Professional Documents
Culture Documents
Q2 2024
REFINING
GAS PROCESSING
PETROCHEMICALS
NAPHTHA
SPLITTER REVAMP AI APPLICATIONS
BIOGENIC
CO-PROCESSING BLUE HYDROGEN
TO MAKE SURE YOU CONTINUE RECEIVING A REGULAR ISSUE OF PTQ
5 ptq&a
94 Technology in Action
Cover
Commercial relevance of technology ranging from refinery FCC assets to chemical plastic waste recycling reflects
investments exceeding $185 billion to expand operations for meeting increased demand.
©2024. The entire content of this publication is protected by copyright. All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system or transmitted in any form or by any means – electronic, mechanical, photocopying, recording or otherwise
– without the prior permission of the copyright owner.
The opinions and views expressed by the authors in this publication are not necessarily those of the editor or publisher and while every
care has been taken in the preparation of all material included in Petroleum Technology Quarterly and its supplements the publisher
cannot be held responsible for any statements, opinions or views or for any inaccuracies.
www.digitalrefining.com
www.decarbonisationtechnology.com
Make every
molecule matter
At Shell Catalysts & Technologies, we understand how small, unseen chemical reactions
can affect the health of our loved ones, neighbours, and the planet at large. That is
what motivates us in our mission to Make Every Molecule Matter. Every member of
our team is focused on the same goal – developing cleaner energy solutions that
enable industries to tackle global climate challenges starting at the molecular level.
Our experienced scientists and expert engineers collaborate with customers to create
solutions tailored to your specific energy transition and performance challenges. And
when they do, they use the knowledge we gained from Shell’s corporate heritage as the
designer, owner and operator of complex industrial process plants around the world.
shell ct_memm_PDQ.indd
002_Q2_Shell.indd 1 1 12/7/2311:16:43
11/03/2024 4:27 PM
ptq
PETROLEUM TECHNOLOGY QUARTERLY
Vol 29 No 3
Keeping refineries
profitable
T
Q2 (Apr, May, Jun) 2024
hroughout every quarterly issue of PTQ, we have seen the need to find new
ways to improve and apply existing technology for changing market forces
and regulatory requirements. So, with adaptation of carbon taxation, refiners
Editor must produce higher yields of fuels per unit of CO2 from conversion units such as the
Rene Gonzalez
fluid catalytic cracking (FCC) unit. One interesting route is to operate FCCs more sus-
editor@petroleumtechnology.com
tainably by supporting the co-processing of alternative feeds. But how many FCC
tel: +1 713 449 5817
units are co-processing alternative feedstocks?
The consensus is that co-processing is feasible if less than 5% of the feedstock
Managing Editor
is from alternative sources. Above 5%, processing unknowns weigh in on the need
Rachel Storry
for additional expertise, such as the use of more active and poison-resistant cata-
rachel.storry@emap.com
lysts and absorbents for efficient conversion of renewable feedstocks to fuels and
Editorial Assistant chemicals, bringing new capabilities to existing facilities seen in Europe and North
Lisa Harrison America.
lisa.harrison@emap.com Elsewhere, project start-ups reflect strategies to increase fuel production and feed-
stock volumes for monetisation to petrochemicals, as seen in the Middle East and other
Graphics countries. For example, Saudi Arabia is building nine liquid-based ethylene steam
Peter Harper crackers to produce a wide range of products, such as a high-margin polymers.
Through joint ventures, steam cracker projects will process byproducts from
Business Development Director crude processing, including naphtha and off-gas, to produce ethylene, such as
Paul Mason with a multi-billion-dollar project between Saudi Aramco and South Korea to pro-
sales@petroleumtechnology.com duce propylene, butadiene, and other basic chemicals by 2026. Petrochemicals
tel: +44 7841 699431 are set to account for nearly half of growth in oil demand to 2050, according to the
International Energy Agency. Petrochemicals are also poised to consume an addi-
Managing Director tional 56 billion cubic metres of natural gas by 2030, the agency recently noted.
Richard Watts The global petrochemical market is projected to be worth roughly $800 billion by
richard.watts@emap.com 2030, according to Precedence Research. At this juncture, there is a lot of pressure
to close refineries in Europe if old assets cannot be modified, such as with the ability
Circulation to co-process alternative fuels. What is happening with refinery growth in Africa
Fran Havard
now could significantly impact European refineries, such as with a refinery in Nigeria
circulation@petroleumtechnology.
that includes a new 600,000 bpd single column crude unit. According to Honeywell
com
UOP’s Keith Couch, speaking at the January NARTC in Houston, “when that unit
starts up (i.e., at the Dangote refinery), the price pressure on European refineries will
EMAP, 10th Floor, Southern House, approach $4 per barrel”. The Dangote refinery has ramped up its stock of imported
Wellesley Grove, Croydon CR0 1XG crude oil, including shipments of US light sweet crude oil, according to trader and
tel +44 208 253 8695 ship tracking data. So, the facility is well on its way to running at full capacity.
Will European refiners be able to stave off imports? Perhaps with government
Register to receive your regular copy subsidies and ESG-focused projects. Elsewhere, the rapid start-up of major capital
of PTQ at https://bit.ly/370Tg1e projects in China and India is taking 36 months, while projects in the Middle East
need perhaps 96 months to complete due to supply chain and labour restrictions.
This makes refineries such as Reliance in India more competitive, even as they
PTQ (Petroleum Technology Quarterly)
(ISSN No: 1632-363X, USPS No: 014-781) are leveraged by a lack of crude oil reserves. In almost every instance, the road
is published quarterly plus annual Catalysis
edition by EMAP and is distributed in the US
to success for refiners involves a shift to petrochemicals, such as with olefins and
by SP/Asendia, 17B South Middlesex Avenue, aromatics.
Monroe NJ 08831. Periodicals postage paid at
New Brunswick, NJ. Postmaster: send address
For ‘older’ refinery facilities, such as in the US, that petrochemical opportunity
changes to PTQ (Petroleum Technology could begin with benzene, as its octane-enhancing importance in the gasoline mar-
Quarterly), 17B South Middlesex Avenue,
Monroe NJ 08831. Back numbers available ket is diminishing due to regulatory restrictions. Propylene is another bankable pet-
from the Publisherat $30 per copy inc postage. rochemical that can be considered with the right process and catalyst upgrades, as
discussed in this issue of PTQ. Further into the year, this narrative will discuss track-
ing a facility’s carbon footprint from new profitable and energy-efficient projects.
Rene Gonzalez
PTQ Q2 2024 3
Process Notes
Q What technologies do you see dominating the long- Q What technology and strategies can resolve the huge
term development of alternative fuels like ammonia, gaps between ethylene and propylene production in cer-
methanol, and hydrogen? tain markets?
A Ujjal Mukherjee, Chief Technology Officer, Lummus A Jeffery Nichols, Solutions Delivery Senior Advisor,
Technology HSB Solomon Associates
I see a variety of opportunities coming forth across a The boom over the past 15 years in demand for olefinic
broader spectrum of timelines. In the medium term, blue derivatives, most notably polyolefin plastics, has ushered
hydrogen combined with carbon capture and storage will in a supply deficit in multiple markets, leaving producers
continue as the dominant technology. With the advance- scrambling to capitalise on incremental margins. While the
ment of electrolysers incentivised by programmes such introduction of new production lines has been realised, with
as the Inflation Reduction Act in the US, the cost of green multiple fleet operators executing major capital projects
hydrogen production will continue to drop. However, scal- since 2010, existing facilities are examining their options
ing to match blue hydrogen capacities will require a level of for squeezing every possible pound from their units.
investment that only an enforceable penalty on CO2 pro- The keys to maximising olefin production from any facility
duction could justify. – whether a new, world-scale complex or a vintage cracker,
I see niche areas with plentiful solar, wind or hydroelec- and regardless of feed or technology – fall into four foun-
tric power as first adopters of large-scale green hydrogen dational areas:
projects, such as Neom in Saudi Arabia. In countries with • Shrinking or eliminating losses
an abundance of nuclear energy, such as Abu Dhabi, I soon • Increasing yield
see the production of pink hydrogen as a reality, especially • Increasing capacity
post-COP28. • Sustaining high availability and minimising slowdowns
Pink hydrogen will be used first in the cement and steel and downtime.
industries and later in the refining and petrochemicals sec-
tor. In the long term, turquoise hydrogen produced by the
direct conversion of methane to hydrogen and solid carbon Robust mechanical maintenance
could become attractive at scale, especially if combined with
an electrical heat source from renewable power.
programmes, including sound
Ammonia and methanol will continue to see growth as predictive and preventive
hydrogen carriers. The current cost of renewable hydrogen maintenance components, are
production remains much higher than producing hydrogen
via steam methane reforming. In the long term, large-scale
critical to increasing unit availability
use of renewable hydrogen will become economically via-
ble driven by governmental policy incentives, maturation
of electrolysers, and carbon dioxide (CO₂) taxation. Back As intuitive as these strategies for maximising produc-
cracking of ammonia using a catalytic process to produce tion may seem, applying them in the real world may not be
hydrogen and nitrogen will become an attractive option. straightforward. In addition, they are likely to involve ini-
Produced hydrogen can be used in fuel cells for power tial costs that hinge on the strategy or strategies adopted.
packs. At scale, these power packs can replace marine fuel. Furthermore, no action exists in a vacuum. In some cases,
Firing of 100% hydrogen is technically feasible and improvement in one area will be complemented by positive
tested, and it avoids the challenges associated with the results in another, but sometimes there are trade-offs.
use of ammonia as fuel. Using ammonia for direct firing Robust mechanical maintenance programmes, including
will require careful consideration of burner types and the sound predictive and preventive maintenance components,
design of the selective catalytic reduction equipment. There are critical to increasing unit availability. Higher availabil-
are also lingering concerns about fugitive ammonia emis- ity, almost without fail, translates into reductions in losses
sions leading to increased particulate matter (PM 2.5 and within a particular plant. Operational risk analysis that takes
PM 10.0) when combined with other pollutants found in into account a facility’s tolerance for approaching critical
acid rain. plant limits or constraints is another essential element. Key
The use of methanol as a marine fuel will rise, especially points to consider include excess utility capacity, redun-
with the growth of green methanol. Green methanol pro- dancy in contaminant removal, and safe operating limits
duced from sustainable biomass will see large-scale adop- for major equipment such as compressors. Consider this:
tion in Asia, especially in India. Europe will see increased For a 1,500-kiloton-per-year facility with a 98.5% service
adoption of methanol from hydrogen produced with renew- factor, a four-day reliability event translates into a produc-
able power and captured CO2 (e-methanol). tion impact of more than 15 kilotons of product. Solomon
REC
Alkyfining®
ALKEMAX™
Sulfuric Acid Alkylation
C4 Isomerization
Q What role does artificial intelligence (AI) play in AI-driven digital solutions can
revamping downstream facilities that are scaling back on be adjusted within equipment,
conventional fuels production while upgrading to capture product specification, and utility
value from new products?
constraints to minimise energy and
A Bradley Ford, Global Process Optimisation Solution CO₂ emissions
Leader – Technology, KBC (A Yokogawa Company)
The growing scarcity of skilled labour is impacting the per-
formance of facilities worldwide. In fact, a study by Deloitte
and The Manufacturing Institute reports that the manufac- leverages AI-based platforms such as tcgmcube. We couple
turing skills gap in the US alone could result in 2.1 million this with rigorous first principle-based process technology
unfilled jobs by 2030, resulting in a projected cost total- tools to take available data, quickly create digital twins, and
ling $1 trillion. As the industry pushes for optimisation, the develop unique hybrid solutions. These AI-driven digital
infrastructure’s increasing complexity poses a challenge. solutions can be adjusted within equipment, product speci-
KBC is observing the emergence of various types of AI fication, and utility constraints to minimise energy and CO₂
technologies that are starting to address these challenges. emissions while maximising production of the most valu-
For example, process simulation technologies are preva- able products. These techniques are applicable to optimis-
lent at nearly all global assets, operating as process digital ing existing operations, designing new plants, or evaluating
twins or online real-time optimisers. However, simulation potential retrofit options.
models that reflect reality still require calibration from engi- Once the products have been maximised within existing
neers. KBC now sees AI handling this critical task to: constraints, new solutions are developed with retrofits. This
u Monitor the asset and models can include the change-out of catalyst systems, the addi-
v Identify when calibration is lost tion of new equipment, and sometimes entire upstream or
w Automatically recalibrate it. downstream process technology. All of these impact the
The critical impact is allowing the available finite human overall product slate while remaining within the turndown
resources to focus on higher-value tasks. constraints of the base complex.
Looking into the next steps, generative AI’s capabilities Examples of such strategies include:
are potentially game-changing in capturing organisational • The introduction of biofeedstocks and waste plastic-
knowledge that is dispersed across silos, contextualising derived pyrolysis oils to existing refineries or petrochemical
that knowledge, and allowing junior staff to use it for idea complexes.
generation. Careful oversight is needed to prevent genera- • The elimination of gasoline production while maximising
tive AI systems from ‘hallucinating’ or producing theoretical jet and diesel production, high-sulphur fuel oil production,
outputs that conflict with the data on which the algorithm and high-sulphur coke production.
has been trained. Hence, training programmes are required • Increasing chemicals production from 10-30% and
to educate staff on how to leverage generative AI to cre- sometimes 50% production of needle coke and anode coke
ate ideas for improvement, which still requires peer reviews from an existing coker, which is a dramatic shift in crude
before implementation. slates and more.
Grace, the global leader in FCC catalysts and additives, introduces MIDAS® Pro catalyst
MIDAS® Pro catalyst, for resid cracking in high iron applications. offers the solution for
This innovation, built on our workhorse MIDAS catalyst platform, proved
®
resid cracking in high iron
its capacity to handle even the worst Fe excursions. In commercial trials environments. Gain feed
with multiple in-unit applications, MIDAS® Pro catalyst demonstrated flexibility with better
sustained bottoms cracking in the face of iron spikes that measured bottoms upgrading.
among the highest in the industry. Diffusivity levels were consistently
high, indicating no transport restrictions with concentration of Fe.
This improved iron tolerance allows refiners to operate at higher iron
levels which increases feed processing flexibility and profitability.
Visit us at
ACHEMA
in Frankfurt
Hall 4
Stand D48
CM
Spherical Catalyst Carriers
MY Excellent Size Control with Unprecedented Formulation Flexibility.
CY
CMY
With Saint-Gobain NorPro, you can count on a secure supply of Accu™ spheres catalyst
K carriers, custom-made to your requirements. Our new spherical carriers can be
prototyped for your intended application, making them ideal for fixed bed reactors
or slurry and moving-bed reactors where coking occurs making constant
regeneration essential. Accu spheres are the missing link in the demand for
small diameter carriers which provide high-geometric surface area and even
flow distribution.
NORPRO
Surabhi Thorat and Vivek Srinivasan Dorf Ketal Chemical (I) Pvt Ltd
Sudarshan Vijayaraghavan Dorf Ketal Chemicals PTE Ltd
E
thylene serves as a predominantly petrochemically • Impact on selectivity: Inefficient furnace operation, com-
derived monomer essential for producing plastics, bined with a high potential for coking, leads to a critical
fibres, and various organic chemicals. These end TMT threshold during mid-run cycles, constraining achiev-
-products find applications across industries such as pack- able severity levels. Consequently, this limitation adversely
aging, transportation, construction, and other industrial and affects the overall selectivity for ethylene make.
consumer markets. Notably, more than half of global ethyl- • Market dynamics: In the competitive petrochemical indus-
ene derivative consumption is attributed to non-durable or try, delays or interruptions in production can affect a compa-
consumable end uses, especially in packaging. ny’s ability to meet customer demands. This can result in lost
The majority of this consumption is associated with poly- market share and potential long-term damage to business
ethylene, a plastic resin that constitutes most ethylene usage. relationships.
Given its status as one of the largest volume petrochemicals • Increased emission and environmental impact: Poorly
globally, the consumption of ethylene is influenced by eco- optimised furnace operations tend to emit higher levels of
nomic and energy cycles. Its extensive and diverse derivative greenhouse gases, thereby exacerbating the carbon foot-
portfolio, covering both non-durable and durable end uses, print, as furnaces are commonly fuelled by fossil fuels. The
positions ethylene as a benchmark for gauging the overall frequent decoking process also introduces further emissions,
performance of the petrochemical industry. compounding an already elevated environmental impact.
To mitigate these financial losses, it is crucial for plant
Furnace efficiency operators to implement effective monitoring, maintenance,
In the complex realm of ethylene production, furnace efficiency and operational strategies to prevent or address issues
plays a pivotal role in determining overall operational success. such as coke formation and elevated TMTs in a timely man-
However, a significant challenge arises from shortened run ner. Regular inspections, proper decoking procedures, and
lengths, increasing downtime and production costs. This issue adherence to best practices in furnace operation can contrib-
is rooted in the increasing tube metal temperature (TMT) of ute to improved efficiency and extended run lengths.
the furnace, a critical factor affecting operational longevity.
A decrease in run length in an ethylene furnace due to coke Differentiator
formation and a rise in TMT can result in significant financial Understanding and addressing elevated TMTs is key to
losses for several reasons. Here are some potential factors overcoming these aforementioned challenges. Dorf Ketal’s
contributing to the financial impact: proprietary artificial intelligence (AI) solution CokeNil offers
• Production interruption: Reduced run length means the solutions to extend run lengths through predictive mainte-
ethylene furnace is not operating at its optimal capacity for nance, optimised process control, and improved fault detec-
the intended duration. This interruption in production can tion. It is a data-driven, deep domain insight-based solution
lead to lower yields of ethylene and other desired products, where every process parameter is evaluated thoroughly in
resulting in lost revenue. the exploratory data analysis phase to ensure the accuracy of
• Increased decoking: Coke formation and elevated TMT the outcome. It is built on advanced deep learning methods
may call for frequent decoking, resulting in higher energy such as long short-term memory (LSTM) networks, Random
costs and reduced capacity utilisation rates. Frequent decok- Forest regression, CatBoost regression, XGBoost regression,
ing can also affect the overall reliability of the tubes, compro- and time series algorithms to help the model identify the pat-
mising the life span of radiant tubes. tern and behaviour of each critical process parameter.
