnery
experience in process optimisation, handling
production units. He holds a MTech in
environmental engineering degree from Indian
Institute of Technology, Madras.
Email: bijug@bharatpetroleum.in
T S Thorat is a Manager (R&D) with Bharat
Petroleum, with expertise in crude evaluations,
crude compatibility and blending, high acid
crude processing, bitumen and bottom of the
barrel processing, biofuels and new product
development. He holds a PhD in heterogeneous
catalysis chemistry from ICT.
Email: thoratts@bharatpetroleum.in
P V C Rao is a Senior Manager (R&D) with
Bharat Petroleum and has 24 years experience
of research in petrochemicals and refning. He
holds a PhD in chemistry from Indian Institute
of Technology, Bombay.
Email: raopvc@bharatpetroleum.in
A e r z e n - o n e s t e p a h e a d
Compressed oilfree gasesl
The requirements for the compression of air and gases are
varied. Especially in the chemical industry where the composition
of the media to be conveyed is completely different.
Aerzener Maschinenfabrik product range with special designs
can fulfll most applications. t does not matter whether screw
compressors for oilfree compression or positive displacement
blowers for conveying of impure or aggressive gases are used
or machinery for statically high system pressures is required -
Aerzen offers the optimum.
Challenge us! Make use of our know-how.
Please take our leafets A1-004, G1-151 and V1-067 for
further information or the nternet under www.aerzen.com
Aerzener Maschinenfabrik GmbH
Reherweg 28
.
31855 Aerzen / Germany
Phone: + 49 51 54 / 8 10
.
Fax: + 49 51 54 / 81 91 91
info@aerzener.de
.
www.aerzen.com
B I o w e r s . . . . C o m p r e s s o r s . . . . G a s m e t e r s B I o w e r s . . . . C o m p r e s s o r s . . . . G a s m e t e r s
120 x 190_PTQ.indd 1 25.05.2011 19:10:37
www.ptqenquiry.com
for further information
rathore bpcl.indd 5 9/6/11 10:44:09
Europes largest independent refner is meeting the
demands of the EUs Excise Movement and Control
System (EMCS) for transborder shipments from its sites
in Germany with a fully automated terminal manage-
ment and automation system. Petroplus sites in
Germany have begun using a version of OpenTAS
from software company Implico that integrates all of
the EMCS processes to deliver a smooth product-
loading operation.
Each day, between 10 and 15 railcars and 300350
tank trucks leave the Petroplus refnery in Ingolstadt.
Of these, one or two railcars and around 100 tank
trucks reach the border with Austria, Switzerland or
the Czech Republic.
The EMCS electronic customs processing system has
been a requirement for cross-border traffc within the
EU since the beginning of 2011. Although there are
many web-based applications available, Petroplus
rejected these as being too slow for its large number of
EMCS-related shipments. Not only that, but these
applications could be prone to error as data has to be
entered manually. Petroplus therefore opted for the
automated EMCS solution in OpenTAS. The refnery
was already using Implicos Terminal Management and
Automation System to automate its loading and ship-
ping processes.
The EMCS solution in OpenTAS carries out the frst
consistency check as soon as a tank truck drives into
the terminal. It checks, for example, whether all neces-
sary permits are present and whether the supplier and
consignee are known to customs. The key beneft is
that if irregularities are detected, the truck is not even
loaded. This reduces waiting times and prevents the
build-up of queues at the exit. If the irregularities were
not picked up until after the tank truck had loaded up
and was about to leave the refnery, the truck would
block the exit because it would not be allowed to leave.
Another advantage of the automated system is that
loading data is logged while the truck is in the loading
bay rather than, as is normally the case, at the exit.
This provides a noticeable time saving for the haulage
frm, since the EMCS declaration reaches customs
along with all the pertinent tax information as quickly
as possible; namely, right after the tank truck has been
loaded. Customs approval for the consignment is
normally received before the driver arrives at the exit,
allowing the driver to leave the refnery straightaway
without having to wait. In addition, the system allows
drivers to enter direct delivery codes using dialogues
in the automatic dispatch system.
EMCS is an electronic haulage and monitoring
Reduce gasoline cutpoint
Venturi steam traps solve condensate
pressure problem
system for the cross-border transportation of goods
subject to excise duty within the EU. This includes
petroleum products. The system replaces the previous
paper-based procedure. The Administrative
Accompanying Document (AAD) is now replaced by the
electronic Administrative Document (e-AD), which
enables shipping to be handled more effciently.
Another advantage is that both the companies involved
and the customs authorities can use EMCS to monitor
the movement of goods in real-time. Cross-border ship-
ments in Germany are registered via a customs web
portal or with certifed software such as OpenTAS.
After installation of the system at Ingolstadt refnery
at the beginning of December, Implico also imple-
mented EMCS at the Petroplus terminal in Passau.
Eni has started work on the frst commercial applica-
tion of its slurry technology (EST) at Sannazzaro de
Burgondi refnery, near Pavia in northern Italy. EST is
Enis proprietary technology for the conversion of
heavy oil residues to products including gasoline and
gas oil.
The project is scheduled for completion with the
startup of the 23 000 b/d plant by the end of 2012. Eni
has been developing the process since the mid-1990s,
starting at its San Donato Milanese laboratories and
continuing at its Taranto refnery, where a 1200 b/d
demonstration plant started operations in 2005. The
design of the new plant began in 2008 and involved
contractor Saipem for engineering activities.
