Refinery Process
Refinery Process
Sponsored by:
Technology Solutions
Processes index - 1 [next page]
Alkylation Desulfurization Hydrocracking (ISOCRACKING)
Alkylation, catalytic Dewaxing Hydrocracking (LC-FINING)
Alkylation--feed preparation Dewaxing/wax deoiling Hydrocracking-residue
Alkylation-HF Diesel-ultra-low-sulfur diesel (ULSD) Hydrodearmatization
Alkylation-sulfuric acid Diesel-upgrading Hydrofinishing
Aromatics Ethers Hydrofinishing/hydrotreating
Aromatics extractive distillation Ethers-ETBE Hydrogen
Aromatics recovery Ethers-MTBE Hydrogenation
Benzene reduction Flue gas denitrification Hydrogen-HTCT and HTCR twin plants
Biodiesel Flue gas desulfurization-SNOX Hydrogen-HTER-p
Catalytic dewaxing Fluid catalytic cracking Hydrogen-methanol-to-shift
Catalytic reforming Fluid catalytic cracking-pretreatment Hydrogen-recovery
Coking Gas treating-H2S removal Hydrogen-steam reforming
Coking, fluid Gasification Hydrogen-steam-methane
Coking,flexi Gasoline desulfurization reforming (SMR)
Crude distillation Gasoline desulfurization, ultra deep Hydroprocessing, residue
Crude topping units H2S and SWS gas conversion Hydroprocessing, ULSD
Deasphalting H2S removal Hydrotreating
Deep catalytic cracking Hydroconversion-VGO & DAO Hydrotreating (ISOTREATING)
Deep thermal conversion Hydrocracking Hydrotreating diesel
Sponsored by:
Technology Solutions
Processes index - 2 [previous page]
Hydrotreating/desulfurization Olefins-butenes extractive distillation Treating-gases
Hydrotreating-aromatic saturation Olefins-dehydrogenation of Treating-gasoline and LPG
Hydrotreating-lube and wax light parraffins to olefins Treating-gasoline desulfurization,
Hydrotreating-RDS/VRDS/UFR/OCR Oligomerization-C3/C4 cuts ultra deep
Hydrotreating-resid Oligomerization-polynaphtha Treating-gasoline sweetening
Hydrotreating-residue Paraxylene Treating-kerosine and heavy naphtha
Prereforming with feed ultra purification sweetening
Isomerization
Pressure swing adsorption-rapid cycle Treating-phenolic caustic
Isooctene/isooctane
Refinery offgas-purification and Treating-pressure swing adsorption
Lube and wax processing
olefins recovery Treating-propane
Lube extraction
Resid catalytic cracking Treating-reformer products
Lube hydrotreating
Slack wax deoiling Treating-spent caustic deep neutralization
Lube oil refining, spent
SO2 removal, regenerative Vacuum distillation
Lube treating
Sour gas treatment Visbreaking
Mercaptan removal
Spent acid regneration Wax hydrotreating
NOx abatement
Spent lube oil re-refining Wet gas scrubbing
NOx reduction, low-temperature
Sulfur processing Wet Scrubbing system, EDV
O2 enrichment for Claus units
Sulfur recovery White oil and wax hydrotreating
O2 enrichment for FCC units
Thermal gasoil
Olefin etherfication
Treating jet fuel/kerosine
Olefins recovery
Sponsored by:
Technology Solutions
Company index
ABB Lummus Global GTC Technology Inc.
Air Products and Chemicals, Inc. Haldor Topsoe
Axens Kobe Steel Ltd.
Bechtel Linde AG
Belco Technologies Corp. Lurgi
CB&I Merichem Chemicals & Refinery Services LLC
CDTECH Process Dynamics, Inc.
Chevron Lummus Global LLC. Refining Hydrocarbon Technology LLC
ConocoPhillips Shaw Stone &Webster
Davy Process Technology Shell Global Solutions International BV
DuPont Technip
ExxonMobil Engineering & Research Uhde GmbH
Foster Wheeler UOP LLC
Gas Technology Products
Genoil Inc.
Goar, Allison & Associates
Sponsored by:
Technology Solutions
ABB Lummus Global
Alkylation
Coking
Fluid catalytic cracking
Hydrotreating
Hydrotreating-aromatic saturation
Air Products and Chemicals, Inc.
Hydrogen-recovery
Olefins recovery
Axens
Alkylation-feed preparation
Benzene reduction
Catalytic reforming
Ethers
Gasoline desulfurization, ultra deep
Hydroconversion-VGO & DAO
Hydrocracking
Hydrocracking-residue
Hydrotreating diesel
Hydrotreating-resid
Isomerization
Lube oil refining, spent
Oligomerization-C3/C4 cuts
Oligomerization-polynaphtha
Spent lube oil re-refining
Bechtel
Dewaxing
Dewaxing/wax deoiling
Lube extraction
Lube extraction
Lube hydrotreating
Lube hydrotreating
Wax hydrotreating
Belco Technologies Corp.
NOx reduction, low-temperature
SO2 removal, regenerative
Wet Scrubbing system, EDV
CB&I
Catalytic reforming
Crude topping units
Hydrogen-steam reforming
Hydrotreating
CDTECH
Alkylation, catalytic
Hydrogenation
Hydrotreating
Isomerization
Chevron Lummus Global LLC.
Dewaxing
Hydrocracking (ISOCRACKING)
Hydrocracking (LC-FINING)
Hydrofinishing
Hydrotreating (ISOTREATING)
Hydrotreating-RDS/VRDS/UFR/OCR
Processes:
Alkylation
Coking Technology Solutions
Gasoline desulfurization Technology Solutions, a division of ConocoPhillips, is a premier provider of technology solutions for
the vehicles of today and the oilfields and energy systems of tomorrow. Backed by modern research facilities
Isomerization and a strong tradition of innovation, we develop, commercialize and license technologies that help oil and
gas producers, refiners and manufacturers reach their business and operational. From enhanced production
methods, to gasoline sulfur removal processes to valuable catalysts that enhance fuel cell operation, Tech-
nology Solutions prepares producers, refiners and consumers alike for a cleaner, more beneficial future.
Strengths of Our Business
• Focused efforts on developing and commercializing technologies that enable refiners to economically
produce clean fuels and upgrade hydrocarbons into higher value products
• Strategic alignment with both Upstream and Downstream energy segments to effectively capitalize on
extensive R&D, commercial and operational expertise
• Strong relationship-building and problem-solving abilities
• Customer inter-facing and advocacy
Industries Served
Technology Solutions supports both Upstream and Downstream energy segments, including:
• Carbon and petroleum coke
• Gasification
• Sulfur chemistry
• Hydrocarbon processing and upgrading
• Upstream technologies
• Enhanced recovery
Corporate Overview
ConocoPhillips (NYSE:COP) is an international, integrated energy company. It is the third-largest integrat-
ed energy company in the United States, based on market capitalization, and oil and gas proved reserves
and production; and the second largest refiner in the United States. Worldwide, of non government-con-
trolled companies, ConocoPhillips has the fifth-largest total of proved reserves and is the fourth-largest
refiner. Headquartered in Houston, Texas, ConocoPhillips operates in more than 40 countries. As of March
31, 2006, the company had approximately 38,000 employees worldwide and assets of $160 billion.
Technical articles: visbreaking, and providing cost-effective solutions for the refining industry.
Services:
• Integrated hydrogen solutions: • Market studies
• Master planning
Combining hydrogen recovery • Feasibility studies
and optimized steam • Concept screening
• Environmental engineering
• Upgrade refinery residuals into • Front-end design (FEED)
value-added products • Project management (PMC)
• Engineering (E)
• Optimize turnaround projects • Procurement (P)
• Drivers for additional delayed • Construction (C) & construction management (Cm)
• Commissioning & start-up
coking capacity in the refining • Validation
industry • Plant operations & maintenance
• Training
• When solvent deasphalting is
the most appropriate technology Our Global Power Group, world-leading experts in combustion technology, designs, manufactures and
for upgrading residue erects steam generating and auxiliary equipment for power stations and industrial markets worldwide, and
also provides a range of after-market services.
Email: ann_hooper@fwhou.fwc.com
Web: www.fosterwheeler.com
Gas Technology Products
H2S removal
H2S removal
H2S removal
Genoil Inc.
Hydrotreating-residue
Goar, Allison & Associates
Sulfur processing
Sulfur recovery
GTC Technology Inc.
Aromatics
Aromatics recovery
Desulfurization
Paraxylene
Haldor Topsoe
Diesel-ultra-low-sulfur diesel (ULSD)
Diesel-upgrading
Flue gas denitrification
Flue gas desulfurization-SNOX
Fluid catalytic cracking-pretreatment
H2S and SWS gas conversion
Hydrocracking
Hydrodearmatization
Hydrogen-HTCT and HTCR twin plants
Hydrogen-HTER-p
Hydrogen-methanol-to-shift
Hydrogen-steam-methane reforming (SMR)
Hydrotreating
Sour gas treatment
Spent acid regneration
Kobe Steel Ltd.
Hydrocracking
Linde AG
O2 enrichment for Claus units
O2 enrichment for FCC units
Lurgi
Biodiesel
Merichem Chemicals & Refinery Services LLC
Treating jet fuel/kerosine
Treating-gases
Treating-gasoline and LPG
Treating-gasoline desulfurization, ultra deep
Treating-gasoline sweetening
Treating-kerosine and heavy naphtha sweetening
Treating-phenolic caustic
Treating-propane
Treating-reformer products
Treating-spent caustic deep neutralization
Process Dynamics, Inc.
Hydrotreating
Hydrotreating-lube and wax
Lube and wax processing
Refining Hydrocarbon Technology LLC
Alkylation
Isooctene/isooctane
Olefin etherfication
Shaw Stone & Webster
Deep catalytic cracking
Fluid catalytic cracking
Refinery offgas-purification and olefins recovery
Resid catalytic cracking
Shell Global Solutions International BV
Crude distillation
Deep thermal conversion
Fluid catalytic cracking
Gasification
Hydrocracking
Hydroprocessing, residue
Thermal gasoil
Visbreaking
Technip
Crude distillation
Hydrogen
Uhde GmbH
Aromatics extractive distillation
Ethers-ETBE
Ethers-MTBE
Hydrofinishing/hydrotreating
Hydrogen
Lube treating
Olefins-butenes extractive distillation
Olefins-dehydrogenation of light parraffins to olefins
Slack wax deoiling
Vacuum distillation
White oil and wax hydrotreating
Processes:
Alkylation (2)
Alkylation-HF For more than 90 years, UOP LLC, a Honeywell company, has been a leader in developing and com-
mercializing technology for license to the oil refining, petrochemical and gas processing industries. Starting
Catalytic reforming with its first breakthrough technology, UOP has contributed processes and technology that have led to ad-
vances in such diverse industries as motor fuels, plastics, detergents, synthetic fibers and food preservatives.
Fluid catalytic cracking UOP is the largest process licensing organization in the world, providing more than 50 licensed processes
Hydrocracking for the hydrocarbon processing industries and holding more than 2,500 active patents.
UOP offices are in Des Plaines, Illinois, USA (a northwest suburb of Chicago). The company employs
Hydrotreating (2) nearly 3,000 people in its facilities in the United States, Europe and Asia.
The petroleum refining industry is the largest market for UOP technology, products and services. UOP
Hydrotreating/desulfurization processes are used throughout the industry to produce clean-burning, high-performance fuels from a vari-
ety of hydrocarbon products. For example, for 60 years our Platforming process has been used to upgrade
Isomerization (3) low-octane naphtha to high-octane unleaded gasoline, a higher performance fuel. Other technologies con-
vert mercaptans to innocuous disulfides, remove sulfur from fuel, and recover high-purity hydrogen from
Mercaptan removal impure gas streams.
Technologies developed by UOP are almost entirely responsible for providing the fundamental raw
Treating-pressure swing materials – benzene, toluene and xylene (BTX) – of the aromatics-based petrochemicals industry. These
products form the basis of such familiar products as synthetic rubber, polyester fibers, polystyrene foam,
adsorption glues and pharmaceuticals. UOP technologies produce such olefins as ethylene and propylene, used in a
range of products from contact lenses to food packaging. UOP has been active in the development of syn-
thetic detergent chemicals since 1947, and today almost half of the world’s soft (biodegradable) detergents
Technical articles: are produced through UOP-developed processes.
UOP’s gas processing technologies are used for separating, drying and treating gases produced from oil
• Concepts for an overall refinery and gas wells and atmospheric gases.
energy solution through novel in- UOP is the world’s leading producer of synthetic molecular sieve adsorbents used in purifying natural gas,
separating paraffins and drying air through cryogenic separation. Molecular sieves also are used in insulat-
tegration of FCC flue gas power ing glass, refrigeration systems, air brake systems, automotive mufflers and deodorizing products.
recovery UOP provides engineering designs for its processes, and produces key mechanical equipment for some of
its processes. It also offers project management, cost estimation, procurement and facility-design services.
• Changing refinery configura- UOP’s staff of engineers provides customers with a wide range of services, including start-up assistance,
tion for heavy and synthetic operating technical services such as process monitoring and optimization, training of customer personnel,
crude processing catalyst and product testing, equipment inspection, and project management.
For more information:
jennifer.wilson@uop.com
Alkylation
Application: The AlkyClean process converts light olefins into alkylate
by reacting the olefins with isobutane over a true solid acid catalyst.
AlkyClean’s unique catalyst, reactor design and process scheme allow ��������� ��������
operation at low external isobutane-to-olefin ratios while maintaining
excellent product quality. ��������������
������� �������
����������
Products: Alkylate is a high-octane, low-Rvp gasoline component used ������ ������������
��������
��� ���
for blending in all grades of gasoline. ��������
�������
Description: The light olefin feed is combined with the isobutane make-
��������
up and recycle and sent to the alkylation reactors which convert the ��������
������������
olefins into alkylate using a solid acid catalyst (1). The AlkyClean process ���
uses a true solid acid catalyst to produce alkylate, eliminating the safety
and environmental hazards associated with liquid acid technologies. Si-
multaneously, reactors are undergoing a mild liquid-phase regeneration
using isobutane and hydrogen and, periodically, a reactor undergoes a
higher temperature vapor phase hydrogen strip (2). The reactor and mild
regeneration effluent is sent to the product-fractionation section, which
produces n-butane and alkylate products, while also recycling isobutane
and recovering hydrogen used in regeneration for reuse in other refinery Installation: Demonstration unit at Neste Oil’s Porvoo, Finland Refinery.
hydroprocessing units (3). The AlkyClean process does not produce any
acid soluble oils (ASO) or require post treatment of the reactor effluent or Reference: “The AlkyClean process: New technology eliminates liquid
final products. acids,” NPRA 2006 Annual Meeting, March 19–21, 2006.
D’Amico, V., J. Gieseman, E. von Broekhoven, E. van Rooijen and
Product: The C5+ alkylate has a RON of 93–98 depending on processing H. Nousiainen, “Consider new methods to debottleneck clean alkylate
conditions and feed composition. production,” Hydrocarbon Processing, February 2006, pp. 65–70.
Economics: Licensor: ABB Lummus Global, Albemarle Catalysts and Neste Oil.
Investment (2006 USGC basis 10,000-bpsd unit) $/bpsd 4,200
Operating cost, $/gal 0.08
Alkylation
Application: Convert propylene, butylenes, amylenes and isobutane to
the highest quality motor fuel using ReVAP (Reduce Volatility Alkylation
�����������������
Process) alkylation.
�
Products: An ultra-low-sulfur, high-octane and low-Rvp blending stock
�������
for motor and aviation fuels.
���������� �
Description: Dry liquid feed containing olefins and isobutane is charged �����
�
to a combined reactor-settler (1). The reactor uses the principle of dif-
ferential gravity head to effect catalyst circulation through a cooler pri- �����������������
or to contacting highly dispersed hydrocarbon in the reactor pipe. The
���������
hydrocarbon phase that is produced in the settler is fed to the main
��������
fractionator (2), which separates LPG-quality propane, isobutane recycle, �����
��������
Description: Plants are designed to process a mixture of propylene, �������
butylenes and amylenes. Olefins and isobutane-rich streams along with ����������
a recycle stream of H2SO4 are charged to the STRATCO Contactor reac- �����
tor (1). The liquid contents of the Contactor reactor are circulated at high
velocities and an extremely large amount of interfacial area is exposed ��������
between the reacting hydrocarbons and the acid catalyst from the acid �����
settler (2). The entire volume of the liquid in the Contactor reactor is main-
tained at a uniform temperature, less than 1°F between any two points
within the reaction mass. Contactor reactor products pass through a flash
drum (3) and deisobutanizer (4). The refrigeration section consists of a
compressor (5) and depropanizer (6).
The overhead from the deisobutanizer (4) and effluent refrigerant Utilities, typical per bbl alkylate:
recycle (6) constitutes the total isobutane recycle to the reaction zone. Electricity, kWh 13.5
This total quantity of isobutane and all other hydrocarbons is maintained Steam, 150 psig, lb 180
in the liquid phase throughout the Contactor reactor, thereby serving to Water, cooling (20°F rise), 103 gal 1.85
promote the alkylation reaction. Onsite acid regeneration technology is Acid, lb 15
also available. Caustic, lb 0.1
Product quality: The total debutanized alkylate has RON of 92 to 96 Installation: Over 600,000 bpsd installed capacity.
clear and MON of 90 to 94 clear. When processing straight butylenes,
Reference: Hydrocarbon Processing, Vol. 64, No. 9, September 1985,
the debutanized total alkylate has RON as high as 98 clear. Endpoint of
pp. 67–71.
the total alkylate from straight butylene feeds is less than 390°F, and less
than 420°F for mixed feeds containing amylenes in most cases. Licensor: DuPont.
Economics (basis: butylene feed):
Investment (basis: 10,000-bpsd unit), $ per bpsd 4,500
Alkylation
Application: The RHT-Alkylation process is an improved method to react ��������������
C3– C5 olefins with isobutane using the classical sulfuric acid alkylation
��������
process. This process uses a unique mixing device — eductor(s) — that
��������� ��
provides low-temperature (25 – 30°F) operations at isothermal condi- ���
�
�������
�
����������� ��������� ������������
tions. This eductor mixing device is more cost-effective than other de-
vices being used or proposed. It is maintenance free and does not re- ���
��������������������� ���������
quire replacement every two to three years. This mixing device can be a
�
retrofit replacement for existing contactors. In addition, the auto refrig- ������� ������� � ����� �
������ ������� ��������� �������� ��������
eration vapor can be condensed by enhancing pressure and then easily ������� �������������
��������
���� ��������� ����������
absorbed in hydrocarbon liquid, without revamping the compressor. ������������
����������������
�����������
Description: In the RHT-Alkylation, C3– C5 feed from FCC or any other �����������
����������
source including steam cracker, etc., with isobutane make-up, recycle
isobutene, and recovered hydrocarbons from the depropanizer bottom ���������������
and refrigeration vapors are collected in a surge drum — the C4 system
(5). The mixture is pumped to the reactor (1) to the eductor suction port.
The motive fluid is sent to the eductor nozzle from the bottom of reac-
tor, which is essentially sulfuric acid, through pumps to mix the reactants
with the sulfuric-acid catalyst.
The mixing is vigorous to move the reaction to completion. The The hydrocarbon is sent to distillation column(s) (7), to separate alkyl-
makeup acid and acid-soluble oil (ASO) is removed from the pump dis- ate product and isobutane, which is recycled. The butane is sent to offsites
charge. The process has provisions to install a static mixer at the pump or can be converted back to isobutane for processing units requirements.
discharge. Some feed can be injected here to provide higher OSV, which The auto refrigeration occurs in the reactor at temperatures 25–30°F. The
is required for C3 alkylation. Reactor effluent is withdrawn from the isothermal condition lowers acid consumption and yields higher octane
reactor as a side draw and is sent to acid/ hydrocarbon coalescer (2) product due to improved selectivity of 2,4,4 trimethylpentane.
where most of the acid is removed and recycled to the reactor (1). The The auto-refrigeration vapor is compressed (or first enhanced the
coalescers are being used by conventional process to reduce the acid in pressure by the ejector and then absorbed in a heavy liquid — alkylate,
the hydrocarbon phase to 7–15 wppm. The enhanced coalescer design which provides a low-cost option) and then condensed. Some liquid is
RHT can reduce the sulfuric acid content in the hydrocarbon phase to sent to depropanizer (6); propane and light ends are removed. The bot-
negligible levels (below <1 wppm). toms are recycled to C4 system and sent to the reactor.
After the coalescer, the hydrocarbon phase is heated and flashed The major advances of RHT process are threefold: eductor mixing
increasing the alkylate concentration in the hydrocarbon, which is sent device, advance coalescer system to remove acid from hydrocarbon (dry
through the finishing coalescer where essentially all of the remaining system), and C4 autorefrigeration vapors recovery by absorption, mak-
acid is removed. ing compressor redundant.
Continued
Alkylation, continued Commercial units: Technology is ready for commercialization.
References:
Economics: For a US Gulf Coast unit 1Q 2006 with a capacity of 10,000 US patent 5,095168.
bpd alkylate unit US Patent 4,130593.
CAPEX ISBL, MM USD 31.2 Kranz, K., “Alkylation Chemistry,” Stratco, Inc., 2001.
Utilities ISBL costs, USD/ bbl alkylate 3,000 Branzaru, J., “Introduction to Sulfuric Acid Alkylation,” Stratco, Inc.,
Power, kWh 4,050* 2001.
Water, cooling, m3/ h 1,950 Nelson, Handbook of Refining.
Steam, kg / h 25,600 Meyers, R. A., Handbook of Refining, McGraw Hill, New York,
1997.
* Power could be less for absorption application
FCC Feed (about 15% isobutelene in C4 mixed stream) Licensor: Refining Hydrocarbon Technologies LLC.
Product properties: Octane (R+M) / 2:94.8 – 95.4
Alkylation
Application: The Alkad process is used with HF alkylation technology to �����������������
�����������������������
reduce aerosol formation in the event of an HF release, while maintain-
����������������
ing unit operability and product quality. The Alkad process is a passive
mitigation system that will reduce aerosol from any leak that occurs
while additive is in the system. ��������
��������
����� ����
Description: The additive stripper sends acid, water and light-acid sol- ������� �������� ������������
uble oils overhead and on to the acid regenerator. Heavy acid soluble ���������� ������
oils and the concentrated HF-additive complex are sent to the additive
���������� ����
stripper bottoms separator. From this separator the polymer is sent to ����
��������������
neutralization, and the HF-additive complex is recycled to the reactor
section. The acid regenerator removes water and light-acid soluble oils
from the additive stripper overhead stream. The water is in the form of ����������
a constant boiling mixture (CBM) of water and HF. ��������������������
��������������������������� �����������������
There is no expected increase in the need for operator manpower.
Maintenance requirements are similar to equipment currently in stan-
dard operation in an HF alkylation unit in similar service.
duction.
Good hydrogen distribution and reactor design eliminate channeling
���������������
while enabling high turndown ratios. Butene yields are maximized, hy- ������� �����������������
drogen is completely consumed and, essentially, no gaseous byproducts
or heavier compounds are formed. Additional savings are possible when
pure hydrogen is available, eliminating the need for a stabilizer. The pro-
cess integrates easily with the C3/C4 splitter.
Alkyfining performance and impact on HF alkylation product:
The results of an Alkyfining unit treating an FCC C4 HF alkylation Annual savings for a 10,000-bpsd alkylation unit:
unit feed containing 0.8% 1,3-butadiene are: HF unit, US$ 4.1 million
Butadiene in alkylate, ppm < 10 H2SO4 unit, US$ 5.5 million
1-butene isomerization, % 70
Butenes yield, % 100.5
Installation: Over 90 units are operating with a total installed capacity
of 800,000 bpsd.
RON increase in alkylate 2
MON increase in alkylate 1 Licensor: Axens.
Alkylate end point reduction, °C –20
The increases in MON, RON and butenes yield are reflected in a
substantial octane-barrel increase while the lower alkylate end point re-
duces ASO production and HF consumption.
Economics:
Investment:
New unit ISBL cost:
For an HF unit, $/bpsd 430
For an H2SO4 unit, $/bpsd 210
Alkylation—HF
Application: HF Alkylation improves gasoline quality by adding clean-
burning, mid-boiling-range isoparaffins and reducing gasoline pool va-
por pressure and olefin content by conversion of C3– C5 olefin compo- �������
�����
nents to alkylate.
Description: The alkylation reaction catalytically combines C3– C5 olefins ��������
with isobutane to produce motor-fuel alkylate. Alkylation takes place in ����
����������
the presence of HF catalyst under conditions selected to maximize alkyl-
ate yield and quality.
The reactor system is carefully designed to ensure efficient contact- �������
����
����
ing and mixing of hydrocarbon feed with the acid catalyst. Efficient heat ������� �������
�������
transfer conserves cooling water supply. Acid inventory in the reactor
system is minimized by combining high heat-transfer rates and lower ��������
total acid circulation.
Acid regeneration occurs in the acid regenerator or via a patented
internal-acid-regeneration method. Internal regeneration allows the
refiner to shutdown the acid regenerator, thereby realizing a utility
savings as well as reducing acid consumption and eliminating polymer
disposal.
Feed: Alkylation feedstocks are typically treated to remove sulfur and
water. In cases where MTBE and TAME raffinates are still being pro-
cessed, an oxygenate removal unit (ORU) may be desirable.
Selective hydrogenation of butylene feedstock is recommended to
reduce acid regeneration requirements, catalyst (acid) consumption and
increase alkylate octane by isomerizing 1-butene to 2-butene.
Efficiency: HF Alkylation remains the most economically viable method
for the production of alkylate. The acid consumption rate for HF Alkyla-
tion is less than 1/100th the rate for sulfuric alkylation units. And un-
like sulfuric alkylation units, HF Alkylation does not require refrigeration
equipment to maintain a low reactor temperature.
Installations: Over 20 UOP licensed HF alkylation units are in operation.
Licensor: UOP LLC.
Alkylation, sulfuric acid
Application: Autorefrigerated sulfuric-acid catalyzed process that com- Propane product
bines butylene (and propylene or pentylene if desired) with isobutane
to produce high-octane gasoline components that are particularly at-
tractive in locations that are MON limited. Technology can be installed 3
2
grassroots or retrofit into existing alkylation facilities. Recycle
Refrigerant Butane
isobutane product
Products: A low-sulfur, low-Rvp, highly isoparaffinic, high-octane (espe-
cially MON) gasoline blendstock is produced from this alkylation process.
1
4
Description: Olefin feed and recycled isobutane are introduced into the 5 6
stirred, autorefrigerated reactor (1). Mixers provide intimate contact be-
Olefin feed
tween the reactants and acid catalyst. Highly efficient autorefrigeration
removes heat of reaction heat from the reactor. Hydrocarbons, vaporized START
Alkylate
Makeup
from the reactor to provide cooling, are compressed (2) and returned to Recycle acid isobutane product
the reactor. A depropanizer (3), which is fed by a slipstream from the
refrigeration section, is designed to remove any propane introduced to
the plant with the feeds.
Hydrocarbon products are separated from the acid in the settler
containing proprietary internals (4). In the deisobutanizer (5) isobutane
is recovered and recycled along with makeup isobutane to the reactor. Steam, lb 200
Butane is removed from alkylate in the debutanizer (6) to produce a H2SO4, lb 19
low-Rvp, high-octane alkylate product. A small acid stream containing NaOH, 100%, lb 0.1
acid soluble oil byproducts is removed from the unit and is either regen-
erated on site or sent to an off-site sulfuric acid regeneration facility to
Operating experience: Extensive commercial experience in both
ExxonMobil and licensee refineries, with a total operating capacity of
recover acid strength.
119,000-bpsd at 11 locations worldwide. Unit capacities currently range
Yields: from 2,000 to 30,000 bpd. The license of the world’s largest alkylation
Alkylate yield 1.8 bbl C5+/ bbl butylene feed unit, with a capacity of 83,000 bpd, was recently announced at Reliance
Isobutane required 1.2 bbl / bbl butylene feed Petroleum Limited’s Export Refinery in Jamnagar, India. A revamp has
Alkylate quality 97 RON / 94 MON been completed at ExxonMobil’s Altona, Australia refinery and a new
Rvp, psi 3 unit at TNK-BP’s Ryazan, Russia refinery is scheduled to start-up in mid-
2006. The larger units take advantage of the single reactor/settler trains
Utilities: typical per barrel of alkylate produced: with capacities up to 9,500 bpsd.
Water, cooling, M gal 2
Power, kWH 9 Continued
Alkylation, sulfuric acid, continued Economic advantages:
• Lower capital investment—Simple reactor/settler configura-
tion, less compression requirements translate into a significant invest-
Technical advantages: ment savings compared to indirect refrigeration systems
• Autorefridgeration is thermodynamically more efficient, allows
• Lower operating costs—Autorefrigeration, lower mixing and
lower reactor temperatures, which favor better product quality, and
compression power requirements translate into lower operating costs
lowers energy usage.
• Better economy of scale —Reactor system is simple and easily
• Staged reactor results in a high average isobutane concentra-
expandable with 9,500 bpsd single train capacities easily achievable.
tion, which favors high product quality.
• Low space velocity results in high product quality and reduced Reference: Lerner, H., “Exxon sulfuric acid alkylation technology,” Hand-
ester formation eliminating corrosion problems in fractionation equip- book of Petroleum Refining Processes, 2nd Ed., R. A. Meyers, Ed., pp.
ment. 1.3–1.14.
• Low reactor operating pressure translates into high reliabil-
ity for the mechanical seals for the mixers, which operate in the vapor Licensor: ExxonMobil Research & Engineering Co.
phase.
Aromatics
������������
Application: The GT-TransAlk technology produces benzene and xylenes ����������
from toluene and/or heavy aromatics streams. The technology features �������
����������
a proprietary catalyst and can accommodate varying ratios of feedstock,
while maintaining high activity and selectivity.
������
Description: The GT-TransAlk technology encompasses three main pro-
������
cessing areas: feed preparation, reactor and product stabilization sec- ����� ������ ��������
��������� ������
tions. The heavy aromatics stream (usually derived from catalytic refor-
mate) is fed to a C10/C11 splitter. The overhead portion, along with any ����
�������� ����������
toluene that may be available, is the feed to the transalkylation reactor �������
section. The combined feed is mixed with hydrogen, vaporized, and fed ���������
to the reactor. The un-reacted hydrogen is recycled for re-use. The prod-
����
uct stream is stabilized to remove fuel gas and other light components. ������������
The process reactor is charged with a proprietary catalyst, which �������
�������������
exhibits good flexibility to feed stream variations, including 100% C9+
aromatics. Depending on the feed composition, the xylene yield can
vary from 27 to 35% and C9 conversion from 53 to 67%.
Process advantages include:
• Simple, low cost fixed-bed reactor design
• Selective toward xylene production, with high toluene/C9 conver-
sion rates
• Physically stable catalyst
• Flexible to handle up to 100% C9+ components in feed
• Flexible to handle benzene recycle to increase xylene yields
• Moderate operating parameters; catalyst can be used as replace-
ment to other transalkylation units, or in grassroots designs
• Decreased hydrogen consumption due to low cracking rates
• Efficient heat integration scheme, reduces energy consumption.
