You are on page 1of 16

PAPER 2006-727

Releasing the Value of Heavy Oil and Bitumen:


HTL Upgrading of Heavy to Light Oil
E. VEITH
Ivanhoe Energy Inc., Bakersfield, California, USA

This paper has been selected for presentation and publication in the Proceedings at the 1
st
World Heavy Oil Conference.
All papers selected will become the property of WHOC. The right to publish is retained by the WHOCs Publications Committee.
The authors agree to assign the right to publish the above-titled paper to WHOC, who have conveyed non-exclusive right to the
Petroleum Society to publish, if it is selected.

ABSTRACT
Ivanhoe Energy Inc.s proprietary heavy oil to light oil
(HTL) upgrading technology is designed to cost effectively
process heavy oil in the field and produce a stable, significantly
upgraded synthetic oil product along with by-product energy
which can be used to generate steam or electricity. Analysis
shows that HTL can virtually eliminate cost exposure to natural
gas and diluent and capture the majority of the heavy to light
price differential for oil producers. HTL accomplishes this at a
much smaller scale and at lower per barrel capital costs
compared with established competing technologies.
The HTL process represents the application of a
commercially proven technology to a new feedstock. Ensyn
Group, Inc. initially developed the technology in the early
1980s and has applied it commercially for over 15 years in the
biomass (wood) industry. Seven commercial Ensyn biomass
processing facilities are in operation in the United States and
Canada.
A petroleum pilot plant built by Ensyn in 1998 has been
used for testing crude oil from around the world. In December,
2004, operations at a 1,000-barrel-per-day commercial
demonstration plant were initiated in the Belridge Heavy Oil
Field in the San Joaquin Valley in southern California.
Ivanhoe Energy and Ensyn merged in April 2005, and
Ivanhoe Energy gained full ownership of the petroleum
upgrading technology for the development of heavy oil reserves
around the world. Ensyn Group spun off its existing biomass
processing business, Ensyn Renewables Inc., to its shareholders
prior to the closing of the merger with Ivanhoe Energy
Since the commissioning of the commercial demonstration
facility (CDF), Ivanhoe Energy has tested a number of crude
oils and vacuum tower bottom feedstocks. Analysis of CDF
performance confirms that the HTL process is capable of
1
delivering high yields of upgraded product. The HTL process
significantly reduces or eliminates most contaminants and the
product typically meets pipeline specifications without further
treatment. CDF run analysis validates data obtained in the
pilot scale experiments. Pilot results and feedstock and
synthetic product properties are presented.

INTRODUCTION
In mid-2005, Ivanhoe Energy Inc. acquired a new patented
process, called Rapid Thermal Processing (RTP), for the
field-located upgrading of heavy oil and bitumen. This patented
process will be used to upgrade heavy oil to light oil (HTL) to
enable the economic development of heavy oil and bitumen
resources worldwide. Ivanhoe Energys acquisition included a
new Commercial Demonstration Facility (CDF) in California
that demonstrates a processing capacity of approximately 1,000
barrels-per-day (bpd) of heavy crude oil.
There are significant accumulations of heavy crude and
bitumen throughout the world, much of it "stranded" or
economically constrained, which can be targeted by this
technology, thereby providing access to reserves. Heavy oil
upgrading is receiving considerable attention as much of the
remaining unexploited petroleum reserves in the world are
heavy and extra heavy (bitumen). Both Canada and Venezuela
have extensive heavy oil reserves which compare in size to
current reserves in the Middle East. As conventional, lighter
crude oil supplies decline, they will need to be replaced by
heavier crudes
1
. As heavy crude production increases deep
conversion capacity will also have to be expanded.
New residue processing capacity could be added to existing
refineries, or it could be built in separate, stand-alone upgrading
facilities. If the oil is too heavy to transport by pipeline, and/or
there is the need for heat or energy at the production site, heavy
oil upgrading in the field is attractive, and may avoid extensive
modifications of existing refineries. Traditional residue
processing such as coking or hydrocracking are very expensive
processes, and require large scale to be viable. The HTL
technology would provide a lower cost, simpler residue
processing option compatible with field development.
The challenges of using conventional technology to develop
heavy oil and bitumen resources are many. Now, with the
advent of HTL upgrading, producers have the ability to protect
themselves from exposure to significant uncertainties in the
production of heavy crude and bitumen: the volatile supply and
price swings of natural gas used for steam generation and
diluent to transport heavy crude and the high differential
between heavy and light crude prices. Additionally, centralized
mega-projects are no longer required to support traditional
upgrading schemes, as the HTL upgrading technology is
economic in field-sited modules as low as 10,000 to 15,000
bpd, thereby allowing producers without downstream
operations the ability to dramatically reduce the risk of heavy
oil development in a capital-efficient manner. HTL provides an
option to build upgrading capacity, synchronized with field
development.

HTL TECHNOLOGY DEVELOPMENT
The development of the rapid thermal processing
technology commenced in the early 1980s when it was
discovered that a broad array of carbonaceous feedstocks (i.e.
wood and heavy oil) could be thermally cracked to obtain
valuable products, at residence times of a few seconds. The
initial commercial focus of the technology, beginning in 1989,
was aimed at conversion of wood and wood residues to value-
added fuels and chemicals. Seven commercial biomass plants
based on this technology have been in operation for many years.
As the biomass side of the business grew and operational
and design parameters were optimized, the focus turned towards
petroleum feedstocks. The petroleum application of the
technology was demonstrated in Ensyns pilot plant in Ottawa
on more than 90 experimental runs using a number of different
crude oils, bitumen, and residue between 1999 and 2002. Since
it was believed that the technology had relatively low capital
and operating costs compared to conventional carbon rejection
technologies, such as delayed coking, the commercialization of
the HTL process was initiated.