EST technology will enable Sannazzaro to become a
zero fuel oil refnery. The technology can valorise the
exploitation of unconventional crudes and will also
enable Eni to evaluate new sources of raw material for
its refning operations.
EST is a hydroconversion process that employs a
slurry nano catalyst to refne different types of heavy
feedstocks, such as residues from the distillation of
heavy and extra-heavy crude oils (including
Venezuelan crude from the Orinoco Belt) or non-
conventional oils (such as bitumen from tar sands),
characterised by high contents of sulphur, nitrogen,
metals, asphaltenes and other polluting species that are
diffcult to manage with traditional refning processes.
The technology can produce gasoline and gas oil with-
out generating coke or fuel oil for continually declining
markets. Apart from its special catalyst, the process
scheme features a supply of hydrogen produced from
methane, effectively enabling the transformation of
methane into liquid fuels via hydrogen production.
Technology in Action
Slurry hydroconversion going
commercial

Automating customs for product
shipments
136 PTQ Q3 2011 www.eptq.com
case studies copy 3.indd 1 13/6/11 14:27:39
Eagle frewall was used at the demarcation point to
Shells process LAN. The backbone is based on
Beldens PROFInet copper and fbre-optic cables, as
well as Lumberg Automation Ethernet M12 connectors
for all Ethernet devices. At locations where wired LAN
access was diffcult, connectivity relies on wireless
point-to- point links.
The wireless project started in 2007, with planning
at Beldens centre in Neckartenzlingen, Germany.
The frst factory acceptance test involved the construc-
tion and operation of the whole system, with testing
of hardware, including all devices, cables and
labels; software, including frmware versions, confgu-
ration, security and management system; and
documentation and certifcation compliance, and the
overall functionality of the system, including its roam-
ing capability.
The entire setup was moved to Singapore to ensure
the system functioned with Yokogawas control system.
Finally, the whole system was transported to the plant
on Bukom Island and, after completing a fnal site
acceptance test, was handed over to Shell.
Air Products is to build a steam methane reformer
(SMR) for hydrogen production to meet a
supply contract for an additional 200 million scfd of
hydrogen for Valero Energy Corporations refneries
in St Charles, Louisiana, and Port Arthur, Texas.
The new plant will be located in St Charles and
is due to come on stream during the second
half of 2013. It will be connected to Air Products Gulf
Coast hydrogen pipeline supply network, which
supplies refnery and petrochemical companies in
the region.
Air Products is working towards increasing its
hydrogen pipeline supply in the Gulf Coast region,
with the intention of making it the worlds largest
hydrogen pipeline network. Air Products announced
plans to construct a new 180-mile long pipeline in
2010. The new pipeline extension, which is in the
project execution phase, will connect its Texas hydro-
gen system to the Louisiana hydrogen system.
Once complete, the hydrogen pipeline supply
network will stretch from the Houston Ship Channel
in Texas to New Orleans, Louisiana. The pipeline
network will unite over 20 hydrogen plants and over
600 miles of pipelines, and will supply Louisiana and
Texas refneries and petrochemical plants with over
1.2 billion scfd of hydrogen. The new Gulf Coast
hydrogen pipeline network is expected to be opera-
tional in 2012.
The St Charles hydrogen plant will be built through
an alliance between Air Products and Technip. The
enhanced SMR design is intended for optimal heat
integration and minimal heat losses, which, in turn,
will lower natural gas consumption.
Invensys Operations Management has installed its
SimSci-Esscor ROMeo optimisation software for China
National Petroleum Corporations (CNPC) Jilin
Petrochemical Company, northeast Chinas largest inte-
grated refning and petrochemical production site.
The software enables CNPC to respond in real-time
to changes in feedstock variation, process conditions
and utilities prices with energy savings. The installa-
tion at Jilin includes off-line operational analysis and
on-line plant modelling for ethylene production,
enabling CNPC to study and validate the feasibility of
various energy-saving strategies, as well as parameters
affecting energy savings before implementation. The
system operates across the 700 000 t/y ethylene
complex, including 10 cracking furnaces, quenching,
compression, cracking and separation equipment. The
outcome is a reduction in the refnerys energy
consumption by 5%, says Invensys.
Shell Eastern Petroleums ethylene cracker complex
(ECC) on the island of Pulau Bukom, Singapore, has
started up equipped with wireless technology that is
being used to gain secure mobile access to process
control systems. The control project, part of Shells
Houdini project at Bukom, features a wireless network
designed by Belden and includes hardware supplied
by Yokogawa.
Shell wanted mobile access to all process data
via wireless local area network (WLAN) at defned
places in the refnery to provide the appropriate eff-
ciencies. The benefts pursued include improved
overall effciency, resulting from faster commissioning
of the petrochemical process control system and
improved maintenance effciency, as well as system
cost reduction because fewer local panel units are
needed.
Belden performed project management, design and
building of the WLAN system, which includes a wire-
less distribution system incorporating 90 Hirschmann
BAT54-F X2 access points for the ECC and multiple
antennas, linked across a redundant, fbre ring
backbone.
Yokogawa supplied all of the process control and
instrumented safety systems for the Houdini project,
while Belden designed and delivered the wireless
infrastructure, which acts as an interface to the process
control systems. All relevant parts of the system had to
be certifed to comply with ATEX/IEC60079 for hazard-
ous areas.