Continued
Aromatics extractive distillation, continued
Emmrich, G., U. Ranke and H. Gehrke, “Working with an extractive dis-
tillation process,” Petroleum Technology Quarterly, Summer 2001, p. 125.
and with low pressure drop in the hydrogen loop, the product is 90 to 100 �
� �
RONC. With its higher selectivity, trimetallic catalysts RG582 and RG682 �
make an excellent catalyst replacement for semi-regenerative reformers.
The second, the advanced Octanizing process, uses continuous cata-
lyst regeneration allowing operating pressures as low as 3.5 kg /cm2 (50
psig). This is made possible by smooth-flowing moving bed reactors (1–3)
which use a highly stable and selective catalyst suitable for continuous
regeneration (4). Main features of Axens’ regenerative technology are: ���������
Licensor: Axens.
Catalytic dewaxing
Application: Use the ExxonMobil Selective Catalytic Dewaxing (MSDW) ����� �����������������������
process to make high VI lube base stock. ���������
����� ������������
Products: High VI / low-aromatics lube base oils (light neutral through ��� ����
��
����������
bright stocks). Byproducts include fuel gas, naphtha and low-pour diesel. ��������
��� ��� ��� ��������
��� ��� ����� �������������
Description: MSDW is targeted for hydrocracked or severely hydrotreated ����
stocks. The improved selectivity of MSDW for the highly isoparaffinic-lube ����� ���
��������� ������ ����
components results in higher lube yields and VIs. The process uses mul- �� ���� �����
tiple catalyst systems with multiple reactors. Internals are proprietary (the �� ��� �����
���
Spider Vortex Quench Zone technology is used). Feed and recycle gases
are preheated and contact the catalyst in a down-flow-fixed-bed reactor. ����������
Reactor effluent is cooled, and the remaining aromatics are saturated in a ��������
post-treat reactor. The process can be integrated into a lube hydrocracker ������������ ���
or lube hydrotreater. Post-fractionation is targeted for client needs. �����
Operating conditions:
Temperatures, ° F 550 – 800
Hydrogen partial pressures, psig 500 – 2,500
LHSV 0.4 – 3.0
Conversion depends on feed wax content Installation: Eight units are operating and four are in design.
Pour point reduction as needed. Licensor: ExxonMobil Research and Engineering Co.
Yields:
Light neutral Heavy neutral
Lube yield, wt% 94.5 96.5
C1– C4, wt% 1.5 1.0
C5– 400°F, wt% 2.7 1.8
400°F – Lube, wt% 1.5 1.0
H2 cons, scf / bbl 100 – 300 100 – 300
Economics: $3,000 – 5,500 per bpsd installed cost (US Gulf Coast).
Catalytic reforming
Application: Increase the octane of straight-run or cracked naphthas for
��� �����
gasoline production.
�������
Products: High-octane gasoline and hydrogen-rich gas. Byproducts may
be LPG, fuel gas and steam.
��������
Description: Semi-regenerative multibed reforming over platinum or bi-
metallic catalysts. Hydrogen recycled to reactors at the rate of 3 mols / ��������
mol to 7 mols /mol of feed. Straight-run and /or cracked feeds are typi- �����
cally hydrotreated, but low-sulfur feeds (<10 ppm) may be reformed
without hydrotreatment. ������� ����
��������
�� ����
Operating conditions: 875°F to 1,000°F and 150 psig to 400 psig reac- �����
��������
tor conditions. ����
��������������������
Yields: Depend on feed characteristics, product octane and reactor pres-
sure. The following yields are one example. The feed contains 51.4%
paraffins, 41.5% naphthenes and 7.1% aromatics, and boils from 208°F
to 375°F (ASTM D86). Product octane is 99.7 RONC and average reactor
pressure is 200 psig.
Economics:
Utilities, (per bbl feed)
Fuel, 103 Btu release 275
Component wt% vol%
Electricity, kWh 7.2
H2 2.3 1,150 scf/bbl
Water, cooling (20°F rise), gal 216
C1 1.1 —
Steam produced (175 psig sat), lb 100
C2 1.8 —
C3 3.2 — Licensor: CB&I Howe-Baker.
iC4 1.6 —
nC4 2.3 —
C5+ 87.1 —
LPG — 3.7
Reformate — 83.2
Catalytic reforming
Application: The CCR Platforming process is used throughout the world ������
in the petroleum and petrochemical industries. It produces feed for an ������� ������
������������ ��������
aromatics complex or a high-octane gasoline blending product and a ������������� �������
��� ��������
significant amount of hydrogen. ����������� �������� ����������
��������
���� �������
Description: Hydrotreated naphtha feed is combined with recycle hy- ���������
drogen gas and heat exchanged against reactor effluent. The combined
feed is then raised to reaction temperature in the charge heater and sent ����������
Conradson
carbon, wt% 20.0 27.6 48
Continued
Coking, continued
Economics (continued):
Utilities, typical/bbl of feed:
Fuel, 103 Btu 123
Electricity, kWh 3.6
Steam (exported), lb 1
Water, cooling, gal 58
Boiler feedwater, lbs 38
Condensate (exported), lbs 24
� ��� �����
Application: Upgrade residues to lighter hydrocarbon fractions using the
Selective Yield Delayed Coking (SYDEC) process.
� �
Description: Charge is fed directly to the fractionator (1) where it com- �������
bines with recycle and is pumped to the coker heater. The mixture is
heated to coking temperature, causing partial vaporization and mild
cracking. The vapor-liquid mix enters a coke drum (2 or 3) for further � �����
cracking. Drum overhead enters the fractionator (1) to be separated into �������������
gas, naphtha, and light and heavy gas oils. Gas and naphtha enter the
vapor recovery unit (VRU)(4). There are at least two coking drums, one
�������������
coking while the other is decoked using high-pressure water jets. The ����
coking unit also includes a coke handling, coke cutting, water recovery
�����
and blowdown system. Vent gas from the blowdown system is recov-
ered in the VRU.
Operating conditions: Typical ranges are: Utilities, typical per bbl feed:
Heater outlet temperature, ºF 900 – 950 Fuel, 103 Btu 120
Coke drum pressure, psig 15 – 100 Electricity, kWh 3
Recycle ratio, equiv. fresh feed 0 – 1.0 Steam (exported), lb 35
Increased coking temperature decreases coke production; increases Water, cooling, gal 36
liquid yield and gas oil end point. Increasing pressure and/or recycle ra-
Installations: Currently, 52 delayed cokers are installed worldwide with
tio increases gas and coke make, decreases liquid yield and gas oil end
a total installed capacity over 2.5 million bpsd
point.
References: Handbook of Petroleum Refining Processes, Third Ed., Mc-
Yields:
Graw-Hill, pp. 12.33 –12.89.
Operation:
“Delayed coking revamps,” Hydrocarbon Processing, September 2004.
Products, wt% Max dist. Anode coke Needle coke
“Residue upgrading with SYDEC Delayed Coking: Benefits & Eco-
Gas 8.7 8.4 9.8
nomics,” AIChE Spring National Meeting, April 23–27, 2006, Orlando.
Naphtha 14.0 21.6 8.4
“Upgrade refinery residuals into value-added products,” Hydrocar-
Gas oil 48.3 43.8 41.6
bon Processing, June 2006.
Coke 29.3 26.2 40.2
Licensor: Foster Wheeler/UOP LLC.
Economics:
Investment (basis 65,000 –10,000 bpsd)
2Q 2005 US Gulf), $ per bpsd 3,000 –5,200
Coking, fluid
Application: Continuous fluid, bed coking technology to convert heavy
hydrocarbons (vacuum residuum, extra heavy oil or bitumen) to full- Reactor products
range lighter liquid products and fluid coke. Product coke can be sold to fractionator
as fuel or burned in an integrated fluid bed boiler to produce steam and Flue gas to CO boiler
power.
START 1
Products: Liquid products can be upgraded through conventional 3
hydrotreating. Fluid coke is widely used as a solid fuel, with particular 2
Net coke
advantages in cement kilns and in fluid-bed boilers.
Air
Description: Feed (typically 1,050°F+ vacuum resid) enters the scrubber blower
(1) for heat exchange with reactor overhead effluent vapors. The scrub- Air
ber typically cuts 975°F+ higher boiling reactor effluent hydrocarbons Cold Hot
coke coke
for recycle back to the reactor with fresh feed. Alternative scrubber con-
figurations provide process flexibility by integrating the recycle stream
with the VPS or by operating once-through which produces higher liq-
uid yields. Lighter overhead vapors from the scrubber are sent to con-
ventional fractionation and light ends recovery. In the reactor (2), feed
is thermally cracked to a full range of lighter products and coke.
The heat for the thermal cracking reactions is supplied by circulating C5+ liquids, wt% 58.1 62.3
coke between the burner (3) and reactor (2). About 20% of the coke
is burned with air to supply process heat requirements, eliminating the Net product coke, wt % 25.7 23.9
need for an external fuel supply. The rest of the coke is withdrawn and Coke consumed for heat, wt% 4.4 3.4
either sold as a product or burned in a fluid bed boiler. Properties of the
fluid coke enable ease of transport and direct use in fuel applications, Investment: TPC, US Gulf Coast, 2Q 2003 estimate including gas pro-
including stand alone or integrated cogeneration facilities. cessing, coke handling and wet gas scrubbing for removing SOx from
the burner overhead
Yields: Example, typical Middle East vacuum resid (~25 wt% Concar- Capital investment, $/bp/sd 3,300
bon, ~5 wt% sulfur):
Recycle Once-Through Competitive advantages:
Light ends, wt% 11.8 10.4 • Single train capacities >100 Mbpsd; greater than other processes
Naphtha (C5-350°F), wt% 11.5 9.5 • Process wide range of feeds, especially high metals, sulfur and CCR
Distillate (350 – 650°F), wt% 14.5 13.1 • Internally heat integrated, minimal use of fuel gas, and lower coke
Heavy gas oil (650°F+), wt% 32.1 39.7 production than delayed coking
Continued
Coking, fluid, continued
• Lower investment and better economy of scale than delayed cok-
ing
• Efficient integration with fluid bed boilers for cogeneration of
steam and electric power.
Description: The crude is preheated and desalted (1). It is fed to a first ������
�������
dry reboiled pre-flash tower (2) and then to a wet pre-flash tower (3).
The overhead products of the two pre-flash towers are then fraction-
ated as required in a gas plant and rectification towers (4).
The topped crude typically reduced by 2/3 of the total naphtha cut is
then heated in a conventional heater and conventional topping column
(5). If necessary the reduced crude is fractionated in one deep vacuum Utility requirements, typical per bbl of crude feed:
column designed for a sharp fractionation between vacuum gas oil, two Fuel fired, 103 btu 50–65
vacuum distillates (6) and a vacuum residue, which could be also a road Power, kWh 0.9–1.2
bitumen. Steam 65 psig, lb 0–5
Extensive use of pinch technology minimizes heat supplied by heat- Water cooling, (15°C rise) gal 50–100
ers and heat removed by air and water coolers. Total primary energy consumption:
This process is particularly suitable for large crude capacity from for Arabian Light or Russian Export Blend: 1.25 tons of fuel
150,000 to 250,000 bpsd. per 100 tons of Crude
It is also available for condensates and light crudes progressive distil- for Arabian Heavy 1.15 tons of fuel
lation with a slightly adapted scheme. per 100 tons of Crude
Economics: Installation: Technip has designed and constructed one crude unit and
Investment (basis 230,000 bpsd including atmospheric and one condensate unit with the D2000 concept. The latest revamp proj-
vacuum distillation, gas plant and rectification tower) $750 to
$950 per bpsd (US Gulf Coast 2000).
Continued
Crude distillation, continued
ect currently in operation shows an increase of capacity of the existing
crude unit of 30% without heater addition.
Products: Diesel is typically the desired product, but kerosine, turbine ������������
fuel and naphtha are also produced. �����
Operating conditions:
Column pressure, psig 0 – 20
Temperature, °F 550 – 650
Products: Deasphalted oil (DAO) for catalytic cracking and hydrocracking ������ �������
�
���
feedstocks, resins for specification asphalts, and pitch for specification ������� ���������
������ � ���������
asphalts and residue fuels.
�����
Description: Feed and light paraffinic solvent are mixed and then ��������
charged to the extractor (1). The DAO and pitch phases, both containing
solvents, exit the extractor. The DAO and solvent mixture is separated � � ���
under supercritical conditions (2). Both the pitch and DAO products are
������� ��������
stripped of entrained solvent (3,4). A second extraction stage is utililized
if resins are to be produced.
����� ���
Economics:
Investment (Basis: 7,000-bpsd feedrate
capacity, 2006 US Gulf Coast), $/bpsd 11,200
Utilities, typical per bbl feed:
Fuel, 103 Btu (absorbed) 160
Electricity, kWh 15
Steam, lb 35
Water, cooling (25°F rise), gal 1,100
References:
Low, G., J. Townsend and T. Shooter, “Systematic approach for the
revamp of a low-pressure hydrotreater to produce 10-ppm, sulfur-free
diesel at BP Conyton Refinery,” 7th ERTC, Paris, November 2002.
Sarup, B., M. Johansen, L. Skyum and B. Cooper, “ULSD Production
in Practice,” 9th ERTC, Prague, November 2004.
Diesel upgrading
Application: Topsøe’s Diesel Upgrading process can be applied for im- ��������������� �����������
provement of a variety of diesel properties, including reduction of diesel ����������
specific gravity, reduction of T90 and T95 distillation (Back-end-shift),
reduction of aromatics, and improvements of cetane, cold-flow prop- ������� ���������
������� ����������
erties, (pour point, clouds point, viscosity and CFPP) and diesel color �������� ��������
reduction (poly shift). Feeds can range from blends of straight-run and �������
cracked gas oils up to heavy distillates, including light vacuum gas oil. ����������
Economics: Plants and their operations are simple. The same inexpen-
sive (purchased in bulk quantities) and long-lived, non-sophisticated cat-
alysts are used in the main reactor section catalytic region of the Catacol
column, if any.
Ethers—ETBE
Application: The Uhde (Edeleanu) ETBE process combines ethanol and
���� ����������� ����� �������������
isobutene to produce the high-octane oxygenate ethyl tertiary butyl ������� ���� ����������
ether (ETBE).
�����������
Feeds: C4 cuts from steam cracker and FCC units with isobutene con-
tents ranging from 12% to 30%.
Products: ETBE and other tertiary alkyl ethers are primarily used in gas- �������������
���������
oline blending as an octane enhancer to improve hydrocarbon com-
bustion efficiency. Moreover, blending of ETBE to the gasoline pool will
lower vapor pressure (Rvp). ������������
base of the riser via proprietary Micro-Jet feed injection nozzles (1). �
Catalyst and oil vapor flow upwards through a short-contact time,
all-vertical riser (2) where raw oil feedstock is cracked under opti-
mum conditions.
Reaction products exiting the riser are separated from the spent line (7). This arrangement provides the lowest overall unit elevation.
catalyst in a patented, direct-coupled cyclone system (3). Product Catalyst is regenerated by efficient contacting with air for complete
vapors are routed directly to fractionation, thereby eliminating non- combustion of coke. For resid-containing feeds, the optional cata-
selective, post-riser cracking and maintaining the optimum prod- lyst cooler is integrated with the regenerator. The resulting flue gas
uct yield slate. Spent catalyst containing only minute quantities of exits via cyclones (9) to energy recovery/flue gas treating. The hot
hydrocarbon is discharged from the diplegs of the direct-coupled regenerated catalyst is withdrawn via an external withdrawal well
cyclones into the cyclone containment vessel (4). The catalyst flows (10). The well allows independent optimization of catalyst density
down into the stripper containing proprietary modular grid (MG) in the regenerated catalyst standpipe, maximizes slide valve (11)
baffles (5). pressure drop and ensures stable catalyst flow back to the riser feed
Trace hydrocarbons entrained with spent catalyst are removed injection zone.
in the MG stripper using stripping steam. The MG stripper efficient- The catalyst formulation can be tailored to maximize the most
ly removes hydrocarbons at low steam rate. The net stripper vapors desired product. For example, the formulation for maximizing light
are routed to the fractionator via specially designed vents in the olefins (Indmax operation) is a multi-component mixture that pro-
direct-coupled cyclones. Catalyst from the stripper flows down the motes the selective cracking of molecules of different sizes and
spent-catalyst standpipe and through the slide valve (6). The spent shapes to provide very high conversion and yield of light olefins.
catalyst is then transported in dilute phase to the center of the re-
generator (8) through a unique square-bend-spent catalyst transfer
Continued
Fluid catalytic cracking, continued
Economics:
Investment (basis: 30,000 bpsd including reaction/regeneration
system and product recovery. Excluding offsites, power recovery
and flue gas scrubbing US Gulf Coast 2006.)
$/bpsd (typical) 2,400–3,500
Utilities, typical per bbl fresh feed:
Electricity, kWh 0.8–1.0
Steam, 600 psig (produced) 50–200
Maintenance, % of investment per year 2–3
Continued
Fluid catalytic cracking, continued
Installation: More than 70 units with a design capacity of over 2.5-mil-
lion bpd fresh feed.
um gas oils and coker streams to resids. This pretreatment process can
������� ��������
maximize FCC unit performance. ����������
Description: A typical amine system flow scheme is used. The feed gas
contacts the treating solvent in the absorber (1). The resulting rich sol-
vent bottom stream is heated and sent to the regenerator (2). Regen-
erator heat is supplied by any suitable heat source. Lean solvent from
the regenerator is sent through rich/lean solvent exchangers and coolers
before returning to the absorber.
Reference: Garrison, J., et al., “Keyspan Energy Canada Rimbey acid gas
FLEXSORB SE solvent is an aqueous solution of a hindered amine.
enrichment with FLEXSORB SE Plus technology,” 2002 Laurance Reid
FLEXSORB SE Plus solvent is an enhanced aqueous solution, which has
Gas Conditioning Conference, Norman, Oklahoma.
improved H2S regenerability yielding <10 vppm H2S in the treated gas.
Adams-Smith, J., et al., Chevron USA Production Company, “Carter
Hybrid FLEXSORB SE solvent is a hybrid solution containing FLEXSORB SE
Creek Gas Plant FLEXSORB tail gas treating unit,” 2002 GPA Annual
amine, a physical solvent and water.
Meeting, Dallas.
Economics: Lower investment and energy requirements based primarily Connock, L., et al., “High recovery tail gas treating,” Sulphur, No.
on requiring 30% to 50% lower solution circulation rates, compared to 296, November/ December 2004.
conventional amines. Fedich, R., et al., “Selective H2S Removal,” Hydrocarbon Engineer-
ing, May 2004.
Installations: Total gases treated by FLEXSORB solvents are about 2 bil- Fedich, R. B., et al., “Solvent changeover benefits,” Hydrocarbon
lion scfd and the total sulfur recovery is about 900 long tpd. Engineering, Vol. 10, No. 5, May 2005.
FLEXSORB SE—31 plants operating, three in design “Gas Processes 2006,” Hydrocarbon Processing, January 2006.
FLEXSORB SE Plus—19 plants operating, nine in design
Hybrid FLEXSORB SE—two plants operating, three in design Licensor: ExxonMobil Research and Engineering Co.
Over 60 plants operating or in design.
Gasification
Application: The Shell Gasification Process (SGP) converts the heaviest
���
residual liquid hydrocarbon streams with high-sulfur and metals content
�������
into a clean synthesis gas and valuable metal oxides. Sulfur (S) is re- �����
������
�����
moved by normal gas treating processes and sold as elemental S.
The process converts residual streams with virtually zero value as ������ ��������
������
fuel-blending components into valuable, clean gas and byproducts. ��� ������������
This gas can be used to generate power in gas turbines and for making
H2 by the well-known shift and PSA technology. It is one of the few ul-
timate, environmentally acceptable solutions for residual hydrocarbon
streams. ������� ������� ���������� ��������
������ �������
Products: Synthesis gas (CO+H2), sulfur and metal oxides. ����������� ����������
������ ��������
Products: Load-sulfur blending stock for gasoline motor fuels.
Description: Gasoline from the fluid catalytic cracker unit is combined
with a small hydrogen stream and heated. Vaporized gasoline is injected
����������
into the fluid-bed reactor (1), where the proprietary sorbent removes ������ �������
��������
sulfur from the feed. A disengaging zone in the reactor removes sus- ������ ����������
pended sorbent from the vapor, which exits the reactor to be cooled. �����
����
Regeneration: The sorbent (catalyst) is continuously withdrawn from the �������� ������������
reactor and transferred to the regenerator section (2), where the sulfur ������� �������
���������
is removed as SO2 and sent to a sulfur-recovery unit. The cleansed sor-
bent is reconditioned and returned to the reactor. The rate of sorbent
circulation is controlled to help maintain the desired sulfur concentra-
tion in the product.
Economics:
Typical operating conditions: Results:
Temperature, °F 750 – 825 C5+ yield, vol% of feed >100%
Pressure, psig 100 – 500 Lights yield, wt% of feed < 0.1
Space velocity, whsv 4–8 (R+M) loss
Hydrogen purity, % 70 – 99 2 <0.3
Total H2 usage, scf / bbl 40 – 60 Operating cost, ¢/gal* 0.9
Case study premises: * Includes utilities, 4% per year maintenance and sorbent costs.
25,000 - bpd feed
Installation: Forty-three sites licensed as of 1Q 2004.
775 - ppm feed sulfur
25 - ppm product sulfur ( 97% removal ) Licensor: ConocoPhillips.
No cat gasoline splitter
Gasoline desulfurization, ultra-deep
Application: Ultra-deep desulfurization of FCC gasoline with minimal
�������������������������
octane penalty using Prime-G+ process.
��������
Description: FCC debutanizer bottoms are fed directly to a first reactor ��������
����������
wherein, under mild conditions, diolefins are selectively hydrogenated
and mercaptans are converted to heavier sulfur species. The selective
�������� ����������
hydrogenation reactor effluent is then usually split to produce an LCN ��������� �������������
(light cat naphtha) cut and an HCN (heavy cat naphtha). ������ ��������������
The LCN stream is mercaptans-free with a low-sulfur and diolefin ���� ���������
concentration, enabling further processing in an etherification or al- ������
����� ��� ��������
kylation unit. The HCN then enters the main Prime-G+ section where
it undergoes in a dual catalyst reactor system; a deep HDS with very
limited olefins saturation and no aromatics losses produces an ultra- ��������
low-sulfur gasoline. ������
Description: The feed is mixed with hydrogen, heated with reactor efflu- �����
ent exchange and passed through a pretreat reactor for diolefin satura- ����������
������������� ������ ����
tion. After further heat exchange with reactor effluent and preheat using �����������
���
a utility, the hydrocarbon/hydrogen mixture enters the main reaction sec- ��������� �������
���� ��������
tion which features ExxonMobil Research and Engineering Co. (EMRE)
proprietary selective catalyst systems. In this section of the plant, sulfur ������� �����������
is removed in the form of H2S under tailored process conditions, which �����������
�����������
strongly favor hydrodesulfurization while minimizing olefin saturation. ��������
The feed may be full-range, intermediate or heavy FCC-naphtha ����������������
fraction. Other sulfur-containing streams such as light-coker naphtha, ���������
steam cracker or light straight-run naphthas can also be processed with
FCC naphthas. SCANfining technology can be retrofitted to existing
units such as naphtha or diesel hydrotreaters and reformers. SCANfining
technology also features ExxonMobil’s proprietary reactor internals such
as Automatic Bed Bypass Technology for onstream mitigation of reactor
plugging/pressure drop buildup. References: Sapre, A.V., et al., “Case History: Desulfurization of FCC
For high-sulfur feeds and/or very low-sulfur product, with low levels naphtha,” Hydrocarbon Processing, February 2004.
of product mercaptans variations in the plant design from SCANfining I Ellis, E. S., et al., “Meeting the Low Sulfur Mogas Challenge,” World
Process to the SCANfining II Process for greater HDS selectivity, or addi- Refining Association Third European Fuels Conference, March 2002.
tion of a ZEROMER process step for mercaptan conversion, or addition
of an EXOMER process unit for mercaptan extraction. Licensor: ExxonMobil Research and Engineering Co.
EMRE has an alliance with Kellogg Brown & Root (KBR) to pro-
vide SCANfining technology to refiners and an alliance with Merichem
Chemicals & Refinery Services LLC to provide EXOMER technology to
refiners.
Application: LO-CAT removes H2S from gas streams and produces el-
��������� ������������
emental sulfur. LO-CAT units are in service treating refinery fuel gas,
hydrodesulfurization offgas, sour-water-stripper gas, amine acid gas,
����
claus tail gas and sulfur tank vent gas. Sulfur capacities are typically less ��������
than 25 ltpd down to several pounds per day. Key benefits of operation � �
are high (99.9%) H2S removal efficiency, and flexible operation, with vir- ��������
tually 100% turndown capability of H2S composition and total gas flow.
��������
Sulfur is recovered as a slurry, filter cake or high-purity molten sulfur.
The sulfur cake is increasingly being used in agriculture, but can also be ����
deposited in a nonhazardous landfill.
��������
����������
Description: The conventional configuration is used to process combus-
tible gas and product gas streams. Sour gas contacts the dilute, propri-
etary, iron chelate catalyst solution in an absorber (1), where the H2S is
absorbed and oxidized to solid sulfur. Sweet gas leaves the absorber for
use by the refinery. The reduced catalyst solution returns to the oxidizer restrictions on type of gas to be treated; however, some contaminants,
(2), where sparged air reoxidizes the catalyst solution. The catalyst solu- such as SO2, may increase operating costs.
tion is returned to the absorber. Continuous regeneration of the catalyst Installations: Presently, 160 licensed units are in operation with four
solution allows for very low chemical operating costs. units under construction.
In the patented autocirculation configuration, the absorber (1) and
oxidizer (2) are combined in one vessel, but separated internally by baf- Reference: Nagl, G., W. Rouleau and J. Watson,, “Consider optimized
fles. Sparging of the sour gas and regeneration air into the specially Iron-Redox processes to remove sulfur,” Hydrocarbon Processing, Janu-
designed baffle system creates a series of “gas lift” pumps, eliminating ary 2003, pp. 53–57.
the external circulation pumps. This configuration is ideally suited for
treating amine acid gas and sour-water-stripper gas streams. Licensor: Gas Technology Products, a division of Merichem Chemical &
In both configurations, sulfur is concentrated in the oxidizer cone Refinery Services LLC.
and sent to a sulfur filter, which can produce filter cake as high as 85%
sulfur. If desired, the filter cake can be further washed and melted to
produce pure molten sulfur.
vessel top, flows over media where H2S is removed and reacted. Sweet ������
gas exits the bottom of vessel. In the single-vessel configuration, when
the H2S level exceeds the level allowed, the vessel must be bypassed,
media removed through the lower manway, fresh media installed and
vessel returned to service.
For continuous operation, a dual “lead-lag” configuration is desir-
able. The two vessels operate in series, with one vessel in the lead posi-
tion, the other in the lag position. When the H2S level at the outlet of
the lead vessel equals the inlet H2S level (the media is completely spent),
the gas flow is changed and the vessels reverse rolls, so that the “lag”
vessel becomes the “lead” vessel. The vessel with the spent media is
bypassed. The media is replaced, and the vessel with fresh media is re-
turned to service in the “lag” position.
Licensor: Axens.
Hydrocracking
��������
Application: Upgrade vacuum gas oil alone or blended with various
feedstocks (light-cycle oil, deasphalted oil, visbreaker or coker gasoil). ���������
Products: Jet fuel, diesel, very-low-sulfur fuel oil, extra-quality FCC �����
�������
feed with limited or no FCC gasoline post-treatment or high VI lube
base stocks. ��������
�
� � � �
������
Description: This process uses a refining catalyst usually followed by
an amorphous and/or zeolite-type hydrocracking catalyst. Main features ������������
of this process are:
• High tolerance toward feedstock nitrogen
• High selectivity toward middle distillates ����
• High activity of the zeolite, allowing for 3–4 year cycle lengths and �����
��������������
products with low aromatics content until end of cycle.
Three different process arrangements are available: single-step/
once-through; single-step/total conversion with liquid recycle; and two-
step hydrocracking. The process consists of: reaction section (1, 2), gas
separator (3), stripper (4) and product fractionator (5). Economics:
Investment: (Basis: 40,000-bpsd unit, once-through, 90% con-
Product quality: Typical for HVGO (50/50 Arabian light/heavy):
version, battery limits, erected, engineering fees included, 2000
Feed, Jet Gulf Coast), $ per bpsd 2,500 –3,500
HVGO fuel Diesel Utilities, typical per bbl feed:
Sp. gr. 0.932 0.800 0.826 Fuel oil, kg 5.3
TBP cut point, °C 405– 565 140 –225 225 –360 Electricity, kWh 6.9
Sulfur, ppm 31,700 <10 <10 Water, cooling, m 3 0.64
Nitrogen, ppm 853 <5 <5 Steam, MP balance
Metals, ppm <2 – –
Cetane index – – 62 Installation: More than 50 references, cumulative capacity exceeding
Flash pt., °C – 40 125 1 million bpsd, conversions up to 99%.
Smoke pt., mm, EOR – 26–28 – Licensor: Axens.
Aromatics, vol%, EOR – < 12 <8
Viscosity @ 38°C, cSt 110 – 5.3
PAH, wt%, EOR <2
Hydrocracking ��������
Application: Convert a wide variety of feedstocks including vacuum
deep-cut gas oil, coker gas oils, de-asphalted oil (DAO), and FCC cy-
cle oils into high-quality, low-sulfur fuels using ExxonMobil Research ������� ����������
and Engineering Company’s (EMRE) moderate pressure hydrocracking ����������
(MPHC) process. ��������
����
�������
Products: Products include a wide range of high-quality, low-sulfur dis- ���
���� ���� ��������
tillates and blending stocks including LPG, high-octane gasoline, high- �����
quality reformer naphtha. Unconverted bottoms product from the MPHC ������ ����������� ������
unit is very low in sulfur and is an excellent feedstock for fluid catalytic ������� ��� ������������
cracking (FCC), lube-oil basestock production, steam cracking and low- ���
sulfur fuel oil.