• Energy consumption: A less efficient furnace may require CokeNil is a unique AI solution for optimising the furnace
more energy to maintain the desired operating conditions. run length. This plant’s distributed control system (DCS)
Higher energy consumption not only leads to increased feeds live operating conditions (such as naphtha feed com-
operational costs but also contributes to environmental con- position, coil outlet temperature, steam-to-hydrocarbon ratio
cerns if the energy source is not sustainable. [SHC], temperature, and fuel flow) to the CokeNil. These are
1100
1050
1050
1000
950
950
850 900
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33
Run length Run length
Actual TMT Predicted TMT Actual TMT Forecasted TMT
processed by the AI model, and optimal values that need model building includes critical process parameters identifi-
to be set in the real process are recommended. A key point cation, preliminary investigations, pattern discovery, anom-
is that the optimal values will never violate the acceptable aly spotting, hypothesis testing, and establishing correlations
operational ranges of suggested critical parameters. between process parameters and run length by considering
AI algorithms equipped with real-time data analysis TMT as the primary target.
capabilities play a crucial role in predicting and proactively AI leverages historical data to train the run length optimi-
addressing potential furnace issues. By dynamically adjust- sation model. Coking in the radiation coil is inevitable at high
ing parameters and ensuring adaptive control, the AI system cracking temperatures. As TMT reaches maximum threshold
prevents runway TMT and mitigates stress on critical com- values, the furnace undergoes decoking, compromising run
ponents. This dynamic optimisation lays the groundwork length. Furnace run length is directly linked to coke forma-
for extended operational runs, minimising the interruptions tion rate, with operational parameters such as TMT, dilution
caused by unplanned downtime. steam ratio, firing rate, feed composition, feed rate, wall and
floor burners in operation, excess oxygen %, coil pressure
AI/digital twins symbiosis ratio (CPR), coil outlet pressure and venturi ratio determining
AI-driven optimisation plays a key role in ensuring stable fur- the end of furnace runs.
nace operation with reduced delta pressure and dilution steam, In response to these intricacies, the furnace is operated, and
as well as an optimised feed ratio. This leads to a more con- parameters are controlled to minimise the rate of increase in
trolled furnace temperature, contributing to operational stead- TMT throughout the run. This proactive approach mitigates
iness. The nuanced control provided by AI not only enhances the rise in TMT, ultimately contributing to longer furnace run
operational efficiency but also contributes significantly to the lengths and improved operational efficiency.
overall stability of the ethylene production process.
The discussion takes an intriguing turn with the introduc- TMT prediction model
tion of advanced process modelling, including the concept of The CokeNil TMT prediction model has demonstrated a com-
digital twins. This innovative approach enables virtual test- mendable accuracy rate of 98.2%, underscoring the efficacy
ing and scenario optimisation, allowing operators to explore of our AI system. This high level of precision substantiates
various conditions without impacting the physical furnace. the model’s reliability and underscores its potential impact
The symbiosis of AI and digital twins not only optimises cur- in practical applications. It is imperative to recognise that
rent operational parameters but also lays the groundwork for model success extends beyond accuracy alone, encompass-
future advancements in ethylene production processes. ing aspects such as generalisation to unseen data and robust
However, challenges persist, and historical data serves performance across diverse scenarios.
as both a hurdle and a guiding light. The initial phase of AI Furthermore, a holistic assessment that includes metrics
like precision, recall, and F1 score contributes to a more com-
prehensive evaluation of the model’s efficacy. The observed
1150
effectiveness of the base model in predicting TMT for opti-
1100
mal parameter recommendation holds promise for positively
1050 influencing associated processes and systems. Continuous
1000 monitoring, evaluation, and potential retraining with new data
950 are pivotal for sustaining and improving model performance.
Figure 1 shows the results of the TMT prediction model.
900
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 Whenever a new furnace run starts, the CokeNil model
Run length captures critical parameters, which are the base for the model,
Actual TMT Forecasted TMT and generates optimal values for subsequent furnace runs.
The generated optimal values are typically in the acceptable
Figure 3 Scenario 2 where furnace run length is extended operational range, which the plant operator can incorporate
by 18 days into the actual process and see the TMT improvement.
INTERES
Optimize crystaphase.com
crystaphase.com
Bring us your gunky animal fats, your nasty
waste oils yearning to be clean. ActiPhase®
active filtration technology is a
proven performer in some of the
world’s biggest renewable diesel
fuel applications. Don’t trust your
reactor to a filtration newcomer.
Party with the pros.
Optimize
crystaphase.com
Figure 4 CokeNil dashboard
• The ‘Total Naphtha Vs Time’ graph shows the amount of rate for the AI’s recommendations regarding TMT, indicating
naphtha (feedstock) that has been used over time. their effectiveness.
• The ‘COT Vs Time’ graph displays the COT of the furnace. • Similar structure for CPR and SHC ratio: These follow the
• The ‘Economics’ section highlights the economic benefits same pattern as TMT, analysing and potentially optimising
of using the AI model, such as increased production and parameters for improved performance.
reduced costs.
Possible implications
CokeNil alerts and alarms • CokeNil’s AI tool plays a crucial role in analysing plant per-
These key elements of the dashboard include: formance, predicting key parameters and suggesting opti-
• TMT prediction and possible alert: Forecasts the TMT mising actions.
and generates alert to take necessary action. • By optimising TMT, CPR, and SHC ratio, CokeNil could
• Alert for critical parameters: If any critical parameters potentially improve process efficiency, reduce waste, and
exceed safe limits, an alert will be triggered, indicating enhance safety.
a potential concern. It is important to monitor and con- • The high recommendation acceptance rate suggests user
trol these parameters to ensure efficient operation of the confidence in the CokeNil insights and its potential to deliver
furnace. positive outcomes.
All triggered alarms due to specific conditions will be
emailed to the relevant authorities. This ensures that poten- Conclusion
tial risks are addressed efficiently, contributing to a seam- It becomes imperative to strategically optimise operations
less and well-managed process. within steam crackers, with an emphasis on enhancing
furnace efficiency to address the economic constraints and
CokeNil what-if analysis ensure sustainable performance. Central to this discussion
The following elements are identified by the what-if analysis: is the continuous learning capability of AI, empowering the
• Plant admin – diagnostic analysis: Explains the overall system to evolve and adapt based on new data and expe-
purpose of the dashboard. riences. In the context of ethylene furnaces, this continuous
• Status of the furnace (online/offline): Indicates if the plant learning contributes to data-driven decision making. The AI
is operational or not, which can affect the data displayed. system becomes not just a static tool but also a dynamic
• Actual vs predicted TMT: Compares real-time TMT with partner in improving furnace performance and supporting
the AI’s prediction, aiming to optimise TMT for efficient and extended run lengths.
safe operation.
• Specific input prompts: These allow the manual input of
specific feedstock or cycle time data for analysis. Surabhi Thorat is Head of Data Science at Dorf Ketal Chemical (I) Pvt Ltd.
• Recommendation and download: The AI suggest adjust- Vivek Srinivasan, is Senior Manager, Global Technical Services at Dorf
ments based on the TMT comparison, which can be down- Ketal Chemicals India Private Limited. Email viveks@dorfketal.com
loaded for implementation. Sudarshan Vijayaraghavan is District Manager at Dorf Ketal Chemicals
• Recommendation acceptance: Shows the user acceptance PTE Ltd.
LEADERSHIP
POSITIONS
ACTIVE
MEMBERSHIPS
VALUE
AMPP AMPP More than just general knowledge as to
EXPERTISE what the standards specify; we provide
API API expert insight as to why.
CSA
LEADERSHIP
Gain deep industry knowledge without
COST the time and cost to attend hundreds of
DOE
POSITIONS SAVINGS industry code committee meetings each
year.
GPA
(Former) We promote code development and
MTI ADVOCACY improvements based on client feedback
API and industry needs.
ASME PVP
ASTM US TAG
OTHER
BECHT.COM/CONTACT
/COMPANY/BECHT
P
art 1 in PTQ Gas 2024 discussed the importance of If strainer isolation valves are not installed, or another
a properly conducted dryout for cryogenic gas plants means to direct flow through the stagnant pipe loop is
during start-up, the most common dryout options, and not provided, portions of the loop will likely see smaller
the general design features needed for a closed-loop recircu- amounts of dryout flow (or no flow at all), potentially leav-
lation dryout. Closed-loop recirculation dryout performs best ing unknown water in the piping and equipment. Any flow
at effectively removing water from the system while easily path that leaves the fractionator and then returns to the
monitoring dryout progress This allows for a quick transition fractionator will be stagnant. Therefore, consider the same
to plant cooldown after dryout is completed. Part 2 will cover dryout approach as for a thermosyphon reboiler loop.
some of the design challenges, the method for implementing
a closed-loop recirculation dryout, and monitoring progress.
If strainer isolation valves are not
Cold plant dryout challenges
Typical cold plant stagnant areas include:
installed, or another means to direct
• Thermosyphon reboiler loops flow through the stagnant pipe loop
• Reflux system (propane recovery plant) is not provided, portions of the loop
• Lower fractionator section (below expander outlet feed).
will likely see smaller amounts of
Thermosyphon reboiler loops dryout flow (or no flow at all)
There is no obvious location to introduce dryout gas flow
into a thermosyphon reboiler loop unless isolation valves
are installed around the strainers for the exchanger. Some Reflux system (propane recovery plant)
cold plant designs include isolation valves for the strain- Many technologies for propane recovery have a more com-
ers to have the capability for removing a plugged strainer plex reflux system and can be difficult to dry out without
without depressuring and clearing a much larger plant sec- proper dryout connections. A section of the reflux system
tion (if not the entire cold plant) of hydrocarbons. This is from Figure 7 in Part 1, which typically is stagnant during
the easiest location to introduce dryout flow because the dryout, is shown in Figure 2. A 2in minimum dryout con-
isolation valves can be used to direct flow where needed. nection is recommended.
For this case, connect the dryout gas source between the The lowest point in the reflux system is the reflux pumps.
exchanger start-up screen isolation valves (see Figure 1). A The reflux system should be designed to free-drain to the
2in minimum dryout connection is recommended. lowest point. The dryout gas source can be connected to
Fractionator
Reflux
accum.
Dryout
gas supply
Dryout
Reboiler gas supply Reflux to
pass fractionator
Strainer
Low-point drain Reflux pump
Liquid product
Low-point drain
Figure 1 Possible dryout connection for reboiler loops Figure 2 Dryout connection for reflux system
Dryout A A
gas supply
Inlet
Liquid product gas
Cold plant
Low-point drain
B B
Lower fractionator section (below expander outlet feed) per million by volume (ppmv). At least one bed should be
The lower section of the fractionator column below the online with one bed in regeneration (assuming a two-bed
expander outlet feed requires an external flow source to design).
remove water from the mass transfer equipment inside • The dehydrator dust filters must be available for service.
the column. A pipe connecting dryout gas to the column A typical design includes two full-flow filters (one in service
bottoms outlet piping should be designed and installed, as and one off-line).
shown in Figure 3. The piping and connection size must • All control valves and manual valves which would restrict
be large enough for this dryout connection to be effective. flow through the main flow path are open, except the J-T
For a 200 MMSCFD (5.4 x 106 Nm3/hr) cold plant, a 4in valve.
connection is recommended. As the cold plant nameplate • The fractionator bottom is isolated from its downstream
capacity decreases, so can the dryout connection size, but equipment.
use at least a 2in connection. • The expander/booster compressor is completely isolated,
For propane recovery plants, it is important to note with flow through the J-T valve and around the booster
that any water remaining inside the de-ethaniser column compressor through its bypass line.
after cooldown will become trapped, eventually freezing • The moisture analyser used to monitor the cold plant
because the column overhead is colder than the freezing water content is in service.
point of water, and the column bottom is warmer than the
boiling point of water. System design limitations
The following design limits must be adhered to keep from
Location to monitor cold plant water content damaging process equipment during dryout:
A sample location for monitoring the wet gas water content • The recommended maximum dryout gas feed tempera-
at the outlet of the cold plant upstream of the residue gas ture to the cold plant is 140°F (60°C), limited by the brazed
compressor must be available (see Figure 4). The wet gas at aluminum heat exchangers (BAHE) design temperature
this location is representative of the amount of water pres- of 150°F (65°C). Warmer dryout gas is better. Adjust the
ent in the main dryout circulation loop. Any water carried louvres on the residue gas compressor discharge cooler to
away by the warm dryout gas that will then be removed by provide a temperature near the maximum.
the dehydrators can be measured at this point. • Understand the flow rate limitations, which would result
in reaching the maximum velocity limitations across major
Before starting dryout process equipment (dehydrators, filters, and mass transfer
Ensure the following before beginning a cold plant dryout: equipment) during dryout because of system flow at lower
• The pressure-reducing device is installed at the desig- pressure. However, while it is prudent to know the plant
nated location. process design limits to keep from damaging equipment,
• The residue gas compressor is operational and running the maximum flow calculated to limit J-T expansion through
in total recycle before starting dryout gas flow through the the cold plant will most likely be less than the process
train. equipment maximum velocity limits.
• All dehydrator beds are available for service in their nor-
mal cycle. Each bed should have already had at least one Dryout method
successful regeneration cycle, and the water content of At this point, all process requirements should be met so
the dry gas exiting each bed should be less than 1 part dryout can begin.
Conclusion
UOP Ortloff has witnessed many cold plant dryouts during
commissioning and initial start-up over the years (see Part
1 in PTQ Gas 2024).
A well-conducted cold plant dryout will help plant com-
missioning and ensure start-up goes smoothly and does
not last any longer than necessary. There are multiple
dryout options which can remove water from the system.
However, there are drawbacks to some of them.
The closed-loop recirculation dryout does the best job at
effectively removing water from the system while easily
monitoring dryout progress and allows for a quick transi-
tion to plant cooldown after dryout is completed.
Process integrated
Scott A Miller is a Principal Process Engineer with Honeywell UOP, solutions for Ejector
supporting Ortloff Cryogenic Gas Processing Technologies. He has
more than 23 years of experience in the petrochemical and gas pro-
Vacuum Systems
cessing industries, of which 15 are in the design and operation of NGL/
LPG recovery plants. Email: scott.a.miller@honeywell.com
GEA steam Jet Vacuum Systems help optimize
David A Jelf is a Principal Instrument and Controls Engineer with
production processes, reducing both costs
Honeywell UOP, supporting Ortloff Cryogenic Gas Processing and carbon emissions.
Technologies. He has more than 40 years of experience in the design Upgrade your system now!
and operation of NGL/LPG recovery plants.
J A Anguiano is a former Engineering Manager with Honeywell UOP,
previously supporting Ortloff Cryogenic Gas Processing Technologies.
He has more than 25 years of experience in the design and operation
of NGL/LPG recovery plants.
Joe T Lynch is a retired Engineering and Licensing Manager from
Honeywell UOP, previously supporting Ortloff Cryogenic Gas
Processing Technologies. He has more than 45 years of experience in
the design and operation of NGL/LPG recovery plants.
www.digitalrefining.com
IMPROVE SUSTAINABILITY:
OUR MISSION
ITW Online Cleaning clean an entire Unit in as little as 24 hours on a feed-out/feed-in basis.
ITW Onstream Cleaning clean an entire Unit during the run.
Our proprietary chemistries dissolve and stabilize any asphaltenic/paraffinic/polymeric deposits, by transforming the same into a fully
reusable/reprocessable liquid.
When applied to tank cleaning, our technologies effectively, safely and quickly recover oil from sludge, thereby eliminating waste generation
and disposal.
Pro-active application of our cleaning technologies reduces CO2 , VOCs and greenhouse gas emissions, while reducing energy consumption
thereby getting additional value by improved Operational Excellence.
ITW proprietary technologies promote safe working, as they eliminate/reduce mechanical cleaning operations as well as working in confined
spaces.
Our Improved Degassing/Decontamination technology has unrivalled performance all in terms of contaminants elimination, quick safe entry
achievement, waste disposal elimination and environmental compliance.
Our patented chemistry does not create any emulsion and fluids can be easily handled by Waste Water Treatment Plant.
Other proprietary technologies also include: Amine Unit Optimization, Sulphur Dissolution, Ethylene Furnace Run Length Increase, Cleaning of
Texas Towers/Packinox, Coke on catalyst Formation Reduction, Diesel/Fuel Oil Particulate and NOx Emission Reduction.
ITW technologies can be applied to all the Oil & Gas Industry, including the Transportation and Energy Production industries.
Sustainability improvement can be achieved by process optimization, without any major investment.
CONTACT ITW TO LEARN HOW IMPROVED SUSTAINABILITY CAN BE ACHIEVED NOW AT ADDITIONAL VALUE
For more information contact: ITW S.r.l.- C.da S.Cusumano - 96011 Augusta - Italy Hiring
Tel. +39 (0931) 766011 Professionals
E-mail: info@itwtechnologies.com Worldwide
www.itwtechnologies.com
Join ITW Team worldwide and send your Curriculum Vitae to : jobs@itwtechnologies.com
E
merging corrosion concerns surrounding the pro- While these issues have been documented, some FCC
cessing of fats, oils and greases (FOGs) are highly WGC sections have recorded little to no problems. Process
paraffinic. They will vaporise and crack easily, but units with no problems have used highly refined or refined/
some naphthenic acid (TAN) may be present, particularly bleached/deodorised (RBD) FOGs. While FCC units have
in tallow or used cooking oil (UCO). TAN has been known co-processed up to 100% FOGs,1 trials seldom exceed
to increase corrosion in distillation towers and hydrotreat- 10% of the total feed and are commonly below 5%.
ing units. For FCC units, it is generally accepted that the
TAN compounds will crack and pose little harm. However, Catalyst’s role in FCC co-processing
little has been documented on the feed section’s corrosion FCC catalysts drive the conversion and selectivity for
effects before the riser. upgrading heavy feeds to value-added fuel and chemical
As a rule of thumb, the presence of TAN in co-processing products. Conventional FCC catalysts are formulated within
feeds follows the order of bio-oils > animal tallow > vege- the constraints of each unit to maximise product value and
table oils. Many refiners have a maximum TAN specification unit objective. Examples include optimisation for slurry
to reduce the corrosion in the main column overhead (MC
OVHD). A careful inspection of the MC OVHD is important
during the turnaround after co-processing. To maximise product value
Some co-processing feeds may contain carboxylic acids within the unit constraints and
and other oxygenated species. When these feeds crack, they
will form smaller, oxygen-containing compounds. Most of
drive towards incorporation of
these compounds will crack into the lighter boiling fractions renewable and circular carbon,
(wet gas compressor [WGC] section). Elevated carbon diox- the co-processing catalyst can be
ide (CO2), carbon monoxide (CO), and low levels of alcohols
are commonly seen. Other compounds observed are esters,
formulated and optimised for the
ketones, phenolics, and acids (acetic, formic, and propionic). specific feed chemistry
In most cases, the appearance of these compounds will
be in low concentrations (wppm or wppb levels). Better
analytical methods to measure them should be developed conversion, metals tolerance, coke selectivity, and maxim-
for co-processing. These acids may increase the corrosion ising gasoline and/or light olefins for a given feed, product
concerns in the MC OVHD section. Elevated methyl acrylate targets, and operational constraints.
gas has been observed in tallow-based feeds in general The same holds true for catalyst technologies for FCC
and UCO feedstocks in particular. Control of deoxygenation co-processing of alternative feedstocks. In addition to the
pathways, as discussed later, is crucial to minimising light different metals in the alterative feeds, the differences in the
oxygenate formation and the impact on these supporting hydrocarbon species present alter the conversion chemis-
operations. try. To maximise product value within the unit constraints
Light molecular weight (MW) products from co-process- and drive towards the incorporation of renewable and cir-
ing will primarily concentrate in the WGC section. The acids cular carbon, the co-processing catalyst can be formulated
will react with ammonia, and the pH of the systems will and optimised for the specific feed chemistry.