WLAN products installed at Bukom include outdoor
access points, together with accessories, antennas,
surge arrestors, managed switches, media converters
and Industrial HiVision. For enhanced security, the
SMR hydrogen will add to Gulf Coast
network
Optimised petrochemicals
Wireless integration of ethylene
cracker
www.eptq.com PTQ Q3 2011 137
case studies copy 3.indd 2 13/6/11 11:43:14
Control valve developments
Emerson has released developments
to its Fisher GX control valve and
actuator system range to make it
suitable for a wider choice of appli-
cations. These updates include the
addition of a Whisper Trim III low-
noise trim and an anti-cavitation
trim. There is also a cryogenic
version that enables use down to
-196C and a design for use in pres-
sure swing adsorption (PSA) skids.
To verify cycle life, the PSA design
has been tested to withstand over
one million high-speed cycles, says
Emerson.
The GX is a globe-style valve
system that can control a range
of process liquids, gases and
vapours. Operations and mainte-
nance savings are delivered
through the use of standardised
constructions as well as the predic-
tive diagnostics and performance of
the integral Fisher Fieldvue
DVC2000 digital valve controller.
Even the cryogenic and PSA
designs, which are both arduous
applications, use mainly standard
parts so that delivery times are
three weeks ex works. Valves with
a range of standard options are
available in one week.
Rotary screw compressors
CompAir has extended its L-RS
range of regulated-speed rotary
screw compressors to cover ratings
from 7.5250 kW. The oil-injected
screw technology is based on the
same platform as CompAirs fxed-
speed units, and energy savings of
up to 25% are achievable, says the
developer.
Each compressor is ftted with an
inverter, with the aim of minimis-
ing energy use; the unit only
produces the appropriate volume of
air required. As a result, no energy
is said to be lost due to the
compressor idling in off-load mode,
resulting in lower power usage, as
well as reduced wear and tear on
components. This inverter is inte-
grated into an electric cabinet and
P
protected from dust by a replace-
able inlet flter.
Compressors from 7.522 kW are
available as an air station, which
includes a refrigerant dryer and air
receiver in a compact footprint. All
models have low noise levels, says
CompAir, so the compressors can
be positioned at the point of use
without the need to house them in
separate spaces or install additional
pipework.
Units are available with two types
of compressor controllers, depend-
ing on the model. As standard, the
L07RS-L22RS range has the inte-
grated Delcos Pro intelligent
controller. This continuously moni-
tors operating parameters by means
of a range of programmable inputs
and outputs.
L30RS-L132RS models are avail-
able with the new Delcos XL
controller, a touchscreen controller
that displays detailed system anal-
ysis without the need for
additional monitoring devices. The
controllers service diagnostics
enable compressor inputs and
outputs to be tested, while provid-
ing identifcation of any faults,
which can result in reduced
downtime.
All serviceable components are
grouped behind a lift-off canopy
section with quick-release catches
or hinged side doors. This allows
engineers easier access, to help
reduce downtime and keep service
costs to a minimum.
Focus: rotating equipment
From research to production
Pfeiffer Vacuum has overhauled its
Oktaline G gas-cooled Roots pump
so it can be used in process applica-
tions as well as for research and
development. Pumping speeds
range from 25012 000 m
3
/h.
In operation, heated gas is cooled
on the pressure side and partially
returned to the suction chamber.
This enables continuous operation
at high pressures, says Pfeiffer. The
use of hammer-shaped rotors also
enables the pump to achieve high
differential pressures, according to
the company.
The most demanding applications
can be realised in combination with
the gas coolers needed for process
adaptation. According to the devel-
oper, energy-effcient drives and the
use of frequency converters lower
energy consumption by up to 20%.
Pumping speeds can be adjusted,
and a gentle soft-start of the Roots
pump is enabled. This is designed
to keep operating costs low, increase
service life and protect the pump.
The gear and bearing chambers in
the Oktaline G are completely
separate from the gas-pumping
chamber.
Oil-free compression
Atlas Copco has launched an
energy-effcient, high-speed three-
stage turbo oil-free compressor. The
ZH350+ is said to be the most
energy-effcient oil-free compressor
built at Atlas Copco.
According to the company, it is
the frst medium-pressure industrial
three-stage centrifugal compressor
to be commercialised with high-
speed motors and without a
gearbox. The ZH 350+ is up to 4%
more energy effcient than the
companys top-performing oil-free
compressors. The new turbo
compressors low package power
consumption is attributable to fve
factors, says the developer.
First, The ZH350+s three-stage
design achieves high thermody-
namic effciency and low power
138 PTQ Q3 2011 www.eptq.com
Energy-effcient
drives and the use of
frequency converters
lower energy
consumption by up
to 20%
ind news copy 9.indd 1 13/6/11 10:02:33
consumption. A three-stage design
is recognised as being the most
effcient design for achieving
compression between 7 and 11 bar
with a turbo compressor. This
design results in average lifecycle
costs that are 3% lower compared
to an oil-free screw design and 18%
when compared to a two-stage
turbo design.
Second, the ZH 350+design elim-
inates the gearbox and thereby the
transmission losses associated with
a gearbox. Transmission losses can
result in a reduction in energy eff-
ciency of up to 9% in compressors
in the 350 kW range.
Third, the ZH 350+ has titanium
impellers, which are more reliable
and permit faster load-unload
cycles. The gain in speed results in
reduced power losses in times of
low air demand.