�������������
Description: The process uses a multiple catalyst system in multi-bed ��������
reactor(s) that incorporates proprietary advanced quench and redistri-
bution internals (Spider Vortex). Heavy hydrocarbons and recycle gas
are preheated and contact the catalyst in the trickle-phase fixed-bed
reactor(s). Reactor effluent is flashed in high- and low-temperature
separators. An amine scrubber removes H2S from the recycle gas be-
fore it gets compressed and re-circulated back to the unit. An opti-
mized, low cost stripper/fractionator arrangement is used for product Yields:
recovery. Naphtha, wt% 4 10 10
When higher-quality distillates are required, the addition of a low- Kero/jet, wt% 6 10 10
cost, highly integrated distillate post-treating unit (PTU) can be incor- Diesel, wt% 22 26 27
porated in the design to meet or exceed high-pressure hydrocracking LSGO (FCC feed), wt% 65 50 50
product quality at lower capital cost and hydrogen consumption H2 consumption, wt% 1.0 –1.5 1.3 –1.8 1.5 – 2.0
Product quality:
Operating conditions and yields: Typical operating conditions on a Mid-
Kero sulfur, wppm 20 – 200 20 – 200 20 – 200
dle East VGO for a once-through MPHC operation are shown:
Kero smoke Pt, mm 13 – 18 15 – 20 17 – 22
Operation conditions: Diesel sulfur, wppm 30 – 500 30 – 300 30 – 200
Configuration MPHC MPHC MPHC Diesel cetane no. 45 – 50 47 – 52 50 – 55
Nominal conversion, % 35 50 50
H2 pressure, psig 800 800 1,250
Continued
Hydrocracking, continued
Utilities, per bbl of feed:
Electric power, kW 4.1 7.2
Fuel (absorbed), Btu 67,100 69,600
Steam, MP (export), lb (15.9) (21.1)
Water, cooling, gal 101 178
Wash water, gal 1.5 2.2
Lean amine, gal 36.1 36.1
are sent to an inline hydrotreating (HT) step and a solid-liquid separation ���������������
als are removed from the residual. The solid-free fraction from the top ������������� �������
of the TI-remover is combined with the heavy fraction from the vacuum �������������
Continued
Hydrocracking, continued
Feed VGO VGO VGO VGO
Product quality
Kerosine smoke, mm 29–32 29–32 29–32
Diesel cetane number 58–64 58–64 58–64
UCO BMCI 6–8
UCO Waxy V.I. 143–145
UCO Dewaxed V.I. 131–133
Operating conditions:
Temperature, °F 770– 820 Economics: Basis 2005 US Gulf Coast
Hydrogen partial pressure, psi 1,600 –1,950 Investment in $ per bpsd 4,500 – 6,500
LHSV, hr –1 0.25– 0.6 Utilities, per bbl of feed
Conversion, wt% 50 – 80 Fuel, 103 Btu 70
Power, kWh 11
Examples: Ural VR feed: a 540°C+ cut from Ural crude is processed at Catalyst makeup, lb 0.2– 0.8
66% conversion to obtain a stable fuel oil containing less than 1%wt
sulfur, 25% diesel and 30% VGO. The diesel cut is further hydrotreated Installation: There are seven H-OilRC units in operation with a total capac-
to meet ULSD specifications using an integrated Prime-D unit. Arab Me- ity of 300,000 bpsd. Two additional references for H-OilDC, the ebullated
dium VR feed: a vacuum residue from a blend 70% Arab Light-30% bed technology for VGO and DAO, add another 139,900 bpsd.
Arab Heavy containing 5.5wt% sulfur is processed at above 75% con- Licensor: Axens.
version to obtain a stable fuel oil with 2wt% sulfur.
Hydrocracking
Application: Desulfurization, demetalization, CCR reduction and hydro-
cracking of atmospheric and vacuum resids using the LC-FINING Pro- ���������������
cess. ����������
��������� �����
Products: Full range of high-quality distillates. Residual products can ������� � �
be used as fuel oil, synthetic crude or feedstock for a resid FCC, coker, �
visbreaker or solvent deasphalter.
�
�
Description: Fresh hydrocarbon liquid feed is mixed with hydrogen and �����������
reacted within an expanded catalyst bed (1) maintained in turbulence by ����
liquid upflow to achieve efficient isothermal operation. Product quality is ����� �
maintained constant and at a high level by intermittent catalyst addition
and withdrawal. Reactor products flow to a high-pressure separator (2), ��������
low-pressure separator (3) and product fractionator (4). Recycle hydro-
gen is separated (5) and purified (6).
Process features include onstream catalyst addition and withdrawal.
Recovering and purifying the recycled H2 at low pressure rather than at high
pressure can reduce capital cost and allows design at lower gas rates.
Operating conditions:
Reactor temperature, °F 725 – 840
Atm. resid Vac. resid
Reactor pressure, psig 1,400 – 3,500 Feed
Gravity, °API 12.40 4.73 4.73 4.73
H2 partial pressure, psig 1,000 – 2,700
Sulfur, wt % 3.90 4.97 4.97 4.97
LSHV 0.1 to 0.6
Ni / V, ppmw 18 /65 39 /142 39 /142 39 /142
Conversion, % 40 – 97+ Conversion, vol% 45 60 75 95
Desulfurization, % 60 – 90 (1,022°F+)
Demetalization, % 50 – 98 Products, vol%
CCR reduction, % 35 – 80 C4 1.11 2.35 3.57 5.53
C5–350°F 6.89 12.60 18.25 23.86
Yields: For Arabian heavy/Arabian light blends: 350 –700°F (650°F) (15.24) 30.62 42.65 64.81
700 (650°F) –1,022°F (55.27) 21.46 19.32 11.92
1,022°F+ 25.33 40.00 25.00 5.0
C5+, °API / wt% S 23.70 / 0.54 22.5 / 0.71 26.6 / 0.66 33.3 / 0.33
Continued
Hydrocracking, continued
Economics:
Investment, estimated (US Gulf Coast, 2006)
Size, bpsd fresh feed 92,000 49,000
$/bpsd typical fresh feed 3,000 5,000 5,800 7,200
Utilities, per bbl fresh feed
Fuel fired, 103 Btu 56.1 62.8 69.8 88.6
Electricity, kWh 8.4 13.9 16.5 22.9
Steam (export), lb 35.5 69.2 97.0 97.7
Water, cooling, gal. 64.2 163 164 248
Typical feedstocks are the effluent from a dewaxing reactor, effluent from
hydrated feeds or solvent-dewaxed feedstocks. The products are highly
stabilized base-oil, technical-grade white oil or food-grade white oil. ����������
As shown in the simplified flow diagram, feedstocks are mixed with ������������� �
�
recycle hydrogen and fresh makeup hydrogen, heated and charged to a ����
�
reactor containing ISOFINISHING Catalyst (1). Effluent from the finishing �����
���������
stock flexibility. ������ ���������
���
������� ����� �������
�����
����� ������� ��������
Description: The generic flowsheet consists of feed pretreatment, pre- ����������
��
reforming (optional), steam-HC reforming, shift conversion and hydro- ��� �����
�������� �������
gen purification by pressure swing adsorption (PSA). However, it is often ��� ����� ����� ����
tailored to satisfy specific requirements.
�������������
Feed pretreatment normally involves removal of sulfur, chlorine and ����� ����������
other catalyst poisons after preheating to 350°C to 400°C.
The treated feed gas mixed with process steam is reformed in a fired
reformer (with adiadatic pre-reformer upstream, if used) after neces-
sary superheating. The net reforming reactions are strongly endother-
mic. Heat is supplied by combusting PSA purge gas, supplemented by
drogen recovery and generation, and recuperative (post-)reforming also
makeup fuel in multiple burners in a top-fired furnace.
for capacity retrofits.
Reforming severity is optimized for each specific case. Waste heat
from reformed gas is recovered through steam generation before the Installations: TECHNIP has been involved in over 240 hydrogen plants
water-gas shift conversion. Most of the carbon monoxide is further con- worldwide, covering a wide range of capacities. Most installations are
verted to hydrogen. Process condensate resulting from heat recovery for refinery application with basic features for high reliability and opti-
and cooling is separated and generally reused in the steam system after mized cost.
necessary treatment. The entire steam generation is usually on natural
circulation, which adds to higher reliability. The gas flows to the PSA unit Licensor: Technip.
that provides high-purity hydrogen product (up to < 1ppm CO) at near
inlet pressures.
Typical specific energy consumption based on feed + fuel – export
steam ranges between 3.0 and 3.5 Gcal / KNm 3 ( 330 – 370 Btu / scf ) LHV,
depending upon feedstock, plant capacity, optimization criteria and
steam-export requirements. Recent advances include integration of hy-
Hydrogen
Application: Production of hydrogen for refinery applications (e.g., ��������������
hydrotreating and hydrocracking) as well as for petrochemical and other ���� �����
industrial uses. ����
���������� � ��������
Feed: Natural gas, refinery offgases, LPG, naphtha or mixtures thereof. ������ �
� �
��
Product: High-purity hydrogen (typically >99.9%), CO, CO2, HP steam
and/or electricity may be produced as separate creditable byproduct.
����������
Description: The plant generally comprises four process units. The feed
is desulfurized (1), mixed with steam and converted to synthesis gas in a ������������
steam reformer (2) over a nickel-containing catalyst at 20 – 40 bar pres- ���� ���
��������
sure and outlet temperatures of 800 – 900°C. �������������� �
zone. These factors combine to give a long catalyst life. Additionally, ������������
mercaptans can react with diolefins to make heavy, thermally-stable sul-
fides. The sulfides are fractionated to the bottoms product. This can
eliminate the need for a separate mercaptan removal step. The distillate
product is ideal feedstock for alkylation or etherification processes.
The heat of reaction evaporates liquid, and the resulting vapor is
condensed in the overhead condenser (2) to provide additional reflux.
Economics: Fixed-bed hydrogenation requires a distillation column fol-
lowed by a fixed-bed hydrogenation unit. The CDHydro process elimi-
The natural temperature profile in the fractionation column results in a
nates the fixed-bed unit by incorporating catalyst in the column. When
virtually isothermal catalyst bed rather than the temperature increase
a new distillation column is used, capital cost of the column is only 5%
typical of conventional reactors.
to 20% more than for a standard column depending on the CDHydro
The CDHydro process can operate at much lower pressure than
application. Elimination of the fixed-bed reactor and stripper can reduce
conventional processes. Pressures for the CDHydro process are typically
capital cost by as much as 50%.
set by the fractionation requirements. Additionally, the elimination of a
separate hydrogenation reactor and hydrogen stripper offers significant Installation: Forty-five CDHydro units are in commercial operation for
capital cost reduction relative to conventional technologies. C4, C5, C6 and benzene hydrogenation applications. Nineteen units
Feeding the CDHydro process with reformate and light-straight run have been in operation for more than five years and total commer-
for benzene saturation provides the refiner with increased flexibility to cial operating time now exceeds 100 years for CDHydro technologies.
produce low-benzene gasoline. Isomerization of the resulting C5 / C6 Twelve additional units are currently in engineering / construction.
overhead stream provides higher octane and yield due to reduced ben-
zene and C7+ content compared to typical isomerization feedstocks. Licensor: CDTECH.
Hydrogen—HTCR and HTCR twin
plants
��������� ����������� ������������������ ����� ���
Application: Produce hydrogen from hydrocarbon feedstocks such as:
natural gas, LPG, naphtha, refinery offgases, etc., using the Haldor Top-
søe Convective Reformer (HTCR). Plant capacities range from approxi-
mately 5,000 Nm3/ h to 25,000+ Nm3/ h (5 MM scfd to 25+ MMscfd) and �����
��
hydrogen purity from about 99.5 – 99.999+%. This is achieved without
��
any steam export.
Description: The HTER-p is installed in parallel with the tubular steam �����������������
methane reformer (SMR) and fed independently with desulfurized �����������
feed taken upstream the reformer section. This enables individual ad- ���������
��������
justment of feedrate and steam- and process steam-to-carbon ratio to
obtain the desired conversion. The hydrocarbon feed is reformed over
a catalyst bed installed in the HTER-p. Process effluent from the SMR ����
is transferred to the HTER-p and mixed internally with the product gas
from the HTER-p catalyst. The process gas supplies the required heat
for the reforming reaction in the tubes of the HTER-p. Thus, no addi-
tional firing is required for the reforming reactions in the HTER-p.
Description: Typical PRISM membrane systems consist of a pretreatment Economics: Economic benefits are derived from high-product recoveries
(1) section to remove entrained liquids and preheat feed before gas en- and purities, from high reliability and low capital cost. Additional ben-
ters the membrane separators (2). Various membrane separator con- efits include relative ease of operation with minimal maintenance. Also,
figurations are possible to optimize purity and recovery, and operating systems are expandable and adaptable to changing requirements.
and capital costs such as adding a second stage membrane separator
Installations: Over 270 PRISM H2 membrane systems have been com-
(3). Pretreatment options include water scrubbing to recover ammonia
missioned or are in design. These systems include over 54 systems in re-
from ammonia synthesis purge stream.
finery applications, 124 in ammonia synthesis purge and 30 in synthesis
Membrane separators are compact bundles of hollow fibers contained in
gas applications.
a coded pressure vessel. The pressurized feed enters the vessel and flows on
the outside of the fibers (shell side). Hydrogen selectively permeates through Licensor: Air Products and Chemicals, Inc.
the membrane to the inside of the hollow fibers (tube side), which is at lower
pressure. PRISM membrane separators’ key benefits include resistance to wa-
ter exposure, particulates and low feed to nonpermeate pressure drop.
Membrane systems consist of a pre-assembled skid unit with pres-
sure vessels, interconnecting piping, and instrumentation and are fac-
tory tested for ease of installation and commissioning.
Hydrogen—steam reforming ���������
�������������
Application: Production of hydrogen for refinery hydrotreating and hydro- ���� ��
cracking or other refinery, petrochemical and other uses.
�
Feedstock: Light hydrocarbons such as natural gas, refinery fuel gas, LPG/bu- �
tane mixed pentanes and light naphtha.
�
Product: High-purity hydrogen (99.9+%) at any required pressure. �
Description: The feed is heated in the feed preheater and passed through
the hydrotreater (1). The hydrotreater converts sulfur compounds to H2S and �
saturates any unsaturated hydrocarbons in the feed. The gas is then sent to
the desulfurizers (2). These adsorb the H2S from the gas.
The desulfurized feed gas is mixed with steam and superheated in the
feed preheat coil. The feed mixture then passes through catalyst-filled tubes in ����� �
the reformer (3). In the presence of nickel catalyst, the feed reacts with steam
to produce hydrogen and carbon oxides. Heat for the endothermic reform-
ing reaction is provided by carefully controlled external firing in the reformer. Economics: Typical utilities per Mscf of hydrogen production based on a
Combustion air preheat is used, if applicable to limit export steam. natural gas feedstock and maximum export steam:
Gas leaving the reformer is cooled by the process steam generator (4). Gas Feed and fuel, MM Btu LHV 0.44
is then fed to the shift converter (5), which contains a bed of copper-promoted Export steam, lb 75
iron-chromium catalyst. This converts CO and water vapor to additional H2 Boiler feedwater, lb 115
and CO2. Shift converter effluent gas is cooled, condensate is separated and Power, kW 0.5
the gas is sent to a PSA hydrogen purification system (6). Water, cooling, gal 10
The PSA system operates on a repeated cycle having two basic steps: ad-
sorption and regeneration. PSA offgas is sent to the reformer, where it provides Installations: Over 175 plants worldwide — ranging in size from less than
most of the fuel requirement. Hydrogen from the PSA unit is sent off plot. A 1 MMscfd to over 120 MMscfd capacities. Plant designs for capacities
small hydrogen stream is recycled to the feed of the plant for hydrotreating. from 1 to 200 MMscfd.
The thermal efficiency of the plant is optimized by recovery of heat
from the reformer flue gas stream and from the reformer effluent process Supplier: CB&I Howe Baker.
gas stream. This energy is utilized to preheat reformer feed gas and generate
steam for reforming and export. The process design is customized for each
application depending on project economics and export steam demand.
Hydrogen—steam reforming
Application: Manufacture hydrogen for hydrotreating, hydrocracking or
other refinery or chemical use. �
����������������
�����
Feedstock: Light saturated hydrocarbons: refinery gas or natural gas,
�����
LPG or light naphtha.
� �
Products: Typical purity 99.99%; pressure 300 psig, with steam or CO2
as byproducts.
Operating conditions: 550°F to 750°F and 400 psig to 1,500 psig reac-
tor conditions.
Economics:
Utilities, (per bbl feed) Naphtha Diesel
Fuel, 103 Btu release 48 59.5
Electricity, kWh 0.65 1.60
Water, cooling (20°F rise), gal 35 42
Description: The light, mid and heavy cat naphthas (LCN, MCN, HCN)
�����
are treated separately, under optimal conditions for each. The full- �������
range FCC gasoline sulfur reduction begins with fractionation of the
light naphtha overhead in a CDHydro column. Mercaptan sulfur re-
��������
acts quantitatively with excess diolefins to produce heavier sulfur com-
pounds, and the remaining diolefins are partially saturated to olefins ���
by reaction with hydrogen. Bottoms from the CDHydro column, con-
taining the reacted mercaptans, are fed to the CDHDS column where
the MCN and HCN are catalytically desulfurized in two separate zones.
HDS conditions are optimized for each fraction to achieve the desired
or to replace catalyst. Typical fixed-bed processes will require a mid FCC
sulfur reduction with minimal olefin saturation. Olefins are concentrat-
shutdown to regenerate/replace catalyst, requiring higher capital cost
ed at the top of the column, where conditions are mild, while sulfur
for feed, storage, pumping and additional feed capacity.
is concentrated at the bottom where the conditions result in very high
levels of HDS. Economics: The estimated ISBL capital cost for a 50,000-bpd CDHydro/
No cracking reactions occur at the mild conditions, so that yield CDHDS unit with 95% desulfurization is $40 million (2005 US Gulf Coast).
losses are easily minimized with vent-gas recovery. The three product Direct operating costs—including utilities, catalyst, hydrogen and octane
streams are stabilized together or separately, as desired, resulting in replacement—are estimated at $0.04/gal of full-range FCC gasoline.
product streams appropriate for their subsequent use. The two columns
are heat integrated to minimize energy requirements. Typical reformer Installation: Twenty-one CDHydro/CDHDS units are in operation treating
hydrogen is used in both columns without makeup compression. The FCC gasoline and 12 more units are currently in engineering/construc-
sulfur reduction achieved will allow the blending of gasoline that meets tion. Total licensed capacity exceeds 1.3 million bpd.
current and future regulations.
Catalytic distillation essentially eliminates catalyst fouling because Licensor: CDTECH.
the fractionation removes heavy-coke precursors from the catalyst zone
before coke can form and foul the catalyst pores. Thus, catalyst life in
catalytic distillation is increased significantly beyond typical fixed-bed
life. The CDHydro/CDHDS units can operate throughout an FCC turn-
around cycle up to six years without requiring a shutdown to regenerate
Hydrotreating �����������
Application: Topsøe hydrotreating technology has a wide range of ap- ������ ����������
��������
plications, including the purification of naphtha, distillates and residue,
as well as the deep desulfurization and color improvement of diesel fuel ��������
������� ����������
and pretreatment of FCC and hydrocracker feedstocks.
Products: Ultra-low-sulfur diesel fuel, and clean feedstocks for FCC and �������
hydrocracker units.
����������
Description: Topsøe’s hydrotreating process design incorporates our in- ���������� �����������
dustrially proven high-activity TK catalysts with optimized graded-bed
loading and high-performance, patented reactor internals. The combi- �����
�����������
nation of these features and custom design of grassroots and revamp ������������� �������������
���������
hydrotreating units result in process solutions that meet the refiner’s ������������
objectives in the most economic way. ���������
In the Topsøe hydrotreater, feed is mixed with hydrogen, heated
and partially evaporated in a feed/effluent exchanger before it enters
the reactor. In the reactor, Topsøe’s high-efficiency internals have a
low sensitivity to unlevelness and are designed to ensure the most Operating conditions: Typical operating pressures range from 20 to 80
effective mixing of liquid and vapor streams and the maximum utili- barg (300 to 1,200 psig), and typical operating temperatures range from
zation of the catalyst volume. These internals are effective at a high 320°C to 400°C (600°F to 750°F).
range of liquid loadings, thereby enabling high turndown ratios. Top-
References: Cooper, B. H. and K. G. Knudsen, “Production of ULSD: Cat-
søe’s graded-bed technology and the use of shape-optimized inert
alyst, kinetics and reactor design,” World Petroleum Congress, 2002.
topping and catalysts minimize the build-up of pressure drop, there-
Patel, R. and K. G. Knudsen, “How are refiners meeting the ul-
by enabling longer catalyst cycle length. The hydrotreating catalysts
tra-low-sulfur diesel challenge,” NPRA Annual Meeting, San Antonio,
themselves are of the Topsøe TK series, and have proven their high
March 2003.
activities and outstanding performance in numerous operating units
Topsøe, H., K. Knudsen, L. Skyum and B. Cooper, “ULSD with BRIM
throughout the world. The reactor effluent is cooled in the feed-ef-
catalyst technology,” NPRA Annual Meeting, San Francisco, March 2005.
fluent exchanger, and the gas and liquid are separated. The hydro-
gen gas is sent to an amine wash for removal of hydrogen sulfide Installation: More than 60 Topsøe hydrotreating units for the various ap-
and is then recycled to the reactor. Cold hydrogen recycle is used as plications above are in operation or in the design phase.
quench gas between the catalyst beds, if required. The liquid product
is steam stripped in a product stripper column to remove hydrogen Licensor: Haldor Topsøe A/S.
sulfide, dissolved gases and light ends.
Hydrotreating
Application: The IsoTherming process provides refiners an economical ��������
means to produce ultra-low-sulfur diesel (ULSD), low-sulfur and low-
nitrogen FCC feedstocks, and other very low-sulfur hydrocarbon prod- � �
����
ucts. In addition, IsoTherming can provide a cost-effective approach to
wax and petrolatum hydrogenation to produce food-grade or pharma-
ceutical-grade oil and wax products, and lubestock hydroprocessing for
sulfur reduction and VI improvement. � � �
Continued
Hydrotreating, continued
Economics: Revamp investment (basis 15,000 –20,000 bpsd, 1Q 2004,
US Gulf Coast) $400/bpsd diesel
Installation: Four units have been licensed for ULSD; two units licensed
for gasoil mild hydrocracking.
removal (6) and recycled to the reactor. The liquid phase is sent to the ���� ���������� ����� ���������
�������� ��������
stripper (7) where small amounts of gas and naphtha are removed and �����������
high-quality product diesel is recovered. ��������
���
Whether the need is for a new unit or for maximum reuse of existing
diesel HDS units, the Prime-D hydrotreating toolbox of solutions meets
the challenge. Process objectives ranging from low-sulfur, ultra-low-sul-
fur, low-aromatics, and/or high cetane number are met with minimum
cost by:
• Selection of the proper catalyst from the HR 500 Series, based on and technical service feedback to ensure the right application of the
the feed analysis and processing objectives. HR 500 catalysts cover the right technology for new and revamp projects.
range of ULSD requirements with highly active and stable catalysts. HR Whatever the diesel quality goals—ULSD, high cetane or low aro-
526 CoMo exhibits high desulfurization rates at low to medium pres- matics—Prime-D’s Hydrotreating Toolbox approach will attain your goals
sures; HR 538/HR 548 NiMo have higher hydrogenation activities at in a cost-effective manner.
higher pressures. Installation: Over 150 middle distillate hydrotreaters have been li-
• Use of proven, efficient reactor internals, EquiFlow, that allow censed or revamped. They include 56 low- and ultra-low-sulfur diesel
near-perfect gas and liquid distribution and outstanding radial tempera- units (<50 ppm), as well as a number of cetane boosting units. Most
ture profiles. of those units are equipped with Equiflow internals.
• Loading catalyst in the reactor(s) with the Catapac dense loading
technique for up to 20% more reactor capacity. Over 10,000 tons of References: “Getting Total Performance with Hydrotreating,” Petroleum
catalyst have been loaded quickly, easily and safely in recent years using Technology Quarterly, Spring 2002.
the Catapac technique. “Premium Performance Hydrotreating with Axens HR 400 Series
• Application of Advanced Process Control for dependable opera- Hydrotreating Catalysts,” NPRA Annual Meeting, March 2002, San
tion and longer catalyst life. Antonio.
• Sound engineering design based on years of R&D, process design Continued
Hydrotreating, diesel, continued
“The Hydrotreating Toolbox Approach,” Hart’s European Fuel News,
May 29, 2002.
“Squeezing the most from hydrotreaters,” Hydrocarbon Asia, April/
May 2004.
Licensor: Axens.
Hydrotreating/desulfurization
Application: The SelectFining process is a gasoline desulfurization tech- �����������
nology developed to produce ultra-low-sulfur gasoline by removing ���������
more than 99% of the sulfur present in olefinic naphtha while: ���������������� ����
• Minimizing octane loss �������
Description: Oil feed and hydrogen are charged to the reactors in a suited to revamp existing RDS/VRDS units for additional throughput or
once-through operation. The catalyst combination can be varied signifi- heavier feedstock.
cantly according to feedstock properties to meet the required product
Installation: Over 26 RDS/VRDS units are in operation. Six units have ex-
qualities. Product separation is done by the hot separator, cold separator
tensive experience with VR feedstocks. Sixteen units prepare feedstock
and fractionator. Recycle hydrogen passes through an H2S absorber.
for RFCC units. Four OCR units and two UFR unit are in operation, with
A wide range of AR, VR and DAO feedstocks can be processed. Ex-
another six in engineering. Total current operating capacity is about 1.1
isting units have processed feedstocks with viscosities as high as 6,000
million bpsd
cSt at 100°C and feed-metals contents of 500 ppm.
Onstream Catalyst Replacement (OCR) reactor technology has been References: Reynolds, “Resid Hydroprocessing With Chevron Technol-
commercialized to improve catalyst utilization and increase run length ogy,” JPI, Tokyo, Japan, Fall 1998.
with high-metals, heavy feedstocks. This technology allows spent cata- Reynolds and Brossard, “RDS/VRDS Hydrotreating Broadens Appli-
lyst to be removed from one or more reactors and replaced with fresh cation of RFCC,” HTI Quarterly, Winter 1995/96.
while the reactors continue to operate normally. The novel use of up- Reynolds, et al., “VRDS for conversion to middle distillate,” NPRA
flow reactors in OCR provides greatly increased tolerance of feed solids Annual Meetng, March 1998, Paper AM-98-23.
while maintaining low-pressure drop.
A related technology called UFR (upflow reactor) uses a multibed Licensor: Chevron Lummus Global LLC.
upflow reactor for minimum pressure drop in cases where onstream
catalyst replacement is not necessary. OCR and UFR are particularly well
Hydrotreating—resid
����
Application: Upgrade or convert atmospheric and vacuum residues us-
ing the Hyvahl fixed-bed process.
Products: Low-sulfur fuels (0.3% to 1.0% sulfur) and RFCC feeds (re-
moval of metals, sulfur and nitrogen, reduction of carbon residue). Thirty �������
percent to 50% conversion of the 565°C+ fraction into distillates. ��������
�������
Description: Residue feed and hydrogen, heated in a feed/effluent ex-
changer and furnace, enter a reactor section—typically comprising of a
guard-reactor section, main HDM and HDS reactors.
The guard reactors are onstream at the same time in series, and they �������
Licensor: Axens.
Hydrotreating—residue
Application: Upgrading or converting atmospheric and vacuum residues �������
using the Genoil GHU process.
Products: Removal of metals, maximize desulfurization (>90%), deni- ��
trogenation (>70%), reduction of carbon residue (>90%) with API in- �� ������� ���������
���������
crease during the conversion process. Up to 90% conversion of 350°C+
fraction into distillates.
Description: Genoil has developed devices that enhance the mixing of
�������� ������������
liquid hydrocarbons as well as highly efficient reactor internals. These
modifications have contributed to the high level of residue conversion,
desulfurization, denitrogenation and turnaround time.
Residue feed and hydrogen are heated in feed effluent exchanger
and furnace and enter the reactor section. The reactor section is typically
Investment: Based on 20,000 bpd, atmospheric or vacuum or VR feeds,
comprised of a guard-section hydrodemineralization (HDM) reactor, and
$3,000 – 5,000/bbl based of Gulf Coast rates.
sulfur-removal section—hydrodesulfurization (HDS).
The guard-reactor section protects the downstream HDS reactors by Installations: Genoil owns and operates a 10 BPD demonstration plant
removing metals and asphaltenes. Catalyst and operational objectives can at Two Hill, Alberta where we have conducted testing on several differ-
be adjusted according to feed metals content through different mechani- ent types of residue and crudes shipped to our facility by various com-
cal means to insure longer runtimes and constant protection for down- panies. We are currently working with several companies to get first
stream reactors. commercial installation of the GHU process.
After the guard section carries out the final removal of metals and
conversion, the HDS section removes the sulfur to design specifications. Reference: Asia Pacific Refining Conference Bangkok, Thailand, Sep-
With these factors in mind, Genoil has come up with a hydroconversion tember, 2005.
process that provides higher conversion, higher desulfurization and de- RPBC Moscow, Russia, April Conference 2006.
nitrogenation rates at lower pressure by using a simple, easy to operate Middle East Refining Conference, Doha, Qatar, May 2006.
process. America Oil and Gas Reporter, October 2005.
The upgrading process can be used as a field upgrader where heavy Energy Magazine, June 2005, Petroleum Technology Quarterly, Jan-
oil can be upgraded to WTI specification, pipeline specifications and pipe- uary 2006.
lined to refineries. Unconverted oil from the GHUunit can be sold as a Oil & Gas Product News, “Surge Global Announcement,” Jan./Feb.
stable, low-sulfur fuel oil or sent to another heavy-oil conversion unit for 2005.
further upgrading.
Licensor: Genoil Inc.
Yields: Net increase in production from 0 –10% naphtha, 1–20% kero-
sine and 21– 47% diesel.
Isomerization
Application: C5 /C6 paraffin-rich hydrocarbon streams are isomerized to
produce high RON and MON product suitable for addition to the gaso-
line pool. ��
������
Description: Several variations of the C5 / C6 isomerization process are ����������
available. The choice can be a once-through reaction for an inexpensive-
but-limited octane boost, or, for substantial octane improvement and as �����
�
an alternate (in addition) to the conventional DIH recycle option, the Ip- ���������
� �
sorb Isom scheme shown to recycle the normal paraffins for their com-
plete conversion. The Hexorb Isom configuration achieves a complete
normal paraffin conversion plus substantial conversion of low (75) oc- ��������
tane methyl pentanes gives the maximum octane results. With the most
active isomerization catalyst (chlorinated alumina), particularly with the �������
Albemarle /Axens jointly developed ATIS2L catalyst, the isomerization
performance varies from 84 to 92: once-through isomerization -84,
isomerization with DIH recycle -88, Ipsorb -90, Hexorb-92.
Description:
C4 olefin skeletal isomerization (ISOMPLUS)
�
A zeolite-based catalyst especially developed for this process pro-
vides near equilibrium conversion of normal butenes to isobutylene at � �
high selectivity and long process cycle times. A simple process scheme
and moderate process conditions result in low capital and operating
costs. Hydrocarbon feed containing n-butenes, such as C4 raffinate, can ���
be processed without steam or other diluents, nor the addition of cata- ������������������
lyst activation agents to promote the reaction. Near-equilibrium con-
version levels up to 44% of the contained n-butenes are achieved at
greater than 90% selectivity to isobutylene. During the process cycle,
coke gradually builds up on the catalyst, reducing the isomerization ac-
tivity. At the end of the process cycle, the feed is switched to a fresh
catalyst bed, and the spent catalyst bed is regenerated by oxidizing the Total installed cost: Feedrate, Mbpd ISBL cost, $MM
coke with an air/nitrogen mixture. The butene isomerate is suitable for 10 8
making high purity isobutylene product. 15 11
30 20
C5 olefin skeletal isomerization (ISOMPLUS) Utility consumption: per barrel of feed (assuming an electric-motor-
A zeolite-based catalyst especially developed for this process pro- driven compressor) are:
vides near-equilibrium conversion of normal pentenes to isoamylene at Power, kWh 3.2
high selectivity and long process cycle times. Hydrocarbon feeds con- Fuel gas, MMBtu 0.44
taining n-pentenes, such as C5 raffinate, are processed in the skeletal Steam, MP, MMBtu 0.002
isomerization reactor without steam or other diluents, nor the addition Water, cooling, MMBtu 0.051
of catalyst activation agents to promote the reaction. Near-equilibrium Nitrogen, scf 57–250
conversion levels up to 72% of the contained normal pentenes are ob-
served at greater than 95% selectivity to isoamylenes. Installation: Two plants are in operation. Two licensed units are in vari-
ous stages of design.