decrease. Methyl acrylate will most likely form solids. Amine
units will have problems with elevated CO and CO2 levels. Maximising deoxygenation
In general, foaming issues may occur in high-pressure pro- As previously stated, most units may only process small
cess units, and corrosion concerns are more common. quantities of FOGs. Typically, these units rely on the base cat-
Many industrial water and process chemical suppliers alyst already in the unit without consideration of the change
have developed improved practices when co-processing. in the feed chemistry, resulting in impacts on unit operations
70%
Influencing deoxygenation during Dehydration Biogenic coke and gum formation
co-processing 20% canola oil over
Hy
50%
dr
Decarbonylation
n
tio
on
n
r
ca
30%
De
Figure 2 Oxygen balances during co-processing 20 wt% canola oil over various grades of ReNewFCC formulated for
maximum gasoline, illustrating the role of altering the catalytically active sites to influence deoxygenation pathways and
the generalised impact on the FCC
1 Gasoline
∆Yield (%)
-1
-1.5
2
Iso-Conversion = 74 wt%
-2.5
Figure 3 Delta yields during co-processing of 20 wt% soybean oil in VGO for Ketjen’s baseline max olefin catalyst
technology and ReNewFCC formulated for max olefins
1.1
1 0.5 0.7
Dry gas LPG LCO Coke CTO Propylene
0
-0.3
Gasoline Bottoms -0.1 -0.3 -0.3 -0.1
-0.5 -0.5 -0.5 C4 Olefins
-1
-1.1 -1.0
-1.6 -1.4 -1.6
-2
-3
3
2.3 2.3 20% WPO effective conversion - VGO @ 75% conversion
20% PS_Corr
2
20% PP_Corr
∆Yield (%)
1.1 1.1
1 CTO 0.9
-1.3
-2 -1.8
-2.4
-3
Figure 4 Delta yields comparing co-processing of 20 wt% PP-WPO in VGO and 20 wt% PS-WPO in VGO to a 100%
VGO baseline using a baseline Ketjen catalyst techology. The top plot uses the standard 430F+ conversion to compare
the changes in the product slate at an iso-conversion of 75%. The bottom plot uses the effective conversion to determine
the impact of introducing naphtha fractions from the WPO into the feed
This conversion subtracts the fraction of naphtha initially However, the PP-WPO, comprised mainly of iso-olefins,
present in the feed, leading to a more accurate interpreta- results in increased C₄= and a reduction in coke.
tion of the impact of catalyst and feed on product distribu-
tion during WPO co-processing. Flexibility
Applying this effective conversion leads to the observa- FCC units offer almost unparalleled flexibility for pro-
tion that the slurry yield (bottom of Figure 4) is lower during cessing a wide diversity of feedstocks. The International
WPO co-processing compared to 100% VGO processing, Sustainability and Carbon Certification (ISCC) is a good
and clear differences in the product slate for the different indicator of current trends for processing renewables and
WPOs emerge. recyclables (R&R). Reliable catalyst solutions to address the
The addition of PS-WPO shows a slight reduction in introduction of a wide variety of R&R are being considered
LPG, LCO, and bottoms, while there is an increase in coke. and investigated for FCC co-processing.
Compared to co-processing PP-WPO, there is an increase in Against this backdrop, ReNewFCC offers the same
LPG that is mainly a result of increased C₄ olefins. The feed degree of flexibility and effectiveness in addressing dif-
impact differences between PS and PP-derived oils can be ferent operations, feeds, and yield objectives featured in
directly related to their hydrocarbon composition. Due to the Ketjen’s established portfolio of more conventional FCC
aromatic nature of the PS-WPO, there is minimal impact catalysts, building on a foundation of industry-leading bot-
on LPG yield, LPG olefins, and increased coke selectivity. toms upgrading and metals tolerance to create solutions
3
Standard Technology-20% PP-WPO vs Standard Technology-VGO
2.5
ReNewFCC CircularMax -20% PP-WPO vs Standard Technology-VGO
2
1.5
∆Yield (%)
0.5
Coke LCO Dry gas H2
0
650F+ C4= C2=
-0.5 C3=
Gasoline LPG
-1
-1.5
Figure 5 Delta yields from processing 100% VGO over a standard light olefin catalyst technology, demonstrating the
comparative impact of co-processing 20 wt% PP-WPO over the standard and ReNewFCC catalyst technologies at a
standard (430F+) conversion of 75%
The highest
sample throughput.
The fewest worries.
Get rapid and continuous testing of up to
12 samples with the new AS Vision autosampler
• Large sample port numbering for mistake-free setup
• Innovative metal distribution block for lowest sample carryover
• Fail-safe operation and corrosion-resistant construction
• Fewer rinsing cycles for significantly reduced cross-contamination
GRABNER INSTRUMENTS® Messtechnik GmbH A-1220 Vienna/Austria Dr. Otto-Neurath-Gasse 1 | +43/1/282 16 27-0 | www.grabner-instruments.com
PETROLAB COMPANY (USA) 2001 North Indianwood Avenue, Broken Arrow, OK, 74012, USA | +1-918-459-7170 | www.petrolab.com
©2023 AMETEK, Inc. All rights reserved.
A
dividing wall column1 (DWC) is a distillation column petrochemical, and chemical applications. Recently, they
combining the operation of two conventional col- have even been implemented for reactive distillation. More
umns in one shell. The separation of a three-compo- complex DWCs (Dual3 DWCs) can be used for the separa-
nent mixture into individual components is made possible tion of a stream in four cuts.
by the installation of a vertical wall in the middle of the col-
umn. The feed enters the column on one side of the wall, DWC: Benefits and limitations
and on the other side of the wall, a high concentration of From a unit operation point of view, a DWC eliminates the
the middle boiling component is achieved. thermodynamic inefficiency of the two-column setup,4
The result is that three components can be sharply where the remixing of the middle and heavy boiling com-
fractionated in one single column. In the conventional ponents at the bottom of Tower 1 cannot be avoided (see
setup, two columns, two reboilers, and two condensers Figures 1 and 2).
are needed: the DWC needs only one vessel, one reboiler, DWCs allows for several benefits5 compared to the
and one condenser. The DWC has shown its advantages two-column setup:
in many applications, typically for liquid mixtures of equal • Up to 40% reduction of investment cost
amounts of three volatile components and for cases • Up to 40% reduction of operational cost
where a high amount of middle boiling component must • Up to 40% additional theoretical stages
be separated from smaller amounts of light and heavier • Up to 50% energy savings
components.2 • Up to 40% reduction of plot plan space and consistent
Rule of thumb: The bigger the side-stream portion of the reduction of greenhouse gas emissions.
feed, the higher is the effectiveness of the dividing wall. DWCs also have limitations compared to the two-column
DWCs have been implemented in a large variety of refinery, setup:
Qc1
Tower T1
T1 Top
A
Qc2 A
T2
B
Feed
A, B, C
Number of trays
Qr1 B
C
B, C
Remixing
Qr2
Bottom
0 0.2 0.4 0.6 0.8 1
C
Composition mole fraction
Number of trays
Number of trays
Feed T1
A, B, C B
B B
C C
Qr
Bottom Bottom
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
C
Composition mole fraction Composition mole fraction
• Operational pressure variation between column sections The column is heat integrated with the FCC unit. The
is not possible heat duty to the column is provided by two reboilers: upper
• Increased temperature difference between reboiler and reboiler, where the heating medium is the light cycle oil
condenser (LCO) coming from the reboiler of the debutaniser, and lower
• Increased pressure drops reboiler, heated with the heavy cycle oil (LCO) coming from
• Increased length and diameter of the vessel the FCC main fractionator. The top condensation system
• Increased complexity for modelling in the process provides the cooling duty to draw off the light naphtha and
simulation generate the internal reflux required for the fractionation of
• Increased complexity for the hydraulic design of the mass the cuts. The feed is preheated at the preheat exchanger
transfer components by means of the side-cut draw before entering the tower in
• Lower operational flexibility. the middle section (see Figure 3).
The three cuts are further processed to the downstream
Existing naphtha splitter MOG and SCANfiner units to meet the quality of the gas-
The existing naphtha splitter is a conventional distillation oline pool.
column with a side draw. It splits the full naphtha coming
from the bottom of a depentaniser located at the GasCon Revamp motivation
of the fluid catalytic cracking (FCC) unit, into three streams: All three cuts were initially used for gasoline. The existing
light naphtha, middle naphtha and heavy naphtha. tower was able to achieve the required performances both
1.2 0.8
XX Pressure kg/cm2g
To De-C5 reboiler
YY Temperature ˚C
CLEAN PROCESS
+ CLEAR PROGRESS You strengthen your plant’s safety, productivity
and availability with innovations and resources.
10 ˚C 121.2
From De-C5
bottom Middle naphtha to
30 ˚C 127
138 Unionfining/Platformer
50 ˚C 134.1
70 ˚C 145.3 LCO from De-C4
reboiler
90 ˚C 163.9
95 ˚C 173.4 To De-C5 XX Pressure kg/cm2g
reboiler
FBP ˚C 181.3
YY Temperature ˚C
Figure 4 Required middle cut quality as feed for the Parex unit
from the fractionation efficiency and the hydraulic capacity separation efficiency and produce three cuts with the fol-
points of view. Several years later, after the first start-up, the lowing distillation boiling points (as D86):
tower was called for a different duty: top and bottom cuts • Light naphtha with a 98°C final boiling point.
were still pro-gasoline, and the middle cut had to be used as • Middle naphtha with 110.5°C initial boiling point, and
feed for the downstream Parex unit after being processed at final boiling point lower than 182°C for Case 1 and lower
the Unionfining and Platformer units (see Figure 4). than 179°C for Case 2.
A sharper separation between the three cuts is required • Heavy naphtha with initial boiling point >144°C at 300 t/h
compared to the original design. To accomplish the new feed flow rate.
service, several options were investigated, among others: • An operating range of 110-50% of design.
Shifting the upper reboiler duty to the bottom reboiler
to maximise utilisation of the stripping section fractionation Major constraints
trays. In attempting to minimise investment costs, the revamp
Increasing overall reboiler and condenser duties by was implemented while considering several major con-
20% over the original design to maximise the reflux ratio in straints: reuse the existing equipment as much as possible,
the attempt to at least meet middle-cut specifications. minimise additional equipment, minimise tower modifica-
Shifting the column’s feed location downward. tions, and minimise modifications at the existing control
Optimising the side-cut draw-off location. system.
None of the above options could be even close to the A very narrow time frame for the project execution was
required qualities of the three cuts: defined: licence, basic, and detailed engineering, manu-
• Insufficient separation between light naphtha and mid- facturing, and delivery of goods at the site in 18 weeks.
dle naphtha, resulting in higher C6 content than desired in Implementation of the project at the site in less than 20
the middle naphtha. working days within the time frame of the planned unit
• Poor separation between middle naphtha and heavy turnaround. All these items were under the responsibility of
naphtha, resulting in higher C10+ content than required in the one single company, GTC, now Sulzer CFCL.
middle naphtha. It would cause over-coking at the down-
stream Platformer reactors with consequent high deactiva- Process simulation results
tion rate of the catalyst and lower yields of valuable aromatic While simulating a middle DWC, several models can be
components. used. The most common are:
Adding another distillation tower would achieve the Four columns interconnected: The first column mod-
target, but it was not even considered due to the high els the section of the tower above the partition baffle. The
investment cost. To minimise the investment cost while second column models the section of the column equipped
maximising energy savings and minimising GHG emissions, with the baffle and facing the feed. The third column mod-
it was decided to convert the tower into a DWC. els the section of the column equipped with the baffle and
facing the middle-cut draw-off. The fourth column models
Revamping scope the section of the column below the partition baffle.
The main revamping target was to increase tower Two columns interconnected (see Figure 5): The first
Z414_S17
501
Feed B
408
S14
!
Qr
Z414_C11A
Z414_C11B 529
S7
C S5
column models the section of the column equipped with transfer components and associated tower internals of a
the partition baffle and facing the feed, also called the pre- DWC.
fractionator. The second column models three sections, It must be underlined that this is the most critical step
from top to bottom: the section of the column above the while designing a DWC. The offset location of the dividing
partition baffle; the section of the column equipped with wall to the axe of the column and the mass transfer com-
the partition baffle and facing the middle-cut draw-off; and ponents at the right and left of the baffle should be prop-
the section of the column below the partition baffle. This erly designed to achieve the performance of the tower, as
second column is also known as the main fractionator. This expected from the process simulation model.
second model is also known as the Petlyuk6 configuration, Ideally, the pressure drop at the left and right of the baf-
in honour of the author who first studied and discussed this fle will be the same so that the vapour coming from the
topic in detail in 1965. section below the baffle will split between the prefraction-
At Sulzer, Petlyuk‘s model is preferred. In addition, an ator and the main fractionator, as expected by the process
in-house optimised model has been developed. It allows for simulation model. In real life, this may not be possible due
a fast, solid, and reliable convergence of the process sim- to the required operational flexibility, not only in terms of
ulation. The output results can be automatically imported operating range (110-50% of design capacity) but also
into a dedicated hydraulic rating tool for the design of mass in terms of cases with different feed composition and or
D 86 % vol
Light naphtha
IBP ˚C 70.6 52.2 110.5 144.7 1.3 to SCANfiner
16.5
5 ˚C 86.4 63.7 113.7 157.2
10 ˚C 90.7 67.4 115.6 163.5 Middle naphtha to
138 Unionfining/Platformer
30 ˚C 105.0 70.7 122.4 172.1
13 161
50 ˚C 126.5 74.3 131.9 178.8
70 ˚C 150.0 78.9 143.6 187.2
90 ˚C 175.7 84.9 159.2 201.5
XX Pressure kg/cm2g
95 ˚C 184.8 89.0 165.6 208.3
FBP ˚C 198.0 97.2 181.0 214.6 YY Temperature ˚C
18 ex41
ZZ Duty MMKcal/h
GAP ˚C 24.7 -8.4
1.5 217
Heavy naphtha to SCANfiner
LCO LCO
ex41
HCO HCO
Heavy naphtha Heavy naphtha
to SCANfiner to SCANfiner
product qualities. Also, the extension of the dividing wall side of the wall and at the prefractionator above the feed.
should be properly designed to secure enough spare theo- Four-pass high-performance trays9 have been installed in
retical stages to mitigate the effects of the deviations from the prefractionator below the feed. A tailor-made chimney
ideality. tray has been installed above the partition baffle to control,
by gravity, the internal reflux split between the prefraction-
Post-revamp process simulation results ator and main fractionator side of the dividing wall.
A simplified process flow diagram for Case 1 is shown in The existing middle-cut draw-off nozzle has been
Figure 6. This includes the material balance, the qualities of blanked off and relocated a few trays above the existing
the three cuts, and the main operating data (pressure pro- one. A chimney tray has been installed below the bottom
file, temperature profile, and duties at both the reboilers). partition baffle to collect the liquid coming from both sides
In particular, the revamped naphtha splitter converted to of the wall and feed the distributor above the bed of struc-
a DWC allows for an almost 25°C gap between light and tured packing. One new reboiler has been added at the bot-
heavy naphtha. The overlap between middle and heavy tom, and a new bottom pump has been added for the heavy
naphtha is as low as 8°C and consistently lower than the naphtha draw-off, with associated piping and fittings.
existing setup. Also, the content of C10+ aromatic compo- No modifications were required at the top condenser sys-
nents is far below the limits required by downstream units. tem, at the upper reboiler or at the feed preheat exchanger.
Two new provisional nozzles have been installed above
Major units modifications the partition baffle and blanked off. The purpose of these
Figure 7 shows the tower before the revamp on the left nozzles is for external reflux control (for future implemen-
side, and the tower after the revamp on the right side. For tation), but only in case the tower could not be operated
easy interpretation, the new and/or modified equipment is smoothly with the supplied internal reflux control system.
red in colour. The existing unit control system has been maintained
The tower has been completely retrofitted from top unchanged. All tower modifications and external equip-
to bottom by replacing existing conventional trays with ment were implemented in 20 working days within the time
high-performance trays7 and high-performance structured frame of the planned unit turnaround.
packing at the bottom of the tower. Two partition baffles
have been installed: one in the middle-bottom portion of Conclusions
the large ID section of the tower, and one at the top of the The DWC transformation allowed for a consistent improve-
small ID portion of the column. To minimise the welding to ment of the fractionation of the three cuts, particularly
the tower wall with the final goal of minimising the instal- between middle and heavy naphtha: from a 40°C overlap
lation time at site, a tailor-made partition baffle has been to a 2°C gap.
implemented, also known as the ‘built-in dividing wall’. The unit ran for more than seven years after the revamp
The built-in dividing baffle is made by a combination of without any significant issues. Last year, during the
thicker plate beams and lightweight bolted panels particu- planned shutdown, the column was opened for inspection,
larly suitable for revamp projects. Two-pass high perfor- and all the tower internals were found in place without any
mance trays8 have been installed in the main fractionator damage.
ÞƷټȵƷژǠȄɫƷȽɋǠȄǒژǠȄژ
FژȽȏǹɓɋǠȏȄȽِ
ɓȵژȄƷɬژƌɲȲȏȵɋً¾ژƷɱƌȽژǠȄɫƷȽɋȂƷȄɋژɬǠǹǹي
UȄƩȵƷƌȽƷژȏɓȵژȽɓȲȲǹɲژȏǑژî°uژדٮɋȏژȽɓȲȲȏȵɋژ
ȵƷ˟ȄƷȵȽژټȲȵȏȲɲǹƷȄƷژȲȵȏưɓƩɋǠȏȄ
-ȄȽɓȵƷژƌƩƩƷȽȽژɋȏژȵƷǹǠƌƨǹƷًژǚǠǒǚٮȴɓƌǹǠɋɲژ
FژƩƌɋƌǹɲȽɋȽژƌȄưژƌưưǠɋǠɫƷȽژ
-ȄǚƌȄƩƷژȲȵȏưɓƩɋǠȏȄژȏǑژȏƩɋƌȄƷژƨƌȵȵƷǹȽژ
ǑȵȏȂژFًژƌǹǵɲژƌȄưژu¾-ژɓȄǠɋȽ
FȏǹǹȏɬژɓȽژȏȄژkǠȄǵƷưUȄژǑȏȵژȂȏȵƷژɓȲưƌɋƷȽٍ
E
nergy consumption is a key cost factor in refinery Rotating equipment issues significantly impacted asset
economics. About half of the refiner’s operating costs availability and throughput. This issue led to decreased
are dedicated to meeting the demands of energy profit margins and hindered the site’s ability to capitalise
consumption.1 Adding to this complexity, a Reuters report on a high-margin environment. The emphasis was on iden-
revealed that 43% of surveyed global workers plan to leave tifying reasons for underperforming asset reliability and
the energy sector by 2025.2 building consensus on corrective actions.