Fourth, optimally sized coolers
signifcantly reduce pressure drop,
affecting the stability of subsequent
compression changes. The design
also features an enhanced heat
transfer area. Finally, the system
blows off before the aftercooler,
which means that minimal
compressed air is lost when air
demand decreases.
The combined reduction in air
losses, pressure drops and transmis-
sion losses result in a compressor
that is 4% more energy-effcient
than the companys top-performing
compressors under comparable
operating conditions, says Atlas
Copco. Further energy savings are
possible with the companys rotary
drum heat-of-compression dryer
(MD/ND) for centrifugal compres-
sors. These dryers are also
compatible with the ZH 350+. The
ZH 350+ was TUV certifed ISO
8573-1 (2010) Class 0 with regard to
oil content.
Atlas Copco recently purchased a
regional distributor of its compres-
sors in the US, Tencarva Machinery
Company, to develop its aftersales
business. Tencarva operates in
Tennessee, Virginia, Mississippi,
Maryland and Delaware. The
acquired business will be incorpo-
rated into the East Region of Atlas
Copco Compressors in the US. The
company will operate from
Tencarvas current locations for a
period of 612 months. After divest-
ing the compressor business,
Tencarva Machinery Company will
continue to sell mainly industrial
pumps.
Labyrinth compressors
for China
Sinopec Wuhan Company has
ordered six Laby (labyrinth piston)
compressors from Burckhardt
Compression for its petrochemical
plant at Wuhan City, Hubei prov-
ince. The Laby has a labyrinth
sealing system on piston and piston
rod glands for oil- and contact-free
compression, resulting in minimum
wear and tear in low-temperature
ethylene boil-off applications, says
Burckhardt.
The compressor is insensitive to
particles and contaminated gases,
and easily manages the compres-
sion of ethylene boil-off gas at
suction temperatures down to
-100C (-150F), says the developer.
The gas-tight design reduces gas
emissions and losses to the environ-
ment to zero.
Two of the compressors on order
will be used for propylene compres-
sion; one as a recycle gas
compressor (ST process) and one as
an off-gas compressor (JPP process).
In addition, four compressors will
be used for ethylene liquefaction:
two for ethylene BOG compression
within the recondensing cycle; and
two for propane compression
within the refrigeration cycle.
The 800 000 t/y ethylene project
at Wuhan City is the largest
integration project in central China.
Laby compressors compress bone-
dry, dirty, abrasive and other gases
with no contamination and without
oil, with a discharge pressure up to
300 bar (4350 psi), fow up to 11 000
Nm
3
/h (6500 scfm) and shaft power
up to 2200 kW (3000 HP).
Kobelco opens US
compressor plant
Kobe Steel subsidiary Kobelco
Compressors America (KCA) is
opening a new $20 million factory
that will double its manufacturing
capacity for compressor systems.
KCA has also moved its headquar-
ters to the new plant.
Formerly known as Kobelco EDTI
Compressors, Inc., KCA is expected
to play a leading role in Kobe Steel
Groups process gas compressor
business covering North, Central
and South America.
Based in Corona, California, KCA
designs, manufactures and sells
compressor systems. Among its
customers are Exxon Mobil
Corporation, Petrleo Brasileiro
S.A (Petrobras) of Brazil, oil refn-
ing companies and petrochemical
frms.
Investments for capacity expan-
sion to meet rising energy demands
in North, Central and South
America are promising over the
medium- to long-term future.
Environmental investments are also
growing due to stricter regulations.
As a result, demand for compressor
systems is anticipated to grow
steadily in markets such as the
petrochemical and oil refning
industries.
Since Kobe Steels initial invest-
ment in KCA in 2003, the company
has expanded its operations. In
2008, KCA completed the No. 2
Plant, increasing production capac-
ity to $50 million. Business
conditions continued to remain
strong in 2010, resulting in orders
totalling $86 million. The facility is
currently in full production, and
even warehouses outside the prem-
ises are being used to store parts.
Construction of the No. 3 Plant
began in August 2009. With the
startup of the new facility, manufac-
turing capacity has doubled to $100
million.
www.eptq.com PTQ Q3 2011 139
Investments for
capacity expansion
to meet rising
energy demands in
North, Central and
South America are
promising over the
medium- to long-
term future
ind news copy 9.indd 2 13/6/11 10:02:44
CO
2
capture in Brazil
A pilot-scale project involving a
consortium of seven energy compa-
nies aims to demonstrate savings of
around 2030% in CO
2
emissions
from a typical refnery. The CO
2
Capture Projects (CCP) oxycom-
bustion capture trial test, at
Petrobrass research complex in
Parana state, Brazil, has been trial-
ling technology capable of capturing
up to 95% of emissions from an
FCC unit, one of the highest CO
2
-
emitting units in a refnery.
The demonstration involves a
full-burn FCC unit. It is expected to
confrm the technical and economic
viability of retroftting an FCC unit
to enable CO
2
capture through
oxycombustion. The project will test
startup and shutdown procedures,
as well as different operational
conditions and process confgura-
tions, providing the consortium
with data for scale-up.
Conventionally, air is used to
regenerate FCC catalyst, by burning
the coke deposited on the surface.
In oxycombustion mode, air is
replaced by pure oxygen, which is
diluted with recycled CO
2
to main-
tain thermal balance and catalyst
fuidisation.