Economics: The ISOMPLUS process offers the advantages of low capital
investment and operating costs coupled with a high yield of isobutylene Licensor: CDTECH and Lyondell Chemical Co.
or isoamylene. Also, the small quantity of heavy byproducts formed can
easily be blended into the gasoline pool. Capital costs (equipment, labor
and detailed engineering) for three different plant sizes are:
Isomerization
Application: Hydrisom is the ConocoPhillips selective diolefin hydroge-
nation process, with specific isomerization of butene-1 to butene-2 and ����������
��������� �������� ����
3-methyl-butene-1 to 2-methyl-butene-1 and 2-methyl-butene-2. The ���������
Hydrisom Process uses a liquid-phase reaction over a commercially avail-
able catalyst in a fixed-bed reactor.
Description: The fresh C5/ C6 feed is combined with make-up and re- ����������
Feed: Typical feed sources for the Par-Isom process include hydrotreated
light straight-run naphtha, light natural gasoline or condensate and light Installation: The first commercial Par-Isom process unit was placed in
raffinate from benzene extraction units. operation in 1996. There are currently 10 units in operation. The first
Water and oxygenates at concentrations of typical hydrotreated commercial application of PI-242 catalyst was in 2003, and the unit has
naphtha are not detrimental, although free water in the feedstock must demonstrated successful performance meeting all expectations.
be avoided. Sulfur suppresses activity, as expected, for any noble-metal
based catalyst. However, the suppression effect is fully reversible by sub- Licensor: UOP LLC.
sequent processing with clean feedstocks.
Yield: Typical product C5+ yields are 97 wt% of the fresh feed. The
product octane is 81 to 87, depending on the flow configuration and
feedstock qualities.
Isomerization
Application: Most of the implemented legislation requires limiting ��������������� ����������������
benzene concentration in the gasoline pool. This has increased the ��������
demand for high-performance C5 and C6 naphtha isomerization tech-
��������
nology because of its ability to reduce the benzene concentration in �����
the gasoline pool while maintaining or increasing the pool octane. The
Penex process has served as the primary isomerization technology for ����������
upgrading C5 / C6 light straight-run naphtha.
in the first and finishing reactors. The process operates at lower pressure �� � �
���������
and provides lower costs for the hydrogenation unit. ������� �
�
Description: The feed is water washed to remove any basic compounds
that can poison the catalyst system. Most applications will be directed �
toward isooctene production. However as olefin specifications are re- ��
�����������������
�������� �
quired, the isooctene can be hydrogenated to isooctane, which is an
excellent gasoline blending stock.
The RHT isooctene process has a unique configuration; it is flexible
and can provide low per pass conversion through dilution, using a new provides the dilution, and reactor effluent is fed to the column at mul-
selectivator. The dual catalyst system also provides multiple advantages. tiple locations. Recycling does not increase column size due to the
The isobutylene conversion is 97–99 %, with better selectivity and yield unique configuration of the process. The isooctene is taken from the
together with enhanced catalyst life. The product is over 91% C8 olefins, debutanizer column bottom and is sent to OSBL after cooling or as is
and 5 – 9% C12 olefins, with very small amount of C16 olefins. sent to hydrogenation unit. The C4s are taken as overhead stream and
The feed after water wash, is mixed with recycle stream, which sent to OSBL or alkylation unit. Isooctene/product, octane (R+M)/2 is
provides the dilution (also some unreacted isobutylene) and is mixed expected to be about 105.
with a small amount of hydrogen. The feed is sent to the dual-bed re- If isooctane is to be produced the debutanizer bottom, isooctene
actor for isooctene reaction in which most of isobutylene is converted product is sent to hydrogenation unit. The isooctene is pumped to the
to isooctene and codimer. The residual conversion is done with single- required pressure (which is much lower than conventional processes),
resin catalyst via a side reactor. The feed to the side reactor is taken as mixed with recycle stream and hydrogen and is heated to the reaction
a side draw from the column and does contain unreacted isobutylene,
selectivator, normal olefins and non-reactive C4s. The recycle stream Continued
Isooctene/Isooctane, continued Economics:
Isooctene Isooctane1
temperature before sending it the first hydrogenation reactor. This reac- CAPEX ISBL, MM USD
tor uses a nickel (Ni) or palladium (Pd) catalyst. (US Gulf Coast 1Q 06, 1,000 bpd) 8.15 5.5
If feed is coming directly from the isooctene unit, only a start-up Utilities basis 1,000-bpd isooctene/isooctane
heater is required. The reactor effluent is flashed, and the vent is sent Power, kWh 65 105
to OSBL. The liquid stream is recycled to the reactor after cooling (to Water, cooling, m /h 3 154 243
remove heat of reaction) and a portion is forwarded to the finishing Steam, HP, kg/h 3,870 4,650
reactor—which also applies a Ni or Pd catalyst (preferably Pd catalyst) — Basis: FCC feed (about 15 –20% isobutelene in C4 mixed stream)
and residual hydrogenation to isooctane reaction occurs. The isooctane 1These utilities are for isooctene / isooctane cumulative.
product, octane (R+M)/2 is >98.
The reaction occurs in liquid phase or two phase (preferably two Installation: Technology is ready for commercial application.
phases), which results in lower pressure option. The olefins in isooctene
product are hydrogenated to over 99%. The finishing reactor effluent is Licensor: Refining Hydrocarbon Technologies LLC.
sent to isooctane stripper, which removes all light ends, and the product
is stabilized and can be stored easily.
Lube and wax processing
Application: Vacuum gas oils (VGOs) are simultaneously extracted and ������� ���������
���� ����
dewaxed on a single unit to produce low-pour aromatic extracts and
lube-base stocks having low-pour points. Low viscosity grades (60 SUS)
to bright stocks can be produced. With additional stages of filtration,
waxes can be deoiled to produce fully refined paraffin waxes.
Products: Lube-base stocks having low pour points (– 20°C). Very low
��������
pour point aromatic extracts. Slack waxes or low-oil content waxes.
���������
Description: Process Dynamics’ integrated extraction/dewaxing technol-
ogy is a revolutionary process combining solvent extraction and solvent
dewaxing onto a single unit, using a common solvent system, for extrac-
tion and dewaxing steps. This process offers the advantage of operating
a single unit rather than separate extraction and dewaxing units; thus, ��������� ��������
reducing both capitol and operating costs. The more selective solvent
system produces lube-base stocks of higher quality and higher yields
when compared to other technologies.
Primary solvent and warm feed are mixed together; temperature is
controlled by adding the cosolvent solvent. Filtrate or wax may be re-
cycled for solids adjustment. Cold cosolvent is added, and the slurry is Installation: Basic engineering package for the first commercial unit has
filtered (or separated by other means). Solvents are recovered from the been completed.
primary filtrate producing an aromatic extract. The wax cake is repulped
with additional solvent/cosolvent mix and refiltered. Solvents are recov- Licensor: P. D. Licensing, LLC (Process Dynamics, Inc.).
ered from the filtrate producing a lube-base stock.
Extraction/dewaxing comparisons of 90 SUS stock
Furfural/MEK Process Dynamics
A B
Raffinate yields, vol% 53 60 71
Dewaxed oil properties:
Viscosity @40°C, cSt 16.5 18.7 20
Viscosity index 92 98 92
Pour pt., °F 5 5 5
Lube extraction
Application: Bechtel’s MP Refining process is a solvent-extraction pro-
�
cess that uses N-methyl-2-pyrrolidone (NMP) as the solvent to selec- ���� � �
tively remove the undesirable components of low-quality lubrication oil,
����� �
which are naturally present in crude oil distillate and residual stocks. �
The unit produces paraffinic or naphthenic raffinates suitable for further
processing into lube-base stocks. This process selectively removes aro- ���� �����
matics and compounds containing heteroatoms (e.g., oxygen, nitrogen ����
and sulfur).
����������
Products: A raffinate that may be dewaxed to produce a high-qual-
�������
ity lube-base oil, characterized by high viscosity index, good thermal
and oxidation stability, light color and excellent additive response. The
byproduct extracts, being high in aromatic content, can be used, in
some cases, for carbon black feedstocks, rubber extender oils and other
nonlube applications where this feature is desirable. Economics:
Description: The distillate or residual feedstock and solvent are contact- Investment (Basis: 10,000-bpsd feedrate
ed in the extraction tower (1) at controlled temperatures and flowrates capacity, 2006 US Gulf Coast), $/bpsd 3,000
required for optimum countercurrent, liquid-liquid extraction of the Utilities, typical per bbl feed:
feedstock. The extract stream, containing the bulk of the solvent, exits Fuel, 103 Btu (absorbed) 100
the bottom of the extraction tower. It is routed to a recovery section Electricity, kWh 2
to remove solvent contained in this stream. Solvent is separated from Steam, lb 5
the extract oil by multiple-effect evaporation (2) at various pressures, Water, cooling (25°F rise), gal 600
followed by vacuum flashing and steam stripping (3) under vacuum. Installation: This process is being used in 15 licensed units to produce high-
The raffinate stream exits the overhead of the extraction tower and is quality lubricating oils. Of this number, eight are units converted from phe-
routed to a recovery section to remove the NMP solvent contained in nol or furfural, with another three units under license for conversion.
this stream by flashing and steam stripping (4) under vacuum.
Overhead vapors from the steam strippers are condensed and com- Licensor: Bechtel Corp.
bined with solvent condensate from the recovery sections and are dis-
tilled at low pressure to remove water from the solvent (5). Solvent is
recovered in a single tower because NMP does not form an azeotrope
with water, as does furfural. The water is drained to the oily-water sew-
er. The solvent is cooled and recycled to the extraction section.
Continued
Lube extraction
Application: Bechtel’s Furfural Refining process is a solvent-extraction
process that uses furfural as the solvent to selectively remove undesir-
�
able components of low lubrication oil quality, which are naturally pres- ���� � �
ent in crude oil distillate and residual stocks. This process selectively re-
����� � �
moves aromatics and compounds containing heteroatoms (e.g., oxygen, �
nitrogen and sulfur). The unit produces paraffinic raffinates suitable for ����
further processing into lube base stocks. ����
����
Products: A raffinate that may be dewaxed to produce a high-qual- �����
ity lube-base oil, characterized by high viscosity index, good thermal ����������
and oxidation stability, light color and excellent additive response. The
�������
byproduct extracts, being high in aromatic content, can be used, in
some cases, for carbon black feedstocks, rubber extender oils and other
nonlube applications where this feature is desirable.
Economics:
Investment (Basis 7,000-bpsd feedrate capacity,
2006 U.S. Gulf Coast), $/bpsd 7,100
Utilitiies, typical per bbl feed:
Fuel, 103 Btu (absorbed) 70
Electricity, kWh 5
Steam, lb 15
Water, cooling (25°F rise), gal 200
Continued
Lube oil refining, spent, continued
Utilities: Basis one metric ton of water-free feedstock
Config. 1 Config. 2
Electrical power, kWh 45 55
Fuel, million kcal 0.62 0.72
Steam, LP, kg — 23.2
Steam, MP, kg 872 890
Water, cooling, m3 54 59
Installation: Ten units have been licensed using all or part of the Revivoil
Technology.
Products: Lube oil raffinates of high viscosity indices. The raffinates con- �������
������
�������
tain substantially all of the desirable lubricating oil components present ������� ���� ��������
in the feedstock. The extract contains a concentrate of aromatics that ��� ������
�������
may be utilized as rubber oil or cracker feed. ���
�������
pounds. In the extraction tower, the feed oil is introduced below the top
at a predetermined temperature. The raffinate phase leaves at the top
of the tower, and the extract, which contains the bulk of the furfural, is
withdrawn from the bottom. The extract phase is cooled and a so-called
“pseudo raffinate“ may be sent back to the extraction tower. Multi-
stage solvent recovery systems for raffinate and extract solutions secure
energy efficient operation.
Utility requirements (typical, Middle East Crude), units per m3 of feed:
Electricity, kWh 10
Steam, MP, kg 10
Steam, LP, kg 35
Fuel oil, kg 20
Water, cooling, m3 20
Installation: Numerous installations using the Uhde (Edeleanu) propri-
etary technology are in operation worldwide. The most recent is a com-
plete lube-oil production facility licensed to the state of Turkmenistan.
Licensor: Uhde GmbH.
Mercaptan removal
����������
Application: Extraction of mercaptans from gases, LPG, lower boiling
fractions and gasolines, or sweetening of gasoline, jet fuel and diesel by ����������������� ��������
in situ conversion of mercaptans into disulfides.
�
Products: Essentially mercaptan sulfur-free, i.e., less than 5 ppmw, and
concomitant reduced total sulfur content when treated by Merox ex- �
���
������������� �
traction technique.
Products: If conditions are appropriate, the flue gas is treated to achieve ���������
NOx reductions of 40% to 70%+ with minimal NH3 slip or leakage. �������������
�������
Description: The technology involves the gas-phase reaction of NO with ��� ������
��������� ���������������
NH3 (either aqueous or anhydrous) to produce elemental nitrogen if con-
�����������
ditions are favorable. Ammonia is injected into the flue gas using steam
or air as a carrier gas into a zone where the temperature is 1,600°F ����
to 2,000°F. This range can be extended down to 1,300°F with a small �������������� ��������������
amount of hydrogen added to the injected gas. For most applications,
wall injectors are used for simplicity of operation.
Yield: Cleaned flue gas with 40% to 70%+ NOx reduction and less than
10-ppm NH3 slip.
Economics: Considerably less costly than catalytic systems but relatively McIntyre, A. D., “Applications of the THERMAL DeNOx process to
variable depending on scale and site specifics. Third-party studies have FBC boilers,” CIBO 13th Annual Fluidized Bed Conference, Lake Charles,
estimated the all-in cost at about 600 US$ / ton of NOx removed. Louisiana, 1997.
Installation: Over 135 applications on all types of fired heaters, boilers Licensor: ExxonMobil Research and Engineering Co., via an alliance with
and incinerators with a wide variety of fuels (gas, oil, coal, coke, wood Engineers India Ltd. (for India) and Hamon Research-Cottrell (for the rest
and waste). The technology can also be applied to full-burn FCCU re- of the world).
generators.
Operating conditions: Ozone injection typically occurs in the flue-gas Reference: Confuorto, et al., “LoTOx technology demonstration at Mar-
stream upstream of the scrubber, near atmospheric pressure and at athon Ashland Petroleum LLC’s refinery at Texas City, Texas,” NPRA An-
temperatures up to roughly 150°C. For higher-temperature streams, the nual Meeting, March 2004, San Antonio.
ozone is injected after a quench section of the scrubber, at adiabatic
Licensor: Belco Technologies Corp., as a sub-licensor for The BOC
saturation, typically 60°C to 75°C. High-particulate saturated gas and
Group, Inc.
sulfur loading (SOx or TRS) do not cause problems.
Economics: The costs for NOx control using this technology are espe-
cially low when used as a part of a multi-pollutant control scenario.
Sulfurous and particulate-laden streams can be treated attractively as no
pretreatment is required by the LoTOx system.
can be achieved via oxygen enrichment in the FCC regeneration. ���������������������������� �������������
In the FCC reactor, long-chain hydrocarbons are cleaved into shorter ������������������������������������� �������������
�������������������������������������� ��������������
chains in a fluidized-bed reactor at 450 –550°C. This reaction produces ��������������������������� ��������������
coke as a byproduct that deposits on the catalyst. To remove the coke ������������������������������������ ���������������������
from the catalyst, it is burned off at 650 –750°C in the regenerator. The
regenerated catalyst is returned to the reactor.
Oxygen enrichment, typically up to 27 vol% oxygen, intensifies cata- products, such as naphtha. Likewise, lower value products increased
lyst regeneration and can substantially raise throughput capacity and/or only 5%, as fuel gas. The net profit increased substantially. Installed cost
conversion of the FCC unit. Oxygen sources can be liquid oxygen tanks, for oxygen enrichment is typically below $250,000.
onsite ASUs or pipeline supply. Oxygen consumption in FCC units fluctu- Operating costs will depend on the cost for oxygen and the duration
ates widely in most cases; thus, tanks are the best choice with respect to of oxygen enrichment. Economical oxygen usage can be calculated on
ease of operation, flexibility and economy. a case-by-case basis and should include increased yields of higher-value
For oxygen addition into the CS air duct, a number of safety rules products and optional usage of lower-value feeds.
must be observed. The oxygen metering device FLOWTRAIN contains all
necessary safety features, including flow control, low-temperature and Installations: Currently, four units are in operation, plus test installations
low-pressure alarm and switch-off, and safe standby operation. All of to quantify the effects of higher capacity and conversion levels.
these features are connected to the FCC units’ process control system.
Reference: Heisel, M. P., C. Morén, A. Reichhold, A. Krause, A. Berlanga,
An efficient mixing device ensures even oxygen distribution in the air
“Cracking with Oxygen,” Linde Technology 1, 2004.
feed to the FCC regeneration.
Contributor: Linde AG, Division Gas and Engineering.
Economics: Oxygen enrichment in FCC regeneration is economically
favorable in many plants. For example, one refinery increased through-
put by 15%. The net improvement was a 26% increase in higher-value
Olefin etherification
Application: New processing methods improve etherification of C4– C7
reactive olefins including light catalytic naphtha (LCN) with alcohol (e.g., �����������������
����������������
���
�������
������������ ��������� ���������
�������
�������
����������
����������������
������
methanol and ethanol). The processes, RHT-MixedEthers, RHT-MTBE, ������������ ��������
RHT-ETBE, RHT-TAME and RHT-TAEE, use unique concepts to achieve the ������� �������������
�����
��������������� ���
maximum conversion without applying cumbersome catalyst in the col- �����������
�����
�� ��
umn. The processing economics provide improvements over other avail-
�
able ether technologies currently available. The technology suite can ���������� �
�
��
�������
� �� �������
be applied to ethyl tertiary butyl ether (ETBE) production in which wet ���������� ����������
������ � �
ethanol can be used in place of dry ethanol. The drier can be eliminated,
�
which is approximately half the cost for an etherification unit. The RHT �����
��� ��������
���������� ��
ethers processes can provide the highest conversion with unique mul-
������������������
tiple equilibrium stages. ������������� ������� ������������
������������������������������������������
Description: The feed is water washed to remove basic compounds
that are poisons for the resin catalyst of the etherification reaction. The
C4 ethers—methyl tertiary butyl ether (MTBE)/ETBE), C5– tertiary amyl
methyl ether (TAME/ tertiary amyl ethyl ether (TAEE) and C6 /C7 ethers
are made in this process separately. The reaction is difficult; heavier
ethers conversion of the reactive olefins are equilibrium conversion of separate the ether and heavy hydrocarbons from C4 or C5 hydrocarbons,
about 97% for MTBE and 70% for TAME and much lower for C6/C7 which are taken as overhead. Single or multiple draw offs are taken from
ethers are expected. the fractionation column. In the fractionation column, unreacted olefins
Higher alcohols have similar effects (azeotrope hydrocarbon/alcohol (C4 or C5) are sent to the finishing reactor (5). This stream normally does
relationship decreases when using methanol over ethanol). The equilib- not require alcohol, since azeotrope levels are available. But, some ad-
rium conversions and azeotrope effects for higher ethers are lower, as is ditional alcohol is added for the equilibrium-stage reaction. Depending
expected. After the hydrocarbon feed is washed, it is mixed with alcohol on the liquid withdrawn (number of side draws), the conversion can be
with reactive olefin ratio control with alcohol. enhanced to a higher level than via other conventional or unconven-
The feed mixture is heated to reaction temperature (and mixed with tional processes.
recycle stream (for MTBE/ETBE only) and is sent to the first reactor (1), By installing multiple reactors, it is possible to extinct the olefins
where equilibrium conversion is done in the presence of sulfonated resin within the raffinate. The cost of side draws and reactors can achieve
catalyst, e.g. Amberlyst 15 or 35 or equivalent from other vendors. pay-off in 6 to 18 months by the higher catalyst cost as compared to
Major vaporization is detrimental to this reaction. Vapor-phase re- other processes. This process could provide 97– 99.9% isobutene con-
active olefins are not available for reaction. Additionally at higher tem- version in C4 feed (depending on the configuration) and 95 – 98+% of
peratures, there is slight thermal degradation of the catalyst occurs. The isoamylenes in C5 stream.
reactor effluent is sent to fractionator (debutanizer or depentanizer) to Continued
catalyst. Distillation is done at optimum conditions. Much lower steam
Olefin etherification, continued consumption for alcohol recovery. For example, the C5 feed case requires
less alcohol with RHT configuration (azeotropic alcohol is not required)
The ether product is taken from the bottom, cooled and sent to the
and lowers lower steam consumption.
storage. The raffinate is washed in extractor column (6) with and is sent
to the OSBL. The water/alcohol mixture is sent to alcohol recovery col- Economics:
umn (7) where the alcohol is recovered and recycled as feed. CAPEX ISBL, MM USD (US Gulf Coast 1Q06,
For ETBE and TAEE, ethanol dehydration is required for most of the 1,000-bpd ether product) 9.1
processes, whereas for RHT process, wet ethanol can be used providing Utilities Basis 1,000 bpd ether
maximum conversions. If need be, the TBA specification can be met by Power kWh 45.0
optimum design with additional equipment providing high ETBE yield Water, cooling m3/ h 250
and conversion. Cost of ethanol dehydration is much more than the Steam MP, Kg / h 6,000
present configuration for the RHT wet-ethanol process. Basis: FCC Feed (about 15–20% isobutylene in C4 mixed stream)
The total capital cost /economics is lower with conventional catalyst
usage, compared to other technologies, which use complicated struc- Commercial units: Technology is ready for commercialization.
ture, require installing a manway (cumbersome) and require frequently
catalyst changes outs.
Licensor: Refining Hydrocarbon Technologies LLC.
The RHT ether processes can provide maximum conversion as com-
pared to other technologies with better economics. No complicated or
proprietary internals for the column including single source expensive
Olefins recovery
Application: Recover high-purity hydrogen (H2) and C2+ liquid products
from refinery offgases using cryogenics.
�
Description: Cryogenic separation of refinery offgases and purges con-
taining 10– 80% H2 and 15 – 40% hydrocarbon liquids such as ethylene, �������
���������
ethane, propylene, propane and butanes. Refinery offgases are option-
ally compressed and then pretreated (1) to remove sulfur, carbon di-
oxide ( CO2), H2 O and other trace impurities. Treated feed is partially � �
condensed in an integrated multi-passage exchanger system (2) against �
Economics:
Consumption per metric ton of FCC C4 fraction feedstock:
Steam, t / t 0.5 – 0.8
Water, cooling ( T = 10°C ), m3/ t 15.0
Electric power, kWh/t 25.0
Product purity:
n - Butene content 99.+ wt.– % min.
Solvent content 1 wt.– ppm max.
cycle and internally generated steam. After preheating, the feed is sent
to the reaction section. This section consists of an externally fired tubular
fixed-bed reactor (Uhde reformer) connected in series with an adiabat-
ic fixed-bed oxyreactor (secondary reformer type). In the reformer, the
Apart from light-ends, which are internally used as fuel gas, the
endothermic dehydrogenation reaction takes place over a proprietary,
olefin is the only product. High-purity H2 may optionally be recovered
noble metal catalyst.
from light-ends in the gas separation section.
In the adiabatic oxyreactor, part of the hydrogen from the intermediate
product leaving the reformer is selectively converted with added oxygen Economics: Typical specific consumption figures (for polymer-grade
or air, thereby forming steam. This is followed by further dehydrogenation propylene production) are shown (per metric ton of propylene product,
over the same noble-metal catalyst. Exothermic selective H2 conversion including production of oxygen and all steam required):
in the oxyreactor increases olefin product space-time yield and supplies
Propane, kg/metric ton 1,200
heat for further endothermic dehydrogenation. The reaction takes place Fuel gas, GJ/metric ton 6.4
at temperatures between 500– 600°C and at 4 – 6 bar. Circul. cooling water, m3/metric ton 170
The Uhde reformer is top-fired and has a proprietary “cold” out- Electrical energy, kWh/metric ton 100
let manifold system to enhance reliability. Heat recovery utilizes process
heat for high-pressure steam generation, feed preheat and for heat re- Installation: Two commercial plants using the STAR process for dehydro-
quired in the fractionation section. genation of isobutane to isobutylene have been commissioned (in the
After cooling and condensate separation, the product is subse- US and Argentina). More than 60 Uhde reformers and 25 Uhde second-
quently compressed, light-ends are separated and the olefin product is ary reformers have been constructed worldwide.
separated from unconverted paraffins in the fractionation section. Continued
Olefins, continued
References: Heinritz-Adrian, M., “Advanced technology for C3/C4 dehy-
drogenation, “ First Russian & CIS GasTechnology Conference, Moscow,
Russia, September 2004.
Heinritz-Adrian, M., N. Thiagarajan, S. Wenzel and H. Gehrke,
“STAR—Uhde’s dehydrogenation technology (an alternative route to
C3- and C4-olefins),” ERTC Petrochemical 2003, Paris, France, March
2003.
Thiagarajan, N., U. Ranke and F. Ennenbach, “Propane/butane de-
hydrogenation by steam active reforming,” Achema 2000, Frankfurt,
Germany, May 2000.
Description: Sulfur components contained in the hydrocarbon feed are ������� �������
converted to H 2 S in the HDS vessel and then fed to two desulfurization
vessels in series. Each vessel contains two catalyst types—the first for
bulk sulfur removal and the second for ultrapurification down to sulfur �������
levels of less than 1 ppb. ���
The two-desulfurization vessels are arranged in series in such a way ���������� ���� ��� ���
that either may be located in the lead position allowing online change ��������������� ��������������� �����������
������ ������
out of the catalysts. The novel interchanger between the two vessels ����������������
allows for the lead and lag vessels to work under different optimized
conditions for the duties that require two catalyst types. This arrange-
ment may be retrofitted to existing units.
Desulfurized feed is then fed to a fixed bed of nickel-based catalyst
that converts the hydrocarbon feed, in the presence of steam, to a prod-
uct stream containing only methane together with H 2, CO, CO 2 and Installation: CRG process technology covers 40 years of experience with
unreacted steam which is suitable for further processing in a conven- over 150 plants built and operated. Ongoing development of the cata-
tional fired reformer. The CRG prereformer enables capital cost savings lyst has lead to almost 50 such units since 1990.
in primary reforming due to reductions in the radiant box heat load. It
also allows high-activity gas-reforming catalyst to be used. The ability to Catalyst: The CRG catalyst is manufactured under license by Johnson
increase preheat temperatures and transfer radiant duty to the convec- Matthey Catalysts.
tion section of the primary reformer can minimize involuntary steam
production. Licensor: The process and CRG catalyst are licensed by Davy Process
Technology.
Operating conditions: The desulfurization section typically operates be-
tween 170 ° C and 420 ° C and the CRG prereformer will operate over a
wide range of temperatures from 250 ° C to 650 ° C and at pressures up
to 75 bara.
Pressure swing adsorption—rapid
cycle
Application: Hydrogen recovery from fuel gas and hydrogen contain-
ing offgas streams in refinery and chemical processes offers many po-
tential benefits, including process uplift, reduced H2 costs, avoided
H2 plant expansion and emissions reductions. It also requires a cost-
effective separation technology to be economical. Rapid-cycle pres-
sure swing adsorption (RCPSA) technology offers a more-compact,
less-expensive and more-energy-efficient solution for H2 recovery
compared to conventional PSA technology. This technology has been
jointly developed by ExxonMobil Research and Engineering Co. (EMRE)
and QuestAir Technologies. The resulting product—the “QuestAir H- ���������������� ���������������
�����������������
6200”— will have its first large-scale commercial application in 2007 �������������������������������������������
in an ExxonMobil Refinery.
Installation: Three units are currently operating. Several units are under
construction, and many units are under design.
ing macrocrystalline (paraffinic) and microcrystalline wax (from residual ���������� �������
oil). Oil contents typically range from 5–25 wt%. ����
����������
Installations: One SRU tail-gas system and two FCCU scrubbing systems.
Reference: Confuorto, Weaver and Pedersen, “LABSORB regenerative
Sour gas treatment
���������
Application: The WSA process (Wet gas Sulfuric Acid) treats all types of ������
��������������
sulfur-containing gases such as amine and Rectisol regenerator offgas,
���
SWS gas and Claus plant tail gas in refineries, gas treatment plants, ���������
petrochemicals and coke chemicals plants. The WSA process can also be ������� ������
���
applied for SOx removal and regeneration of spent sulfuric acid.
Sulfur, in any form, is efficiently recovered as concentrated commer- �����������
���
cial-quality sulfuric acid. ���������
Installation: Ten units have been licensed using all or part of the Revivoil
Technology.
Economics: D’GAASS achieves 10 ppmw combined H2S/H2Sx in product Reference: US Patent 5,632,967.
sulfur without using catalyst. Elevated pressure results in the following Nasato, E. and T. A. Allison, “Sulfur degasification—The D’GAASS
benefits: low capital investment, very small footprint, low operating cost process,” Laurance Reid Gas Conditioning Conference, Norman, Okla-
and low air requirement. Operation is simple, requiring minimal opera- homa, March, 1998.
tor and maintenance time. No chemicals, catalysts, etc., are required. Fenderson, S., “Continued development of the D’GAASS sulfur de-
gasification process,” Brimstone Sulfur Recovery Symposium, Canmore,
Installations: Twenty-one D’GAASS units in operation. Twenty-six ad-
Alberta, May 2001.
ditional trains in engineering and construction phase with total capacity
over 25,000 long ton per day (LTPD). Licensor: Goar, Allison & Associates, Inc.
Sulfur recovery
Application: The COPE Oxygen Enrichment Process allows existing Claus ������������
��� �����������
sulfur recovery/tail gas cleanup units to increase capacity and recov- ������������� ��� ���
ery, can provide redundant sulfur processing capacity, and can improve ���� ���
������ ������ ���
combustion performance of units processing lean acid gas. � �� � ��
�����
Description: The sulfur processing capacity of typical Claus sulfur re- ��� �� �������
���
��
covery units can be increased to more than 200% of the base capacity ���������� ��� ��
��� ��
through partial to complete replacement of combustion air with pure ��� ���
oxygen (O2). SRU capacity is typically limited by hydraulic pressure drop. ��� ���
���
As O2 replaces combustion air, the quantity of inert nitrogen is reduced ���
allowing additional acid gas to be processed. � �� �������������������
�� � ��
The process can be implemented in two stages. As the O2 enrich-
��
ment level increases, the combustion temperature (1) increases. COPE ��
Phase I, which does not use a recycle stream, can often achieve 50% ��� �� ��� ��
capacity increase through O2 enrichment to the maximum reaction fur-
nace refractory temperature limit of 2,700ºF – 2,800ºF. Higher O2 enrich-
ment levels are possible with COPE Phase II which uses an internal pro-
cess recycle stream to moderate the combustion temperature allowing
enrichment up to 100% O2. Operating costs are a function of O2 cost, reduced incinerator fuel, and
Flow through the remainder of the SRU (2, 3, and 4) and the tail reduced operating and maintenance labor costs.
gas cleanup unit is greatly reduced. Ammonia and hydrocarbon acid
gas impurity destruction and thermal stage conversion are improved at Installations: Twenty-nine COPE trains at 17 locations.
the higher O2 enriched combustion temperatures. Overall SRU sulfur Reference: US Patents 4,552,747 and 6,508,998.
recovery is typically increased by 0.5% to 1%. A single proprietary COPE Sala, L., W. P. Ferrell and P. Morris, “The COPE process—Increase
burner handles acid gas, recycle gas, air and O2. sulfur recovery capacity to meet changing needs,” European Fuels Week
Operating conditions: Combustion pressure is from 6 psig to 12 psig; Conference, Giardini Naxos, Taormina, Italy, April 2000.
combustion temperature is up to 2,800ºF. Oxygen concentration is from Nasato, E. and T. A. Allison, “COPE Ejector—Proven technology,”
21% to 100%. Sulphur 2002, Vienna, Austria, October 2002.