Amidst this financial and labour landscape, a mid-sized While historical underinvestment played a role in current
refinery confronts rising energy costs, workforce shortages, performance levels, the primary issue stemmed from a lack
and the pressing need to meet net zero global commit- of root cause analysis and follow-up actions, combined
ments by 2050. In addition, this journey unfolds in a setting with a lack of start-up readiness.
of diverse perspectives and agendas among the leadership In Europe, rising energy costs have heightened the focus
team and board of directors. Despite these obstacles, the on energy optimisation for many assets.3 A strategic energy
refinery committed to cutting Scope 1 and 2 emissions by and maturity review identified that the refinery overlooked
30%, benchmarked against a 2019 baseline, as depicted in the latest technology, monitoring tools, and efficient prac-
Figure 1. tices for energy management. Relying primarily on the
Operating in a rapidly changing business environment, operators’ tacit knowledge, the asset fell short of industry
the board sought to balance profitability by understanding benchmarks, including KBC’s proprietary Best Technology
how the precise product mix, customer base, feedstocks, Index. However, through comparative analysis, the team
and technologies interact. To reinforce this initiative, the identified quick-win opportunities that required minimal or
refinery’s Chief Operating Officer (COO) emphasised, “We no capital expenditures to swiftly improve the asset’s per-
need to strategically invest for the long term, but we must formance. These opportunities not only delivered immedi-
also remain profitable in the here and now. If we cannot ate value but also contributed to the programme’s overall
generate a positive cashflow today, how can we possibly funding.
raise the funds for the investment we need to make?” Additionally, long-term, mid-to-high Capex improve-
Achieving a balance between present performance ments were identified and fed into the decarbonisation
and long-term strategic investments required alignment, workstream. Addressing no or low Capex opportunities
commitment, and a clear and executable plan. Optimising was integral to the refinery’s success in optimising energy
current performance while ensuring long-term success usage and achieving its goals.
required the refiner to keep one foot firmly planted in the
present and the other in the future.
120%
The present – Project Catalyst 100%
95%
100%
The refinery’s leadership team conducted a comprehensive 90%
85%
‘wall-to-wall’ operations assessment. This evaluation uncov- 80%
85%
ered several key focus areas, including soaring energy costs, 70% 55%
60%
reliability issues, and team demographic challenges. Over a 60%
six-week period, the following challenges were addressed: 40% 45%
• Asset reliability to overcome underinvestment Scope 1 & 2
20% Scope 3
• Energy efficiency for quick wins
0%
• Yield optimisation with digital twin software 0%
2019 2025 2030 2035 2040 2050
• Turnaround optimisation for decarbonisation planning
• Recruitment and organisation to mitigate challenges
• Technology for digital transformation. Figure 1 Scope 1, 2, and 3 emissions reduction targets
By creating a digital twin of the refinery, yield optimisation address this, several highly facilitated programme design
opportunities could be thoroughly reviewed. This assess- workshops were held. These workshops were deliberately
ment revealed numerous areas where operating parame- not high tech. The team intentionally used basic tools such
ters and procedures could be improved. Furthermore, the as marker pens, sticky notes, a long sheet of brown paper,
focus on no or low CapEx opportunities played an integral and coloured voting dots.
role to the programme benefits and funding, similar to Prioritising specific interventions was based on the cor-
energy efficiency opportunities. porate goals, legislative frameworks (such as maintaining
Refinery margin erosion is largely attributed to turna- a safe and compliant operation) and individual benefits
rounds.4 This programme aimed to review both the overall aligned with opportunities (such as maximising the trans-
turnaround philosophy and the detailed scope of the next formation investment). Engaging team members across
two turnarounds. The focus was to identify activities that the organisation led to the development of a practical pro-
could be descoped, moved off the critical path, and opti- gramme where each workstream was assigned monthly
mised. Given two scheduled turnarounds between the deliverables that were captured on a 4m-long poster. This
assessment and 2030, the turnaround review was inextri- poster, shown in Figure 2, served as a key management
cably linked to the decarbonisation planning activities. tool to communicate the programme’s progress and ensure
Many refiners struggle with dual challenges: an ageing that everyone from the control room to the board room
workforce and difficulty recruiting young engineers and understood the project.
skilled operators.1,5 In addition, the oil and gas industry
is frequently perceived as contributing to climate change The future – Project Horizon 2050
rather than being part of the solution in the energy tran- Building on the Project Catalyst programme, a similar exer-
sition journey.3 The scale of the problem was assessed, cise was conducted to ensure the refiner met their decar-
which has led to developing mitigation plans to address bonisation targets. This initiative resulted in Project Horizon
potential gaps over one-, three-, five-, 10-, and 15-year 2050. It should be noted that the ‘One Foot in the Present
time horizons. – Project Catalyst’ and ‘One Foot in the Future – Project
Technology is crucial for streamlining operations, rev- Horizon’ workshops were held concurrently. The COO’s
olutionising work processes, and supporting necessary vision was to transform the refinery into a future-ready
productivity improvements to mitigate ongoing resourcing asset capable of meeting customer and societal demands
issues. During this programme stage, the technology and in a sustainable, affordable, and reliable way, including a
digitalisation vision were outlined for the operating assets, decarbonisation roadmap for reaching net zero with tech-
consistent with the overall organisational vision. Having nology leading the way.
defined the vision, a gap analysis was conducted, leading The main objective of Project Horizon 2050 was to
to the development of a draft target application architec- develop an agreed-upon decarbonisation roadmap that
ture. This strategic approach ensured the technology was balanced short-, medium-, and long-term initiatives to
leveraged optimally to achieve the refinery’s goals. achieve the refiner’s decarbonisation goals. To secure the
After completing the assessment phase, the next success of the roadmap and achieve net zero emissions by
step involved defining the resultant programme, Project 2050, three pillars – strategy, operations and technology –
Catalyst. The process of defining the programme involved were defined, along with interim milestones.
overcoming the primary challenge of getting multiple The implementation of the refiner’s roadmap needed
stakeholders with differing perceptions regarding priorities, to be anchored on technology and aligned with the other
timescales, and resource requirements into agreement. To two pillars to ensure sustained actions and results. This
Johnson
Q2 PTQ Screens.indd
StaySafe Full page Ad 1
v4 Blurred Hero Image.indd 1 11/03/2024 11:27:33
3/8/2024 3:07:22 PM
approach meant balancing current operations with a for- utility production and supply, and process and utility-gen-
ward-looking strategy. erated emissions. Powered by a process simulator with
As outlined in Project Catalyst, the roadmap started with necessary modules, it addressed present and future energy
examining operations and implementing energy efficiency transition challenges. Moreover, the IP3EM model also
to provide quick wins while balancing emissions reduction assessed the economic impact of the different alternatives
with profitability. In this case, it involved reviewing existing being evaluated and guided the refinery throughout its
assets and operations to ensure optimal functionality. This entire decarbonisation journey.
would achieve around a 15% emissions reduction, at best The IP3EM revealed key benefits such as analysing refin-
20%. However, to reach the desired mid-30% emissions ery margins, return on investment using emissions reduc-
reduction target and net zero by 2050, a clear strategy, tion scenarios, and accurately identifying and quantifying
outside-the-box thinking, and alternative solutions were major sources of Scope 1 greenhouse gas (GHG) emissions.
needed to drive organisational change. Several projects were considered. The IP3EM was key to
To achieve this first milestone of 30% emissions reduction, assessing the impact of Scope 1 and 2 emissions, energy
the refiner built a green hydrogen plant to replace part of the consumption, and economics regarding furnace electrifica-
hydrogen produced at their steam methane reformer. This tion, hydrogen firing, a green hydrogen plant (electrolyser),
step required various options to be evaluated to determine and oxygen firing. Moreover, the analysis also included
the impact each alternative would have on energy, emis- future business cases that accounted for Scope 1 and 3
sions, and economics. Besides developing a solid business emissions. These projects included carbon capture and uti-
case and a set of digital tools to analyse the green hydrogen lisation (CCU) for sustainable aviation fuel, CCU urea pro-
plant, additional actions and opportunities were explored to duction, and a potential renewable hydroprocessing plant.
achieve net zero emissions, such as electrification, carbon The second technology involved implementing a digital
capture storage and utilisation, and hydrogen firing. energy management system to serve as the programme’s
Moreover, testing and verifying results were essential foundation. This software optimised the energy system in
to achieve top performance. As a result, tools capable of terms of economics and emissions to sustain long-term
incorporating the latest site modifications were essential value.
for real-time energy and emissions optimisation. By incor- Simultaneously cutting costs and GHG emissions started
porating these measures, the refinery worked towards in a predominantly centralised, hydrocarbon-based energy
achieving its emission reduction goals and contributing to a generation setting. During the energy transition, processes
more sustainable future. improved, renewable energy vectors were introduced, and
Technology leads the way. Against this backdrop, and assets were decentralised.
to ensure alignment between strategy and current opera- Ultimately, the goal is to achieve net zero GHG emis-
tions, two main technologies were adopted: the Integrated sions in a largely renewable energy system by maintaining
Process, Energy, Emissions and Economics Model (IP3EM) decentralised, low CO2 emissions energy generation, trans-
and digital energy management system. portation, as well as storage and utilisation.
The first technology implemented was the IP3EM. This The case in Figure 3 demonstrates a utility system that
tool optimises renewable power options to reduce Scope 2 blends traditional with renewable energy sources. This
emissions while incorporating future Scope 3 projects. scenario emphasises the importance of aligning short-term
As shown in Figure 2, this refinery-wide model compre- scheduling with real-time operations for maintaining feasi-
hensively considered energy demanded from processing, bility and profitability.
492 Kg/h
Electrolyser
H2
50+
VARIABLES
in engineering an
efficient scrubber/quench system
C
CM
MY > Absorbers
CY > Gas Quench
CMY > Gas Wash Systems
K
> Spray Dryers
> Injectors
> Fuel Burners
www.halliburton.com/en/production/specialty-chemicals
© 2023 Halliburton. All Rights Reserved.
S
ustainable aviation fuels (SAF) have been at the fore- or hydrocracker (for example) to produce renewable fuel.
front of the fuel industry’s agenda for several years, Co-processing of renewable feedstocks, such as virgin oils
with increasing attention driven largely by regulatory or waste oils and fats, in hydroprocessing units has been
pressures. Across the globe, from the European Union and common practice for over a decade, primarily to produce
the UK to Japan, Brazil, India, and Singapore, SAF man- renewable diesel.
dates are being implemented, sometimes accompanied by The amount of renewable feedstock added is typically at
significant penalties for non-compliance, as exemplified by relatively low levels of less than 5%. Such a low ratio gen-
the ReFuelEU aviation initiatives. These mandates are set erally only necessitates a catalyst change, which makes the
to significantly accelerate the deployment of SAF produc- process relatively straightforward and makes co-process-
tion, and projections indicate that bolstered by such meas- ing a simpler alternative to major revamps.
ures, SAF production could reach approximately 13 million In cases where the co-processing ratio is increased to
tons by 2030 and potentially soar to around 120 million 10% or higher, revamps are typically required to address
tons by 2050.1 common challenges, such as managing high exotherms or
Simultaneously, regulators are incentivising SAF supply mitigating corrosion issues.2
through attractive credit schemes. In the US, SAF suppliers As written in the first quarter of 2024, most refineries
can benefit from Renewable Identification Number (RIN) in the EU have become familiar with co-processing, and
credits, Low Carbon Fuel Standard (LCFS) credits, Inflation few consider it a significant obstacle. Coupled with the
Reduction Act (IRA) production tax credits, and state-based numerous benefits of co-processing for both producers
credits from states like Washington, Minnesota, and Illinois. and off-takers, this is becoming an increasingly attractive
However, there are significant challenges to over- option for renewable fuel production.
come to increase the supply of SAF. Production of SAF in
standalone plants is capital intensive, and the process of Regulations on co-processing
revamping an existing unit or constructing a new one can As mentioned in the introduction, regulatory initiatives
span several years. This is where co-processing in a ker- supporting the supply or demand of SAF are being imple-
osene hydrotreater presents a compelling alternative. Not mented globally. This section outlines the key regulatory
only is it a low-capital expenditure (Capex) approach, but initiatives introduced across various regions, as well as the
it also allows for the rapid deployment of SAF production regulatory fuel compliance standards for co-processing.
within a few months (given the right conditions). In the EU, the ReFuelEU initiative targets a 70% SAF
With more than 50 cycles of units co-processing renewa- share by 2050, split evenly between biogenic jet fuel and
ble feedstocks in various hydroprocessing units worldwide eSAF. Non-compliance penalties enforce this obligation on
and with a wide range of renewable feedstocks, Topsoe has fuel suppliers. The UK is developing a similar SAF regula-
accumulated extensive experience in this field. This article tory framework, potentially including a revenue certainty
is intended to offer a starting guide to anyone considering scheme.
going down the co-processing path by comprehensively In the US, the SAF Grand Challenge promotes SAF
covering the regulations to be aware of, why co-processing production, bolstered by incentives like RIN credits from
in the kerosene hydrotreater is the preferred choice for SAF the Renewable Fuel Standard (RFS), LCFS credits from
production, and the challenges and solutions. California, Federal production tax credits (IRA), and addi-
tional credits from specific states. British Columbia, Canada,
What is co-processing? has also issued SAF mandates, targeting 1% SAF by 2028
In co-processing, even small amounts of fossil feedstock, and 3% by 2030.
such as crude oil, can be replaced with renewable feed- Other countries are also embarking on the SAF journey.
stock, such as vegetable oil, animal fat, and used cooking Japan aims for 10% SAF by 2030, India targets 1% SAF
oil (UCO). These are processed together in the hydrotreater by 2027 initially for international flights, Brazil seeks a 1%
Sustainability No food and feed crop Indirect ILUC criteria via 14 sustainability criteria
criteria No high ILUC feedstocks the minimum GHG savings (including ILUC)
Minimum GHG savings Min. 65% GHG saving Min. 50% for D4/D5 RIN; Min 10% GHG saving
Min. 60% for D3 RIN
Fossil comparator 94 gCO2eq/MJ _ 89 gCO2eq/MJ
LCA GHG calculation model RED II, Annex V GREET model CORSIA model
Table 1
reduction in GHG emissions by 2026, potentially rising to • Waste listed in Annex IX part B
10% by 2037, and Singapore announced that it is aiming • Other biogenic feedstocks not listed in Annex IX, pro-
for a 1% SAF target from 2026 with plans to raise it to vided they are not food and feed crops, intermediate crops,
3-5% by 2030. palm fatty acid distillate, palm and soy-derived materials, or
Furthermore, discussions on mandates and incentives soap stock and its derivatives. To comply, these feedstocks
are underway in numerous countries, including the United must achieve a minimum of 65% GHG emission savings.
Arab Emirates (UAE), Turkey, New Zealand, South Korea, In the US, the choice of feedstocks depends on the type
Australia, China, Chile, Colombia, Thailand, Malaysia, of credits sought, such as RIN in the RFS scheme, LCFS in
Indonesia, and more. California, or Clean Fuel Production Tax Credit (PTC) cred-
From a fuel compliance perspective, the production of its. Each scheme has specific requirements and should be
SAF is guided by two key ASTM standards: ASTM D7566, examined individually. Both the RFS and IRA schemes also
which currently outlines eight SAF pathways for standalone require GHG savings, albeit with different methodologies.
plants, and ASTM D1655, which details the three current Under the CORSIA, waste feedstocks are listed, and
co-processing pathways. The co-processing pathway dis- default GHG values are provided for selected feedstocks.
cussed in this article involves the co-processing of mono-, The ICAO has also developed guidance to include addi-
di-, and triglycerides, free fatty acids (FFAs), and fatty acid tional materials in the positive list.
esters (FAEs), allowing co-processing up to 5 vol% with
fossil feedstocks (further details can be found in Annex Kerosene hydrotreater: Preferred choice for SAF
A1.2.2.1 of ASTM D1655). production
Regulatory considerations extend beyond fuel compli- Co-processing can be carried out in various refinery units,
ance to also include questions about permissible feedstocks, including fluid catalytic cracking (FCC), diesel hydrotreat-
minimum greenhouse gas (GHG) savings, and GHG calcu- ers, hydrocrackers, and kerosene hydrotreaters. However,
lation methodologies. These complex issues are addressed the optimal choice often hinges on the principle of selectiv-
in various forums, but in this article, we will focus on regula- ity and optimal cost of production: Which unit will yield the
tions related to co-processing. In the EU-27, the Delegated highest SAF output from the renewable feedstock at the
Act on co-processing, issued on June 5, 2023, clarified the most optimal cost of production?
methodologies for calculating the share of renewable fuels In this context, a kerosene hydrotreater emerges as the
via co-processing.3 most preferred choice. With the right catalysts, this unit can
This regulation supersedes the methodologies previously maximise the conversion of biogenic carbon into the jet fuel
established by specific member states like Spain and the fraction, thereby optimising SAF production. Low-pressure
Netherlands, which were developed in the absence of a com- reactor operation (~25-30 bar g) makes co-processing in
mon EU approach. The Carbon Offsetting and Reduction the kerosene hydrotreater option unique with respect to the
Scheme for International Aviation (CORSIA) scheme devel- high recovery of biogenic carbon.
oped by the International Civil Aviation Organization (ICAO)
also covers SAF by co-processing, with the co-processing Co-processing challenges and their solutions
methodology described in the methodology of a life-cycle While co-processing in general comes up with some
assessment (LCA) and default LCA values.4 challenges, the limited catalyst volume, lower hydrogen
partial pressure, and use of relatively simple metallurgy
What types of feedstocks? in a kerosene hydrotreater make this challenge even more
The feedstocks eligible for SAF production via co-process- interesting.
ing vary by jurisdiction. In the EU, three types of biogenic If the end product is to obtain SAF certification, there
feedstocks are permitted (according to ReFuelEU Aviation): are three main challenges to be aware of when producing
• Waste listed in Annex IX part A SAF by means of co-processing in your kerosene unit. In
0 Hydrocracking
Number of carbon atoms: 18
-25 Boiling point: 317 ˚C
Melting point: 28 ˚C
-50
Figure 2 Hydroisomerisation as a more efficient pathway for lowering the freezing point
O O
H H H H Octadecane
Water Water
O O
HDO route H H H H Octadecane
Propane Water Water
+ 16 H2
O O
H H H H Docosane
Water Water
O Triglyceride Reverse water gas shift
O C O + H H O + +O C-
O O 9c-oleic acid H H
O Methanation
O 13c-eurcic acid H H
+O C- + H H CH4 + O
H H
O H H
9c 12c-linoleic acid
Decarboxylation O C O
route (DCO) Carbon dioxide Hepadecane
+ 7 H2 O C O
Propane Carbon dioxide Hepadecane
O C O
Carbon dioxide Henicosane
Figure 4 Two possible reaction routes for removing oxygen atoms from lipidic feedstocks converting free fatty acids and
triglycerides into n-paraffins
Beta Ionsation Low investment Low background lab and EN1 6640 (Annex D) 0.2-5% deviation
equipment gas purification device required
Liquid scintillation Fast/lower cost/easy LSC-Benzene: applicable D6866 (Method C); Maximum total error
counters (LSC) sample preparation/ to clear liquids only EN 16640 (Annex C); of +/-3% (absolute)
Referred in fast analysis (5-10h) DIN-51637
ASTM D1655
Accelerator mass Applicable to solid, Complex/time-consuming/ D6866 Instrumental error within 0.1-0.5%;
Spectrometry (AMS) gas, and liquids high cost/laborious (Method B); Interlaboratory total uncertainty
Referred in sample preparation EN 16640 +/-3% (absolute);
ASTM D1655 (Annex E) Lower detection limit 0.4%
Table 2
hydrotreaters is the best choice for SAF production with Technology and a Diploma in Business Development and Management
respect to lower operating cost, lower capital investment, from the Technical University of Denmark.
higher recovery of biogenic carbon in the SAF product,
short implementation time, and low payback period. This Sylvain Verdier is Senior Strategy Manager at Topsoe, Denmark, advis-
method can save up to two years of construction time com- ing internal and external stakeholders on how to manage the energy
pared to building new facilities. transition with a focus on technologies, regulations, sustainability, and
The main operational challenges of using a kerosene feedstocks. He holds an MSc in chemical engineering from ENSGTI and
hydrotreater as a co-processing unit to produce SAF are a PhD in chemical engineering from the Technical University of Denmark.