Established in 2000, the CCP
aims to advance the technologies
that will underpin the deployment
of industrial-scale CO
2
capture
and storage (CCS) in the oil and
gas industry. The CCP identifed
oxycombustion as one of the most
promising capture technologies to
take forward for demonstration,
from over 200 options evaluated. It
initially conducted an economic
assessment of oxyfring and post-
combustion amine absorption for
CO
2
capture from the FCC regener-
ator. Both processes were able to
achieve the required specifcations
and recovery level. Although the
post-combustion option had a
lower capital cost, the lower
operational costs for oxyfring
delivered a lower overall capture
cost.
P
Hydrocracking selection
for Jazan
Saudi Aramco has selected hydro-
cracking technology from Chevron
Lummus Global (CLG) for its Jazan
refnery project, which is currently
at the front-end engineering design
(FEED) stage. CLGs Isocracking
technology will be used to convert
106 000 b/d of vacuum gas oil into
diesel fuel meeting Euro V quality
requirements at the new 400 000
b/d Jazan refnery.
CLG will provide an engineering
package, including the hydrocrack-
ing reactors, Isomix internals,
Isocracking catalysts, follow-up tech-
nical support during the detailed
engineering design, training prior to
startup, and startup support during
the commissioning of the new
refnery.
Oman Refneries and Petrochemi-
cals Company (ORPC) will also
license Isocracking technology as
part of its Sohar refnery expansion
project. The Sohar hydrocracker
will process 74 000 b/d of vacuum
gas oil and light cycle oil, to
increase naphtha production for the
refnerys petrochemicals complex
and produce ultra-low sulphur
diesel and kerosene.
The expansion of Sohar refnery is
being undertaken to increase the
total refning capacity to 187 000
b/d when the project is completed
in 2015, from 116 000 b/d at
present. The expansion, at an esti-
mated cost of $1.5 billion, is
expected to boost petrochemical
production in the area. Following
the expansion, Sohar refnery is
expected to meet the additional
feedstock requirements of Aromatics
Oman and Oman Polypropylene,
both located in Sohar.
Aromatics Oman, commissioned
in 2010, uses naphtha supplied by
Sohar refnery as well as imported
material, and has the capacity to
produce 818 000 t/y of paraxylene
and 198 000 t/y of benzene.
Oman Polypropylene, which
began operations in 2007, is
Industry News
intended as the seed for the
development of a downstream
polypropylene industry in Oman,
using feedstock from Sohar refnery.
The additional production of petro-
leum products would meet local
demand, and enhance Omans abil-
ity to export petroleum and
petrochemical products.
The engineering, procurement
and construction contract for Sohar
refnery expansion is expected to be
foated in Q1 2012. This will culmi-
nate in the commissioning and
beginning of on-stream operations
of the refnery units by 2015.
Partners in simulation
Invensys Operations Management
and Axens have formed a partner-
ship in which Axens will become
a sub-licensor of Invensyss
SimSci-Esscor Pro/II software, a
steady-state process simulator used
to design, analyse and improve
chemical processes.
Pro/II software will be delivered
as a component of Axenss webPOS,
a web-based process operations
simulator, which includes tools to
perform functions including off-line
engineering studies to monitor the
impact of operating variables and
feed quality on product yields and
qualities; evaluating bottlenecks for
new feedstock or new product qual-
ity requirements; and making
strategic and operational decisions
regarding production and product
distribution.
According to Invensys, the part-
nership will strengthen its role in
the petrochemical and refning
industries by embedding Pro/II in
webPOS.
Maximising polypropylene
New FCC technology developed to
achieve higher levels of propylene
production from FCC operations is
nearing commercial operation. The
technology, called High Severity
Fluidized Catalytic Cracking (HS-
FCC), is the result of process
development in Japan and Saudi
140 PTQ Q3 2011 www.eptq.com
ind news copy 8.indd 1 13/6/11 10:15:21
Arabia, and is said to produce
higher yields of propylene and
other light products than conven-
tional FCC units can achieve. The
technologys developers have
selected Shaw Group and Axens to
promote and license the technology
worldwide.
HS-FCC technology evolved from
a 15-year development effort.
During phase one of the process,
Japans JX Nippon Oil & Energy
Corporation (JX) and Saudi Arabias
King Fahd University of Petroleum
and Minerals (KFUPM) formed a
research venture. JX, which leads
the technology developers,
provided technical research, and
KFUPM provided the location for
initial laboratory testing facilities.
During phase two, Saudi Aramco
joined JX and KFUPM to continue
developing the technology. The
expanded team designed, built and
operated a 30 b/d demonstration
unit at Saudi Aramcos Ras Tanura
refnery.
JX embarked on the third phase
of development, including scaling-
up of a demonstration unit to a
3000 b/d pre-commercial demon-
stration unit, which is being built at
JXs refnery in Mizushima, Japan.
Shaw and Axens provided engi-
neering services for the unit, which
is expected to be operational this
year.
By shortening the contact time
period of the process, over-cracking
and other secondary reactions are
curtailed. Since the reaction takes
place at a high temperature, the
technology uses at least twice the
C/O ratio used by conventional
FCC. A high C/O ratio magnifes
the contribution by catalytic crack-
ing and has the effect of curtailing
thermal cracking, which becomes
predominant at high temperatures.