Licensor: Goar, Allison & Associates, Inc., and Air Products and Chemi-
Economics: Expanded SRU and tail gas unit retrofit sulfur processing
cals, Inc.
capacity at capital cost of 15% – 25% of new plant cost. New plant sav-
ings of up to 25%, and redundant capacity at 15% of base capital cost.
Thermal gasoil
Application: The Shell Thermal Gasoil process is a combined residue and
waxy distillate conversion unit. The process is an attractive low-cost con- ���
version option for hydroskimming refineries in gasoil-driven markets or �������
for complex refineries with constrained waxy distillate conversion capac-
ity. The typical feedstock is atmospheric residue, which eliminates the � �����
������
need for an upstream vacuum flasher. This process features Shell Soaker
Visbreaking technology for residue conversion and an integrated recycle
heater system for the conversion of waxy distillate.
����
� � ����������
Description: The preheated atmospheric (or vacuum) residue is charged
to the visbreaker heater (1) and from there to the soaker (2). The con- ����� �
�
version takes place in both the heater and soaker and is controlled by
the operating temperature and pressure. The soaker effluent is routed ������ � �������������
to a cyclone (3). The cyclone overheads are charged to an atmospheric ���������������
fractionator (4) to produce the desired products including a light waxy
distillate. The cyclone and fractionator bottoms are routed to a vacuum
flasher (6), where waxy distillate is recovered. The combined waxy distil-
lates are fully converted in the distillate heater (5) at elevated pressure. Utilities, typical consumption/production for a 25,000-bpd unit,
Yields: Depend on feed type and product specifications. dependent on configuration and a site’s marginal economic values for
steam and fuel:
Feed atmospheric residue Middle East
Fuel as fuel oil equivalent, bpd 675
Viscosity, cSt @ 100°C 31
Power, MW 1.7
Products, % wt. Net steam production (18 bar), tpd 370
Gas 6.4
Gasoline, ECP 165°C 12.9 Installation: To date, 12 Shell Thermal Gasoil units have been built. Post
Gasoil, ECP 350°C 38.6 startup services and technical services for existing units are available
Residue, ECP 520°C+ 42.1 from Shell Global Solutions..
Economics: The typical investment for a 25,000-bpd unit will be about Licensor: Shell Global Solutions International B.V., and ABB Lummus
$2,400 to $3,000/bbl installed, excluding treating facilities. (Basis: West- Global B.V.
ern Euope, 2004.)
Treating—jet fuel/kerosine
Application: NAPFINING / MERICAT II / AQUAFINING systems eliminate ���
�������������������
naphthenic acids and mercaptans from kerosine to meet acid number ���������
and mercaptan jet fuel specifications. Caustic, air and catalyst are used ��������
along with FIBER-FILM Contactor technology and an upflow catalyst im- �����
pregnated carbon bed saturated with caustic.
Description: In a MERICAT system, the caustic phase flows along the fi- �������������
bers of the FIBER-FILM Contactor as it preferentially wets the fibers. Prior
to entering the FIBER-FILM Contactor the gasoline phase mixes with air
through a proprietary air sparger. The gasoline then flows through the
caustic-wetted fibers in the Contactor where the mercaptans are ex- ��������
�������������
tracted and converted to disulfides in the caustic phase. The disulfides
are immediately absorbed back into the gasoline phase. The two phases
disengage and the caustic is recycled back to the FIBER-FILM Contactor
until spent. �������������
Competitive advantages:
• Minimal caustic and catalyst consumption
• Operating simplicity
• Minimal capital investment
• Recausticizing of the carbon bed without interruption of treating.
The FIBER-FILM section keeps organic acids from entering the carbon
bed. This conserves caustic and avoids fouling of the bed with sodium
naphthenate soaps. Competitive downflow reactors need frequent car-
bon bed hot water washings to remove these soaps whereas MERICAT
II does not require hot water washes.
The MERICAT II onstream factor is 100% while competitive systems
requiring periodic cleaning have unpredictable onstream factors.
Contactor.
content.
���������������������
Competitive advantages: �����
����������� �����
• Minimizes spent caustic disposal cost �����������
• Reduces the phenol content of spent caustic and increases the ��������
�������
phenol content of sweetened gasoline thus adding value
• Operates over a wide pH range ��������
�������
• Simple to operate
• No corrosion problems due to the buffering effect of CO2.
Continued
Treating—pressure swing adsorption, continued
Installation: Since commercialization in 1966, UOP has provided over
700 PSA systems in more than 60 countries in the refining, petrochemi-
cal, polymer, steel and power-generation industries. The Polybed PSA
system has demonstrated exceptional economic value in many appli-
cations, such as hydrogen recovery from refinery off-gases, monomer
recovery monomers in polyolefin plants, hydrogen extraction from gas-
ification syngas, helium purification for industrial gas use, adjustment
of synthesis gas for ammonia production, methane purification for pet-
rochemicals production, and H2/CO ratio adjustment for syngas used in
the manufacture of oxo-alcohols. Feed conditions typically range from
(7–70 kg/cm2g) (100 to 1,000 psig), with concentrations of the desired
component from 30 – 98+ mol %. System capacities range from less than
1 to more than 350 MMscfd (less than 1,100 to more than 390,000
Nm3/h).
Product: Vacuum distillates of precisely defined viscosities and flash points ���������
���
(for lube production) and low metals content (for FCC and hydrocracker ������������
units) as well as vacuum residue with specified softening point, penetra- ��� ����������
tion and flash point. ����������
ing units for wax treatment. Standard sizes are 500, 1,000, 2,000 and �
3,000-bpsd feedrate.
�� � �
The core of the unit is standardized; however, individual modules are
modified as needed for specific client needs. This unit will be fabricated
to industry standards in a shop environment and delivered to the plant �����������
site as an essentially complete unit. Cost and schedule reductions of at
least 20% over conventional stick-built units are expected. The standard
licensor’s process guarantees and contractor’s performance guarantees
(hydraulic and mechanical) come with the modules.
Economics:
Investment (Basis 2,000-bpsd feedrate capacity,
2006 US Gulf Coast), $/bpsd 7,300
Wet gas scrubbing (WGS) ���
condensed SO 2 stream.
Description: The flue gas enters the spray tower through the quench
section where it is immediately quenched to saturation temperature.
It proceeds to the absorber section for particulate and SO2 reduction. Economics: The EDV wet scrubbing system has been extremely suc-
The spray tower is an open tower with multiple levels of BELCO-G- cessful in the incineration and refining industries due to the very high
Nozzles. These nonplugging and abrasion-resistant nozzles remove scrubbing capabilities, very reliable operation and reasonable price.
particulates by impacting on the water/reagent curtains. At the same Installation: More than 200 applications worldwide on various pro-
time, these curtains also reduce SO 2 and SO 3 emissions. The BELCO- cesses including 43 FCCU applications, 5 heater applications, 1 SRU
G-Nozzles are designed not to produce mist; thus a conventional mist tailgas unit and 1 fluidized coker application to date.
eliminator is not required.
Upon leaving the absorber section, the saturated gases are direct- Reference: Confuorto and Weaver, “Flue gas scrubbing of FCCU re-
ed to the EDV filtering modules to remove the fine particulates and ad- generator flue gas—performance, reliability, and flexibility—a case his-
ditional SO3. The filtering module is designed to cause condensation of tory,” Hydrocarbon Engineering, 1999.
the saturated gas onto the fine particles and onto the acid mist, thus Eagleson and Dharia, “Controlling FCCU emissions,” 11th Refin-
allowing it to be collected by the BELCO-F-Nozzle located at the top. ing Technology Meeting, HPCL, Hyderabad, 2000.
To ensure droplet-free stack, the flue gas enters a droplet separa-
tor. This is an open design that contains fixed-spin vanes that induce a Licensor: Belco Technologies Corp.
cyclonic flow of the gas. As the gases spiral down the droplet separa-
tor, the centrifugal forces drive any free droplets to the wall, separat-
ing them from the gas stream.
White oil and wax hydrotreating
��������������� �����
Application: Process to produce white oils and waxes.
�������
Feeds: Nonrefined as well as solvent- or hydrogen-refined naphthenic or ������� ��� �������
�����
paraffinic vacuum distillates or deoiled waxes. �� �������� ��
������� �������
Products: Technical- and medical-grade white oils and waxes for plasti-
cizer, textile, cosmetic, pharmaceutical and food industries. Products are
in accordance with the US Food and Drug Administration (FDA) regula-
���
tions and the German Pharmacopoeia (DAB 8 and DAB 9) specifications. �������
����������
Description: This catalytic hydrotreating process uses two reactors. Hydro- �����
���������
gen and feed are heated upstream of the first reaction zone (containing a
���� ������������
special presulfided NiMo/alumina catalyst) and are separated downstream
of the reactors into the main product and byproducts (hydrogen sulfide ����������� ������������ ���������������������
������������������� ������������������ �����������������������
and light hydrocarbons). A stripping column permits adjusting product ������������������� ���������������
specifications for technical-grade white oil or feed to the second hydro-
genation stage.
When hydrotreating waxes, however, medical quality is obtained
in the one-stage process. In the second reactor, the feed is passed
over a highly active hydrogenation catalyst to achieve a very low level
of aromatics, especially of polynuclear compounds. This scheme per-
Installation: Four installations use the Uhde (Edeleanu) proprietary tech-
nology, one of which has the largest capacity worldwide.
mits each stage to operate independently and to produce technical- or
medical-grade white oils separately. Yields after the first stage range Licensor: Uhde GmbH.
from 85% to 99% depending on feedstock. Yields from the second
hydrogenation step are nearly 100%. When treating waxes, the yield
is approximately 98%.
Presented By:
James D. Fleshman
Senior Principal Engineer
Foster Wheeler USA
Corporation
Houston, TX
Mario Campi
Process Manager
Foster Wheeler
Milan, Italy
The design of a hydrogen supply system involves much more than specifying a plant to
produce a particular amount of hydrogen. In order to remain competitive in today’s
economy, it is necessary to optimize how hydrogen is produced, recovered and
compressed, particularly for a complex refinery.
This article addresses the methods for meeting these challenges and focuses on an
innovative, cost-effective solution developed by Foster Wheeler for the NIS Pancevo
Refinery in Serbia.
AM-06-27
Page 1
Hydrogen Management
The most economic solution to a hydrogen supply problem is to make best use of
existing resources. Adding a new unit to supply the entire shortfall without
optimization is rarely the low cost solution.
The final evaluation also factors in capital and operating cost, as well as
operability and reliability of supply.
AM-06-27
Page 2
Hydrogen Balance
Revamp Opportunities
There are a number of options for upgrading existing hydrogen production units.
Depending on the age of the unit, new technologies are available in virtually all
areas of the plant. These include improved alloys for reformer catalyst tubes,
improved catalysts, and more efficient CO2 removal solutions or Pressure Swing
Adsorption (PSA) adsorbent.
AM-06-27
Page 3
exhausted, and a new hydrogen production unit identified, the new facilities need
to be integrated into the existing system.
Because of economics, there has been an increased focus on the best way to
integrate available hydrogen-rich refinery offgas streams into hydrogen
generation capacity. Aside from the actual hydrogen balance, it is necessary to
consider purification, compression, and the impact on the refinery fuel balance.
• Combine refinery offgas (ROG) with SMR gas upstream of a common PSA.
• Combine the ROG with feedstock upstream of the SMR.
• Send the ROG to a dedicated hydrocarbon based PSA unit where tail gas is
recompressed for either fuel export or utilized as reformer feed.
The first configuration, use of a common PSA unit, often appears the simplest
and can at least theoretically offer a cost-effective solution. There are, however,
a number of adsorbent and fuel balance problems with this configuration, which
limit its application.
The impurities present in SMR gas – CO2, CO and N2 – require a different set of
PSA adsorbents than are required by the hydrocarbons present in ROG. A PSA
unit with mixed feedstock requires adsorbents to remove both sets of impurities.
In that case there is a risk that the heavier hydrocarbons from the refinery gas
will reach the adsorbents for SMR gas and be permanently adsorbed, damaging
AM-06-27
Page 4
the bed and reducing its capacity. This makes the combined system very
sensitive to the flow and to the composition of the ROG, and puts the system at
risk of permanent damage if the flow or composition of the ROG varies.
This results in the following constraints on the combined PSA system, which
limits it usefulness:
- The ratio of ROG to SMR gas needs to be kept constant for optimum bed
usage without risk of bed contamination.
- If the SMR is shutdown, the PSA operation on ROG alone may result in
adsorbent contamination by heavy hydrocarbons moving too high in the bed.
- In many cases the offgas stream often contains less H2 and significantly
higher concentrations of heavy hydrocarbons than originally predicted. This
is difficult to foresee and can lead to irreversible adsorbent de-activation if it
goes undetected and adequate measures are not taken to adjust the PSA
cycle time.
- Because of the heavier hydrocarbons, the combined PSA tail gas has a much
higher calorific value, which is often in excess of the steam reformer fuel
requirements. A tail gas compressor is then required to route the excess gas
into the fuel gas network. Moreover, the excess tail gas contains CO2 from
the SMR, which can upset the refinery fuel system.
If the amount of ROG is small, it may still be feasible to combine SMR and refinery gas
in a single PSA unit. However, as the quantity of hydrogen contained in the ROG
increases, it becomes more practical to either use the ROG as SMR feedstock, which
eliminates the tail gas problem, or to install a separate PSA to purify the ROG streams.
Where separate PSA units are used, tail gas from the SMR PSA is returned to the
SMR as a dedicated fuel stream, while tail gas from the ROG PSA unit can be
compressed and used in the refinery fuel system.
AM-06-27
Page 5
Case History – NIS Pancevo
Foster Wheeler recently had the opportunity to review various integration options at
the NIS Pancevo refinery in Serbia.
NIS-Petrol has two refineries in Serbia: one in Pancevo and the other in Novi Sad,
with a total installed capacity of 5 million metric tons/annum (MMTPA). The
Pancevo refinery is undergoing a development program aimed at:
• Production of gasoline and diesel in accordance with European standards (EU
2010+)
• Meeting domestic and European environmental protection standards
• Increasing conversion
• Generating export products
• Energy optimization
• Increasing and securing refinery profitability
The hydrogen demand for the new units was 64,000 Nm3/h of high purity hydrogen.
Two hydrogen-rich offgas streams were available, one from the CCR unit and a
purge stream from the new MHC/DHT unit. Table 1 shows the composition of the
CCR and MHC/DHT offgas streams. The hydrogen recovered from these two
streams was to be supplemented with ”on purpose” hydrogen produced in a new
SMR. Natural gas was available as primary feed to the SMR with LPG available as
supplemental feed.
AM-06-27
Page 6
Table 1
Hydrogen-Rich ROG Stream Properties
AM-06-27
Page 7
The hydrogen available in the two offgas streams was significant - approximately
45% of the total hydrogen demand for the new facilities. Based on this, the following
possible configurations were evaluated for integration of the ROG purification with
the SMR:
In this case the offgas from the CCR and MHC / DHT units is purified in a dedicated
ROG PSA unit, and the tail gas is compressed to the refinery fuel gas header
pressure of 6 bar-g. Part of this tail gas is used as supplemental fuel in the SMR,
with the remainder exported to the refinery.
Natural gas from battery limits is compressed and used as feed in the SMR. A
separate PSA unit is used to purify the SMR gas, and tail gas from the SMR PSA at
0.3 bar-g is used directly as fuel for the reformer, supplemented by tail gas from the
ROG PSA unit.
In each case a small amount of hydrogen is recycled from the PSA outlet to the
SMR feed purification section for the hydrogenation and removal of organic sulfur.
The flows shown here neglect this recycle.
AM-06-27
Page 8
CCR MHC / DHT
22,493 4650
16
22,800 15
ROG 64,166
Nm3/h of 100 % H2
PSA
0.3
Pressure, barg
6 41,366
4,343
17 35 45,963
NATURAL
SMR SMR
GAS
PSA
4,597
FUEL
FG to
Header
2,271 2,072
AM-06-27
Page 9
Case B – ROG Tail Gas to SMR Feed
As in the first case, the refinery offgas is purified in a dedicated PSA unit. However
in this case the tail gas is compressed and sent to the SMR to be used as feed.
Since hydrogen in the tail gas is recovered as SMR feed instead of being burned as
fuel, it is not as critical to maximize the hydrogen recovery in the ROG PSA unit. A
higher tail gas pressure can then be used to reduce the requirement for tail gas
compression: 1.0 bar-g vs. the 0.3 bar-g used in Case A. This increased pressure is
particularly valuable in this case, since the tail gas must be compressed to the
natural gas pressure of 17 bar-g. Since the PSA unit performance is tied inversely
to the tail gas pressure, the increased pressure reduces the hydrogen recovery in
the PSA unit from 84 to 81.5%. Natural gas from battery limits is then mixed with the
tail gas from the offgas PSA, compressed, and sent to the SMR as feedstock. As in
Case A, the SMR product is then purified in a dedicated PSA, with the tail gas used
as SMR fuel. In this case the supplemental fuel for the SMR is natural gas.
AM-06-27
Page 10
CCR MHC / DHT
22,493 4,650
16
64,166
22,122 15
ROG
Nm3/h of 100 % H2
PSA
1.0
Pressure, barg
42,044
5,021
17 35 46,716
NATURAL SMR
GAS SMR
PSA
0.3 4,672
FUEL
NATURAL
GAS
AM-06-27
Page 11
Case C – ROG Directly to SMR Feed
In this case, the ROG PSA unit is eliminated and offgas from the CCR and
MHC/DHT unit is sent directly to the SMR feed compressor together with the
natural gas. The tail gas from the SMR PSA at 0.3 bar-g is used directly as fuel for
the reformer, with natural gas as the supplemental fuel.
Nm3/h of 100 % H2
Pressure, barg 15
64,166
16
17 35 71,295
NATURAL SMR
GAS SMR PSA
7,129
FUEL
NATURAL
GAS
AM-06-27
Page 12
Evaluation of the Different Cases
The assessment of the three cases, A, B and C, took into account expected
availability as well as capital and operating costs.
The direct utilization of the ROG as SMR feedstock in Case C eliminates one set of
compressors and a PSA unit. This provides lower complexity and fewer processing
steps, resulting in improved operability. Complexity of the other two cases is similar.
From an availability viewpoint, Cases A and B are preferable since high purity
hydrogen can be generated when either the ROG PSA or the SMR system are
unavailable.
The non-availability of ROG has approximately the same impact in all three cases,
particularly since adequate margin was built in to the design of the SMR to partially
compensate for the loss of ROG.
A loss of the ROG PSA tail gas compressor impacts only Cases A and B.
AM-06-27
Page 13
TABLE 2
Operating Issues
Issue A B C
Complexity More More Less
H2 Availability High High Normal
No availability of ROG 34% loss of H2 35% loss of H2 38% loss of H2
Loss of ROG PSA tail
Loss of fuel 10% loss of of H2 No impact
gas compressor
Loss of feed gas
66% loss of H2 65% loss of of H2 Total loss of H2
compressor or SMR
Table 3 shows the differences in the capital cost of the three cases using Case A as
the base case. The costs are referred to the first quarter 2005, Western European
basis.
Table 3
Capital Cost Differences
$MM
Case A B C
Feed Compressors --- +0.1 +0.1
Steam Reformer --- --- +8
SMR PSA --- --- +1
ROG --- -0.6 -2
Tail Gas Compressors --- +0.2 -1
Total Cost Difference --- -0.3 +6
AM-06-27
Page 14
Operating Costs
The main process parameters - the resulting feed and fuel consumption, export
steam flows and related operating costs - are reported in the following tables.
TABLE 4
MAIN PROCESS PARAMETERS
A B C
H2 Recovered from ROG Nm3/hr 22,800 22,122 -
H2 Produced from SMR Nm3/hr 41,366 42,044 64,166
Total Hydrogen Production Nm3/hr 64,166 64,166 64,166
TABLE 5
FEEDSTOCK AND UTILITY CONSUMPTION
A B C
Natural Gas Feed Gcal/hr 131.2 69.0 90.8
Natural Gas Fuel Gcal/hr - 27.3 13.4
ROG Feed, Gcal/Hr Gcal/hr 115.6 115.6 115.6
ROG Tail Gas Export Gcal/hr 27.9 - -
HP Steam Export MT/hr 34.7 34.6 40.0
AM-06-27
Page 15
TABLE 6
FEEDSTOCK AND UTILITY COSTS - $MM/YR
A B C
Total Feed Plus Fuel
(Natural Gas plus ROG, 28.58 24.90 26.33
less ROG Tail Gas Export)
HP Steam Export (6.24) (6.23) (7.20)
Other
(Electricity, Demin water, 3.32 3.51 3.88
cooling water, LP steam)
In Summary
• Case A and B show lower investment cost
• Case B and C show lower operating cost
• Case B shows the lowest combination of investment and operating cost
In addition to developing the system configuration, the project also included design
of the hydrogen plant to meet the objectives of NIS Pancevo, including:
• Flexibility to use natural gas or LPG feedstock
• Efficient and reliable hydrogen supply
• Variable cost as low as possible without incurring excessively high
capital expenditure
• Ability to produce export steam cost effectively
NIS Pancevo’s main objectives were met by the installation of a Foster Wheeler
Terrace Wall® Steam Reformer with microalloy catalyst tubes and state-of-the-art
catalyst.
AM-06-27
Page 16
Because of the requirement for feedstock flexibility, a pre-reformer was integrated
with the Foster Wheeler Terrace Wall® primary reformer. This allowed extended
operation on heavier feedstocks, as well as improved energy efficiency. Since
heavier hydrocarbons are not present in the effluent from the pre-reformer, this
stream can be preheated to a higher temperature before it enters the reformer
radiant section. This reduces the load on the reformer radiant section and reduces
fuel requirements.
The other main features of the Terrace Wall ® furnace which insured that NIS-
Pancevo’s targets would be met are as follows:
1. The Terrace Wall ® design uses a single row of catalyst tubes located inside each
radiant cell. Specially selected burners are arranged in terraces at two levels along
a firing wall to create a nearly uniform heat flux to the double-fired tubes. Slightly
sloping walls assist in providing uniform heat distribution since they slope towards
the tubes as the gas cools. Since the burners are at two levels, flux adjustments can
be made along the tube length as the catalyst ages.
2. The entire Terrace Wall reformer, including radiant and convection sections, can be
brought to the site in modules. This reduces field labor as well as construction cost
and duration. These modules can be shipped over the road to virtually any location.
3. Since the selected reformer at Pancevo used natural draft, no fans were required
and reliability was improved. The Terrace Wall furnace also allows the ability to
install combustion air preheat, and switch to natural draft operation in case of fan
failure.
4. Burners are arranged in two firing levels, which allow optimization of the heater
efficiency and the hydrogen production in accordance with the type of feed, catalyst
AM-06-27
Page 17
life and unit load. There is virtually no limitation in turndown and the steam reformer
can maintain a high efficiency at reduced load.
5. The isolation and replacement of radiant tubes are simplified, since the inlet pigtails
are fully accessible while the outlet pigtails are easily accessible through removable
panels on the outlet manifold box. Each single tube can be removed from the top
without any additional work except the removal of the roof rain cover. In case of
tube failure the tube can be crimped and isolated with the heater in operation.
6. Loading or unloading of the catalyst requires only the removal of the top flange.
Loading of a complete charge of catalyst can be completed in two to three days.
In order to supply the maximum amount of steam to the refinery, combustion air preheat
was not selected and heat recovery was by maximizing steam generation. The Terrace
Wall® design, with the convection section placed on the top of the radiant section,
allowed natural draft operation. This eliminated the induced and forced draft fans,
which further improved the expected availability and reduced operating costs.
AM-06-27
Page 18
This paper has been reproduced for the author or authors as a courtesy by the National
Petrochemical & Refiners Association. Publication of this paper does not signify that the
contents necessarily reflect the opinions of the NPRA, its officers, directors, members, or staff.
Requests for authorization to quote or use the contents should be addressed directly to the
author(s)
Reprinted from:
June 2006 issue, pgs 57 –62
Used with permission.
SPECIALREPORT
www.HydrocarbonProcessing.com
Article copyright © 2006 by Gulf Publishing Company. All rights reserved.
T
he MOL Plc Danube refinery in Szazhalombatta,
Hungary, faced several operations challenges: 1)
declining residual fuel-oil markets and 2) a need to
increase clean transportation distillates output. The refinery
considered a residue upgrading project that would increase
conversion of crude oil to distillates, eliminate residual
fuel oil yields and improve refinery margins. After study-
ing alternative paths, MOL selected delayed coking as the
primary conversion process.
In 1997, the MOL board resolved to implement a resi-
due upgrading project based on delayed coking. This proj-
ect also included new facilities for a delayed coker, hydro-
gen plant and sulfur recovery unit, as well as, revisions to
the distillate hydrotreater and conversion of a vacuum
gasoil (VGO) hydrotreater to a licensed mild hydrocracker. FIG. 1 The MOL Plc Danube refinery, Szazhalombatta, Hungary, with the
new delayed coker complex.
��������������
�� ��� �� �� ��� ��� ���� ���� ����
��������
���������������������� ���������������������
���������������������
Products obtained from the delayed coker included: ��� ��� ���� ����
• Treated coker fuel gas (FG) ��������
• Propylene (99.5% purity) ���������������������� ���������������������
• Propane
• Treated C4 LPG FIG. 5 Specific gravities of coker products-design vs. actual.
• Light coker naphtha (LCN)
• Heavy coker naphtha (HCN)
HYDROCARBON PROCESSING JUNE 2006
SPECIALREPORT PROCESS AND PLANT OPTIMIZATION
TABLE 1. Design features of the delayed coker TABLE 3. Coke quality—design vs. actual
• Advanced coke drum design for optimum life. Specifications included drum Design feed prediction Actual feed operation
plate chemistry and thickness and fabrication instructions. VCM, wt% 8 –11 10.2
• Side-wall double-fired coker heater fitted for online spalling. This design Ni+V ppmw 1,026 1,195
provides long heater run length by minimizing reaction time in the heater.
Heating value, kJ/kg 35,674 35,800
• Safety interlock systems for coke drum isolation, coke cutting system and
HGI 50– 80 64
heater operation. An enclosed operator shelter was specified for the decoking
operation on the top deck of the coke drum structure. Moisture, wt% 12– 15 9.3
• The heater was supplied with air preheat and low NOx features
• Enclosed blowdown system for recovering unconverted oils and vapors TABLE 4. Refinery products before and after coker
without impacting the environment. Result: No scheduled flaring due to drum project
switching and turnover
• Decoking water systems designed for 100% recovery and recycling of water Before After
coker, % coker, %
• Sludge injection system to process refinery sludges and convert them to
product Gasoline 22.4 24.0
• Gas plant with 90%+ recovery of C3 streams; amine scrubbing of coker Naphtha 6.8 8.4
product gas prior to refinery export and fuel use in the coker heater; amine LPG 4.5 5.3
treating of the C3/C4 LPG stream and a naphtha splitter Middle distillates 37.7 42.7
• Licensed sections from others include a C3/C4 LPG caustic treater and Heavy fuel oil 14.6 1.2
propane-propylene splitter
Coke 3.4
Other 14.0 15.0
TABLE 2. Feed quality variance 100.0 100.0
hydrotreating units. The actual sulfur distribution is considered during the design phase. This project is profitable and allowed
environmentally favorable. Fig. 5 illustrates the differing gravities MOL to economically reach its goals of exiting the uncertain
of the liquid products. Table 3 lists the design estimate for coke heavy fuel-oil market, improving competitiveness and improving
quality with the actual coke. environmental conditions. The success of the coker operation
Coke disposal is 30% to local steel and cement industries and is due to an effective teamwork between refiner and licensor to
70% export to purchasers in Europe. After accounting for vari- define the design, control the execution and start up the unit
ances in operation and feed quality, the design basis for yields and without incident. HP
product quality was confirmed by the operating data.
Impact of coker on operations and profitability. As Miroslav Kovac is a supervisor of the delayed coker unit
in MOL Plc’s Danube Refinery. He has 12 years of experience in
a result of the coker project, the MOL Szazhalombatta refinery petroleum refining in the fields of process engineering and opera-
was able to effectively exit the heavy fuel-oil market. Table 4 lists tion. Previously, he worked as a research fellow at the University of
the percentage shift in refinery products before and after the coker Veszprem. He holds an MS degree in chemical engineering from
project. the University of Veszprem.
Article copyright © 2006 by Gulf Publishing Company. All rights reserved. Printed in U.S.A.
Not to be distributed in electronic or printed form, or posted on a website, without express written permission of copyright holder.
The refining industry has for some time been facing weak and volatile refining margins and asset
expansion is likely to remain limited. Refiners have historically had access to very limited capital
expenditure budgets with investment limited to ‘stay in business’ expenditure, generally driven by the
need to meet increasingly stringent product quality legislation. Certainly, where more strategic
investment has proceeded, companies have been under pressure to achieve shorter payback periods.
Bottom-of-the-barrel processing is increasingly likely to be considered, certainly in USA and Europe, for
a number of reasons:
Global transportation fuels demand is significantly influenced by global events affecting economic
activity and confidence. Hence demand growth has historically been erratic. World consumption of
refined products grew by an annual rate of 1.4% during the 1990s. However, if the Former Soviet Union
and Eastern Europe are excluded, the growth rate is higher at 2.4% per annum.
In this period all light products (transportation fuels and petrochemical intermediates) exhibited
stronger growth, with global demand expected to average approximately 2% per annum. We are seeing
a shifting demand/supply balance, especially for middle distillates. In Europe, for example, the deficit
for road diesel is forecast to be around 45 million tonnes by 2010.
-5
-10
million tonnes per year
-15
-20
-25
-30
-35
-40
-45
2000
2005
2010
Page 1 of 4
Residual markets
For most refiners fuel oil is usually an undesirable byproduct. Unlike light products (such as gasoline
and middle distillates), fuel oil usually commands a price below the cost of crude thereby effectively
depressing overall refinery margins.
Residual fuel oil demand declined by 20% during the 1990s even though bunker fuel use increased over
the same period. At a recent bottom of the barrel conference (ref 1) Nexant/Chem Systems reported
that a global 50% fall in residual fuel markets would necessitate around 200 heavy residue destruction
projects with current crude slates. In Europe, legislation regarding the sulfur content of fuel oil may
reduce the market size by 50%, leaving bunker markets as the main outlet. If the pace of market
decline were to accelerate, many refiners might face ‘stay in business’ investment decisions,
converting residuals by means of capital intensive processes with unexciting rates of return.
Some forecasters, however, believe that residual fuel oil demand has now leveled out and could even
increase by around 0.5% per annum. This is being caused by a combination of US national gas
shortages, the early phasing out of nuclear power in certain countries and growth in demand for
bunkers. However, this growth rate is four times lower than the forecast global growth in light product
demand.