References
1 SkyNRG, SkyNRG’s Sustainable Aviation Fuel Market Outlook May
2023, https://skynrg.com/safmo2023.
2 Nystrom S, Michaelsen N, Egeberg R, Improving the production of renew-
able diesel by co-processing, www.digitalrefining.com/article/1000060/
improving-the-production-of-renewable-diesel-by-co-processing
3 C_2023_3513_1_EN_ACT_part1_v3_0.pdf, https://energy.ec.eu-
ropa.eu/system/files/2023-06/C_2023_3513_1_EN_ACT_part1_
v3_0.pdf
4 CORSIA Eligible Fuels, www.icao.int/environmental-protection/
CORSIA/Pages/CORSIA-Eligible-Fuels.aspx
5 www.iscc-system.org/wp-content /uploads/2022/08/ISCC-
Guidance-Document-203-01_Co-processing-requirements.pdf
We recover and refine precious metals from hydrocarbon and petroleum processing catalysts.
Also, our proprietary Pyro-Re® process offers the only pyro-metallurgical recovery of rhenium in
the industry. With it, we recover total rhenium content from spent semi-regenerative and cyclic
fixed-bed hydrocarbon processing catalysts—and get you a full return. With best-in-class
techniques and over seven decades of experience, we deliver the highest possible metal
returns for our customers.
R
aw olefinic C3/C4 streams in integrated petrochem- Front-end hydrogenation processes are characterised by
ical refineries, mainly the product of naphtha steam an excess of hydrogen (H2) present in the raw product
crackers or FCC units, usually contain low-weight stream from the steam cracking or FCC unit. In contrast,
percentages of acetylene derivatives, known as MAPD tail-end selective hydrogenation, situated behind the
compounds. MAPD stands for methylacetylene (propyne, de-methaniser and de-ethaniser columns, is controlled by
MA) and propadiene (PD), which are the most common feeding a stoichiometric amount of hydrogen to the hydro-
impurities; vinylacetylene (but-1-en-3-yne, VA) and ethy- carbon stream. Whereas for front-end hydrogenation, NiCo
lacetylene (1-butyne, EA) can also be formed. These side catalysts are common, tail-end hydrogenation catalysts are
products must be removed to protect downstream pro- precious metals, mainly based on promoted palladium (Pd).
cesses from severe catalyst poisoning and unwanted side Commercial process licensors and catalyst manufacturers
reactions of those highly reactive substances. For instance, for selective hydrogenation include Axens, UOP, Evonik,
polymer-grade propylene must not exceed an MAPD impu- Lummus, Linde, KBR, and BASF.
rity level of 1 ppmw. Tail-end liquid state selective hydrogenation often incor-
The predominant method for removing these compounds porates multistage operation. In the first stage especially, a
is selective hydrogenation of the triple bonds while preserv- recycle stream of the product can be integrated to control
ing the olefinic species. Whereas selective hydrogenation the acetylene concentration and, in turn, the exothermicity
of acetylene in the C2 fraction is performed in the gase- of the reaction.1-4
ous phase, C3/C4 selective hydrogenation is predominantly For operators of tail-end selective hydrogenation pro-
carried out in a liquid state. This is for reasons of energy cesses, many aspects of the catalyst and process perfor-
efficiency and the smaller reactor volume. mance are important to investigate, such as the long-term
Another benefit of a liquid state reaction is that greenoil stability of the catalyst, the acetylene derivatives conversion
formed as a byproduct during the hydrogenation reaction level as a function of feed composition, the optimal operating
is washed out of the catalyst bed. Greenoil is an unwanted window of the process, and the control of greenoil formation.
side product in C3-C4 selective hydrogenation, formed by
the oligomerisation of olefins to various heavier unsatu- Experimental set-up
rated hydrocarbons. This greenoil can lead to catalyst deac- The experimental unit was equipped with a flexible
tivation as well as contamination in the unit. up-and-downflow configuration. Liquid feed was pumped
Commercial selective hydrogenation processes can be by a two-piston syringe pump, whereas the hydrogen flow
differentiated into front-end and tail-end configurations. was controlled by a mass flow controller (see Figure 1). The
GC
Heater/cooler
H2 Reactor
Reaction feed N2
Upflow
configuration
Compound 1,3- 4-vinylcyclo- n-butane 1-butene c-2-butene Vinyl- Methyl- Ethyl- Propa- Propylene
butadiene hexene acetylene acetylene acetylene diene
wt% 37.6 7.25 5.90 4.59 0.088 0.56 0.34 0.075 0.29 Balance
Table 1
pressure was controlled using a backpressure regulator sit- selectively poisoned by lead (Pb) on a calcium carbonate
uated after (liquid phase operation) or before (gas phase carrier, Merck).5 The reactor with an inner diameter of 8 mm
operation) the reactor. The resulting product mixture was was filled with a 2:3 (by volume) mixture of the Lindlar cat-
diluted with N2 downstream, whereas any greenoil formed alyst and silicon carbide as inert material. The total catalyst
was kept liquid and thus separated from the product stream. zone length was 77 mm. Reference conditions for testing
The temperature of the catalyst bed was controlled by a were a reactor temperature of 35°C, a pressure of 24 barg,
heating/cooling mantle connected to a thermostat equipped a molar hydrogen/VA ratio of 2, a hydrocarbon feed-based
with a parallel electrical heating circuit. The accessible tem- liquid hourly space velocity (LHSV) of 22 h-1, and an upflow
peratures ranged from sub-ambient to 600°C. configuration of the reactor system.
With this flexible set-up, it was possible to run the reac- This approximates a typical operating window of com-
tion at various pressures and temperatures in both liquid mercial units with a hydrogen dosage just below the sto-
and gas phase. Although most commercial selective hydro- ichiometric amount with respect to all acetylenes and PD.
genations of these C3/C4 species are performed in the liquid These parameters were systematically varied during the
phase, the experiment additionally compared liquid and gas testing program, as will be further discussed.
phase performance to demonstrate the unit’s flexibility to Under these bespoke standard conditions, the following
accommodate different light olefin selective hydrogenation conversions were observed: 78.6 wt% for VA, 49.4 wt%
reactions. In liquid phase operation, the impact of upflow vs for EA, and 54.1 wt% for MA. The higher conversion of
downflow configuration was investigated. VA compared to MA and EA is characteristic of Pd-based
The hydrocarbon feedstock simulates a C3-C4 olefin catalysts. Hydrogen was quantitatively consumed, which is
stream. It should be noted that the feedstock contains also typical for a tail-end selective hydrogenation process.
7.25 wt% 4-vinylcyclohexane as a dimerisation product of However, containing 5 wt% Pd, the commercially available
1,3-butadiene, extending the educt spectrum up to C8. Lindlar catalyst provided a fivefold higher active metal load
The effluent stream was analysed with an Agilent gas than a typical industrially available catalyst for this reaction
chromatograph (GC) equipped with a flame ionisation detec- and, hence, was significantly more active. This explains the
tor (FID) and an RTX-Alumina BOND column. Conversion high BD loss of 9.53 wt%.
was calculated based on the weight and molar educt, and
product flow rates, with the educt flow rates defined by LHSV variation
repeated bypass measurements between the experiments. Starting from the standard conditions, the hydrocar-
bon-based LHSV varied in a broad range between 7.2 h-1
Results and discussion and 82.4 h-1. Due to the high activity of the Lindlar cat-
Operation under standard conditions alyst used, the hydrogen conversion remained at nearly
A commercial Lindlar catalyst was used (5 wt% Pd, 100 mol% over the whole LHSV range. At higher temper-
atures, a temperature increase of the catalyst zone could
be detected, with the zone temperature ranging from
T = 16.7 ˚C T = 35.8 ˚C T = 50.1 ˚C 80 35.5°C at LHSV = 7.2 h-1 to 36.7°C at LHSV = 82.4 h-1. The
11
heater/cooler system was able to keep the catalyst zone
75
Acetylenes conversion (%)
65 reaction.
10 60
Temperature variation
55 Temperature was set to 15, 35 and 50°C, while the other
9.5 50 parameters were kept at their standard values (see Figure 2).
The highest acetylene conversion was observed at 15°C,
45
with a corresponding minimised butadiene (BD) loss. When
9
raising the temperature to 35 and 50°C, the acetylene con-
VA conversion MA conversion EA conversion BD loss
version decreases, with slightly rising butadiene conver-
sion. Quantitative hydrogen consumption was observed at
Figure 2 Acetylene conversion (right axis) and butadiene all temperatures, limiting the overall hydrogenation activity.
loss (left axis) vs reactor temperature (upflow, liq.), p = 24 It is noteworthy that the relative shifts in BD conversion are
barg, H2:VA = 2 mol/mol, LHSV = 22 h-1 much smaller than the shifts in acetylene conversion due to
Influence of variable process conditions operation at a significantly shorter contact time with the
Operating the unit under standard conditions as previously catalyst. Correspondingly, the BD loss was significantly
described, the upflow configuration of the unit was tested lower (see Figure 5).
against a downflow configuration. In downflow operation, It is noteworthy that at the lowest H2/VA ratio, VA is
a significant loss was observed in overall acetylenes con- selectively hydrogenated to 1,3-butadiene, thus caus-
version (see Figure 3). The most probable reason for this is ing a negative butadiene conversion. This high selectiv-
that in the downflow regime, some gas-filled voids remain ity is also targeted in commercial processes to maximise
between the catalyst particles, preventing the formation of 1,3-butadiene yield. The most likely reasons for the higher
a uniform feed phase in the catalyst bed. selectivity of gas phase operation are less restricted mass
By reducing the reactor pressure to 0.6 barg, all hydro- transfer in the gas phase, plus the lower tendency for dou-
carbons could be evaporated, and the reaction could be ble bonds to adsorb on the Pd active sites at lower pres-
run in the gaseous phase (see Figure 4). As for the liquid sure. Furthermore, in a liquid state, H2 first needs to be
state experiments, the H2/VA ratio was systematically solved in liquid hydrocarbon; then, an equilibrium H2 con-
changed at constant a hydrocarbon-based LHSV of 22 h-1. centration must be established in the hydrocarbon phase,
At equal H2/VA ratios, much higher acetylene derivatives permitting the H2 concentration available at the active site
conversions could be observed compared to liquid phase to be limited.
Conversions vs. H2/VA ratio, liquid phase upflow Conversions vs. H2/VA ratio, gas phase
36.8 36.8
100 100
34.6 34.6
80 80
Reactor temperature (˚C)
36.4 36.4
Conversion (%)
60 60
36.2 36.2
40 36 40 36
35.8 35.8
20 20
35.6 35.6
0 0
35.4 35.4
2 4 6 2 4 6
H2/VA Ratio in feed (mol/mol) H2/VA Ratio in feed (mol/mol)
Figure 4 Acetylene conversion vs H2 to vinylacetylene ratio (left axis), catalyst bed temperature (right axis); left: p = 24
barg, LHSV = 22 h-1 (liq., upflow), right: p = 0.6 barg, LHSV = 22 h-1 (gas phase)
20
0
MA conversion (gas) EA conversion (gas) VA conversion (gas)
40
20
0
50 100 50 100 50 100
Conversion of acetylenes (wt. %)
H2/VA ratio (liq.) = 0.99 H2/VA ratio (liq.) = 4.12 H2/VA ratio (liq.) = 0.29 H2/VA ratio (liq.) = 0.47 H2/VA ratio (liq.) = 6.91
H2/VA ratio (liq.) = 2.03 H2/VA ratio (liq.) = 7.2 H2/VA ratio (liq.) = 2.03 H2/VA ratio (liq.) = 1.44
Figure 5 1,3-butadiene loss vs acetylene conversion, liquid phase upflow (24 barg, 35°C, LHSV =22 h-1) vs gas phase
(0.6 barg, 35°C, LHSV =22 h-1)
Carbon monoxide (CO) dosage feed and further downstream nitrogen (N2) dilution. Figure
As for MAPD selective hydrogenation, front-end and 6 shows a significant drop in all acetylene conversions at
tail-end configurations are possible in acetylene selective the onset of the CO dosage. Also, H2 was no longer quanti-
hydrogenation in an ethylene matrix. In the case of front- tatively consumed, with H2 consumption instead dropping
end hydrogenation, an excess of H2 is present in the feed to about 75 mol%. A marginal temperature drop could be
gas, plus low levels of CO, which moderate the activity of observed in the catalyst zone. A shift of conversion levels
the hydrogenation catalyst towards acetylene. In tail-end toward triple bond hydrogenation was not observed, with
configuration, the H2 content can be controlled at stoichio- the BD conversion staying virtually stable.
metric levels, so it is therefore desirable to operate with an
intrinsically selective catalyst system.6 Greenoil formation and handling
To simulate a CO impurity in the feedstock, typical When running selective hydrogenation reactions in an
for front-end hydrogenation, a capillary for dosing CO, olefinic matrix, the control and handling of greenoil are
designed to dose 100 ppm CO in H2 at standard conditions, important aspects. During the operations, it was crucial to
was installed. The CO could not be detected by online GC, as separate greenoil from the C1-C8 product fraction before
the H2 itself became heavily diluted by excess hydrocarbon online GC analysis. This could be achieved by downstream
90
80
Onset of VA conversion 35.9
CO dosage BD conversion
70 Temperature 35.85
Reactor temperature (˚C)
60
35.8
Conversion (%)
50
40 35.75
30 35.7
20
35.65
10
35.6
0
580 585 590 595 600 605 610 615 620 625 630 635
TOS (h)
Figure 6 Effect of a 100 ppm level CO dosage on VA and BD conversion at standard conditions, 24 barg, 35 °C, H2:VA =
2, mol/mol, LHSV = 22 h-1.
80
1.5 60
Area (%)
40
20
1
0
0 100 200 300 400 500 600 700 800
0.5
0
0 100 200 300 400 500 600 700 800
Temperature (%)
N2 dilution followed by pressure regulation and controlled selective hydrogenation processes in the gas or liquid
evaporation of the main reaction products through a dome phase, like acetylene in a C2 matrix, with stoichiometric or
pressure regulator. excess levels of hydrogen. The influence of temperature,
Greenoil could then be separated from lighter products in H2/acetylenes ratio, liquid state upflow vs downflow, gas
a liquid trap. Keeping a 50°C temperature at the pressure phase vs liquid phase operation, and CO poisoning has
regulator and the liquid trap during this step was of vital been tested and discussed.
importance, as at higher temperatures, increased greenoil
formation and carry-over led to downstream plugging of Acknowledgement
pipes. At lower temperatures, lighter products were not We thank our colleagues at hte for their support in this study. Special
quantitatively carried over into the gas phase. thanks to Tobias Zimmermann for his significant contributions to the
Samples of greenoil were collected and analysed via sim- study within his occupational activities at hte.
ulated distillation (SimDist: see Figure 7). Results showed
a boiling point range of 200°C up to >700°C, with a sig- Literature
nificant peak for C16 species. It should be noted that the 1 Mohundro E L, Overview on C2 and C3 Selective Hydrogenation in
Ethylene Plants, AIChE 15th Ethylene Producers Conference, 2003.
greenoil was stored at 50°C for several days before the
2 Butadiene, IHS Markit Report, 2021.
analysis, potentially allowing for further oligomerisation. It
3 Propylene, Technoeconomics – Energy and Chemicals, Nexant
does not necessarily resemble the greenoil composition out
Market Report, 2018.
of the catalyst bed. The reaction network behind greenoil
4 Iselborn S, BASF selective and full hydrogenation technology Selop®,
formation is not yet fully understood and is still a topic of
Handbook of Petrochemicals Production Processes, Chap. 2.3, 2nd
debate in the available literature.1
ed., 2019, edited by R A Meyers, New York, McGraw-Hill Education.
5 Lindlar H, Ein neuer Katalysator für selektive Hydrierungen, Helv
Conclusions Chim Acta 1952, 35, pp.446-450.
In this case study, the selective hydrogenation of tail-end 6 McCue A J, Anderson J A, Recent advances in selective acetylene
acetylene derivatives and propadiene was successfully hydrogenation using palladium containing catalysts, Front. Chem. Sci.
downscaled in a highly unsaturated hydrocarbon matrix, Eng. 2015, 9(2), pp.142-153.
including 1,3-butadiene, in the laboratory environment.
The set-up allows for the testing of commercial catalyst
shapes under industrially relevant conditions. The high flex-
Edgar Jordan is Application and Project Manager focused on light ole-
ibility of the unit allows for screening in a broad parameter
fins conversion processes at hte GmbH, Heidelberg, Germany. He has
range for many types of commercial MAPD catalyst. An more than five years of experience in high throughput and pilot-scale
optimum operating window was found to maximise MAPD catalyst testing and holds a PhD in chemistry from the University of
purification at low BD loss. Münster, Germany.
All trend data and performance data were measured in Charlotte Fritsch is Business Development Manager focused on
parallel and automatically linked to corresponding online catalyst testing and process development at hte GmbH, Heidelberg.
GC data and can be reported on request. This unit is one She holds a PhD in chemistry from Karlsruhe Institute of Technology,
building block of olefin conversion capabilities in support- Germany.
ing the needs of refineries in integrating petrochemical Joachim Haertlé studied chemical and bioengineering at the University
complexes. of Erlangen and has been working as a project coordinator at hte since
The experimental concept can be transferred to other 2018.
FTC filters provide consistent effluent quality with the most predictable
Keep your cash cow flowing with reliable pre-treatment of your catalyst feed.