HS-FCC features a 550650C reac-
tion temperature and a contact time
period of 0.5 to several seconds.
In the development of catalyst for
the process, emphasis was put more
on olefn selectivity than on crack-
ing activity, and the amount of acid
in the catalyst was adjusted accord-
ingly. In addition, an additive for
the increased production of olefns
was developed. By combining these
reaction conditions, catalyst and
additive, high olefn and gasoline
yields could be obtained.
Contract seals upgrade
completion
KBC Advanced Technologies and
Polish refner Grupa Lotos have
agreed a contract to develop a refn-
ing excellence partnership (REP).
This will focus on business processes
and activities within Grupa Lotos to
enable a quicker response to chang-
ing petroleum market economics
and planning requirements, as well
as variations in individual process
unit performance. It will also ensure
the refners major upgrade
programme, Project 10+, completed
at the beginning of 2011, boosts
refnery proftability.
The 10+ programme is the largest
capex project to have been under-
taken by the Lotos Group. Its goal
was to increase the throughput
capacities and depth of conversion
of the Gdansk refnery, and conse-
quently to improve the companys
competitive position. Completion
and launch of a number of key
units constructed as part of the 10+
programme was achieved in 2010.
By the end of December 2010, the
project had been completed within
schedule. The scope of engineering,
procurement and construction work
was completed with respect to all
key and ancillary units.
The mild hydrocracking unit,
the largest item of equipment
constructed within the programme,
was assigned ready for start-up in
October 2010. The frst product
streams obtained from this facility
in mid-January 2011 achieved a
conversion ratio of 70%, compared
with a design conversion ratio of
60%. The main purpose of the MHC
unit is to intensify oil processing
and obtain a higher yield of higher
quality fuel components from each
barrel of crude. The residuum oil
supercritical extraction (ROSE) unit,
the last to be constructed under the
10+ programme, was completed on
schedule in December 2010.
Within the new partnership,
KBCs software models will be used
to ensure planning activities meet
the requirements of the new confg-
uration at Gdansk refnery.
Petronas complex planned
Petronas is to develop a $20 billion
integrated refnery and petrochemi-
cal complex in southern Johor state.
The project, to be called Refnery
and Petrochemicals Integrated
Development (Rapid), aims to
develop the companys operations
in high-value product development.
The capacity of the Rapid project
will exceed the combined output of
the Kerteh and Gebeng integrated
petrochemical complexes in the east
coast of Peninsular Malaysia.
With an estimated cost of $20
billion, Rapid is expected to be
ready by mid-2016. It will include
the construction of a crude oil refn-
ery, a naphtha cracker and an
associated petrochemical plant on a
1000 ha site in Pengerang. A
detailed feasibility study is sched-
uled to be completed by September
this year, while a fnal investment
decision is to be made by the end
of 2012.
Hydrotreating for Venezuela
Venezuelas PdVSA has signed up
Haldor Topse to supply a 85 000
b/d hydrotreating unit within the
parastatals Centro de Refnacin
Paraguan project. In view of the
large capacity of the hydrotreater,
the project will include two trains
with four hydrotreating reactors in
total, which will produce ultra-low-
sulphur diesel containing less than
10 ppm sulphur.
The diesel hydrotreater is part of
an optimisation project in PdVSAs
largest refning complex in
Paraguan, Falcon State. Equipped
with the new unit, the refnery will
be able to meet new restrictions on
sulphur content in diesel for local
www.eptq.com PTQ Q3 2011 141
FCC technology
developed to achieve
higher levels of
propylene production
from FCC operations
is nearing commercial
operation
ind news copy 8.indd 2 13/6/11 10:15:34
142 PTQ Q3 2011 www.eptq.com
and export markets. The planned
diesel product quality corresponds
to Euro V specifcations.
Topse will supply an engineer-
ing design package for the
hydrotreaters. The scope of supply
includes design of the reactors,
basic engineering, catalysts and
detailed design of the proprietary
reactor internals. Erection of the
unit is expected to begin in 2014,
and PdVSA expects to begin
production by the end of 2015.
Coal-to-gasoline plant in
progress
TransGas Development Systems has
begun construction of a coal-to-
gasoline complex, the USs largest,
at the Adams Fork Energy plant in
Mingo County, West Virginia. The
plant will convert regional coal into
18 000 b/d of gasoline.
The project will employ Uhdes
Prenfo gasifcation technology,
based on the Koppers-Totzek coal
gasifcation process developed 70
years ago. The Adams Fork Energy
coal-to-gasoline plant will be the
frst licensed Uhde Prenfo gasifca-
tion application in the US.
TransGas and Uhde Corporation
of America signed the licence agree-
ment in December 2008 and an air
permit was issued in February 2010.
The plant will consist of coal prepa-
ration, coal gasifcation, gas
cleaning, methanol synthesis and a
methanol-to-gasoline technology.
The complex is expected to be
completed by 2015.
The Prenfo process has been used
in an IGCC power plant in
Puertollano, Spain, which is in
operation and said to have one of
the highest effciencies in the world
for an IGCC plant.
Koreas biggest utility, Kepco, and
Uhde have formed a joint venture
to develop projects in synthetic
natural gas production and
combined cycle gasifcation based
on Uhdes Prenfo process. Kepco
will have a 66% share of the
venture, with Uhde holding the
balance.