Directionally, world crude supply is expected to become heavier and more sour.
Light crude (34 API or higher) represents around 50% of crude currently processed. Although production
of light sour crude is expected to increase by 9 million b/d by 2015, the production of light sweet
crude is expected to increase by only 1-2 million b/d over the same period.
The pace of decline in light sweet availability will be influenced by production levels in West and North
Africa, the North Sea and elsewhere. The availability of light condensates could also have a major
influence on light feedstock supplies. Refiners have to some degree been able to avoid investment in
upgrading by crude substitution, especially in market conditions in which reduced light/heavy price
differentials have prevailed. However, fuel specifications are becoming so stringent that this solution is
not sufficient to meet the new specifications and residue upgrading is required.
Heavy crude release to the market has been held back by adherence to OPEC quotas (which encourage
sales of higher price light crudes). Heavy sour crude makes up only a small percentage of current
production but is increasing.
Light/heavy crude differentials can be explained largely by two key factors - supply, estimated by the
relative proportions of these grades in the output slate (the supply side), and demand for heavy
products, particularly residual fuel oil (the demand side). These factors pull in different directions,
with spreads widening as the relative output of heavy crude rises and narrowing with stronger fuel oil
demand. There is no reason to suspect that the volatility in differentials will disappear and
directionally, one would expect differentials to widen.
Page 2 of 4
Much of the investment in coking in the USA in the past 10-15 years has been driven by the wish to
process lower value crudes from Latin America. In broad terms about 50% of the potential economic
benefit from a coker comes from using cheaper crude (the other 50% comes from the higher value of
the upgraded products).
The bigger the spread between heavy and light products, the more incentive refiners have to invest in
the necessary hardware to process cheaper feedstocks to gain competitive advantage. The heavy
crude oils produce a higher portion of heavy, highly contaminated materials, which historically have
been disposed of to fuel oil. However, these stocks can be upgraded to more valuable products using
mature technologies such as fluid catalytic cracking (FCC), hydrocracking and delayed coking, thereby
helping to correct finished product imbalances to improve refinery margins. Typically, a delayed coker
will convert 65% of its feed into transportation fuels, the other 35% being coke.
Coke market
Fuel grade coke is an internationally traded product and competes with coal in the market place. Its
higher heating value over coal (typically 20% higher) can encourage its use, with traditional power
stations able to blend around 20% into coal feed without the need for significant modifications. The
high sulfur content of fuel grade petroleum coke usually means that it can only be used on power
plants with flue gas desulfurisation. Coke traders consider that the future market for fuel grade coke
is strong although it is generally regarded as a low value, distress product.
Anode grade coke, used in aluminum smelting, is a robust market and fetches a far higher price than
fuel grade ($300+ per tonne). To a large extent the production of anode grade coke is feedstock driven
requiring the processing of a low sulphur, low metals crude. Overall refinery economics, in general,
dictate whether purchasing this higher quality, more expensive crude, to make anode coke is
justifiable.
The demand for needle coke, used in the steel industry, has been fairly flat over the last 20 years. It is
unlikely any new needle cokers will be built for economic reasons.
Delayed coking is a mature process which remains for many the preferred residue upgrading option
because of its ability to handle the heaviest contaminated crudes. Globally, approximately one third of
installed residue upgrading plant is by delayed coking.
Today, around 50% of the worldwide delayed coking capacity is concentrated in the US with more than
2,000,000 B/D of installed capacity. In the last 15 years, delayed coking capacity has grown by 56% in
the USA, followed by hydrocracking (37%) and FCC (14%).
A recent paper by Foster Wheeler (Ref 2) compared the economics of a number of residue upgrading
schemes:
• Delayed coking/fluid catalytic cracking
• Integrated Gasification Combined Cycle (IGCC)
• Atmospheric residue hydrotreatment and residue catalytic cracking (RFCC)
In summary, if transportation fuels command a higher value to the refiner than power, the refiner is
most likely to opt for a delayed coking or RFCC-based scheme.
Schemes based on FCC/delayed coking, hydrocracking/delayed coking appeared to provide the lowest
capital and operating costs. All three schemes offered a positive internal rate of return.
Page 3 of 4
Coking offers refiners much more flexibility in feedstock and crude selection, allowing the refinery
scope to take advantage of spot markets. Coking offers the refinery the opportunity to move towards
zero fuel oil production whilst meeting future product demand growth from low-value residuals rather
than additional crude processing.
Ref 1) EPC 2nd Bottom of the Barrel Technology Conference, Istanbul, 8-9 October 2002.
Keynote address by Nexant/Chem Systems.
Ref 2) Advances in residue upgrading technologies offer refiners cost-effective options for
zero fuel oil production. Graham Phillips, Fang Liu, Foster Wheeler 2002 European
Refining Technology Conference, Paris.
BY:
Graham Phillips
Technology Manager Refining, Foster Wheeler
Michael Stewart
Senior Planning Consultant, Foster Wheeler
Randy Wolf
Regional Vice President, Foster Wheeler
Page 4 of 4
WHEN SOLVENT DEASPHALTING IS THE MOST
APPROPRIATE TECHNOLOGY FOR UPGRADING RESIDUE
Edward J. Houde - UOP LLC: edward.houde@uop.com
Michael J. McGrath - Foster Wheeler USA: michael_mcgrath@fwhou.fwc.com
The fit for a particular location is particularly ∗∗ the production of road bitumen, and have
dependant on finding a use for the SDA pitch and included both two-product (deasphalted oil
various options are discussed. (DAO) and pitch) and three-product
(deasphalted oil, resin and pitch) process
configurations.
Background
While solvent deasphalting (SDA) has been used
for more than fifty years to upgrade non-volatile This merging of FW and UOP solvent
residues, the technology continues to evolve over deasphalting technologies provides our clients
time. It is a robust economical process that uses with the widest range of experiences, process
an aliphatic solvent to separate the typically more features, engineering know-how, technical
valuable oils and resins from the more aromatic support, and most importantly, a more efficient
and asphaltenic components of its vacuum residue and lower cost design.
feedstock. The earliest commercial applications of
solvent deasphalting used propane as the solvent Technology Overview
to extract high-quality lubricating oil bright stock
from vacuum residue. These applications were Solvent deasphalting, whether for the production
called propane deasphalting (or propane of lubricating oil or cracking stocks, uses a light
deresining when used to separate high molecular hydrocarbon solvent specifically tailored to ensure
weight resins from Pennsylvania-grade vacuum the most economical deasphalting design. For
residue). Solvent deasphalting process have example, propane solvent may be specified for a
gradually extended to include the preparation of low deasphalted oil yield operation such as lub
feedstocks for catalytic cracking, hydrocracking, production, while a solvent containing
and hydrodesulfurization units, as well as the hydrocarbons as heavy as hexane may be used to
production of specialty asphalts. obtain a high deasphalted oil yield for the
production of additional conversion unit feedstock.
In 1996, UOP and Foster Wheeler USA Plant designs have been developed using heavy
Corporation (FW) entered into a collaboration solvents at elevated temperatures in order to
agreement to merge their solvent deasphalting maximize the yield of usable deasphalted oil and
technologies. Both companies had extensive minimize the yield of pitch having a softening point
backgrounds in solvent deasphalting, and both of 350°F or higher.
companies had recently entered into collaboration
covering other residue upgrading technologies UOP/FW SDA technology is unique by different
such as visbreaking and delayed coking. A from other solvent deasphalting technologies in
collaboration covering solvent deasphalting was that it is not just one technology, but rather a
considered a logical fit that would benefit the combination of technology features and options
IDTC Conference
London, England Page 1 of 11
February 2006
WHEN SOLVENT DEASPHALTING IS THE MOST
APPROPRIATE TECHNOLOGY FOR UPGRADING RESIDUE
Edward J. Houde - UOP LLC
Michael J. McGrath - Foster Wheeler USA Corporation
which have been developed by both companies. by incorporating both stripping and
These features and options allow UOP/FW to rectification of the oil feed. Superior
offer the optimum process design for any solvent quality DAO is obtained from the RDC
deasphalting application. even at DAO recovery rates exceeding 85
percent, with an even greater difference in
For example, UOP has predominately focused its quality being seen at lower DAO yields.
technology on downstream conversion unit
applications. As a result, UOP’s solvent ∗∗ Supercritical solvent recovery. Supercritical
deasphalting experience has principally focused recovery of the solvent allows more efficient
on the use of butane and heavier type solvents utilization of the system’s thermodynamic
that can obtain higher DAO recoveries. There is a characteristics while also reducing the unit’s
distinct advantage to the use of supercritical operating costs.
separation for the recovery of the solvent and
DAO when using these types of solvents. ** Multiple product recovery designs that take
Consequently, UOP developed supercritical advantage of the changes in liquid-liquid
solvent-DAO separation technology. The other equilibrium that result from changes in
area that UOP focused its development efforts operating conditions between those utilized
involved minimizing the solvent to oil ratio while during extraction and those used for solvent
still producing a reasonably high quality DAO. recovery.
FW’s SDA technology development emphasis was ∗∗ Lower solvent requirements used to achieve
initially more focused on lower lift, very high quality processing objectives. Although increasing
applications, such as the production of lube oil the amount of solvent used in the extraction
feedstocks for hydrocracking and further solvent improves the extraction efficiency, it also has
refining. Consequently its technology originally a major impact on the unit’s operating costs.
focused around propane/butane deasphalting Consequently, the lowest solvent-to-oil ratio
using optimized extraction techniques for those necessary to achieve the desired product
specific applications. Additionally FW has made separation is typically specified.
available its detailed design and construction
experiences from a multitude of SDA projects. ∗∗ Optimal design of heat exchange systems.
UOP and FW’s combined design experience
Technology Advantages in optimizing SDA heat exchange systems
allows the SDA user to select a multitude of
The UOP/FW SDA technology has several distinct heat exchange options, depending on the
advantages that ensure the refiner obtains the project specific objectives and opportunities.
most efficient, economical, and flexible SDA
process. These include: Extraction Devices
∗∗ Novel extraction devices tailored to the The efficiency of the extraction process is the key
specific application: equipment design variable impacting both the
capital and operating costs of SDA. The
- UOP/FW/Sulzer’s Structured Packing and extractor’s role in SDA is to separate the
proprietary internals in both the multi- precipitate (pitch phase) from the continuous fluid
stage counter current extractor as well as stream (DAO/solvent).
the DAO and resin separators. This
technology provides state-of-the-art Both single-stage co-current extraction, where the
contacting and separation devices to bulk of the solvent is mixed with the feed prior to
maximize extraction efficiency as well as the extractor, as well as multi-stage counter-
optimal recovery of clean products. current extraction, where the bulk of the solvent
enters the bottom of the extractor separate from
- FW’s multi-stage rotating disk contactor the feed, have been used commercially.
(RDC). The RDC is specifically designed
to achieve high product yields and quality Structured packing or RDCs used in multi-stage
IDTC Conference
London, England Page 2 of 11
February 2006
WHEN SOLVENT DEASPHALTING IS THE MOST
APPROPRIATE TECHNOLOGY FOR UPGRADING RESIDUE
Edward J. Houde - UOP LLC
Michael J. McGrath - Foster Wheeler USA Corporation
counter-current extraction provide both the The impact of solvent-to-oil ratio on both capital
contacting area and time required for extraction as and operating costs is conservatively reflected in
well as segregation of the stages to reduce back Tables 2 and 3. These tables summarize an
mixing below the feed stage. Above the feed internal study undertaken by UOP and FW to
stage, structured packing or coils provide for the assess the relative costs of different solvent-to-oil
coalescing of entrained droplets of feed or pitch. ratios. The cost of fuel has significantly increased
since this study was performed and the savings at
Supercritical Solvent Recovery lower solvent ratios would be even more
pronounced. While the cost benefits at a lower
Although, often referred to as supercritical solvent ratio can be substantial, the solvent-to-oil
extraction, it is the solvent separation, not the ratio also has an impact on DAO quality (see
extraction that is carried out in the supercritical Figure 1). Consequently, the optimal solvent ratio
region of the solvent. The use of supercritical is determined based on the DAO’s downstream
solvent recovery results in a simpler process flow. processing requirements.
Gone is the need for multiple flash towers and
condensers associated with conventional multiple Figure 1: Effect of Solvent to Oil Ratio
effect evaporative type solvent recovery systems. 70
3/1 5/1
8/1
PPM Metals in DAO
60
Supercritical solvent recovery allows for more 50
efficient utilization of the system’s thermodynamic 40
characteristics. Presented below is a comparative 30
tabulation of the utility requirements of 20
conventional subcritical evaporative and 10
supercritical solvent recovery systems. Note that 0
the difference from subcritical to supercritical 20 25 30 35 40 45 50 55 60 65 70
IDTC Conference
London, England Page 3 of 11
February 2006
WHEN SOLVENT DEASPHALTING IS THE MOST
APPROPRIATE TECHNOLOGY FOR UPGRADING RESIDUE
Edward J. Houde - UOP LLC
Michael J. McGrath - Foster Wheeler USA Corporation
Table 2
UOP/FW SDA Advantage
Capital Costs
Option: 1 2 Delta
Case: High Solvent to Low Solvent to _
Oil Ratio Oil Ratio
Solvent-to-oil Ratio (S/O) 8 5 3
Solvent C4 Mix C4 Mix -
Size (BPD) 28,000 28,000 -
% Equip. Assoc. with S/O 60 60 -
Equip. Assoc. with S/O $ 20,520,000 $ 15,700,000 $ 4,820,000
Other Equipment Cost $ 15,680,000 $ 15,680,000 $ -
Installed Cost, $MM $ 36,200,000 $ 31,380,000 $ 4,820,000
Unit Cost, $/BBL 1,290 1,120 170
Table 3
UOP/FW SDA Advantage
Operating Costs
Option: 1 2 Delta
Case: High Solvent to Low Solvent to -
Oil Ratio Oil Ratio
Solvent-to-oil Ratio 8 5 3
Solvent C4 Mix C4 Mix -
Average Utility Consumptions
(per barrel of feed)
Fuel, MMBTU 0.075 0.056 0.019
Steam, lbs 12.0 10.5 1.5
Power, kW 2.67 1.77 0.90
Incremental Cost
($Bbl Feed)
Fuel 0.293 0.206 0.087
Steam 0.00006 0.00005 0.000
Power 0.179 0.119 0.060
Total 0.342 0.248 0.094
Yearly Cost (28,000 BPD) $3,600,000 $2,500,000 $1,100,000
Unit Cost Reference
Fuel, MMBTU $3.82
3
Steam, 10 lbs $4.83
Power, $/kW $0.067
IDTC Conference
London, England Page 4 of 11
February 2006
WHEN SOLVENT DEASPHALTING IS THE MOST
APPROPRIATE TECHNOLOGY FOR UPGRADING RESIDUE
Edward J. Houde - UOP LLC
Michael J. McGrath - Foster Wheeler USA Corporation
Process Description
The recovered solvent streams from the DAO and
In non-lube oil production applications, regardless pitch recovery sections are cooled/condensed and
of whether a two-product or three-product recycled back to the extraction section for reuse.
configuration is employed, the unit’s design would
Figure 2: Supercritical SDA Process
employ structured packing, supercritical solvent
recovery and the flexibility to utilize a range of
solvent types to achieve the desired separation.
IDTC Conference
London, England Page 5 of 11
February 2006
WHEN SOLVENT DEASPHALTING IS THE MOST
APPROPRIATE TECHNOLOGY FOR UPGRADING RESIDUE
Edward J. Houde - UOP LLC
Michael J. McGrath - Foster Wheeler USA Corporation
IDTC Conference
London, England Page 6 of 11
February 2006
WHEN SOLVENT DEASPHALTING IS THE MOST
APPROPRIATE TECHNOLOGY FOR UPGRADING RESIDUE
Edward J. Houde - UOP LLC
Michael J. McGrath - Foster Wheeler USA Corporation
difficult to maintain a stable operation at these extracts the more paraffinic components from
conditions, the extractor temperature is typically vacuum residues while rejecting the condensed
maintained below the solvent’s critical ring aromatics. The deasphalted oil product
temperature. quality is characterized by:
The quantity of solvent contained in the ∗∗ Reduced metal content - DAO with as little
solvent/residue mixture that is charged to the as 1 wppm nickel plus vanadium content
extractor vessel has an impact on extraction has been produced even from Venezuelan
efficiency. As shown in Figure 1, increasing the residues containing 700 to 1,000 wppm of
amount of solvent in the extractor while these metals.
maintaining a constant DAO yield improves the
degree of separation of the individual components ∗∗ Reduced carbon residue - The carbon
and results in the recovery of a higher-quality residue in deasphalted oils is significantly
DAO. lower than for distilled oils of equivalent
viscosity or mid-boiling point.
However, since the quantity of solvent which is
recirculated within the unit is significantly greater The deasphalted oil product yield-quality relation-
than the amount of feedstock being processed, ships obtained when solvent deasphalting typical
any improvement in product quality which results vacuum residues are shown in Figure 3. This data
from an increased solvent recirculation rate must is based on UOP and Foster Wheeler’s extensive
be balanced against the additional operating costs library of pilot plant and commercial solvent
associated with the increased solvent recirculation deasphalting data.
and solvent recovery requirements, and the
increased capital costs associated with the larger Figure 3: SDA Quality Selectivity
equipment sizes. Once the required solvent-to-oil
ratio is established, however, it is usually not
adjusted unless the feed rate is increased and the 100
solvent circulation becomes the limitation on unit
Appearing in Deasphalted Oil, %
90
Sulfur, Nitrogen and Metals
TABLE 6
ESTIMATED UTILITY REQUIREMENTS
IDTC Conference
London, England Page 8 of 11
February 2006
WHEN SOLVENT DEASPHALTING IS THE MOST
APPROPRIATE TECHNOLOGY FOR UPGRADING RESIDUE
Edward J. Houde - UOP LLC
Michael J. McGrath - Foster Wheeler USA Corporation
It is analogous to vacuum distillation in that it only A recent project that has started up in the Far East
provides separation of products not conversion of is a good example of the applicability of solvent
products. If you change to a lower quality feed it deasphalting in an existing refinery. In this case,
will result in a lower quality of DAO or a lower yield the refiner desired to recover additional cracking
or both. It is not analogous in that it separates by stock from a residual fuel stream. The solvent
molecular types rather than volatility. The result in deasphalting unit was designed to process 6,000
general is that higher boiling components are BPSD (approximately 40,000 kg/hr) of a Middle
recovered in the unit’s pitch product and the lower Eastern crude blend to recover a 50% yield of
boiling components are recovered in the unit’s DAO. UOP/FW/Sulzer’s latest state-of-the-art,
DAO product. Due to solvent selectivity, however, proprietary internals and structured packing were
the pitch will contain low-boiling, highly aromatic used in both the unit’s extractor and DAO
components while the DAO will contain high- separator.
boiling, paraffinic components. For a fuels type
solvent deasphalter, the pitch production will be The addition of the solvent deasphalting unit
minimized for a specific cracking stock (DAO plus allowed the refiner to increase the amount of
VGO) quality by maximizing the lift in the vacuum transportation fuels produced from the refinery.
unit to the limit of VGO quality. This also The recovered pitch, along with some clarified
minimizes the size of the solvent deasphalter. slurry oil, (≤ 20 Liq Vol % of the blend) was used
as a high viscosity residual fuel by an existing
Solvent deasphalting is less expensive to build nearby IPP to produce power. In order to
and operate than other residue conversion compensate for the higher viscosity of the IPP’s
processes, although as noted earlier, it does not fuel blend, for this application, the fuel system and
provide any actual conversion. Therefore, it is burner temperatures were higher than from a tyical
most applicable to recovering the large quantity of solvent deasphalting operation.
high quality oils in light residues while rejecting the
small quantity of asphaltenes and impurities such
UOP/FW Technical Service
as metals and those components that contribute to
carbon residue. In addition, unlike residue
conversion units which benefit from economy of In addition to the benefits provided by UOP and
scale, solvent deasphalting can be economically FW’s broad commercial experience bases,
applied at very small feed rates. licensees have full access to the widest range of
Finally, solvent deasphalting has good applicability support resources available to the refining
where the demand for a low-value residual fuel is industry. This allows UOP/FW to offer licensees
significantly smaller than the production of residual high-quality support from initial conception,
fuel oil based on vacuum tower bottoms through unit design and construction, and
production. continuing on through unit start-up and monitoring
of the operating unit’s performance. These
Utilization of Pitch services include:
∗∗ Pilot plant testing to establish design basis
The utilization of SDA pitch is very much
∗∗ Technical services during unit checkout and
dependant on the local market. The primary outlet
start-up
for the pitch is as fuel, mainly as a blend
component in the residual oil market. Another ∗∗ Ongoing operations monitoring
significant market is as a blend component in the ∗∗ Engineer and operator training programs
IDTC Conference
London, England Page 9 of 11
February 2006
WHEN SOLVENT DEASPHALTING IS THE MOST
APPROPRIATE TECHNOLOGY FOR UPGRADING RESIDUE
Edward J. Houde - UOP LLC
Michael J. McGrath - Foster Wheeler USA Corporation
IDTC Conference
London, England Page 10 of 11
February 2006
WHEN SOLVENT DEASPHALTING IS THE MOST
APPROPRIATE TECHNOLOGY FOR UPGRADING RESIDUE
Edward J. Houde - UOP LLC
Michael J. McGrath - Foster Wheeler USA Corporation
TABLE 7
SOLVENT DEASPHALTING
PILOT PLANT vs. COMMERCIAL RDC OPERATION
Commercial
Pilot Plant Operation
Feed Inspections
Pitch Inspections
Penetration @ 77ºF 8 10
IDTC Conference
London, England Page 11 of 11
February 2006
Annual Meeting
March 19-21, 2006
Grand America Hotel
Salt Lake City, UT
Presented By:
Keith Couch
Manager, FCC Process
Development
UOP LLC
Des Plaines, IL
Leonard Bell
Process Sales Specialist
UOP LLC
Des Plaines, IL
John Yarborough
Mechanical Specialist
UOP LLC
Des Plaines, IL
INTRODUCTION
Refiners are more focused today than ever on improving utility consumption and reducing stack
emissions. One area receiving significant interest is power recovery from the FCC flue gas,
especially since this power is “clean” in that no additional CO2 is produced or emitted.
While much work has been done over the past 40 years to improve the reliability and operability
of FCC flue gas power recovery systems, the process has remained largely unchanged; that is,
until now. Traditionally, the FCC flue gas power recovery system has all too often been treated
as an “accessory”, tacked on only to higher capacity, higher pressure FCC units in areas of high
electrical cost. In order to make this technology useful for a wider range of FCC operators, UOP
has developed some innovative improvements to the way power recovery systems are
incorporated into the FCC unit. These innovations significantly reduce the capital cost per unit
of energy recovered from FCC unit flue gas in an environmentally friendly manner. These
innovations can potentially double the ROI for a power recovery system when compared to
traditional installations. This has greatly increased the application range of power recovery
systems to FCC capacities for which it was previously considered uneconomical.
AM-06-10
© 2006 UOP LLC. All rights reserved.
Page 1
In this paper UOP will discuss the history of FCC flue gas power recovery, show the economics
associated with implementation of a traditional power recovery system, discuss some of the
recent advancements in process design and the impact they can have on power recovery
economics.
While there are a few exceptions, typical FCC regenerator designs include two stages of
cyclones inside the regenerator. It was originally believed that this efficiency was high enough to
protect the expander from excessive erosion. This proved not to be the case.
In 1963, Shell Oil solved the flue gas fines problem by placing an additional catalyst separator
outside the regenerator at Norco, Louisiana and Oakmont, Canada refineries. This additional
stage of catalyst separation became known as a Third Stage Separator (TSS), and FCC flue gas
power recovery became a sustainable reality.
The regenerator cyclones substantially reduce both the catalyst loading and the particle size
distribution in the flue gas. This allows the TSS cyclone elements to be designed at much higher
velocities, thus higher efficiency, without a concern for erosion of the TSS. Depending on
regenerator design and catalyst systems, a modern TSS is capable of reducing the catalyst in the
flue gas to less than 1 wt-% catalyst larger than 10 microns. This is more than sufficient to
provide long-term reliable protection of the power recovery expander.
There are several possible Power Recovery System configurations that can be incorporated into
both new and existing FCC Unit designs. Selection of the specific equipment type and
arrangement is always refinery specific, and is generally a balance of utility requirements,
process optimization, and preference between different configurations.
Amongst the various configurations, there are basically two types of power recovery applications
to consider: 1) a four or five body power recovery train in which the expander is directly coupled
to the main air blower to provide direct shaft power, and 2) a two, three or four body power
AM-06-10
Page 2
recovery train in which the expander is coupled to a generator to produce electrical power. These
configurations will be discussed in turn.
Between 1963 and 1981, 18 power recovery applications were commissioned industry-wide.
These were typically five-body trains; consisting of a hot gas expander, main air blower, steam
turbine, motor/generator and gear box as necessary. A five-body train is shown in Figure 1. This
configuration was historically the most common power recovery system for new unit
installations.
FIGURE 1
Traditional 5-Body Power Recovery Train
In this configuration, the expander is coupled to the main air blower and provides a direct
transfer of energy to the shaft. The direct transfer of energy to the main air blower minimizes
power transfer losses, and is the most energy efficient configuration.
The steam turbine is used to start up the train. The shaft speed is increased to match the electrical
frequency of the motor (the “synchronous speed”) to that of the power grid and the electrical
breaker is closed. Once the breaker is closed, the shaft speed of the train is fixed to the frequency
of the power grid. The air flow rate is controlled by adjusting the inlet guide vanes. Generally,
the combination of a steam turbine and motor can provide the required power to operate the air
blower at design conditions, with the expander out of service. After enough flue gas is present,
and process conditions are stable, the expander can be commissioned. An overview schematic of
the traditional power recovery train is shown in Figure 2.
AM-06-10
Page 3
FIGURE 2
Traditional Power Recovery System – “Five-Body Train”
The motor/generator imports or exports power as required to maintain a constant train shaft
speed. If more energy is supplied by the expander than is required by the blower there is a
surplus of electricity generated and exported to the power grid. Conversely, if the blower power
requirement is higher than the expander is providing, electricity is consumed by the motor to
maintain the train at normal speed. With the expander coupled to the main air blower, in the
event of a breaker disconnect the shaft power requirement of the MAB acts as over-speed
protection for the expander.
With a power recovery system, butterfly valves are used in the flue gas line to control the
differential pressure between the reactor and regenerator. A dedicated, high speed “power
recovery control system” performs all process control functions. Fundamentally, the regenerator-
reactor pressure differential controller (PDIC) adjusts the expander inlet valve to control the
differential pressure, and only opens the bypass valve when the expander is out of service or
expander maximum throughput is reached. This control scheme maximizes the power generation
potential by minimizing the amount of flue gas that is diverted around the expander.
AM-06-10
Page 4
With a traditional five-body power recovery train (PRT) if the expander requires any repair or
maintenance, the entire FCC Unit has to be shut down. The business interruption costs associated
with shutting down the entire FCC unit can be substantial, and rapidly negate the economic
advantage associated with the power recovery system. With these concerns in mind, there was a
desire by many refiners to decouple the power recovery expander from the main air blower shaft.
This was initially a very challenging problem that was solved in the early 1980s, and ushered in
the Gen Set power recovery system.
The gen set PRT is a “stand-alone” system in which the expander is connected to a generator and
the main air blower is installed as a separate machine. An overview schematic of a modern Gen
Set power recovery system is shown in Figure 3. By removing the main air blower from the
power recovery train, the shaft load associated with the blower is eliminated. The main concern
with this configuration was that in the event of a breaker-disconnect the shaft load drops
essentially to zero and the expander could speed up in an uncontrolled manner, resulting in a
potential over-speed. These concerns led to the development of fast acting control valves, high
speed electro-hydraulic actuators, and improved instrumentation and control systems that rapidly
divert flue gas out of the expander to decelerate the train. In 1983, one of the first gen set power
recovery systems was commissioned at Saras, S.p.A., in Sardinia, Italy.
AM-06-10
Page 5
FIGURE 3
Gen Set Power Recovery System
Split
HSS Range Flue
> Gas
Orifice
Chamber
PIC PDIC
Butterfly
Third
Valves
Stage
Separator Flue Gas Electrostatic
Combustor Cooler Precipitator
Style
Regenerator Critical
Flow
Nozzle
Gear
Box Generator
To / From
Reactor
Expander
Gear Motor
Box
Main Air
Blower
With a modern Gen Set power recovery application, FCC unit down time associated with either
maintenance or failure of the power recovery expander is essentially eliminated. Operation of the
FCC unit can be maintained while the PRT is isolated to complete any repair or routine
maintenance work.
The decision to couple the power recovery expander to the MAB shaft or install a Gen Set PRT
is a question that each refiner must address. With the current cost of FCC unit downtime, more
refiners are seeing the potential economic advantage of the Gen Set power recovery system.
AM-06-10
Page 6
FIGURE 4
Historical Power Costs & PRT Installations
12.00 10
Total New PRT Instillations
Coal Price, x10 $/short ton
9
Natural Gas Price, Wellhead ($ / k Cubic Feet)
10.00 Average Retail Price of Electricity, Industrial (Cents/KWH)
8
8.00
6
6.00 5
4
4.00
3
2
2.00
1
0.00 0
1973 1978 1983 1988 1993 1998 2003
Although natural gas prices have been rising sharply in recent years, since about 1993 coal has
dramatically outpaced all other fuels for electrical power generation. Both the availability and
low price of coal has maintained electrical rates at historic lows since 1999, but are only recently
starting to increase as shown in Figure 4.
In contrast with the current low electrical rates, UOP has observed a recent resurgence in the
industry interest in FCC flue gas power recovery systems. This interest appears to be primarily
driven by refiners focusing on direct returns by lowering their operating cost, and indirect returns
on investor confidence by improving the Energy Intensity Index (EII) of their operations and
reducing environmental emissions.
AM-06-10
Page 7
Environmental Emissions
The application of an FCC flue gas power recovery system is “green” with respect to electrical
power generation. No additional CO2, SOx or NOx are created in association with the power
generated. This can provide both permitting and economic benefits to the refiner. The economic
benefit is going to be discussed later in this paper.
The impact on the refiner operating with a lower EII and good neighbor emissions stewardship
improves investor confidence that the refinery is being properly managed. As a result, many
refiners will accept lower returns on capital projects focused on energy optimization. However,
the projects still have to be economically attractive. With the recent low electrical rates this has
proved particularly challenging for traditional FCC flue gas power recovery and has focused
UOP on inventing new ways to improve the economic viability for power recovery systems and
improving the reliability of the system.
With renewed interest in FCC flue gas power recovery, UOP reviewed many of the historical
designs and realized that very little had changed since the early 1970s. Opportunities arising
from recent advancements in component equipment designs (expander, actuators, instruments,
and TSS) had not been utilized.
As emission requirements have tightened, particulate matter in the FCC flue gas has been
controlled by the installation of an electrostatic precipitator, wet gas scrubber or barrier filter.
Each of these emission control devices requires considerable capital and operating expense, but
has no financial Return on Investment (ROI) beyond supporting a permit to operate.
The remainder of this paper will discuss some of UOP’s recent FCC flue gas power recovery
innovations. With each innovation a supportive economic case study will be shown that
progresses one advancement to the next, starting with the traditional maximum electrical power
generation configuration. Although the economic analyses are cited as case studies, they are
closely based on recent commercial projects.