7 13.8 49.0849
ftc-houston.com
Richard Stambaugh
Merichem Technologies
M
ore than a hundred different crude oils are traded on within five minutes, and concentrations exceeding 700
the market today. The most desirable crudes are light ppm can cause immediate collapse and death within just
and sweet, containing less than 0.5% (by weight) one or two breaths.
sulphur. They are highly efficient, leave minimal residue in
the refinement process, and the low density makes them Spent caustics
easy to distil and transport. Demand for light, sweet crude When the malodorous and potentially dangerous sulphides
grades rose exponentially across Europe, Africa, and the US and mercaptans are expended from sweetened hydro-
as refineries adjusted their slates ahead of the International carbons, the residual caustics are considered spent. The
Maritime Organization (IMO) 2020 lower sulphur cap. complexities and numerous processes involved in treating
In contrast, heavy, sour crudes contain more than 0.5% spent caustic are challenging. It is one of the most prob-
sulphur. They are generally extracted from the Gulf of lematic industrial wastes in terms of disposal. When man-
Mexico, Mexico, South America, Canada, and the 24 OPEC aged poorly, spent caustic can cause significant harm to
Member Nations. Sour crudes are more complicated to health and the environment. Since 1997, the Environmental
refine into lower-sulphur fuels, and the refiners who can Protection Agency (EPA), under the Resource Conservation
process them have a distinct advantage over those who
cannot, particularly considering IMO 2020. The impetus
has been on refineries to determine the best methods of The complexities and numerous
separating sulphur compounds from sour crude. In doing
so, US refineries can take in almost any crude oil and turn it
processes involved in treating spent
into clean-burning fuel. caustic are challenging. It is one
of the most problematic industrial
The problem
Sitting midway to 2030 net zero goals, it has become
wastes in terms of disposal
clear to scientists and environmentalists that incremen-
tal and fragmented change is insufficient to achieve the
Sustainable Development Goals (SDGs) in the seven years and Recovery Act (RCRA), has classified spent caustics as
remaining. Meeting (net zero goals) requires transforma- hazardous waste, with strict handling and disposal require-
tive and accelerated actions, outlined in the 2023 Global ments because of the high levels of contaminants.
Sustainable Development Report (GSDR), a new phase of Spent caustics typically have a high pH value (>12) and high
accelerated progress by UN member states towards the sulphide concentrations (2-3% w/w), not accounting for the
Sustainable Development Goals leading up to 2030. It out- phenols, mercaptans, amines, or other organic compounds
lines the shifts required to transform the energy sector, and that are emulsified in the caustic. Spent caustic is classified
the increasingly stringent fuel specifications have made as an EPA Hazardous Waste Number of DOO3, hazardous
crude oil desulphurisation obligatory. waste under the US Resource Conservation and Recovery
The sulphur-containing compounds in sour crudes are Act (RCRA), and a Class 8 (corrosive) hazard throughout
among the most undesirable. Hydrogen sulphide (H2S) is United Nations (UN) participating countries use the UN
toxic to humans and particularly dangerous because, at suf- Globally Harmonized System of Classification and Labeling
ficiently high concentrations, it paralyses olfactory nerves of Chemicals (GHS), which is defined as substances that
even when exposed to it for only a few minutes. According can cause significant harm to living tissue and/or corrode
to the National Institute for Occupational Safety and Health steel and aluminium upon contact. Spent caustic is also on
(NIOSH), H2S environmental concentrations of 100 ppm the Right to Know Hazardous Substance List cited by the
are immediately dangerous to life or health. Concentrations Occupational Safety and Health Administration (OSHA),
greater than 500 ppm can cause a person to collapse American Conference of Governmental Industrial Hygienists
Fresh
TC
FC AC Cooling caustic
water
MP steam
Cooling water Neutralised
FC
brine
Spent FC
caustic
M
Sulphuric acid
demand (COD) and biochemical oxygen demand (BOD). This The simple neutralisation process is streamlined and
allows the treated brine to be sent to a wastewater treatment efficient. Spent caustic and acid are injected continuously
facility for final polishing before release to the environment. into a treatment unit, mixed, and then entered into a three-
Treatment technologies typically utilise sulphuric acid, phase separator. The acid flow is regulated by sensing the
atmospheric oxygen, nitrogen, fuel gas, electricity, steam, pH of the released brine. Any acid gases or oils generated
and/or cooling water – all elements readily available at most by the reaction will separate. If necessary. to lower the COD
refineries. After treatment, a neutral pH, saltwater effluent and odour, nitrogen or fuel gas can be added to strip away
is produced that has minimal odour, presents no health dissolved acid gases.
and safety issues, and is easily manageable in wastewater When low levels of unregenerated sulphide, hydro-
treatment plants. sulphide, and mercaptide salts are present, a variation of
The most frequently used technologies include sim- simple neutralisation adds a stripping column to the three-
ple neutralisation, deep neutralisation, wet air oxidation phase separator where the nitrogen or fuel gas can effi-
(WAO) and incineration. Although caustic neutralisation ciently remove these dissolved acid gases. The degree of
and oxidation systems have been used for decades, their removal of these dissolved acid gases is highly dependent
seemingly simple processes are surprisingly complex. The on the treatment pH, with a lower pH releasing more of
type and concentration of impurities affect the temperature the gases. Where high levels of these impurities are pres-
and pressure requirements, how many treatment steps are ent, the pH must be lowered beyond the normal discharge
required, the mixing and residence times for the reaction range, requiring the need for ‘deep neutralisation’.
and separation steps, and the selection of the metallurgy to
protect equipment from severe corrosion. Deep neutralisation
When spent caustic contains high concentrations of unre-
Simple neutralisation generated sulphides, mercaptides and organic compounds,
Simple neutralisation is the most straightforward and least or the brine product specifications are more demanding,
expensive treating option for regenerated caustic streams deep neutralisation is utilised. This treatment method can
coming from light hydrocarbon streams such as LPG where treat most sulphidic, phenolic, and naphthenic spent caus-
most of the toxicity and COD result from excess NaOH, tics. ‘Deep’ refers to acidification of the spent caustic to a
thiosulphate, and dissolved organics. low pH followed by re-neutralisation with caustic.
A simple acid, such as sulphuric or hydrochloric acid, is This process reconstitutes all the original acids that were
added to the spent caustic to neutralise any free NaOH present in the treated hydrocarbon streams: sulphides,
remaining and drop the pH to an acceptable wastewater mercaptides, and organic compounds, such as phenolics,
discharge value, within the preferred range of 6-9, although cresylics, and naphthenic acids. The organics form an oil
some publicly owned treatment works allow higher or phase that separates from the brine. If deemed beneficial or
lower pH ranges. The acids are conveniently available and re-blended with another hydrocarbon stream, the organics
add no COD to the treated brine. If the resulting sulphate or can be decanted for reclamation.
chloride levels are too high, organic acids, such as formic or The deep neutralisation treatment system generally con-
acetic acid, can be substituted. sists of a low pH reactor, a three-phase separator, acid gas
PC
HP steam Oxidation
O2
PC AC
FC
M TC FC Off-gas
Spent
caustic
FC
M
Neutralisation
LC
Caustic
AC Cooling water
Oxidised caustic
pH
Acid
Brine Neutralised brine
Figure 2 Merichem Mericon II spent caustic treatment system employing wet air oxidation
FC
Fuel gas
Oil or nitrogen
Acid
LC
FC pH
AC Brine LC
M
Acid
M
Steam or Acid oils to
CWS refinery
electric
heater pH
FC AC
Spent
caustic
FC
Spent
caustic FC
Neutralised
FC caustic
Process
water
Visit www.arielcorp.com/pledge
Rajib Talukder
Saudi Aramco
L
everaging a rich pool of expertise from pre-commis- optimisation in the FEED stage ensures that the system
sioning and commissioning phases, this initiative is aligns with relevant standards and codes.
an effort to elevate front-end engineering design • Early issue identification: Identifying and addressing
(FEED) processes. It highlights the importance of a proactive potential design issues early in the project life cycle is more
approach to design and standardisation, ensuring engineer- cost-effective than making corrections during the detailed
ing solutions are robust and efficient from the start. engineering or operational phases. FEED stage optimisation
The narrative explores the process of revisiting and refin- allows for early issue identification and resolution.
ing FEED deliverables, leveraging the author’s experience • Ease of operation: A well-optimised design makes the
in spearheading a FEED standardisation initiative within compressor system easier to operate and maintain. This is
a distinguished engineering firm. This initiative aimed to important for the personnel managing the equipment, as
streamline FEED deliverables by integrating the extensive it reduces the complexity of operation and maintenance
knowledge and expertise of the engineering community. procedures.
Optimising compressor dry gas seal (DGS) line design dur- In summary, optimising compressor DGS line design
ing the FEED stage is crucial for several reasons: during the FEED stage is critical for achieving reliability,
• Reliability and performance: Proper design optimisation safety, environmental compliance, and cost effectiveness
ensures that the DGS system operates reliably and effi- throughout the lifecycle of the compressor system. It allows
ciently. This is essential for maintaining the performance of for proactive decision-making and sets the foundation for a
the compressor and avoiding unplanned downtime. well-performing and efficient facility.
• Safety: A well-designed DGS system enhances safety by Key insights were gained from the collective experience
minimising the risk of seal failure and gas leakage. Safety is a in pre-commissioning and commissioning, particularly in
critical consideration in the oil and gas industry, and a flawed enhancing compressor deliverables through modifications
design could lead to hazardous situations. in FEED documentation. Two primary recommendations
• Environmental compliance: Efficient DGS systems help in emerged:
minimising fugitive emissions, contributing to environmental
compliance. Reducing leaks and emissions is essential for P&ID note enhancement
meeting environmental regulations and sustainability goals. It was proposed to include a specific note on the compres-
• Operational costs: Optimisation in the FEED stage can sor’s piping and instrumentation diagram (P&ID) concerning
lead to cost savings during the operational phase. A well-de- the seal gas line. This note would emphasise the importance
signed system may require less maintenance, resulting in of minimising pockets, ensuring proper slopes, and providing
reduced operational and maintenance costs over the lifespan adequate heat tracing and insulation. The goal is to maintain
of the compressor. temperatures above the dew point, thereby pre-empting the
• Energy efficiency: Properly designed DGS systems con- need for modifications during the engineering, procurement,
tribute to the overall energy efficiency of the compressor. and construction (EPC) phase, which could lead to change
This can lead to lower energy consumption and reduced orders.
operating costs.
• Long-term asset integrity: The design decisions made Compressor datasheet improvement
during the FEED stage have a long-term impact on the Another recommendation focused on the compressor data-
integrity of the compressor and associated equipment. A sheet, suggesting the inclusion of precise dew point values
well-optimised design helps in preserving the integrity of or graphs. This contrasts with the more generic directive
assets, extending their operational life. of maintaining temperatures ‘20°C above the dew point’.
• Compliance with standards and codes: Adhering to Providing EPC contractors with specific, actionable infor-
industry standards and codes is essential for ensuring the mation from the outset is intended to facilitate a smoother
reliability and safety of the compressor system. Design project execution.
RT Dew Point
RGS Out Ex HP Amine Entrained WTR RGC Suc RGC Discharge Sat Watr CompDis& Seal DE Sealgas To Seal
Vapour fraction 1.0000 1.0000 0.0000 0.9999 1.0000 0.0000 1.0000 1.0000 0.9966
Temperature C 70.00 70.00 70.00 70.00 95.40 70.00 95.40 95.40 60.22
Pressure bar_g 49.00 49.00 49.00 49.00 60.00 49.00 60.00 60.00 57.40
Molecular weight 3.788 3.877 18.02 3.878 3.878 18.02 3.878 3.878 3.878
Mass flow kg/h 20,834.72 21,456.72 10.00 21,466.72 21,466.72 622.01 21,294 171.73 171.73
Phase comp mass flow (overall-H2O) kg/h 0.00 622.01 10.00 632.01 632.01 622.01 626.95 5.06 5.06
HC dew point 59 59 <empty> 59 63 <empty> 63 63 62
Water dew point <empty> 70 264 70 75 264 75 75 74
However, these suggestions were perceived as exceed- an RG gas scrubber treated with lean amine solution, was
ing the traditional scope of FEED, which typically deals with identified as a key area of concern.
pipeline sizes of 4in and larger. The consideration of details
for a DGS line, which is only 1in in diameter, was viewed Analysis: The analysis commenced with the FEED contrac-
as overly granular for the FEED stage. The argument was tor’s recommendation, anchored in API-692 guidelines,
that detailing such specifics in the datasheet might unduly which was added to the compressor mechanical datasheet.
prolong FEED activities, with these aspects being more This recommendation emphasised the need for the seal gas
appropriately addressed during the EPC phase. line to be traced and insulated, potentially incorporating a
While reviewing the EPC deliverables for a diesel hydro- heater to maintain the temperature at least 20°C above the
treater (DHT) recycle gas (RG) compressor, the author, draw- dew point of the seal gas. The EPC contractor was then
ing upon extensive experience in project oversight, identified tasked with carrying out detailed dew point calculations at
critical areas of improvement. This analysis aimed to ensure the compressor discharge, from where dry seal gas tapping
that the design and operational integrity of the compressor is taken.
system met the highest standards of efficiency and relia- The results from the Hysys simulation were pivotal in
bility. The following sections detail specific observations understanding the dew point dynamics of the DGS system.
made during the review, an analytical process undertaken to The Hysys simulation result shows the dew points of the
understand the implications of these findings, and mitigation dry gas, as seen in Table 1.
measures adopted to address these gaps effectively. Simulation output (see Figure 1) coupled with consider-
ations for entrained water, as per GPSA Chapter 7 recom-
Dry gas seal line arrangement mendations, called for a comprehensive analysis. The RG
Findings: The review of EPC deliverables for a DHT recycle compressor’s upstream knock-out drum (KOD) is equipped
gas compressor revealed significant gaps in the DGS line with a demister to enhance gas purity.
arrangement: A critical finding from the analysis was the gap between
• The 1in seal gas line lacked tracing and insulation, a devi- the compressor discharge temperature (95°C) and the
ation from the vendor’s P&ID, which showed tracing and water dew point (75°C). Heat loss calculation indicated a
insulation post-seal gas filters. significant risk of temperature drop in the DGS line, poten-
• EPC P&ID documentation failed to specify the need for tially down to 60°C at the filter inlet. The analysis of heat
sloping the 1in line back towards the tapping point. loss, considering ambient temperatures ranging from 21°C
• Additionally, the EPC contractor’s approach to dew point to 30°C and wind speeds varying from 3 to 15 m/s, empha-
temperature, set at 70°C based on compressor suction from sised the considerable effect of wind velocity on heat dissi-
pation. This significant relationship is depicted in Figure 2.
Compressor suction and discharge dew points
Mitigated measure: In response to the comprehensive
Compressor Compressor analysis, the EPC contractor agreed to trace and insulate
suction discharge the DGS line, aligning with the detailed dew point and heat
Hydrocarbon dew point (°C) 59 63 loss calculations.
Water dew point (°C) 70 75
15 m/s
500 been decided to maintain the tracing temperature at 120°C.
400
9 m/s This approach ensures the dry gas temperature remains at
least 45°C above the dew point, effectively achieving a sim-
300
3 m/s ilar outcome to that of a gas conditioning system.
200
100
Dry gas filter set-up
20 21 22 23 24 25 26 27 28 29 30 Findings: During the evaluation of the compressor sys-
Ambient temperature (˚C) tem’s components, it was discovered that the vendor had
3 m/s 6 m/s 9 m/s 12 m/s 15 m/s not incorporated a separator downstream of the seal gas
filter. This omission meant that any liquid accumulating
Figure 2 Heat loss (w/m) in relation to wind speed (m/s) within the filter could only be removed by isolating the fil-
and ambient temperature (°C). ter, and the filter is isolated when the differential pressure
is high.
lacked a gas conditioning system, as recommended by
API-692. The EPC contractor expressed concerns about Analysis: The absence of a separator posed significant
installing such a system, citing potential impacts on the pro- operational challenges. The filter’s small size and the pres-
ject’s schedule and cost. Additionally, the requirement for sure differential across it meant that detecting and remov-
a gas conditioning system was not explicitly stated in the ing liquid entrainment effectively was problematic. Liquid
FEED deliverables, contributing to the decision against its that condenses and settles at the bottom of the filter can-
implementation. not be adequately measured or separated from the seal gas
Analysis: A gas conditioning system typically involves under these conditions. This situation could lead to opera-
cooling the gas to condense out water and hydrocarbon, tional inefficiencies and potential damage to the DGS due
separating the condensed liquid, and then reheating the to the presence of liquid in the gas stream.
dry gas. The proposed configuration would ensure that Mitigation: To address this critical gap, negotiations with
the temperature across the heater raises the gas to at least the vendor led to an agreement to install a separator on
above the compressor discharge temperature, thereby the filter drain line. This addition is designed with capabil-
maintaining a substantial margin above the dew point. ities for proper liquid level detection and a draining option,
The dew point improvement observed is at least 45ºC, as significantly enhancing the system’s ability to manage and
depicted in Figure 3. Nonetheless, there is a risk of cooling remove entrained liquids. The implementation of a separa-
water leakage from the cooler, which could affect dry gas tor ensures that the DGS operates under optimal conditions
quality. Implementing a cooler, separator, and heater, along by minimising the risk of liquid contamination.
RT Dew Point
Ex HP Entrained CompDis To Dry
RGS Out Amine WTR RGC Suc RGC Out Sat Watr & Seal DE Sealgas Cooler Out Sel Gas V Seal Gas L Gas Seal
Vapour fraction 1.0000 1.0000 0.0000 0.9999 1.0000 0.0000 1.0000 1.0000 0.9927 1.0000 0.0000 1.0000
Temperature C 70.00 70.00 70.00 70.00 95.40 70.00 95.40 95.40 50.00 50.00 50.00 95.00
Pressure bar_g 49.00 49.00 49.00 49.00 60.00 49.00 60.00 60.00 60.00 60.00 60.00 60.00
Molecular weight 3.788 3.877 18.02 3.878 3.878 18.02 3.878 3.878 3.878 3.552 48.14 3.552
Mass flow kg/h 20,834.72 21,456.72 10.00 21,466.72 21,466.72 622.01 21,294.99 171.73 171.73 156.15 15.59 156.15
Phase comp mass kg/h 0.00 622.01 10.00 632.01 632.01 622.01 626.95 5.06 5.06 1.60 3.45 1.60
flow (overall-H2O)
HC dew point 58.60 58.56 <empty> 58.56 62.97 <empty> 62.97 62.97 62.97 50.00 193.5 50.00
Water dew point <empty> 70 263.9 70.37 75.05 263.9 75.05 75.05 75.05 49.99 244.4 49.99
All from
This make-up hydrogen, delivered at a lowing measures are recommended
substantially higher pressure than the for inclusion in FEED deliverables
compressor discharge, originates from related to centrifugal compressors:
one
a make-up gas compressor, which is • P&ID specifications: It should be
generally a lubricating type recipro- explicitly noted that the seal gas line
Source!
cating machine. must be designed to minimise pockets
and ensure a slope back to the tapping
Analysis: Concerns were raised with point from its highest point, with heat
the EPC contractor regarding the tracing and insulation to maintain the
operational risks linked to the make-up temperature well above the dew point.
hydrogen line. It was highlighted that if • Compressor datasheet enhance-
Fast Delivery
operators fail to close this line by mis- ments: The datasheet should include
take, the make-up hydrogen, despite a dew point curve as illustrated in Worldwide
its very low dew point, contains lubri- Annex-B of API-692, providing clear
cant oil, presenting a contamination guidelines for managing temperature
risk to the DGS system. and condensation risks. Duranit®
Catalyst Support
The make-up hydrogen’s movement • External source dry gas use: When
Material
at a speed of 60 m/s, typical for gas dry gas is sourced externally, the data-
velocity within piping systems, signif- sheet and P&ID must highlight that it
Support Plates /
icantly increases the likelihood of the should not preferably be sourced from Grids
entrained lube oil being aerosolised. any lubricating type of compressor.