Buyer wanted for France refnery
LyondellBasell is looking for a buyer
for its 105 000 b/d refnery at Berre,
France, operated by its subsidiary,
www.eptq.com PTQ Q3 2011 141
Compagnie Ptrochimique de Berre.
LyondellBasells former manage-
ment bought the refnery in 2008 but
has not met expectations.
Alongside its crude refning oper-
ations, the Berre site has a steam
cracker and polypropylene and
polyethylene plants, all owned and
operated by a subsidiary of
LyondellBasell. The company is not
looking to sell the olefns and poly-
olefns units at Berre.
Bunker operations in UAE
Shell Marine Products has begun
supplying marine fuels at Jebel Ali
Port in Dubai, United Arab Emirates.
Shell has acquired storage oil tanks
adjacent to one of the main termi-
nals at Jebel Ali Port, and has set up
a high-capacity bunker barge to
deliver fuels and services. The 8000-
tonne barge, which was previously
deployed in Singapore, is one of the
frst vessels to use mass fow meters
in the Middle East. Through the
introduction of the latest fow meter
technology, Shell says it aims to
provide increased transparency in
the quantity measurement of marine
fuels delivery.
Poor demand hits biofuels
refnery
Slow demand for biofuels has hit
Europes largest biofuels refnery,
the 300 million Ensus plant on
Teesside, UK, which has halted
production through the summer
months. Ensus blamed the shut-
down, which took effect at the end
of May, on lower-than-expected
demand for bioethanol across
Europe because of slow implemen-
tation of a renewable energy
directive by European Union
member states and on ethanol from
the US, which benefts from tax
breaks, fooding the European
market. The directive says that a
proportion of transport fuel must
come from renewable sources to
help combat global warming. Ensus
said delays in member states agree-
ing sustainability criteria for green
fuel had also been a factor.
The company said the shutdown,
expected to last between two and
four months, was temporary.
However, suspension of production,
which takes effect at the end of this
month, has implications for the
supply chain. The plant is one of
the worlds largest cereal grain
biorefneries.
The plant at Wilton began operat-
ing early in 2010. It uses more than
a million tonnes of wheat a year to
produce about 400 million litres of
ethanol, sold to Shell for use in
petrol; 350 000 tonnes a year of
high-protein animal feed, sold to
Glencore; and 350 000 tonnes of
carbon dioxide, liquefed at Wilton
and sold into the food and drinks
industry.
Wastewater treatment upgrade
Siemens Water Technologies will
provide a system to treat wastewa-
ter at Sinopecs Anqing refnery, in
Anhui Province, China. The project
will include a powdered activated
carbon treatment (PACT) system, a
Zimpro wet air regeneration (WAR)
hydrothermal unit, and a Hydro-
Clear sand fltration system. The
three-tier system will be used to
treat salty and oily wastewater from
refning and petrochemical produc-
tion activities from existing and
upgraded units. The system will
become operational in 2012.
Wastewater from the refnery
needs to meet local specifcations
for surface discharge. The 1000
m
3
/h system aims to help Sinopec
discharge a cleaner effuent and
also generate approximately 500
m
3
/h of reusable water. The refn-
ery will reuse the water in its
cooling tower system, which will
help to further offset the strain on
surface water supplies.
The PACT system combines
biological treatment and carbon
adsorption to remove organics. The
solids are pumped as slurry to the
Zimpro WAR unit, where the
carbon is regenerated and biological
solids are destroyed. Since it is said
to generate virtually no sludge that
requires landfll disposal or inciner-
ation, the WAR system is expected
to lower operating costs for the
Anqing refnerys wastewater treat-
ment. The Hydro-Clear flter system
makes it possible to flter water and
return it to the cooling circuit,
resulting in savings in the amount
of fresh water required and thus
further reducing costs. The
ind news copy 8.indd 3 13/6/11 10:16:01
treatment systems small footprint design will also
reduce land requirements, says Siemens.
Technology awards for Iraq refneries
Refning companies have selected Honeywell companies
to supply technology for major upgrade and new-build
projects. Iraq State Company for Oil Projects (SCOP) has
signed up UOP to provide technologies to process
300 000 b/d of domestic crude oil into gasoline and diesel
fuel at a new plant in Nassiriya, while North Refneries
Company (NRC) has appointed Honeywell Process
Solutions as the main engineering, procurement and
construction contractor to upgrade the automation
systems at its refnery in Baiji.
UOP will provide reforming, isomerisation, FCC and
selective hydrotreating technologies for the Nassiriya
project. As part of the overall technology package, it
will provide basic engineering, technology licences,
catalysts and speciality equipment for the new refnery.
Design will begin in Q2 2011. SCOP has made the
implementation of this new refnery a priority and is
currently offering the project as an opportunity to
potential investors.
The technologies to be provided to SCOP include
CCR Platforming and Penex processes, which are used
to produce high-octane gasoline, as well as FCC and
SelectFining technologies, which enable high-yield
production of ultra-low-sulphur diesel fuel and
gasoline.
SCOP is part of the Iraqi Ministry of Oil and responsi-
ble for the processing of all petroleum products in Iraq.
Refning capacity in Iraq is expected to reach 1.6 million
b/d by 2017, more than twice the capacity of its existing
refneries. This is expected to double again by 2030,
driven by an increased demand for transportation fuel
within Iraq.
UOP previously had awards in 2010 from South
Refneries Company and North Refneries Company
for new refneries in Maissan and Kirkuk, Iraq, each
rated at 150 000 b/d.