AM-06-10
Page 8
Flue Gas Basis
To start this series of case studies, a relatively common flue gas basis was chosen which meets a
“middle of the road” FCC operation. The following conditions were used.
Utility Costs
The utility costs used for the case studies were based on 2005 U.S. Gulf Coast costs.
These costs are represented as total erected costs, and as such, exclude all project costs (licensor
basic design, royalties, and T&K fees, spare parts, start-up services, training, and owners costs),
as well as project contingencies, as these costs are unit-specific and carry a wide range of
variability.
Nevertheless, every attempt has been made to develop each of these case specific erected costs
on a consistent design and installation basis to ensure comparative accuracy and effectively
support the resultant ROI representations.
AM-06-10
Page 9
that must be made up in the boiler house. A debit has been applied to the economics reflecting a
pound-for-pound shift in HP steam production from the FCC unit to the boiler house.
The economic analysis for the installation of a traditional power recovery system is shown in
Table 1. The ROI for this project is presented for 25, 30, and 35 percent discounted cash flow
factors, and shows a very marginal return between 9.1 – 10.6%. This is an example of why it has
been difficult in recent years to economically justify the installation of a power recovery system
on moderate to small sized units.
TABLE 1
Utility Analysis and ROI - Traditional FCC Power Recovery System
The economics of major capital projects often improve with larger size units due to economies of
scale. This is true for the application of FCC flue gas power recovery systems. Figures 5 & 6
show a comparison of erected costs and ROIs for various capacity units, pegged to the conditions
for the 50,000 BPSD case study of this paper and a scale-up to a 125,000 BPSD unit. The two
lines in each Figure show the trend lines for a traditional power recovery and the new UOP
power recovery system. As the size of the unit is decreased, a minimum erected cost level is
reached, which can result in falling ROI levels with smaller units. Conversely, on larger units,
the incremental increase in erected costs leads to higher ROIs.
The remainder of this paper is going to discuss some of the aspects of the new UOP power
recovery system, and how we have been able to reduce the total erected cost and improve energy
integrations for a step change improvement in ROI over base.
AM-06-10
Page 10
FIGURE 5
Relative Erected Cost Impact with Varying Size FCC Units
Base
Relative Erected Cost
Base
-45%
-30%
FIGURE 6
Relative ROI Impact with Varying Size FCC Units
AM-06-10
Page 11
CASE STUDY #2: (TRADITIONAL PRT WITH TSS INTEGRATED EXPANDER
BYPASS LINE)
Advancements in TSS design provided the first power recovery system optimization opportunity
by reducing the capital costs of the TSS, and associated flue gas duct and structure. The new
UOP TSS is about 40 percent smaller than other TSS offerings for the same capacity, making it
less expensive to fabricate, easier to install, and better suited where plot space is a premium.
Figure 7 shows an equal capacity comparison of the older radial flow TSS and the new UOP TSS
design.
The most significant improvement in the design is that the UOP TSS utilizes axial flow for
catalyst/gas separation. The flue gas flow is maintained essentially in one direction - in the top
and out the bottom of the unit. Axial flow distribution minimizes the potential for solids re-
entrainment resulting from changes in directional gas flow and resultant eddy current formations.
More importantly, reducing the directional flow changes minimizes pressure loss across the TSS
by 0.25 to 0.50 psi. This translates directly into more power recovery potential across the
expander.
In addition to being smaller, the clean gas outlet is located on the lower side of the vessel. This
minimizes the amount of steel structure and hot wall flue gas duct (typically 304H stainless steel)
between the TSS and the expander inlet.
FIGURE 7
Third Stage Separator – Equal Capacity Comparison
AM-06-10
Page 12
Proper design of the expander inlet / outlet lines is extremely critical to the overall operation and
reliability of the power recovery expander. The allowable nozzle loadings on the expander inlet
and outlet nozzle are extremely small. Great care must be taken in the detailed engineering process
to ensure that the nozzle loadings are maintained within allowable parameters across the entire
operating range as well as transient conditions of the system.
In older power recovery system designs, the expander bypass line was typically designed as hot-
wall pipe connected directly to the expander inlet line “minimum distance” from the outlet of the
TSS. See Figure 8. The objective of this configuration was to minimize the loading effects that the
bypass line imparts on the expander inlet nozzle during intermittent or transient use. The bypass
line can be in service during all modes of operation; startup, shutdown and normal operation. As
the bypass valve opens, the flow of hot flue gas causes the flue gas duct to heat up and thermally
expand. The resultant duct expansion imposes a great deal of force loading and rotational moment
on the expander inlet line, making the piping system both difficult to design and costly to install.
FIGURE 8
Traditional Expander Bypass Line Installation
TSS Underflow
AM-06-10
Page 13
To alleviate this problem, UOP has added a second clean gas outlet nozzle to the TSS and attached
the expander bypass line as a lower cost cold-wall line, directly to the TSS vessel. See Figure 9.
FIGURE 9
UOP Modified Expander Bypass Line Installation
TSS Underflow
Where: TSS = Third Stage Separator, E = Expander, G = Gear, M/G = Motor Generator
This configuration has many significant benefits that increase the overall reliability, operability,
and cost effectiveness of the system. In this design, the TSS acts as a fixed anchor point in the duct
design for both the expander inlet and bypass lines.
The line from the TSS to the inlet of the expander becomes a very clean, minimum impact design,
allowing for shorter duct length between the expander and the TSS. The transient loads applied to
the expander inlet nozzle associated with intermittent or normal use of the bypass line are
minimized, essentially reduced to only the axial pressure thrust with varying gas flows. The bypass
line also becomes much shorter in length and can be of cold wall design. Both of these provide for
lower overall design and installation costs as well as operational and maintenance benefits.
The combined implementation of an axial flow UOP TSS with a side connected clean gas outlet
and integral cold-wall bypass line reduces the detailed piping design requirements, structural
steelwork, large diameter piping, pipe supports, and expansion joints for the system. This lowers
the total erected cost for this case study from $28.9 MM to $27.1 MM, with a corresponding
improvement in ROI as shown in Table 2.
AM-06-10
Page 14
TABLE 2
Utility Analysis and ROI – UOP TSS and TSS Anchored Bypass Line
Although the economic benefits seen here with 1) the addition of a PRT and 2) the installation of a
UOP TSS are relatively small, it leads into a progression of additional technology improvements
and costs reductions that greatly improve the economics of flue gas power recovery.
The integration of a steam turbine as a supplemental driver to assist with startup of the PRT is
relatively common. In traditional operation, the flue gas flow is maximized to the expander and the
steam flow to the turbine is reduced to the minimum required to keep the turbine in operation. At
this point, the steam turbine becomes a marginally utilized asset until the next FCC shutdown.
Integration of the steam turbine provides a means for the refiner to supplement electrical power
generation from the PRT. However, in designs of the past, the generator has been sized according to
the rated capacity of the expander. As such, using the steam turbine to supplement electrical power
generation has been limited to times when the FCC unit is in turndown operations and not making
the design power output. In considering ways to maximize the use of existing assets, the PRT steam
turbine has been a significantly under-utilized piece of equipment.
Most refineries operate a boiler house that generates a single level of high pressure steam. Lower
levels of steam are supplied by successive letdown stations to the medium- and low-pressure steam
headers. Additional LP steam is also often generated by exhausting steam turbines into the LP
header. A typical steam letdown configuration is shown in Figure 10.
AM-06-10
Page 15
FIGURE 10
Typical Steam Letdown Configuration
When steam is let down across a control valve, the potential energy of the steam is reduced without
work being done. With the addition of a steam turbine in the FCC power recovery train as shown in
Figure 11, there is a way to capture a significant amount of this lost energy.
FIGURE 11
UOP Steam Letdown Configuration
HP Steam Header
(600 or 450# typ)
Boiler
House
FCC
Power Recovery Train
Gear
G
E T
PIC
MP Steam Header
(150# typ)
LP Steam Header
Refinery PIC
(50# typ)
Turbine
Exhaust
AM-06-10
Page 16
In this configuration, HP or MP steam can be let down efficiently across the turbine into the lower
pressure steam headers. The energy transferred to the generator shaft is used to produce
supplemental electrical power. To optimize this configuration, the PRT motor/generator needs to be
sized to accommodate the proper steam let-down requirements to meet the refinery’s needs. As
depicted, multiple levels of steam letdown can be accommodated through a single turbine. This
utility integration is a process design that extends beyond the battery limits of the FCC unit, and
allows the refiner to optimize the economics of operating their facility-wide steam and electrical
systems.
When installing an FCC flue gas power recovery system, most of the auxiliary equipment is already
required; i.e. the generator, 13.8 kVa cable, switches gear, foundation, electrical controls and
substation. The incremental cost of adding the steam letdown turbine to the power recovery train is
low compared to the potential energy recovered, and can significantly increase the return on
investment for installation of a power recovery system in the FCC unit.
If we focus on the battery limits of the FCC unit, there are additional energy integration
opportunities available than can be used to help maximize use of available assets; in this case, the
PRT steam letdown steam turbine. There are several MP steam flows to the reactor including the
riser lift steam, feed distributor dispersion steam, spent catalyst stripping steam, and reactor
fluidization steam. The steam that is injected into the reactor is heated up to the reactor operating
temperature, and is one of the loads on the overall heat balance. From a process standpoint, we do
not want the additional heat input that the superheat of the MP steam provides. However, LP steam
is simply too low in pressure to be used for these applications.
With the integration of a flue gas power recovery system, the normal FCC process steam can be
routed through a letdown turbine on the PRT as shown in Figure 12. In this manner, the excess
superheat and pressure energy of the steam is transferred as shaft power to the PRT and used to
either supplement the blower power requirement or to produce electricity in the generator.
AM-06-10
Page 17
FIGURE 12
UOP Process Integrated Steam Turbine
Products to
Main Column
HP/MP
Stripping Steam Outlet
Steam
Feed Gear
Distributor
Steam G
E T
Steam
Outlet
PIC
LP Steam
Lift
Steam
PIC
PIC
MP Steam
~100 psig
MP Steam Header Steam Letdown
In application, the higher the steam pressure supplied to the turbine, the greater the economic return
on integrated electrical power generation. The turbine exhaust pressure is a variable with which the
operator can control the amount of superheat remaining in the steam; the lower the exhaust pressure,
the lower the remaining superheat. As more energy is removed, the higher the electrical power
generation from the PRT and the higher the heat load on the reactor. While the temperature of the
steam is a relatively small heat load on the reactor, the lower the steam temperature, the higher the
catalyst-to-oil ratio in the unit for improved product selectivities.
The addition of a steam letdown turbine within the power recovery train increases the total
erected cost of the system from $27.1 MM to $28.4 MM as shown in Table 3. With the steam
conditions used, integrating just the reactor, riser, and stripper steam letdown covers the
additional cost of the turbine and steam integration. If the steam source is changed from MP to
HP steam, the ROI is improved by +1.7 percent.
AM-06-10
Page 18
TABLE 3
Utility Analysis and ROI – Integrated Process Steam Letdown
With the integration of a steam turbine, multiple levels of letdown can be simultaneously
incorporated into the system. To provide a basis for the economic impact that integrated steam
letdown can have, Table 4 shows the additional power generation and economic improvement
per 10,000 pounds of steam letdown for three different levels of steam.
TABLE 4
Utility Analysis and ROI – Per 10,000 lb/hr Steam Letdown
For example, if 30,000 lb/hr of steam is let down from the 600# HP to the 150# MP steam header
for a 25 percent DCF bracket, the subsequent increase in ROI and recovered electrical power
would be:
ROI = 13.0 + (3 * 0.5) MW = 15.6 + (3* 0.39)
ROI = 14.5% and, MW = 16.77
AM-06-10
Page 19
CASE STUDY #4: TEMPERATURE CONTROLLED EXPANDER INLET
With the traditional approach to power recovery, electrical power generation is maximized by
directing the highest temperature, highest pressure flue gas to the inlet of the expander. The energy
recovered across the expander results in a flue gas temperature reduction of 150 - 400°F and a
minimum exhaust pressure. The flue gas is then routed to a low pressure flue gas cooler for residual
energy recovery in the form of steam production. However, due to the temperature reduction, the
quantity of the steam generation is lower than before the expander was placed into service. Even
with the most efficient cooler designs, installing a power recovery expander upstream of a flue gas
cooler to maximize electrical power generation can result in a 20-30 percent reduction in high
pressure steam production that must be financially off-set with the value of electrical power
generation for the installation of a power recovery system to be economically attractive.
Considering the dynamic balance between steam and electricity costs, UOP evaluated several
options to improve the economics for installing a power recovery system. In a traditionally applied
power recovery system, the operating temperature of the regenerator dictates the inlet
temperature to the expander. The higher flue gas temperature at the expander inlet duct requires
the use of expensive stainless steel duct.
In the new temperature-controlled expander inlet design shown in Figure 13, the expander is
placed downstream of a high pressure flue gas cooler, reducing the metallurgy requirement of the
entire power recovery system to lower cost carbon steel, resulting in a total erected cost that is
potentially 30–40 percent lower than that of a traditional system design, depending on the capacity
of the unit.
This provides the refiner with another means to help optimize the economics of their overall
refinery utility systems between maximum HP steam generation and maximum electrical power
generation at 1050°F maximum expander inlet.
AM-06-10
Page 20
FIGURE 13
UOP Temperature Controlled Power Recovery System
HP Steam BFW
HP
Steam Economizer Stack
Regenerator Gen
Steam Inlet
Gear
Air Inlet TSS
E T G
Blower Motor
In this new system, the flue gas from the regenerator is first routed through a flue gas cooler,
then to the power recovery train. A bypass line is installed around the flue gas cooler to provide a
means to control the inlet temperature to the expander. This allows the refiner to optimize energy
production from the FCC unit between steam and electrical power. The control system is
configured to provide load-following variable peak response control of both the refinery HP steam
system and electrical distribution. If the refinery HP steam requirement drops, or if the refinery is
close to an electrical surcharge threshold, the flue gas cooler bypass line is opened to direct
additional hot flue gas to the expander to produce additional electrical power. This allows the
refiner the capability to optimize the HP steam generation and electrical power generation as
utility economics shift, independent of the operation of the FCC reactor / regenerator.
The lower flue gas temperature downstream of the flue gas cooler allows the entire power recovery
system (vessels, control valves, expansion joints, piping, and duct work) to be designed and
installed with lower cost carbon steel materials as opposed to the higher cost stainless steel and cold
wall refractory lined duct work required by the traditional system. The lower temperature system
design results in less thermal movement of the flue gas duct, reducing the size, type, and quantity of
expansion joints required. This further reduces the erected cost of the system. Even though the
electrical power generated with the lower temperature expander inlet is reduced from that of the
traditional system, as shown in Table 5, the substantially lower cost of the system far exceeds the
reduction in electrical power generation, resulting in a significant step change increase in ROI.
AM-06-10
Page 21
TABLE 5
Utility Analysis and ROI – Temperature Controlled Expander Inlet
In addition to the economics presented in Table 5, the lower inlet temperature to the expander
increases the long term reliability of the system and helps minimize expander blade erosion and
power recovery loss over time. The cooler catalyst particles that pass over the expander blades are
much less apt to fuse into catalyst deposits on the blades and casing, further improving the system
reliability as a function of reducing expander blade tip erosion and tip-rub-induced shaft vibration.
TABLE 6
Value of Emissions Reductions – $ / metric ton
$ / metric ton
Carbon Dioxide (CO2) 3–5
Sulfur Dioxide (SO2) 300 – 600
Nitrogen Oxide (NOx) 3,000 – 10,000
AM-06-10
Page 22
To assess the value of generating “green” power with a power recovery system, the marginal fuel
for electrical production was considered to be natural gas. Every kW-hr produced in the power
recovery system results in a kW-hr lower requirement from a cogeneration plant. With the use of
natural gas, the SOx reduction is near zero, the CO2 reduction is proportional to the fuel
consumption, and the NOx reduction is based on the use of low NOx burners with an emission of
40 ppm NOx in the flue gas. The value of emission reductions used for the economic analysis is
based on the average of the ranges shown in Table 6. With a fuel gas heat rate of 9,090
BTU/kW-hr, the resultant value for emissions reductions is tabulated in Table 7.
TABLE 7
Emissions Reductions– metric ton/MW-hr
The economic impact of the emissions reduction in association with installation of a power
recovery system is noticeable, and as shown in Table 8, further improves the ROI for a power
recovery system.
TABLE 8
Utility Analysis and ROI – Emissions Credit
AM-06-10
Page 23
SUMMARY
Increased global focus on reducing energy consumption and emissions are working together to
make FCC flue gas power recovery more attractive. Even in this environment, the economics
associated with a traditional power recovery system can be marginal for an average sized FCC
Unit. However, UOP has developed a series of novel improvements to the traditional scheme that
make it an attractive investment across a broader range of FCC capacities at the current price of
electricity.
The improvements discussed herein, while novel in application, are all supported by proven
technologies and serve to reduce capital cost or improve efficiency and availability. The ‘TSS
Integrated Bypass Line’ reduces capital cost of the expander inlet line and is made possible by
UOP’s commercially proven new TSS design. The ‘Reactor Riser Steam Letdown Turbine’
utilizes existing turbine technology to improve efficiency. The ‘Temperature Controlled
Expander Inlet’ utilizes existing cooler technology, along with a turbine, to reduce capital cost,
improve efficiency and improve availability.
While an abundant supply of low cost coal has helped keep electricity prices in check, there are
signs that the price of coal is on the rise. An increase in electricity prices to the inflation-
adjusted average of 1973 to 1988, with all else equal, would result in a 70 percent increase in the
ROI for all cases considered. Because UOP believes that the long-term factors that drive energy
efficiency are on the rise and will remain so for the foreseeable future, UOP remains committed
to improving the process behind power recovery from FCC flue gas systems.
The concepts presented in this paper provide a glimpse at some of the recent work UOP has been
performing on FCC flue gas power recovery. New ideas and new opportunities are being
developed that build upon the recent technology advancements. UOP is currently working on a
new power recovery system that further reduces total erected cost and increases the overall
power recovered to further improve ROI. The newer system significantly reduces required plot
space and allows the refiner to potentially meet current and future particulate matter stack
emission requirements.
AM-06-10
Page 24
ACKNOWLEDGEMENTS
The authors of this paper would like to express their thanks to the following companies and
individuals, for their assistance in providing data and/or support that have helped make this paper
a reality.
2. UOP – FCC Engineering Tech Center – Daniel N. Myers –– Assistance with Flue Gas
Cooler steam estimates, and historical perspectives on FCC flue gas power recovery.
3. UOP - FCC Engineering Tech Center - John Yarborough – Assistance with pipe stress
analyses, large bore piping design, proper equipment layout, and 3-D graphics used in the
presentation.
REFERENCES
1. V. J. Memmott, and B. Dodds, “Innovative Technology Meets Processing and Environmental
Goals: Flying J Commissions New MSCC and TSS”, National Petrochemical & Refiners
Association (NPRA) Annual Meeting, Paper AM-03-13, March 2003
2. K. A. Couch, K. D. Seibert, and P. J. Van Opdorp, “Controlling FCC Yields and emissions –
UOP Technology for a Changing Environment”, National Petrochemical & Refiners
Association (NPRA) Annual Meeting, Paper AM-04-45, March 2004
4. Colin High, “Air Emissions Reductions and Value from Green Power”, Ninth National Green
Power Marketing Conference, Albany, New York, Oct 4-6, 2004
5. K. J. Reading, “Expander Controls Past, Present and Future”, Third International Expanders
Users' Council, Houston, Texas, June 1995
AM-06-10
Page 25
6. K. J. Reading, S. Rubino, and T. Kociuba, “The Application of Large Induction Generator to
a Fluid Catalytic Cracking Power Recovery Train”, IEEE / IAS Annual Meeting, Denver, CO
October, 1986
UOP LLC
25 East Algonquin Road
Des Plaines, IL 60017-5017
Page 26
Annual Meeting
March 19-21, 2006
Grand America Hotel
Salt Lake City, UT
Presented By:
Gary Brierley
Technology Manager
UOP LLC
Des Plaines, IL
Visnja Gembicki
Marketing Manager,
Refining
UOP LLC
Des Plaines, IL
Tim Cowan
Senior Development
Engineer
UOP LLC
Des Plaines, IL
INTRODUCTION
Reduced availability of light conventional crudes in the future will create demand for new crude
sources that will necessitate refinery configuration changes. The production of heavy crudes,
synthetic crudes, and bitumen blends is growing, and the supply of bitumen-derived crudes is
expected to reach almost three million barrels per day by the year 20151. A plethora of synthetic
crudes and bitumen blends have become available, all of which pose different challenges for
today’s refiners. Some crudes are both higher in contaminant levels and have a composition that
makes them more difficult to upgrade. Coupled with the demand for increased production of
ultra-clean diesel and gasoline, innovative refinery configuration changes will be needed to
accommodate these new feedstocks. The potential processing schemes under consideration range
from simple hydrotreating for contaminant removal, to hydrocracking and fluid catalytic
cracking for conversion of gas oil to high-quality transportation fuels. It is the integration of
these process technologies, however, that offers the greatest economic potential. This paper
focuses on the processing of heavy and synthetic crude blends using innovative process
integration across several technology platforms to produce clean fuels.
MARKET SITUATION
World oil demand is projected to continue increasing, at a rate of about 1.5% per year, with
increased growth of transportation fuels coupled with a relatively flat heavy oil demand. The
Energy Information Administration recently predicted that the demand for crude oil in the United
States will increase at an average rate of 1.1% through to the year 20302. Most of the 400,000-
barrel annual increase in crude consumption in the United States will be used for transportation
fuels, with gasoline representing about 45% of total petroleum consumption. The demand for
distillate fuels is growing at a faster rate than the demand for gasoline. These changes in demand
AM-06-16
© 2006 UOP LLC. All rights reserved.
Page 1
have been accompanied by well-known ultra-low-sulfur regulations for diesel and gasoline,
forcing a significant capital investment through the end of decade.
The production of light and medium crudes from the Western Canadian sedimentary basin is
dropping by about 20% every five years. Given the increasing demand for oil, and declining
production of conventional crudes in both the United States and Canada, the Canadian oil sands
will become a key source of future crude supply. The volume of oil in place in the various
deposits is estimated at 1.7 trillion barrels, of which 174 billion barrels is recoverable with
existing technology. This places the size of the established reserves second only to Saudi
Arabia. As recovery technologies improve, the size of those recoverable reserves could increase
significantly. The proximity of the source, security of supply, and competitive pricing will drive
the refinery investment needed to accommodate these crudes.
Today, the level of production has reached one million barrels per day, and it is expected to
increase to almost three million barrels per day by 2015. Oil sands crudes are expected to
represent more than 75% of the crude produced in Western Canada. While most of the imports
into the United States are in PADDs II and IV, further pipeline expansions will increase
penetration in PADD II and the Pacific Northwest (Northern PADD V). New pipeline systems
are expected in the future, reaching Texas (PADD III) and a new sea port in British Columbia.
From this new port, marine shipments to both California (Southern PADD V) and the Far East
are expected.
As will be shown later, the composition and contaminant levels of bitumen-derived crudes does
not make them an easy replacement for conventional crudes, especially since most existing
refineries have limited capacity to accept poorer quality feedstocks. These crudes are
fundamentally different, so refiners need to understand them and also be prepared for the
changes needed to process them.
Bitumen-derived Crudes
The term “synthetic crude” has never been strictly defined, but it has come to mean a blend of
naphtha, distillate, and gas oil range materials, with no resid (1050°F+, 565°C+ material).
Canadian synthetic crudes first became available in 1967 when Suncor (then Great Canadian Oil
Sands) started to market a blend produced by hydrotreating the naphtha, distillate, and gas oil
generated in a delayed coking unit. The light, sweet synthetic crude marketed by Suncor today is
called Suncor Oil Sands Blend A (OSA). Syncrude Canada Ltd. started production in 1978,
marketing a fully-hydrotreated blend utilizing fluidized-bed coking technology as the primary
upgrading step. Today, this product is referred to as Syncrude Sweet Blend (SSB). Husky Oil
started up a heavy conventional crude upgrader in 1990 using a combination of ebullated-bed
hydroprocessing and delayed coking technologies. Their sweet synthetic crude is traded as
Husky Sweet Blend (HSB). The Athabasca Oils Sands Project (AOSP) started producing a
sweet synthetic crude in 2003 called Premium Albian Synthetic (PAS) using ebullated-bed
hydroprocessing technology. There are also small volumes of two synthetic crudes produced at
the Consumers’ Co-op refinery called NSA and NSB.
The quality of the kerosene and diesel in these synthetic crude blends has been a major concern
in the past. Bitumen is itself a very aromatic feed, and the choice of primary upgrading
AM-06-16
Page 2
technology has an effect on the final distillate quality. Husky’s HSB and AOSP’s PAS crudes
are produced using an ebullated-bed resid hydrocracking technology, and hence have full-range
diesel cetane numbers above 40. Suncor’s OSA crude is produced using delayed coking, giving
the full-range diesel a cetane number very close to, or just below 40. Syncrude’s SSB crude is
produced using the more-severe fluidized-bed coking technology, and the diesel has a cetane
number of about 33. The kerosene cut of SSB has a smoke point of just 13 mm. Starting in
June of 2006, Syncrude will start to produce Syncrude Sweet Premium (SSP), where the
distillate has been further upgraded to give a full-range diesel cetane number of 40, and a
kerosene smoke point of 19 mm.
Table 1 compares the basic composition and quality of Syncrude SSB against Brent crude3. The
SSB crude has lower sulfur and no resid (as shipped), it contains significantly less naphtha-range
material, and more distillate and VGO. Note that due to pipeline contamination, synthetic crudes
like SSB can have some resid component when actually received in the refinery.
SSB SSB
Brent Produced Received*
Gravity, API 38.6 31.8 33.5
Sulphur, wt% 0.29 0.1 0.2
RVP, psi 8.2 4.6 N/A
Yield, vol%
C4- 2.9 3.4 N/A
C5- 350°F 28.6 15 21
350 - 650°F 29.6 44 40.5
650 - 350°F 29.8 37.6 31.5
1050°F plus 9.1 0 3
Properties
Kerosene smoke pt, mm 25 13 ---
350 - 650°F
Cetane 50 30 ---
Pour -5 -55 ---
While sweet synthetic blends make up the majority of the synthetic crudes on the market, some
sour synthetic blends are also available. Suncor markets a range of sour synthetic blends, each
tailored to meet specific refinery processing capabilities. Suncor OSE crude is a blend of
hydrotreated coker naphtha with non-hydrotreated coker distillate and coker heavy gas oil.
Suncor’s OSV crude is a blend of hydrotreated coker naphtha with straight-run distillate and
straight-run VGO. There are several other sour blends available (OSH, OCC, etc), each with its
own processing characteristics. While these sour crude blends still contain no resid fraction, they
are generally sold to medium and heavy sour crude refineries. Note that AOSP markets a heavy
sour blend called Albian Heavy Synthetic (AHS) which is a blend of their sweet PAS crude with
AM-06-16
Page 3
unconverted oil from their ebullated-bed resid hydrocracking unit. Since this crude does have a
resid component, it is not really a synthetic crude as it has been defined here, but neither is it a
bitumen blend, like those described below.
Together, the various synthetic crudes make up the majority of the bitumen-derived crudes on
the market today. However, over 400 KB/D of bitumen are produced and shipped to market
without having been upgraded. The bitumen must be diluted with a lighter hydrocarbon stream
to meet the specifications required for shipping in pipelines. DilBits are blends of bitumen and
condensate, typically natural gas condensate. They normally contain 25 – 30 lv% condensate
and 70 – 75 lv% bitumen. The most common streams are Cold Lake Blend (CLB), Bow River
(BRH), and various Lloyd blends (LLB, LLK, WCB). Since the majority of condensate is C5 to
C12 material, and the majority of bitumen is C30+ boiling range material, these crudes have
become known as “dumbbell crudes.” There is a lot of material boiling at each end of the
boiling point curve, but little in the middle.
Natural gas condensate is in short supply in Northern Alberta where the bitumen is produced.
Condensate sells at a significant premium to light sweet crudes for this reason, and some
condensate is actually being shipped by rail back to Alberta from the United States. To address
the shortage of diluent, and the problem with dumbbell crudes, producers have started to market
SynBits, blends of sweet synthetic crude (typically OSA) and bitumen. SynBits have a more
continuous boiling point curve than DilBits, with a significant portion of distillate-range material
in the blend. However, since the synthetic crude diluent has a much lower API gravity than
condensate, more diluent is needed, so SynBits are typically 50 lv% synthetic crude and 50 lv%
bitumen. The most common SynBits on the market today are Christina Lake Blend (CSB) and
MacKay Heavy (MKH), both of which are blends of bitumen produced by Steam Assisted
Gravity Drainage (SAGD) and OSA crudes.
SynDilBits are actually blends of condensate, hydrotreated synthetic crude, and bitumen. They
typically contain about 65 lv% bitumen, with the remaining volume split between the two diluent
streams. The most common of these streams are Wabasca Heavy (WH) and Western Canadian
Select (WCS).
Light sweet synthetic crudes, heavy sour synthetic crudes, DilBits, SynBits, and SynDilBits all
target different refineries. Figure 14 on the next page summarizes how bitumen is upgraded, or
blended with other streams to make various types of crude blends. Light sweet synthetic crudes
are usually sold to light crude refineries, while heavy sour synthetic crudes and DilBits are
normally processed in heavy crude refineries. SynBits and SynDilBits are usually sold to
medium crude refineries, or blended with additional synthetic crude for processing in a light
crude refinery.
AM-06-16
Page 4
Figure 1 – Disposition of Bitumen-derived Crudes
Refining
Condensate
DilBit Heavy
Crude
Heavy Refineries
SCO SynDilBit
Bitumen SynBit Medium
Production Crude
Refineries
Light
(Mining or In-Situ) SCO SCO/SynBit Blend
Light
Upgrading Crude
Light Refineries
Synthetic
Crude Oil
Source: Purvin and Gertz, 2005
(SCO)
WCS is a somewhat unique SynDilBit in that it has a proprietary formula developed by EnCana,
Talisman, Canadian Natural Resources Limited (CNRL), and Petro-Canada. They wanted to
reduce the large number of heavy crudes being marketed from Western Canada, and achieve
consistency in the heavy crude blends being shipped from Canada. Each batch contains
specified amounts of the following crudes; LLW, LLC, CLB, CSB, MKH, and BR6. As such,
each batch contains condensate, hydrotreated synthetic crude (OSA), heavy conventional crude,
medium conventional crude, Cold Lake bitumen, and Athabasca bitumen. Each batch is blended
to meet the following specifications; API gravity of 19 - 22°, carbon residue of 7 – 9 wt%, sulfur
of 2.8 – 3.2 wt%, and a total acid number (TAN) of 0.7 – 1.0 mg KOH/g.
WCS may well become the new marker heavy crude from Western Canada, and efforts are being
made to have it traded on the New York Commodities Exchange. Production of WCS started in
January of 2005, with shipments currently at 250 KB/D, but expected to increase to more than
500 KB/D by 2008.
Each synthetic crude and bitumen blend has its own unique processing characteristics. The
compositions of three conventional light crudes, two heavy conventional crudes, two bitumen
blends, and one synthetic crude are compared and contrasted in Figure 2. Compared to the
marker West Texas Intermediate (WTI) crude, a typical synthetic crude has no resid, 50% more
VGO, 50% more distillate, and only half the naphtha. DilBits and WCS have about three times
the volume of resid material than WTI, 50% more VGO, but only half the distillate range
material, and half the naphtha.