This high-speed gas flow can mechan- If such gas is utilised, its use should Droplet Separators /
ically fragment the oil into extremely be confined to start-up phases only, Demisters
fine droplets, which may then be con- controlled through a pressure control
veyed as an aerosol through the DGS valve to mitigate risks of contamina- Feed Devices:
filter. The potential for these aerosol tion during normal operation. Gas / Liquids
particles to be smaller than 1 micron, Implementing these proactive meas-
the size for which the seal filter is ures at the FEED stage could poten- Liquid Distributors /
designed, poses a direct threat to sys- tially result in significant cost savings Collectors
tem cleanliness and functionality. for a DHT unit for each shutdown.
Random Packings
Each shutdown event, triggered by
- Ceramic
Mitigation: To address this issue, liquid condensation in the dry gas
- Metall
a decision was made to enhance seal, can lead to the tripping of the RG
- Plastic
the system’s control mechanisms. compressor and result in a production
A pressure control valve (PCV) was downtime of up to 72 hours for the Software and
introduced alongside the procedural DHT unit. Therefore, these recom- Consulting
controls. This PCV is set to activate at mendations not only aim to enhance
a pressure lower than the minimum operational reliability and efficiency www.vff.com
discharge pressure of the compres- but also significantly contribute to the
sor, ensuring the make-up hydrogen overall economic performance of the
line is automatically sealed off when project.
not needed. This additional layer of
control significantly minimises the risk Please
of operator error and the subsequent Rajib Talukder is a Process Specialist in visit us in
contamination by lubricant oil. Aramco’s Global Manufacturing Excellence
Hall 4.0
department, with more than 30 years of expe-
Conclusion rience in process engineering activities work-
Stand F66
The extensive discussions and negoti- ing on numerous projects around the world.
ations, while crucial, have underscored He is a chemical engineer who graduated from
the significant potential costs in terms NIT Tiruchirappalli, India. Vereinigte Füllkörper-Farbriken GmbH & Co.KG
P. O. Box 552, 56225 Ransbach-Baumbach,
of both project schedule and budget. Email: rajib.talukder@aramco.com Germany, +49 26 23 / 895 - 0, info@vff.com
NICHE SOLUTIONS
FOR THE
PETROCHEMICAL
INDUSTRY
Helping our customers to grow profitability
and transform towards circular economy in
the petrochemical market.
Our Solutions Provide the
Following Benefits:
Unrivaled feedstock characterization
C
strategies.
CM
MY
CY
CMY
Narendra Verma
HMEL (HPCL-Mittal Energy Ltd.)
P
etrochemicals are expected to be the largest driver years and then gradually decline. Gasoline will experience
of world oil demand growth, surpassing gasoline the greatest impact because it is primarily used for light-
or diesel by 2030.1 Integrating refining and petro- duty passenger vehicles, and the market for these vehicles
chemicals increases the gross refining margin (GRM) from is shifting toward electric.
value-added products produced by petrochemical units. In contrast, demand for petrochemical feedstocks will
In addition, advantages such as feed security and reliable continue to grow. The major oil-derived petrochemical feed-
petrochemical units facilitate the blending of petrochemical stocks are ethane, liquid petroleum gas (LPG), and naphtha.
byproducts into refinery fuel products, leading to lower cost These are primarily used in the production of polymers for
of conversion and reduced Opex due to shared utilities, ser- plastics, synthetic fibre, and other petrochemical intermedi-
vices, and infrastructure. ates. Demand for these products will continue to grow with
Major refining and integrated oil and gas players reconfig- rising global wealth.
uring their plants to maximise more petrochemicals, rather These two developments pose a dual challenge for the
than just fuel, can extract more value from new products world’s 600 plus refiners. Lower overall demand means
rather than just augmenting the existing product slate. less need for refining capacity. At the same time, remaining
While standalone refineries are gradually exploring inte- refining capacity must evolve to match a shift in product mix
gration opportunities with petrochemical complexes, new to meet petrochemical demand. Refiners will need to find
refineries should feature such integration from the start. The ways to make much less gasoline, marginally less diesel,
extent of integration depends on technical feasibility and the and more jet fuel and petrochemical feedstocks.
resulting economic benefits. This is a complex discussion The world’s refiners must fundamentally rethink how
and requires further detailed study on a case-by-case basis. refineries are designed and operated. Globally, refiners have
On the one hand, the energy transition will reduce the capacity to process nearly 100 million barrels of crude
demand for oil products due to hybrid and energy-efficient oil per day. As global demand declines, refinery utilisation is
vehicles, along with switching from fossil fuels to renewa- expected to drop in the key markets of Western Europe and
bles and carbon footprint minimisation. On the other hand, Asia. By the middle of this decade, utilisation in those mar-
it increases opportunities to capture the growing demand kets could drop from the current rate of 85% to percentages
for petrochemicals. Forward-looking refiners are already in the low 70s.4 Refinery utilisation in North America will
looking for opportunities to adjust or modify their produc- be slightly stronger due to location advantages for export
tion modes to capture growing demand for petrochemicals, markets and crude supply. However, all markets will see a
such as by increasing output of naphtha, propylene, and significant contraction in profit margins due to the general
reformate. Against this backdrop, a case study of HMEL’s decline in utilisation.
refinery and petrochemical integrated Bathinda complex is Overall, the drop in utilisation and profitability could
reviewed. result in capacity closures that will affect the least efficient
The Bathinda base refinery has a capacity of 9.0 MMTPA plants and those less able to adapt to new demands. This
and 5% crude oil-to-chemicals processing capability. scenario leaves most current plants operating but with
Integration with a new petrochemical complex comprising a growing pressure to adapt to new conditions despite nar-
steam cracker and downstream polymer block will increase row margins and decreased cash flows to fund changes. As
this to 20%. Moreover, plans to increase chemicals to 25% demand declines for traditional refined crude oil products,
include diesel cracking as one option besides fuel gas, naph- refiners and chemical companies should consider how to
tha, LPG, and kerosene as feed to the cracker. take advantage of the growing oil-to-chemicals opportu-
nity. To develop the right strategy, refiners and chemical
Existing refinery challenges incumbents will have to reflect on their current strengths,
Analysts forecast that the world’s energy transition will soon capabilities, and positioning.
peak in the use of oil-based fuels, followed by a decline. The Internal competition for higher GRM due to uncertainty in
transportation sector is at the forefront of this trend, with product cracks, fluctuating crude oil prices, and the geopo-
total global demand expected to peak in the next one to two litical situation remains challenging. Also, stricter emission
10000 6000
8000
4000
6000
4000 2000
2000
0
18
15
14
16
17
19
20
21
22
23
24
25
0
20
20
20
20
20
20
20
20
20
20
20
20
18
15
14
16
17
19
20
21
22
23
24
25
-2000
20
20
20
20
20
20
20
20
20
20
20
20
Figure 1 India – ethylene/propylene capacity vs total downstream demand2
norms have been regulating refinery processes. Today, most switching from one strategy to another as profit margins
brownfield projects are solely driven by these regulations. change. Figure 1 shows a huge gap between ethylene and
While costs are associated with these projects for improving propylene production (installed capacity) vs demand in India,
fuel quality, only a small fraction can be recovered from fuel which calls for petrochemical capacity additions.
product price hikes. Besides, market demand and new tech- When considering the projected strong demand for pet-
nologies are setting a platform for refineries to integrate with rochemical products in India (vs installed capacity), Table 1
petrochemical units to generate more value-added products. shows that by 2025/2030, there will be a shortfall in demand
by as much as 8/31.8 MMTPA (vs installed capacity).3
Petrochemicals industry status
Petrochemicals will likely remain the mainstay of all industrial Major challenges
and consumer-based products consumed by the world daily. Major petrochemical industry challenges in India include:
To maintain sustainable profitability, the refinery of the future • Lack of advantageous feedstock, like ethane and pro-
must respond quickly to challenging market conditions, pane, due to limited gas production. India still imports 55%
of required feedstocks and depends upon other
Future petrochemical demand in India countries.
• Surpluses that include ‘scattered’ liquid feed-
S.No. Product Nameplate Demand Shortfall Demand Shortfall
stock, like naphtha and kerosene. Currently, there
capacity 2025 2025 2025 2030 2030
10,048 4,233 13,701 7,886
is no common pool.
1 Polypropylene 5,815
2 LLDPE 2,275 4,154 1,879 5,853 3,578 • Alternate route via gasification demands high
3 HDPE 2,735 4,155 1,420 5,466 2,731 Capex.
4 PVC 1,550 5,147 3,597 6,573 5,023 • High utility and fuel cost; even fuel is required
5 MEG 1,919 3,249 1,330 3,969 2,050 for generation of utilities.
6 Formaldehyde 1,339 1,960 621 2,580 1,241 • Non-allocation of depleting domestic gas, and
7 PX 5,565 5,333 (-232] 6,378 813
there is no new allocation of reservoirs.
8 PTA 7,180 7,910 730 9,620 2,440
• Logistics, especially for cryogenics and natural
9 Butadiene 532 411 (-121] 568 36
10 LOPE 615 1,291 676 1,652 1,037 gas liquid (NGL), currently through ships only.
11 Acetic acid 215 1,256 1,041 1,442 1,227 • Plastic disposal remains a major concern since
12 PET 1,756 1,701 (-55) 2,320 564 it is not biodegradable. There is a need to educate
13 Styrene 0 899 899 1,082 1,082 people, and more effort is required to recycle the
14 Polyether polyol 70 401 331 496 426 waste.
15 Butyl acrylate 180 326 146 413 233
16 LAB 530 734 204 791 261
17 Glycol ethers 0 23 23 31 31 Shortfall MMTPA Shortfall MMTPA
18 Propylene glycol 20 110 90 134 114 Yr 2025 Yr 2030
19 Ethanolamine 30 128 98 170 140 • Commodity 11.7 • Commodity 23.2
20 EPR 30 69 39 90 60 • Niche 6.3 • Niche 8.6
21 Phenol 282 608 326 710 428 • Total 18.0 • Total 31.8
22 Acetone 182 247 65 268 86
23 Hexene-1 0 16 16 31 31 Estimated domestic demand of 25 major petrochemicals
24 Polycarbonate 0 195 195 242 242 products is 50 MMTPA and 65 MMTPA respectively in
25 SAP 0 55 55 71 71 years 2025 and 20303
Table 1
Butene-1
LLDPE/HDPE
Ethylene
Alpha Alpha
olefins alcohol
HDPE
Ethylene Ethylene
Off-gases oxide glycol
Cracker Poly-
Propylene propylene
Himmat Singh
Research Scientist
T
he global blue hydrogen (H2) market size is anticipated but without capturing CO2. It creates around 10 tons of CO2
to grow at a compound annual growth rate (CAGR) of for every ton of hydrogen produced. Therefore, there is a
14.8% in the present decade. The growing emphasis need to find a way to produce clean hydrogen that is less
on clean hydrogen energy with low carbon content, rising carbon intensive.
usage of hydrogen fuel as an active propulsion system in There are two ways to move toward cleaner hydrogen pro-
the automotive industry, and speculation of blue hydrogen duction. One is applying carbon capture and storage (CCS)
working as an enabler of green hydrogen are some reasons for fossil fuel-based hydrogen production processes. Natural
behind the growth of the market. Blue hydrogen has a strong gas-based hydrogen production with CCS is referred to as
role to play in the energy transition by helping to build a blue hydrogen – a low-carbon energy carrier. CO2 storage
hydrogen market while continuing to lower emissions. is typically accomplished by injecting the gas into geologi-
Like the green variety, blue hydrogen at present is expen- cal formations such as saline aquifers or depleted oil fields.
sive to produce compared to the traditional carbon-intensive Green hydrogen is produced by using electricity generated
production processes used today. By reducing blue hydro- from renewable sources, such as wind and solar, to produce
gen’s costs, companies could speed up hydrogen’s much- hydrogen via electrolysis.
vaunted replacement of fossil fuels. European countries have The IEA estimates that the demand for hydrogen today
been working on several projects to boost the adoption of is about 90 mtpa, almost all of which is used for ammonia
blue hydrogen in various regions. production and refining and is forecast to reach about 200
Part 1 is an overview of hydrogen and blue hydrogen in mtpa by 2030 and more than 500 mtpa by 2050. There are
terms of their role in energy mix, estimated demand in the other estimates as well, but they are at variance with the IEA.
coming decades, and committed huge investments. The Meeting this demand will require an unparalleled transfor-
overview also covers emerging and matured blue hydro- mation in how clean hydrogen is produced.
gen technologies, namely steam methane reforming (SMR), Europe and China have committed huge investments
autothermal reforming (ATR), natural gas decomposition through to 2030 to lower carbon via blue and green hydro-
(NGD), and the newly introduced Shell SGP process. gen projects, for an estimated production capacity of more
Part 2 concludes in PTQ Q3 2024 with a detailed compar- than 10 mtpa by 2030. However, this is far below the
ative assessment of cost, greenhouse gas (GHG) emissions, demand forecast, leaving a considerable need for further
and other parameters relating to three mature technologies, projects and investments. While green hydrogen may be the
with a brief appraisal of blue hydrogen market dynamics, better economic option in some locations, blue hydrogen has
end-use insights, and concluding remarks. an advantage in others, and therefore both are needed in the
short and medium term. In short, both types of hydrogen
CO2-to-H2 ratio reinforce each other’s strength.
Hydrogen is the most abundant element in the universe, To meet their Paris Agreement commitments, many coun-
and it could play an essential role in tomorrow’s energy mix, tries are turning to blue hydrogen projects as a medium-term
from fuelling all modes of transport to generating electricity solution for hydrogen mass production while also developing
and powering industry. Colours of hydrogen are increasingly green hydrogen for future production. The UK, for example,
used to distinguish different production methods and as a has prioritised driving the growth of blue hydrogen as part of
proxy to represent the associated environmental impact. its Ten Point Plan to reach its net zero ambitions.
Today, close to 95% of hydrogen production comes from Blue hydrogen shares similarities with grey hydrogen, the
fossil resources. As a result, the carbon dioxide (CO2) emis- key difference being that instead of releasing CO2 into the
sions from hydrogen production are quite high. Grey, black, atmosphere, it is captured and stored. Blue hydrogen has
and brown hydrogen refer to fossil-based production. Grey is been a clear leader in the evolution of the hydrogen industry.
the most common form of production and comes from natu- Carbon capture technologies can be retrofitted onto existing
ral gas, or methane (CH4), using steam methane reformation hydrogen processes or integrated into new plants by design.
Figure 1 Technologies identified and covered by IDTechEx for blue hydrogen production Source: IDTechEx2
Notable plans for hydrogen projects include Air Product’s readiness level. SMR and autothermal and POX processes
announced plans for Europe’s largest blue hydrogen pro- are the mature processes, followed by one modified version
duction plant, and RAG Austria says it has commissioned of SMR and two modified versions of ATR. The remaining
“the world’s first 100% hydrogen storage facility in a porous seven technologies are still at a low level of readiness.
underground reservoir”. The global blue hydrogen market is The Hafenstorm report includes a comparison of the six
projected to grow at a CAGR of 8.0% by 2033. main blue hydrogen production technologies, its 10-year
Further discussions relate to different ways to produce market forecasts for 2023-2033, and supply chains, along
blue hydrogen using different technological paths, taking with seven application areas, major innovations and pro-
into consideration major innovations and challenges, with jects. The report also examines applicable carbon capture,
brief appraisal of blue hydrogen market dynamics, end-use utilisation, and storage (CCUS) technologies and discusses
insights, and concluding remarks (including future outlook). the prospects and challenges of producing blue hydrogen.
Blue hydrogen is going to grow due to global decarbonisa-
Blue hydrogen technologies tion efforts in hard-to-abate sectors, such as oil refining and
Producing hydrogen in a carbon-neutral manner is chal- ammonia production.
lenging (and potentially expensive), but the many colourful IDTechEx2 forecasts that the global blue hydrogen mar-
means of producing hydrogen provide exciting opportuni- ket will grow to reach $34 billion by 2033. Overviewing the
ties. Once available, it can be used in almost every vertical production methods indicates SMR is the most developed
in the energy space. This includes power generation, energy and widespread hydrogen production technology (grey
storage, e-fuels production for aviation and heavy road and hydrogen) used throughout the world. Coal gasification (CG)
rail transport, as well as cement and steel manufacturing, is another popular technology used to produce hydrogen
along with applications in other carbon-intensive industries. (black/brown hydrogen), especially in China. Other processes
Blue hydrogen, produced from fossil fuels with CO2 capture, include partial oxidation (POX), which is useful in converting
is currently viewed as the bridge between the high-emissions waste oil/refining products to hydrogen, as well as the more
grey hydrogen and the limited-scale zero-emission green recently developed ATR of methane. This self-heating steam
hydrogen. Two recent reports on blue hydrogen production reforming process that is more cost-effective than SMR for
technologies highlight different commercially deployed and producing blue hydrogen. Air Liquide and Topsoe are the
emerging hydrogen production processes from fossil fuels major players in this field.
and biofuels, along with recent advancements in hydrogen The IDTechEx report further provides coverage of meth-
storage and transport. ane pyrolysis, which produces hydrogen and solid carbon
Progressive Energy1 for the Bacton Energy Hub (BEH) products, the latter being carbon black in most cases. This
Hydrogen Supply Special Interest Group (SIG) documents technological path is mostly occupied by start-ups and small-
technologies for blue hydrogen production from natural gas er-to-medium enterprises (SMEs), some of which are quickly
with CCS, along with the auxiliary process stages required. commercialising their technologies termed ‘promising tech-
As many as 13 technologies have been listed, along with their nologies’. Other processes fall under the categories of novel
Figure 2 The SBHP and advantages of integration with other technologies Source: Shell Catalysts & Technologies3
4, is then used, as in previous technologies, to increase the According to the Shell Blue Hydrogen Process (SBHP), POX
hydrogen yield and maximise CO2 available for capture in the technology provides substantial savings compared with ATR
process stream. Heat released from this exothermic reaction – a 22% lower levelised cost of hydrogen that derives from
is used to generate steam for the downstream process and lower Capex. This saving comes from a lower Capex owing
more general preheating of other processes: to the potential for a higher operating pressure, leading to
smaller hydrogen compressors, CO2 capture and CO2 com-
CH4 + ½ O2 CO + 2H2 (3)
pressor units, and lower Opex.
CO + H2O CO2 + H2 (4)
Partha S Mondal
Fluor Daniel India Pvt. Ltd
E
ngineers designing protective systems for a process 2oo3, 1oo2D, 2oo2D, and more. The term 1 ‘hardware fault
plant often encounter and justify the use of redun- tolerance’ of N means that N+1 is the minimum number of
dancy. The questions encountered are: faults that could lead to a safety loss.