The contract for the upgrading of NRCs existing
control system at Baiji will improve operational eff-
ciencies, reliability and safety at the refnery.
Honeywell systems to be used include the Experion
process knowledge system and safety manager, to
fully automate the plant and replace its 30-year-old
single loop instrument control system. In addition to
improving safety and security at the plant, the new
automation investment will enable NRC to improve
productivity, with the ability to install future
upgrades, according to the supplier.
Honeywell Process Solutions technology will manage
a range of processes, with the aim of optimising yields
and reducing maintenance costs by up to 30%. It will
also provide technical training to NRC employees.
In 2010, Honeywell opened a full-service offce in
Baghdad and has plans to open other offces in south-
ern and northern Iraq over the next few years.
Honeywells Baghdad offce focuses on supplying tech-
nology, including process automation equipment, to
Iraqs oil and gas industry.
The acknowledged rst choice global event
for professionals in sulphur, sulphuric acid
and related industries
7-10 November 2011, Houston, Texas
www.sulphurconference.com
Ofcial publications:
Media Partners:
www.energyglobal.com
Energy Global
Sponsored by:
ptq
PETROLEUM TECHNOLOGY QUARTERLY
Sulphur | Sulphuric Acid | Sulphur Dioxide
SULPHUR
International Conference
& Exhibition
Why attend Sulphur 2011
in Houston?
Hear from the leading voices in the
sulphur/sulphuric acid industry
Network with the largest gathering of
technical/commercial delegates
Benet from expert industry knowledge
at the longest established sulphur event
www.eptq.com
ind news copy 8.indd 4 13/6/11 10:16:15
ABB Global Consulting 99
Aerzener 135
Albemarle Catalysts Company 14
Amistco Separation Products 34 & 39
AP Moller Maersk Lines 121
Arca Regler 37
Axens OBC
Baker Hughes 11
BASF Catalyst 19
Borsig 47
Bryan Research & Engineering 63
Burckhardt Compression 91 & 95
CB&I 7
Chevron Lummus Global 9
Curtiss-Wright Flow Control Corporation 20
Det Norske Veritas 103
DuPont Belco Clean Air Technologies 53
Elliott Group 108
Enersul 111
Euro Petroleum Consultants 44
ExxonMobil Research and Engineering Company 129
Foster Wheeler 41
GE Energy 88
Grabner Instruments 10
Haldor Topse IFC
Hoerbiger Kompressortechnik Holding 65
Hydrocarbon Publishing Company 13
Intercat 69
ITW 66 & 76
Johnson Matthey Catalysts 26
Johnson Screens 74
KBC Advanced Technologies 2
KBR 4
Koch-Glitsch 43
KTI Corporation 127
Linde 96
Lurgi 48
MAN Diesel & Turbo 29
Michell Instruments 16
Mogas Industries 58
MPR Services 17 & 57
National Petrochemical & Refners Association IBC
Newtons 107
OHL Gutermuth Industrial Valves 18
Paharpur Cooling Towers 118
Prosim 83
Rentech Boiler Systems 32
Sabin Metal Corporation 71
Samson 25
Sandvik Steel 101
Selas Fluid Processing Corporation 124
Spectro Analytical Instruments 105
Sulphur 2011 143
Sulzer Chemtech 81
Thermo Scientifc 115
Turbomachinery Symposium 84
UOP 51
Vereinigte Fllkorper-Fabriken 22
World Refning Association 130
Worley Parsons 33
Yokogawa Europe 86
Alphabetical list of advertisers
For more information on these advertisers, go to www.ptqenquiry.com
144 PTQ Q3 2011 www.eptq.com
ad index copy 10.indd 1 13/6/11 14:40:31
.!4) /.!,0%42/#(%-) #!,2%&) .%23!33/#) !4) /.
Meet the Right People
f or Busi ness Success
NPRAs Q&A and Technology Forum is the industrys most comprehensive and useful meeting on renery
operations and process technology. This event assembles representatives from throughout the industry
including reners, petrochemical producers, catalyst and chemicals suppliers, plant automation specialists,
industrial control specialists, process licensors and engineering rms. Attendees will discuss challenges
reners face today and gain valuable knowledge from expert Q&A panelists and other attendees.
Todays rening environment requires attention to plant safety, superior technology, innovation and
excellence in operations. Benet from the shared knowledge of experts throughout the industry. The Q&A
and Technology Forum will keep you moving in the right direction as you optimize your plants operation.
Register at npra.org/meetings or call us at (202) 457-0480 for more information.
Join colleagues and customers at
NPRAs Q&A and Technology Forum
October 9-12 << JW Marriott San Antonio Hill Country Resort & Spa << San Antonio, TX
NPRA_Q&A_Tech_Forum_PTQ.indd 3 5/27/11 10:12:01 AM
npra.indd 1 8/6/11 12:45:26
Your objectives
in focus
Make the most of todays and tomorrows
challenges with leading-edge solutions from Axens
- Clean and alternative fuel technologies
- Petrochemicals
- Energy efficiency
- High performance catalysts & adsorbents
- Revamps
Single source technology and service provider
ISO 9001 ISO 14001 OHSAS 18001
www.axens.net
210/297.qxd:A4 13/08/10 12:12 Page 1
www.ptqenquiry.com
for further information
axens.indd 1 13/9/10 13:27:18