AM-06-16
Page 5
Figure 2 – Crude Composition
100%
80%
vol-%
60%
40%
20%
0%
WTI Arab ANS Lloyd Bow DilBit WCS SCO
Light Blend River
Resid Vacuum Gas Oil Distillate Naphtha/LPG
PROCESSING IMPLICATIONS
The processing implications to the refinery will be a function of the type of bitumen-derived
crude imported. First and foremost, refiners used to processing WTI will have a three-fold
increase in the amount of resid coming to the refinery if they replace WTI with WCS. Beyond
the implications to the crude and vacuum unit, extra resid conversion capacity will certainly be
required. Refineries already seem to be anticipating this need. There is about 200 KB/D of
incremental delayed coking capacity in various stages of planning in PADD II, PADD IV, and
northern PADD V1. The addition of coking capacity can bring a new set of issues to these
refiners. Not only must the refinery deal with the coke disposal issue, they must deal with the
cracked products a coker generates. Coker naphtha typically has a high sulfur and nitrogen
content, but is also rich in olefins and diolefins. Coker naphtha cannot simply be added to the
straight-run feed to the naphtha hydrotreater protecting the catalytic reforming unit. The coker
distillate is also high in sulfur and nitrogen, and has a very low cetane number due to its high
aromatic content. The heavy gas oil produced in a coker from bitumen has a particularly high
aromatic content, and therefore makes a poor FCC feedstock. Each of these streams will also
contain differing levels of silica from the antifoam agent used in the coke drums which can
poison the catalyst in downstream hydroprocessing units. Coupled with the high VGO
component of all synthetic crudes and bitumen blends, there is a twofold impact on the FCC unit.
Higher throughputs are required to process the additional VGO, and the feedstock is poor
quality.
Refiners used to processing the VGO from a sweet conventional crude, like Western Canadian
Mixed Sweet (MSW) crude, will see a significant shift in their FCC yield pattern if they start to
process VGO from most bitumen-derived crudes. Table 2 shows commercial data from a
Canadian refinery, and the impact of switching their FCC feed from a VGO from 100% MSW
crude, to a VGO from SSB crude3. Conversion in the FCC unit dropped by more than 20% as
the yields of LCO and decant oil tripled.
AM-06-16
Page 6
Table 2 – FCC Yield Impact of Processing Synthetic VGO
Western
Canadian Synthetic Crude
Yield, vol% FF
Propylene 11.7 6.2
Butylene 14.0 6.9
Gasoline 61.1 51.3
LCO 9.6 27.6
Decant 2.7 9.3
C3 + Liquid Yield 116 110
Properties
Gasoline
RON 91.9 96.5
MON 80.9 84.3
Most refiners will not see such a drastic change, as they will most likely replace only a portion of
their current crude diet with synthetic crudes or bitumen blends. The impact will be closer to
that shown in Figure 3. This graph shows the FCC yields for the VGOs from three crude blends;
100% Brent crude, 75% Brent plus 25% of a sweet synthetic blend, like Syncrude SSB, and
75% Brent plus 25% of a sour synthetic blend, like Suncor OSE. As the sweet synthetic and the
sour synthetic blends are added to the FCC feed, the yields of LPG and gasoline drop off, while
the yields of light cycle oil and slurry oil increase. This has been one of the historical problems
refiners have experienced while trying to process synthetics. The loss of FCC conversion has
adversely affected the value of the synthetic blends. This adverse impact increases dramatically
as the percentage of synthetic crude in the diet increases. All the VGO-range material in OSE
crude is coker heavy gas oil. This underscores the impact coker gas oil has on FCC
performance, so if a refiner is also adding a new coker to convert the surplus resid in a bitumen
blend like WCS, the overall impact on FCC yields could be even more pronounced.
AM-06-16
Page 7
Figure 3 – Impact of 25% Synthetic on FCC Yields
80%
70% LPG Gasoline LCO CSO
60%
vol-%
50%
40%
30%
20%
10%
0%
Brent Brent + Sweet Brent + Sour Blend
Blend
Crude Blend
Most aspects of the refinery operation will be affected by the shift to synthetic crudes or bitumen
blends. If a refiner decides to import a sweet synthetic crude like SSB, their reformer feed will
become richer. With more naphthenes in the reformer feed, hydrogen production will increase.
They will, however, have limitations blending distillate fuels. Depending on the other crudes
being processed, the refinery could be limited to just 20% SSB in their crude diet if running for
maximum jet production (smoke point limit), or about 35% if running for maximum diesel
production (cetane number limit). These limits will be relaxed when Syncrude starts producing
SSP in June of 2006. There are also fewer distillate blending constraints with other sweet
synthetics like OSA, HSB, or PAS. The production of ULSD is more difficult than indicated by
the low sulfur level of the distillate cuts of the crude. The sulfur and nitrogen species left in the
kerosene and diesel cuts are the most refractory, difficult-to-treat species that could not be
removed in the upgrader’s relatively high-pressure hydrotreaters. FCC conversion and gasoline
yield will drop significantly when using any of the currently-available synthetic crudes, and the
production of lube base stock may be impossible due to the aromatic nature of the synthetic
VGO. The large percentage of VGO-range material in these crudes may result in the FCC unit
capacity becoming a bottleneck. Since synthetic crudes contain no resid, they can be used to
increase refinery throughput without having to increase resid conversion capacity.
If a refiner chooses to process one of the sour synthetic crudes, all of the benefits and limitations
discussed above for a sweet synthetic crude still apply, but much more severe hydrotreating will
be required at the refinery to produce ULSD or acceptable-quality FCC feed. The sulfur and
nitrogen content of these crudes is very high compared to a conventional light sweet crude.
Additional sulfur plant capacity may be required. The TAN number of these crudes may also
become an issue. Crudes with straight-run VGO components from bitumen, like OSV, have
higher TAN numbers. Metallurgy upgrades may be required to handle these crudes. Sour
synthetic crudes produced from coker products are more aromatic in nature, but the TAN
numbers are much lower.
AM-06-16
Page 8
DilBits have both a high resid content and a high sulfur content. They are relatively low in
distillate content, and high in VGO content. Significant resid conversion capacity and
hydrotreating capacity would have to be added to a light sweet crude refinery to process a DilBit.
The FCC unit will again be limited by the large volume of the FCC feed. Although better quality
than a coker-derived VGO, the straight-run VGO from bitumen is still very aromatic, and makes
a poor FCC feed. The TAN number of these crudes can be high, depending on the source of the
bitumen. For example, Athabasca bitumens are reported to have higher TAN numbers. The
large volume of condensate in the blend may pose a problem for some refiners. All the light
naphtha may not be able to be blended into the gasoline pool without exceeding RVP
specifications. As already mentioned, some refiners are already shipping the condensate back to
Alberta by rail, both to alleviate any light ends constraint, but also to realize the higher netbacks
for condensate. Depending on the gasoline to diesel (G/D) ratio of the refinery involved, there
may be insufficient material for the distillate pool.
SynBits generally have a lower resid content than DilBits, but more distillate and VGO-range
material. Since the majority of the distillate comes from a sweet synthetic crude, with the
remainder from bitumen, the distillate quality is only marginal. Some hydrotreating will be
required to achieve a 40 cetane number in the full-range diesel. The VGO is still a poor-quality
FCC feed due to the high aromatic content. The TAN number will be lower than that found in a
DilBit since about half the feed has already been severely hydrotreated. The sulfur content of a
SynBit is also much lower than that of a DilBit.
The aromatic nature of bitumen-derived VGOs makes them poor FCC feedstocks. The quality of
three bitumen-derived VGOs are compared to the VGO from an Arab Light crude in Table 3.
The bitumen-derived fractions have lower API gravities and lower hydrogen contents, consistent
with higher levels of sulfur, nitrogen, and aromatics. A brief overview of the chemistry of
aromatic conversion in an FCC unit is useful to give more insight into reasons why these
feedstocks are more difficult to process.
AM-06-16
Page 9
Table 3 – Bitumen-derived VGO Quality
Figure 4 illustrates how a three-ring aromatic compound with several alkyl side chains reacts in
an FCC unit. Methyl and ethyl groups will tend to stay attached to the aromatic compounds.
Alkyl side chains with a carbon number of three or greater will cleave off close to the aromatic
ring. The removed alkyl side chains will initially become olefins, and may crack again into
smaller components depending on the length of the chain. Paraffinic compounds with a carbon
number of five or less tend not to crack. Paraffinic compounds with a carbon number of six
crack slowly, and paraffinic compounds with a carbon number of seven or more will crack fairly
quickly.
FCC cracking will not open the aromatic ring structures. Depending on the number of short
alkyl side chains remaining, the compound in Figure 4 could end up in the FCC light cycle oil
product, but would most likely be produced as part of the heavy cycle oil pool or the decant oil.
AM-06-16
Page 10
Figure 4 – FCC Conversion of Multi-ring Aromatic Compounds
R1
C
C C2 C2
C FCC Cracking
C
C2 R2 C2
+
R1=C
+
R2=C
If that same three-ring aromatic compound was partially saturated in an FCC feed pretreater
before being fed to the FCC unit, the resulting products would be significantly different (Figure
5). Notice it takes only four moles of hydrogen to saturate two of the three aromatic rings.
Saturated ring structures crack open far more easily in an FCC unit than aromatic ring structures.
While some of the compounds would follow the upper path and partially dehydrogenate back to
a two-ring aromatic compound, the majority of these partially saturated ring structures would
follow the lower pathway. The longer alkyl side chains would first be cleaved off as before. The
saturated rings would then crack open, leaving a single-ring aromatic structure and other
paraffinic and iso-paraffinic molecules.
R1 R1
C C g
a c kin C
C C2 + 4H2 C C2
C Cr 2
+
FC
Hydrotreating R1 = C + R2 = C + i-C5
C C
C R2 C R2
C2 C2
FC
CC
rac
ki ng
C2
+
R1 = C + R2 = C + i-C4 + C6=
The single-ring aromatic structures created in this way are almost never benzene. The structures
formed will usually be toluene and xylenes, and thus have high octane number. The iso-paraffins
created will continue to crack to lighter compounds as dictated by the carbon number.
AM-06-16
Page 11
If that same three-ring aromatic compound was now fully saturated in an FCC feed pretreater
before being fed to the FCC unit, the resultant products would again be significantly different
(Figure 6). It would take seven moles of hydrogen to fully saturate the three-ring structure. As
before, some of the structures would partially dehydrogenate back to a single-ring aromatic
structure, generating several paraffinic molecules. The majority of these saturated ring structures
would crack open completely, generating numerous normal paraffins, iso-paraffins, and olefins.
Again, the paraffins would continue to crack to lighter compounds, depending upon their carbon
number. Normal and iso-paraffinic molecules have much lower octane numbers than aromatic
compounds with the same carbon number. By fully saturating the initial compound in the FCC
feed pretreater, the gasoline yield would be reduced in favor of a slightly higher LPG and gas
yield, and the octane number of the gasoline created would be lower.
+ R1 = C
R1 R1 C2
C C i ng +
C C2 + 7H2 C C2 rack
C C R2 = C + i-C5 + C5=
FC
Hydrotreating
C C
C R2 C R2
C2 C2
FC
C Cr
ack R1 = C + R2 = C +
i ng i-C5 + C4= + C5=
+ C3 + C3=
In general, the FCC cracking of saturated and aromatic ring structures can be summarized by
Table 4 below. Cracking of multi-ring aromatics produces high cycle oil yields but low gas,
LPG and gasoline yields. The gasoline produced would consist mainly of paraffins and olefins.
FCC cracking of single-ring aromatic compounds produces low gas, LPG and cycle oil yields,
but a high gasoline yield. The gasoline produced is mostly aromatic, and therefore has a high
octane number. FCC cracking of saturated ring structures produces a low yield of cycle oil, a
reasonably high gasoline yield, but a higher yield of gas and LPG than cracking of single-ring
aromatics. The gasoline produced would again consist mainly of paraffins and olefins, and
would therefore have a lower octane number.
AM-06-16
Page 12
Table 4 – FCC Cracking of Aromatic Rings (Summary)
From this, it can be concluded that if the refiner’s objective is maximum gasoline yield from his
FCC unit, the feed pretreater conditions should be set to ensure that all the multi-ring aromatic
compounds are saturated down to single-ring aromatics. Hydrogen addition beyond this point
would only serve to lower gasoline yield and gasoline octane. Higher conversions would, of
course, be achieved, but product value would not be maximized. If a refiner were trying to use
his FCC unit to produce olefins for alky feed or MTBE or other chemical production, obviously
a higher feed hydrogen level would increase the production of olefins.
The curve shown in Figure 7 illustrates the FCC gasoline yield as a function of the feed
hydrogen content. VGOs from more aromatic crudes are near the bottom left-hand end of the
curve, while VGOs from more paraffinic crudes would be located near the top right-hand end of
the curve. The trend shows that gasoline yield increases with increasing feed hydrogen. Gasoline
yield will not continue to rise, however, as the feed hydrogen content is increased. At some
point, the gasoline yield will drop off, as the conversion to gas and LPG continues to increase.
Overall, conversion will continue to increase, but the gasoline yield will drop. This drop off is
not a detriment if the refiner is trying to make olefins for alky feed or petrochemical feedstock
with the FCC unit, but it is detrimental if the FCC unit is designed to achieve the maximum
gasoline yield. This yield response is represented by the dashed portion of the curve, although
the exact inflection point is a function of the feedstock and other variables.
AM-06-16
Page 13
Figure 7 – FCC Gasoline Yield Controlled by Feed Hydrogen Content
72
Vacuum
70 Gas Oil
68
Gasoline Yield, vol-%
66
64 LC-Finer Hydrocracker
Gas Oil Bottoms
62
60 Coker
58 Gas Oil
56
54
52
50
48
46
11.4 11.6 11.8 12 12.2 12.4 12.6 12.8 13 13.2 13.4 13.6 13.8
Feed Hydrogen Content, Wt-%
Yui et al5 took several intermediate VGO streams derived from Athabasca bitumen, hydrotreated
them at a constant severity, and then processed them in an FCC pilot plant to determine product
yields and qualities. The gasoline yield data from that study is also shown in Figure 7. Note that
the hydrotreated coker gas oil had a hydrogen content of 11.5wt%, and gave a gasoline yield just
above the UOP curve. The gas oil produced in the ebullated-bed resid hydrocracking unit had a
higher hydrogen content after hydrotreating, and generated a gasoline yield above the curve,
while the hydrotreated VGO distilled directly from bitumen produced a gasoline yield well
above the curve. There is a valid reason as to why these bitumen-derived VGO streams all
generated FCC gasoline yields above the standard UOP curve. One of the characteristics of
these VGOs is the very low paraffin content. Some bitumen derived coker VGO samples
analyzed at UOP have actually been shown to contain no paraffins of any kind. All the
molecules were aromatics, naphthenes, or sulfur species which were almost all aromatic
compounds. Without the paraffins to crack to gas and LPG, bitumen-derived VGOs can produce
very high gasoline yields, if the hydrogen content is raised to the appropriate level.
Many people believe that the unconverted oil from a hydrocracker is the best FCC feedstock
available. The Yui study obtained a sample of hydrocracker bottoms from a Canadian refinery
when they were processing 100% synthetic heavy gas oil in their hydrocracker. It is clear that
very high conversions and gasoline yields can be achieved by first hydrocracking the bitumen-
derived heavy gas oil streams. Since the data points fall to the right of the conventional crude
curve, it suggests that hydrocracker bottoms are less efficient gasoline feedstocks than
conventional crudes. The important question becomes, “Is the extra hydroprocessing severity
required to raise the hydrogen content from about 12.3 wt% all the way to 13.7 wt% worth an
extra five or six percent in FCC gasoline yield?” UOP would suggest that hydrogen is most
effectively used to increase the hydrogen content to the 12.3 to 12.5 wt% range. Beyond this
level, more and more naphthenes and paraffins would be created in the FCC feed, decreasing the
gasoline yield and increasing the LPG and gas yield. Since LPG has a high value to many
refiners as alky feed or chemical plant feedstock, the most economic hydrogen content will be
different for each application, and must be carefully determined.
AM-06-16
Page 14
Table 5 – Required FCC Feed Pre-treat Severity
Recognizing that the optimum hydrogen content for bitumen-derived heavy gas oil streams may
be between 12.0 and 12.5 wt%, UOP has estimated a rough set of operating conditions necessary
using UOP’s Unionfining™ process technology to increase the hydrogen content of each of the
raw feeds to those levels. As might have been expected, the vacuum heavy gas oil would require
the least severe conditions, where heavy gas oil product from the fluidized-bed coking unit
would require the most severe conditions. These pressure and liquid hourly space velocity
combinations are thought to be close to the economic optimum, but the same results can be
achieved at lower pressure if the space velocity is lowered sufficiently.
Many refiners use the UOP K Factor to estimate the hydrogen content of an FCC feedstock. The
UOP K Factor works very well when trying to distinguish between aromatic and paraffinic
VGO's. However, when the FCC feed is hydrotreated, the UOP K factor-hydrogen content
relationship begins to break down. Thus, if the FCC feed has been severely hydrotreated, the
refiners should focus on API gravity, nitrogen content, and most importantly, hydrogen content
of the FCC feed.
The penalty associated with the quality of the VGO from a bitumen-derived crude changes
rapidly once the FCC feeds have been hydrotreated. Figure 8 summarizes the FCC yield pattern
for the same Brent/synthetic VGO blends shown in Figure 3, but after the feeds had been
hydrotreated to a constant sulfur content. The yield of LPG and gasoline is up in all cases, with
the expected reduction in light cycle oil and slurry oil yields. The magnitude of this yield shift is
not consistent between cases. There is a much greater increase in LPG and gasoline yield with
synthetic VGOs in the FCC feed, especially the sour synthetic blend. When this same analysis
was completed substituting the Brent crude component with the more aromatic Arab Light crude,
the increases in LPG and gasoline yields were even greater, to the point that two of the three
cases had gasoline yields greater than the corresponding Brent crude cases. The most difficult
to treat feeds gained the most advantage by hydrotreating.
AM-06-16
Page 15
Figure 8 – FCC Yields on Hydrotreated VGO Feeds
80%
80%
70%
70% LPG
LPG Gasoline
Gasoline LCO
LCO CSO
CSO
60%
60%
50%
50%
vol-%
40%
40%
30%
30%
20%
20%
10%
10%
0%
0%
Brent
Brent Brent
Brent +
+ Sweet
Sweet Brent
Brent +
+ Sour
Sour Blend
Blend
Blend
Blend
Crude
Crude Blend
Blend
In 2000, UOP completed an internal study that compared the economics of FCC feed pretreating
with post-treating of FCC products. UOP studied the impact of blending 25% of two different
synthetic crudes into a conventional crude diet. The FCC yield patterns discussed above were
part of that study. One of the conclusions from that study was confirmation that the economics
for pretreating the FCC feed improved as the quality of the raw FCC feed became worse. The
incremental operating costs, capital costs, and net present values for each case are summarized in
Figure 9 below. While all the capital and product prices used for this study are now six years out
of date, the conclusion is still valid; the tougher the feed, the higher the NPV.
700
Inc Op Costs, $MM/yr
600
$MM (Year 2000)
400 Feed V
ghe
r NP
300 u
To r
200 ghe
i
H
100
0
-100
Brent Brent + Brent + Arab Light Arab + Arab +
Sweet Syn. Sour Syn. Sweet Syn. Sour Syn.
Blend Blend Blend Blend
AM-06-16
Page 16
The economics of FCC feed pretreating projects are driven by two factors: better yields from the
FCC unit itself, and incremental conversion. Whenever a VGO-range feed is severely
hydrotreated, about 10 – 15 lv% of the feed is converted to distillate and lighter products. Some
of this conversion is simply a shift in the boiling point curve as sulfur and nitrogen is removed
from the molecules, but some hydrocracking to lighter products does occur. Assuming the
distillate and lighter products are fractionated from the hydrotreated VGO before it is fed to the
FCC unit, hydrotreating the feed effectively debottlenecks the FCC unit. If a refinery has an
FCC unit sized for 50 KB/D, after hydrotreating, there will only be about 43 KB/D of
hydrotreated VGO left to feed to the FCC unit. This allows the refiner to acquire additional
crude if the other processing units have surplus capacity, or at a minimum, it allows the refiner to
purchase surplus VGO to keep the FCC unit full. In either case, it is the incremental conversion
in the FCC feed pretreater that makes the economics so attractive.
UNICRACKING™ PROCESSES
As previously mentioned, all the sweet and sour synthetic crudes, as well as the bitumen blends,
have a high VGO content relative to conventional crude. Hydrotreating alone can turn these
low-quality VGOs into premium FCC feeds and reduce the volume of the feed available, but
most refiners importing bitumen-derived crudes for the first time will still have surplus VGO that
cannot be processed in their FCC unit. This problem will be made worse by the addition of the
heavy gas oil generated in the coker, for those refiners choosing to process a bitumen blend.
Additional VGO conversion capacity must be added. Refiners may consider an FCC expansion
project, or even an entirely new FCC unit, but since they are probably going to construct a high-
pressure hydrotreater to process the FCC feed, it only makes sense to consider converting the
incremental VGO barrels with hydrocracking technology. Hydrocracking has the advantage in
that, unlike an FCC unit, it can be used to produce both gasoline and high-quality distillate fuels.
In order to avoid increased utility cost and unnecessary quality give-away caused by excess
hydrogen consumption, efficient hydrogen consumption is a critical parameter in
hydroprocessing unit design and operation. With growing demand and more stringent
specifications for fuels, it is recognized by industry experts that hydroprocessing technologies
will be key in the future to meeting the refinery conversion capacity and quality needs. Recent
advances in UOP’s hydrocracking technology portfolio, such as the Advanced Partial
Conversion Unicracking (APCU) process, and the other Unicracking configurations discussed
below, were designed for optimal treatment of the distillate and unconverted product fractions,
resulting in more efficient hydrogen utilization.
When most refiners think about partial-conversion hydrocracking, they think of a simple once-
through Unicracking process unit. This configuration has the lowest capital cost of the various
Unicracking process configurations. The unit shown in Figure 10 is sized for a fresh feed rate of
60 KB/D with an unconverted oil rate of 30 KB/D, meaning the unit is designed to give a gross
conversion of 50 lv%. If the cracking reactor was loaded with a high-activity naphtha catalyst,
such as UOP’s HC™ 29 catalyst, the yield pattern would be very close to that shown in Figure
10. The unit would produce about 5 KB/D of naphtha, 12 KB/D of kerosene with a relatively
low smoke point, and 17 KB/D of heavy diesel with a reasonable cetane number. Note that if the
AM-06-16
Page 17
kerosene and heavy diesel were combined to form a full-range diesel, the cetane number would
be close to 40.
One key consideration with this simple configuration is the quality of the unconverted oil
produced as FCC feed. Unconverted oil from this unit would have high hydrogen content,
similar to the 13.7 wt% seen in the unconverted oil from the Canadian refinery discussed earlier.
Again, if a refiner is trying to make alky feed or petrochemical feedstock with their FCC unit,
this would make an excellent feed, but if the FCC unit is being run for maximum gasoline
production, the hydrogen content of the unconverted oil may be too high.
H2
R-2
Naphtha
61.5 5
R-1
60
Kerosene
12 16 mm Smoke Point
64
Heavy Diesel
17 45 Cetane No
Unconverted
30 Oil
Feed
The quality of the distillate fuels produced in a low conversion once-through Unicracking unit
can be poor. Figure 11 is a plot of both kerosene smoke point and diesel cetane index plotted
against conversion in the Unicracking unit. While these plots are based on a VGO from
conventional Arab Light crude, the message is the same for bitumen-derived VGOs. Kerosene
smoke point and diesel cetane index both increase with increasing conversion. If a refiner wants
to generate high-quality distillate fuels from a poor-quality VGO stream, they need to run the
Unicracking unit at a high conversion level.
AM-06-16
Page 18
Figure 11 – Distillate Quality as a Function of Conversion
30 80
Kerosene Smoke Point, mm
25 t
Poin 70
5 40
0 30
0 20 40 60 80 100
Conversion, wt-%
Conventional VGO Feed
Both the distillate quality and the unconverted oil quality can be controlled by using a different
Unicracking unit configuration. In UOP’s separate-hydrotreat configuration shown in Figure 12,
the severity in the R-1 pretreat reactor would be set to achieve the desired hydrogen content in
the unconverted oil, in this example a hydrogen content of 12.5 wt%. The R-1 effluent would be
routed directly to the product fractionator. About 30 KB/D of unconverted oil would be fed to
the cracking reactor, this time running at 80 lv% crack-per-pass. A bed of pretreat catalyst could
be added to the cracking reactor to get the organic nitrogen down to the optimum level for the
cracking catalyst, about 100 wppm or less. The effluent from the cracking reactor would also be
routed directly to the product fractionator.
The fractionator in the separate-hydrotreat configuration would be larger than that in the once-
through configuration, but the high-pressure equipment in the cracking reactor loop would be
50% smaller, now sized for only 30 KB/D. At the high conversion level of 80 lv% crack-per-
pass, the unit would generate higher yields of both naphtha and kerosene, but less heavy diesel.
The quality of both distillate streams would be significantly higher than the once-through unit.
Only about 3 KB/D of the 30 KB/D of unconverted oil produced in the unit would have been
processed to a high hydrogen content in the cracking reactor, thereby reducing overall hydrogen
consumption. The separate-hydrotreat configuration effectively allows independent control of
both the unconverted oil quality and the distillate fuel quality.
AM-06-16
Page 19
Figure 12 – Separate-Hydrotreat Unicracking Unit
H2
R-2
Naphtha
61.5 5
R-1
60
Kerosene
12 16 mm Smoke Point
64
Heavy Diesel
17 45 Cetane No
Unconverted
30 Oil
Feed
Typical once-through hydrocrackers operating at low to moderate conversion levels may not
produce USLD-quality distillate. The unit pressure is often set by the need to produce low-
aromatic, high-cetane diesel. This higher design pressure results in the production of
unconverted oil with a high hydrogen content, and a higher overall hydrogen consumption. In
the APCU unit flow scheme depicted in Figure 13, the unit is designed to produce ULSD and
partially hydrotreated FCC feedstock as primary products, while operating at a significantly
lower pressure. The refiner achieves two goals with the addition of this unit; increased
production of ULSD, as well as improved quality of the FCC feedstock.
H2
Raw Co-feed
Feed
AMINE
HT PT
Rx Rx
Naphtha
SEP
HC E F.G.
Rx H
S
ULSD
FCC Feed
AM-06-16
Page 20
The benefits of the APCU process compared to a conventional low-conversion hydrocracking
process are reflected in more efficient use of hydrogen due to lower pressure operation, and
staging of reactions in a way that minimizes over-processing and product quality giveaway.
Separation of cracked products from the desulfurized FCC feed in the Enhanced Hot Separator
(EHS), followed by an additional Distillate Unionfining process step, allows the distillate quality
to be controlled independently of FCC feed quality. Over-treating of FCC feed is avoided and
excess hydrogen consumption is minimized. The APCU process can achieve hydrogen savings
of five to ten percent compared to a conventional mild hydrocracking scheme. The integration of
a separate Unionfining reactor in the process enables post-treating of other refinery middle
distillate streams. For refiners processing a bitumen blend following the installation of a coker,
the coker diesel stream must be severely hydrotreated to meet ULSD specs. With the APCU
process, that stream could be processed as a co-feed in the Distillate Unionfining reactor, using
the hydrogen and heat from the Unicracking reactor, resulting in utility cost savings. The level
of hydrotreating of these streams can be independently controlled to add just the right amount of
hydrogen, an added benefit of APCU process.
The typical operating conditions and the resultant FCC feed quality for FCC feed pretreating,
the once-through Unicracking process, the separate-hydrotreat Unicracking process, and the
APCU Unicracking process are summarized in Table 6. Since the quality of the VGO from an
Arab Light crude is so different from that of a WCS VGO, the operating conditions to treat both
feeds are shown for comparison. Note that this comparison represents model predictions for
representative feed properties and that the units are sized for typical run lengths. The table
should be used as a qualitative guide to differentiate between the various processing options.
This is not meant to be a substitute for a customized hydroprocessing solution.
+
Plus additional volume for diesel co-feed
AM-06-16
Page 21
Many refiners believe they can treat a portion of their VGO in a Unicracking unit to generate
some high-quality VGO, and then blend it with lower quality VGO, like coker gas oil, to get an
FCC feed with hydrogen content near the 12.5 wt% UOP believes to be close to the optimum
level. While this may be true mathematically, the FCC unit in actuality is concerned only with
the types of molecules being fed to it, and not the average hydrogen content. The three- and
four-ring aromatics in the coker gas oil will still end up in the cycle oil or the slurry oil, and the
naphthenes and paraffins in the hydrocracker bottoms will still be mostly converted to gas and
LPG. Some hydrogen donor reactions will occur in the FCC unit to transfer hydrogen from the
hydrogen-rich stream to the hydrogen-deficient stream, but the overall yields will still be
significantly worse compared to those obtained if the entire feed stream has been hydrotreated to
a hydrogen content of 12.5 wt%. Some refiners have reported significant synergies when
processing Unicracking unit unconverted oil with lower-quality VGOs in their FCC unit.
CONCLUSIONS
The supply of Canadian light and medium sweet crudes is declining. There is a vast oil sands
resource in Canada which up until recently, has not been exploited to any great degree.
Synthetic crudes and bitumen blends will become the dominant crudes in PADD II, PADD IV,
and Northern PADD V in the next ten to fifteen years. Refiners planning to process one of the
bitumen blends will have to install additional resid conversion capacity. Refiners planning to
process any bitumen-based crude will have to install FCC feed pretreating capacity to maintain
acceptable FCC yields. Many refiners will have to install hydrocracking capacity to convert the
large volume of VGO-range barrels present in these crudes, and to meet the increasing demand
for distillate fuels.
The hydrogen content of an FCC feedstock is the critical parameter to control when trying to
optimize the performance of all the VGO processing units. UOP has developed novel
Unicracking process flow schemes that allow the refiner to control the hydrogen content of the
FCC feed independently from conversion level or distillate quality.
AM-06-16
Page 22
REFERENCES
1. Purvin & Gertz Inc., “Global Markets for Canadian Oil Sands Crudes”, December, 2005
2. Energy Information Administration, “The AEO2006 Early Release”, www.eia.doe.gov,
January 2006
3. Halford, T.L., McIntosh, A.P., and Rassmussen, D., “A Canadian Refiner’s Perspective
of Synthetic Crudes”, Proceedings of the 1997 NCUT Conference, Directions in Refining
and Marketing of Synthetic Crude Oil (SCO) and Heavy Oil, September, 1997
4. Purvin & Gertz Inc. “Fuel Options for Oil Sands Development”, September, 2004
5. Yui, S., Matsumoto, N., and Sasaki, Y., “Athabasca oil sands produce quality FCC
feeds”, Oil & Gas Journal, January 19, 1998 Journal
6. www.crudemonitor.ca – a website sponsored by the Canadian Association of Petroleum
Producers
UOP LLC
25 East Algonquin Road
Des Plaines, IL 60017-5017
AM-06-16
© 2006 UOP LLC. All rights reserved. Page 23 UOP 4535A