• How are hardware fault tolerance (HFT) and redundancy The relationship between redundancy (MooN) and HFT
related, and how are they applied in design? is expressed by the formula M-N. For a 1oo2 redundant
• How much redundancy is required to achieve a targeted architecture, the HFT will be 1; for 2oo2, the HFT will be
safety integrity level (SIL)? 0; and for 2oo3, the HFT will be 1. It is important to note
• Why and when can we use route 2H to assess HFT? that redundancy is not the same as HFT. Table 1 provides
• How is redundancy achieved using multiple devices with examples of HFT and redundancy to illustrate the concept.
similar technology or diversified technology?
• What type of redundancy is required in system design Redundancy required to achieve SIL
(1oo2, 2oo2, 2oo2D, 2oo3)? While evaluating SIL of a safety instrumented function
• How can redundancy affect the PFD value during SIL (SIF), major factors deciding achieved SIL include:
calculation, and how does selected architecture affect SIL u Architectural constraints (redundancy)
verification? Target PFDavg or RRF to be achieved
The primary objective of redundancy is to avert any inter- Requirement of systematic capability (SC).
ruption in system operation in the event of a technical failure IEC 61508 and IEC 61511 both define the minimum HFT
in one of the systems. This implies that if a single sensor fails (architectural constraint) requirement, which is required to
to meet performance requirements, leading to a technical fail- meet the target SIL.
ure, redundant or multiple sensors are available without a loss IEC 61508:2010 provides two routes to satisfy the archi-
of functionality. Redundancy is not solely designed to ensure tecture constraints to meet a particular SIL for a particular
plant safety but also to forestall false trips or false alarms and safety instrumented function:1
ensure availability. Parameters to consider when implement- • Route 1H: It is based on safe failure fraction and hardware
ing redundancy in any system architecture will be reviewed fault tolerance of each element.
to help system designers configure and justify redundancy. • Route 2H: It is based on component reliability data from
Table 1 Table 21
Achieving redundancy4
Table 3 When incorporating redundant sensors, design engineers
must account for the impact of common cause failures. To
field feedback/based on data collected in accordance with mitigate this common cause of failure, redundant sensors
international standards (such as IEC 60300-3-2 or ISO may be physically separated (for example, separate tap-
14224). ping) and electrically isolated (for example, wiring with
It is important to note that if route 2H is selected, the relia- separate junction boxes and cables). This serves to dimin-
bility data uncertainties will be considered when calculating ish common environmental stress.
the target failure measure (PFDavg or PFH), and the system An alternative approach to reducing common causes of
will be improved until there is confidence greater than 90% failure in redundant architecture involves using devices
that the target failure measure is achieved. As per route from different manufacturers. While this approach miti-
1H of IEC 61508, the HFT constraint to SIL is described in gates common design and manufacturing defects to some
Table 2. extent, it is important to note that, as the sensors share the
Hence, as per Table 2, if route 1H is selected for type B same technology, they will respond similarly to external
element, SIL 2 can be achieved by HFT=0 and SFF greater disturbances.
than 90%. Diversity in technology presents another avenue for min-
If IEC 61508:2010 route 2H is followed, Table 3 can be imising common mode failures. In this scenario, different
constructed. This is the same as IEC 61511-2016. sensor technologies are employed to measure the same
Example: Two level transmitters are used to design a SIF. variable in a redundant configuration.
The logic solver (PLC) is designed to trip if either transmit- This helps decrease common faults but introduces new
ter detects a dangerous condition (1oo2). To what SIL can challenges, such as differing calibration procedures, varied
this configuration qualify per IEC 61511 or 61508 Route repair methods, potential data mismatches due to distinct
2H? Since the HFT of this configuration is ‘1’, it means if digital rounding practices, diverse maintenance cycles, dis-
one transmitter fails, the other transmitter can still perform parate spare part requirements, and increased operational
the safety function. As per Table 3, the sensor configuration complexity.
can qualify for SIL 3 for any mode. Therefore, when selecting various sensor technologies
The Type A device (valve) can be better understood or manufacturers, it is crucial to weigh multiple trade-
with the following example: If we follow Route 1H and if offs, considering their impact on operations and mainte-
SFF is <60%, then as per Table 2, we require three valves nance processes at sites, as indicated by the Site Safety
in a series to achieve SIL 3, and if SFF is >= to 60%, then Index (SSI). The SSI, a straightforward five-level model, is
we require two valves in series. Hence, designers and designed to evaluate the influence of operations and main-
equipment manufacturers have always tried to prove that tenance processes at a given site.
SFF>=60% to reduce the cost of having an additional valve
while achieving SIL 3. Redundancy and selected architecture affect
Route 2H is a method to calculate the target failure meas- How does redundancy affect PFD value during SIL calcu-
ure (PFDavg) based on the reliability data uncertainty for the lation, and how does selected architecture affect SIL veri-
entire element according to IEC 61508. It is based on the fication? To understand different types of voting logic and
historical data of the device, where the confidence level their impact, simple mathematical equations for PFDavg and
is more than 90%. Instead of determining the safe failure spurious trip rate (STR) can be used.
fraction (SFF), Table 3 can be used to determine the max- These simple approximate equations do not include the
imum possible SIL against each hardware fault tolerance. common cause, diagnostic, proof test coverage, and other
So, if the confidence level can be demonstrated, then HFT key variables.
REFCOMM 2024
®
See you
in Galveston
I picked up a lot of good information. I would
definitely encourage those that are interested
to participate and not to miss this conference.
Gregg Lorimor, Sr. Engineering Specialist,
HollyFrontier Tulsa
Reasons to attend:
Learn from 50+ technical presentation in the multi-track agenda
Agenda covers coking, cat cracking, sulfur and SDA 642
Network with a large audience of refiners and technical experts ATTENDEES
80+
Platinum Sponsors
EXHIBITORS
Bronze Sponsors
1oo3
Input Logic Output
Sensor
circuit solver circuit
2oo2
Input Logic Output
Sensor
circuit solver circuit
Input Logic Output
Sensor
circuit solver circuit
JOIN THE
DISCUSSIONS IN
BARCELONA
“ This event was a great
opportunity to meet world
wide industrial professionals
Now in its 40th year, the event continues to be
an essential annual forum for the global sulphur
and acid community to learn, connect and do
in sulfuric industry, goods & business.
service providers as well as “
operationals, in the same time You can expect an expanded market outlooks
and in the same place. agenda, including expert insights from CRU’s
Ismaili Abdessamad, OCP, analysis teams on major supply and demand
Operations Senior Manager Sulphur + markets, including sulphur, sulphuric acid and
Sulphuric Acid attendee
phosphates, plus additional industry updates
from key players from across the supply chain.
1oo2D voting
The 1oo2D architecture (see Figure 6) resembles 2oo2D
voting, with the key distinction being that the diagnostic
A A B
unit of one channel can can also diagnose the other chan-
nel. To facilitate this, additional control lines were incorpo-
rated, allowing each unit to control both its own diagnostics
and that of the other channel. This significantly enhances
B C C safety compared to 2oo2D voting architecture. However, it
is important to note that the effectiveness of 1oo2D voting
relies heavily on reliable diagnostics.
Based on all reviewed architectures, Table 5 can be for-
mulated to describe different types of architectures.
Figure 3 2oo3 architecture⁷ and 2oo3 voting circuit
Hybrid-based architectures⁶
(short-circuited), the remaining switch ‘B’ is still in a healthy Hybrid fault tolerance/diagnostic-based architectures
condition to de-energise, open, and bring the system to a incorporate various combinations of architectures, such as
safe state when demand occurs. If one has failed safely (for 2oo(1oo2D). For instance, hybrid architecture is a combi-
example, ‘A’ switch is open circuit), the final element is still nation of 1oo2 and 1oo1D. This hybrid approach ensures
energised as ‘B’ and ‘C’ are healthy and continue to provide high safety integrity with a minimal number of modules.
power supply to the final element. The following examples are derived from a hybrid system
PFDavg and STR can be calculated using the equations that offers flexibility in architecture with multiple variations.
in Table 4. In this architecture, both the spurious trip and Configuration 1 (see Figure 7) aims to achieve maximum
PFDavg improve significantly but are still no better than
1oo2 voting in terms of PFDavg and 2oo2 voting (in terms
of spurious trip). Microcomputer engineers have introduced
2oo2D
automatic diagnostics in the redundancy architectures,
Diagnostic circuit
as discussed in the following examples of these types of
redundancy configurations.
Input Logic Output
Sensor
circuit solver circuit
1oo1D
Cristian S Spica
OLI Systems, Inc
T
he call for the energy transition underscores the need identification of plant bottlenecks and the development of
to decarbonise our lifestyles. Numerous pathways mitigation strategies. This includes material of construction
have been identified to contribute to the decarboni- selection, integrity operating window identification, corro-
sation of the transportation sector. It is imperative to incor- sion inhibition package definition, and wash water injection
porate existing assets into these initiatives to swiftly attain design. By accurately predicting corrosion and fouling risk
decarbonisation benefits in the transportation fuel sector. well ahead of introducing renewable feedstock in the unit,
The refining industry has responded to the target of these tools can validate the actual risks post-introduction.
reducing fossil carbon emissions by integrating renewable
feedstock components into refinery operations, aiming to Co-processing impact on hydrotreatment units
lower the carbon intensity of the resulting fuels. Refiners Hydrotreating units (HDTs) are key facilities in upgrading
are employing diverse strategies for processing renewa- renewable feedstocks, such as vegetable oils (for example,
ble feeds. Some are undertaking the construction of new palm, soybean, rapeseed, sunflower, corn, and jatropha)
units dedicated to renewable diesel or sustainable aviation and alternative or waste-based oils, such as used cooking
fuel (SAF). Simultaneously, others are actively engaged in oil (UCO), waste cooking oils (WCOs), fatty acid methyl
adapting their existing facilities – hydrotreaters or fluid cat- esters (FAMEs), free fatty acid (FFA), palm fatty acid distil-
alytic crackers (FCCs) – to co-process a portion of renew- late (PFAD), palm oil mill effluent (POME), tall oil fatty acid
able feeds. (TOFA), into high-quality biodiesel and renewable diesel.
The introduction of new feed components triggers While these feedstocks may vary in appearance and
entirely new reactions, resulting in the formation of new contain differing levels of impurities, such as alkalis and
products that may introduce fresh challenges. Thus, before phosphorus, they are provided in the form of triglycerides
introducing even a minor number of new feedstocks into (TG), which can be viewed as the condensation of glycer-
an existing facility, it is crucial to understand the potential ols and three carboxylic acids. The desired reaction is the
implications and have a clear strategy for mitigating any deoxygenation of the glycerides and free fatty acids in the
associated risks. In such a scenario, having access to a presence of hydrogen to form linear paraffins, according to
robust tool becomes paramount, aiding in identifying the the mechanism in Figure 1.
type and safe percentage of renewable feed that can be There are two pathways for the main reaction. The first
incorporated into the existing feed. favours high yields (increased return), while the second
Ionic modelling tools offer a solution by enabling the minimises hydrogen consumption (decreases costs). The
first pathway (1) involves complete hydrogenation to form
six moles of water, one mole of propane (C3H8), and three
H2–C–OO–C moles of normal paraffins with the same chain length as the
H2 fatty acid chains (n-C18 and n-C22 in the case of rapeseed
H–C–OO–C oil) per mole of reacted triglyceride. This pathway is usually
termed hydrodeoxygenation (HDO).
H2–C–OO–C The second pathway (2) involves a decarboxylation step,
where three moles of carbon dioxide (CO2), one mole of
1 2 1 Hydrogenolysis / Hydrogenation
TG + 12 H2 3 C18H38 + 6 H2O + C3H8
propane, and three moles of normal paraffins with a chain
2 Decarboxylation / Hydrogenation length that is one carbon atom shorter than the fatty acid
TG + 3 H2 3 C17H36 + C3H8 + 3 CO2 chains (n-C17 and n-C21 in the case of rapeseed oil) are
3 Methanation / WGS produced.
CO2 + 4 H2 CH4 + 2 H2O
CO2 + H2 CO + H2O Since both CO2 and carbon monoxide (CO) are produced,
two additional reactions need to be considered, as shown
Figure 1 Triglycerides reaction pathways in Figure 1 (3). Hydrotreating catalysts are known to be
HDS Amine
reactors Wild naphtha
Additional H2 absorber
consumption Sour
H2 Water
gas Gasoil
stripper
H2
Stripper
reboiler
Wash water Increased CO2
injection concentration
HP separator
Corrosion risk
Product
H2 LP separator
Improved cetane number
Lower density
No impact on CFP
Feed
active for both reverse water gas shift (CO2 + H2 CO + • Increased H2 consumption (lipids HDT reactions).
H2O) and methanation (CO + 3H2 CH4 + H2O). The rela- • Reactors cycle length reduction and performance
tive extent of these two reactions determines the observed decrease:
distribution between CO, CO2, and methane (CH4). Catalyst deactivation and reactor pressure drops
If all triglycerides undergo the decarboxylation route, increase:
seven moles of H2 will be consumed, compared to the 16 Production of CH4/CO/CO2/H2O by conversion of
moles of H2 consumed when all triglycerides are converted lipids followed by water gas shift (WGS) equilibrium
via the HDO route. In other words, there would be a 63% and production of C3H8 by removal of the glycerol
reduction in hydrogen (H2) consumption. However, if all the group.
CO2 produced is shifted to CO, and all the CO formed is CoMo catalyst active sites partial Inhibition due to
subsequently converted into CH4, a total of 19 moles of H2 CO.
will be consumed by the decarboxylation route, resulting in H2 partial pressure decrease.
a 19% increase in H2 consumption. Indeed, co-processing Catalyst active sites partial deactivation due to Ca
biofeeds in an HDT (see Figure 2) can have various effects, and Mg phosphate.
both positive (pros) and negative (cons): Fouling issues (phospholipids, peroxides, olefins and
di-olefins, metals).
Pros: • Reactor effluent air cooler (REAC) – fouling, uniform cor-
• Co-processing biofeeds can lead to the production of rosion and pitting corrosion risk by chlorides in the effluent
intermediate and linear paraffins, thereby improving prod- reactor side:
uct properties. Increased organic and inorganic chloride concentration
LP separator
HC Wild
HP
Stripping Diesel naphtha
separator LP separator
water water steam product
* For long-term service to be deemed acceptable, the corrosion rate must remain below 0.125 mm per year. The corrosion rates were determined based on the operational fluid velocity,
which consistently remained below 6 m/s.
Table 3
balance was carried out on OLI Systems’ proprietary leg- Another factor contributing to NH4HS corrosion is the
acy process simulator, Flowsheet ESP, by reproducing the presence of oxygen, which can accelerate corrosion. It is
DHT process scheme and simulating its behaviour at aver- recommended that wash water be oxygen-free, with a
age process and contaminant level conditions (see Table suggested oxygen limit of <15 ppbwt.
1). Each stream was then analysed in the proprietary OLI From the first phase evaluation, it is evident that the unit
Studio: Corrosion & Stream Analyzer to gain insights into is already experiencing some issues, with corrosion and
the electrolyte chemistry and assess the corrosion and foul- fouling, particularly chloride salt deposition, being the most
ing risk of each of the streams. problematic. Therefore, any increase in chloride intake does
It is worth noting that the unit is equipped with both inter- not seem feasible. Additionally, the chloride content in the
mittent and continuous wash water injection facilities. These fossil feed should be limited. Alternatively, a new intermit-
are situated upstream and downstream of HEX 10, respec- tent wash water injection facility with a capacity of 15-20
tively, with an overall injection rate of 15-20 t/h of wash
water. The wash water mainly originates from the vacuum
distillation unit (VDU), which explains the elevated concen- Another factor contributing to
trations of ammonia, sulphides, and observed oxygen. NH4HS corrosion is the presence
The result of the ionic survey evidenced the following
issues:
of oxygen, which can accelerate
• Fouling issue: Dry NH4Cl β-solid formation in HEX 08-09 corrosion
tube bundles with a RH of 0.7%.
• Fouling issue: NH4Cl α-solid formation in HEX 09-10
tube bundles (the NH4Cl β-crystalline form transition to the t/h should be installed upstream of HEX 08 to dissolve the
α-form at 185⁰C) with a RH of 1.0%. dry chloride salts formed. Additionally, an upgrade in the
• NH4HS corrosion: AC 01 outlet – HP and LP separator material of construction (MoC) for the carbon steel HEX tube
outlet (NH4HS concentration 3.0-5.5%wt ). bundles and piping should be implemented. This upgrade is
NH4Cl deposition in HEX 08-09 is a commonly over- necessary to protect HEX 08-09 from chloride stress cor-
looked occurrence, with no proactive measures taken to rosion cracking due to intermittent water washing and to
address it (for example, feedstock quality control, feed resist NH4HS corrosion in REAC downstream cold areas.
pretreatment, wash water or salt dispersant injection, or
operational parameter change). However, it is widely rec- Going forward
ognised in both literature and practical plant experience The second phase of the project to be discussed in PTQ Q3
that there is a significant risk of pitting corrosion when rel- 2024 assesses the behaviour of the unit when incorporat-
ative humidity exceeds 10-15% for carbon steel material. ing different biogenic feedstocks and different rates, with
While NH4Cl is not corrosive and primarily causes fouling a minimum target of 10 wt%. The feasibility study results
when dry, it becomes corrosive when exposed to moisture revealed that achieving the targeted minimum biofeedstock
due to its hygroscopic nature. incorporation rate of 10% was not feasible with the cur-
The NH4HS corrosion issue is relatively mild, extending rent DHT configuration, necessitating significant capital
from the AC01 outlet to the LP separator cold and diesel expenditure (Capex) for implementation.
stripper overhead areas (see Table 3). This corrosion is sen-
sitive to both concentration and velocity. While engineering Cristian Spica is a Senior Application Engineer with OLI Systems,
best practices suggest that a concentration of 2%wt and a Inc. He has 12 years of experience in chemical engineering activities
velocity of 6.0 m/sec are the safe operation limits for carbon including process engineering, modelling and simulation, engineering,
steel, rather than solely relying on these practices, we con- design and construction, project management, technical support and
ducted an analysis of corrosion rates and the risk of pitting training. He holds a chemical engineering degree from the University
corrosion to which the unit is exposed by surveying the of L’ Aquila, Italy (MSc equivalent) and is an Executive MBA candidate
streams highlighted in purple in the proprietary OLI Studio: at Luiss Business School in Rome.
Corrosion Analyzer (see Figure 3). Email:cristian.spica@olisystems.com
ptq
Q3 2023
REFINING
GAS PROCESSIN
PETROCHEMICAL
G
S
gas
2023
TR AN PE RFORRE SS OR
19/06/2023 10:00:57
ROLE SIT IO NA L MA NC E
PTQ supplem
ent NATU RA
TO AM MOL GAS
gas cover.indd
1
NI A
13/03/2023
11:14:26
Visit our
website
for more
information AND MANY MORE...
BOOK
TO
JOIN
100%
of 2023 attendees said they are
likely to attend this event again
refiningindia.com
PLATINUM PARTNER GOLD PARTNERS BRONZE PARTNER
BROUGHT TOBROUGHT
BROUGHT TO
YOU BYBYTO YOU BY
YOU
with
withsupport
support
from IDS
from IDS
RefiningIndia.indd
Refining India 210x297
1 12324.indd 1 12/03/2024
13/03/2024 16:40
10:10:23