HTL PROCESS DESCRIPTION
HTL upgrading uses a continuous, short contact time
thermal conversion process (Rapid Thermal Processing or
RTP), which takes place at moderate temperatures and at
atmospheric pressure. It uses a circulating transported bed of
hot sand in the system to quickly heat the feedstock and convert
it to more valuable products, in the absence of air. The
following description is representative of the two general
2
commercial process configurations (High Yield and High
Quality) applicable to heavy oil or bitumen upgrading.

Pre-fractionation
In a commercial HTL facility, the whole heavy oil feed is
sent to a series of distillation towers, where material with a
liquid boiling point below 985-1050 F is removed in a vacuum
tower. The lighter material is later recombined with the
upgraded liquid product to form transportable synthetic crude.
The vacuum tower bottoms (VTBs) are routed to the RTP
Reactor. The general process flow is shown in Figure 1

RTP Reaction Section
The heavy oil residue feed or VTBs are atomized and mixed
in the reactor with hot circulating silica sand and transported up
through the reaction zone by transport or carrier gas. See Figure
2. Recycle product gas (or by-product gas) is used as the
transport gas. Rapid mixing of the oil and sand promotes
effective heat transfer and thermal cracking of heavy oil
feedstocks. Carbon is rejected as the long hydrocarbon chains
are cracked and coke is deposited on the sand during thermal
conversion. When long chain hydrocarbons are cracked, the
boiling point is reduced and the cracked product is vaporized.
The coke covered sand, vapor (cracked VTBs), transport gas
and feed that survive the first pass in the reactor are separated in
a high temperature cyclone system. The fluids are quenched
rapidly and vapors condensed with light oil to maximize liquid
yield and minimize undesirable secondary thermal cracking
reactions. A very short residence time is required between the
feed injection point and the quench point. Typically, residence
times less than a few seconds are targeted.

Product Fractionation
Following quenching, the liquid product can be routed to the
product tank for blending (High Yield) or recovered in a
product distillation tower for further separation for recycle
operations (High Quality). Part of the non-condensable gas is
recycled to the reactor system to be used as transport gas for the
circulating sand, with the balance, or net product gas, used to
raise steam and/or power in a Waste Heat Boiler (WHB). All
liquid products, including the light liquids from the whole crude
which are separated in the pre-fractionation step, are blended to
make a synthetic crude product. In the High Yield mode of
operation the resulting quenched product is routed to the
product tank where it is blended. The process flow diagram is
shown in Figure 2. In the High Quality mode of operation the
bottoms product from the product distillation tower is routed
back to the reactor to further convert any VTBs that survived
the first pass through the reactor. See Figure 3.

RTP Reheater Section
The coke covered sand from the RTP Reactor is separated
from the product vapor stream via a high-efficiency cyclone and
directed to a fluidized bed reheater for carbon removal. Air is
used to fluidize the sand and facilitate combustion in the
reheater. See Figure 4. The temperature of the regenerated sand
is adjusted to the proper reactor temperature in a sand cooler
where excess heat is recovered prior to circulation back to the
reactor. High pressure steam can be generated from the heat
recovered in the sand cooler. See Figure 5. Operation of the
circulating sand system is similar to a conventional fluid
catalytic cracker (FCC) and residue fluid catalytic cracker
(RFCC).

By-product Heat, Flue Gas Treatment and
Solids Handling System
Sorbent is added to reduce contaminants in the heavy oil
feed. A portion is added to the heavy oil feed to the Distillation
Tower section to reduce the total acid number (TAN) of the
feed and to capture sulphur released in the reactor section. The
rest is added to the reheater in order to capture SO
2
which is
formed as the normally high sulphur coke is burned off of the
sand grains. The sorbent also acts as a site for the metals to be
deposited.
Flue gas from the reheater is directed through a cyclone
system to remove ash, spent sorbent, and any sand fines from
the gas prior to injection in a power recovery turbine. See
Figure 5. High pressure steam production is possible by energy
recovery fromthe hot flue gas, plus burning excess by-product
gas, in a WHB. This approach maximizes the useful energy
output from the process, either as high pressure steam or as
power available for onsite facility use with the excess exported
for use in heavy oilfield production operations.
3
Since sorbent is added to the feed and reacts with sulphur
released in the reactor section, the by-product gas is effectively
sulphur free and can be routed to a fuel gas system and used for
another purpose if the steam and or power requirements can be
met by the combustion of the by-product coke.
Flue gas polishing to meet SO
2
emission requirements
through a flue gas de-sulphurization unit (FGDSU) is also
included as part of the flue gas system. The ash, spent sorbent
and sand fines are routed to an ash cooler and collected in a
hopper, ready for disposal as a non-hazardous solid waste.

HTL PATENT STATUS
Since 1998, core patents related to the HTL technology
were successfully obtained. Other patent applications related to
heavy oil upgrading, viscosity reduction, and total acid number
(TAN) reduction are currently pending.
Ivanhoe Energy believes that it has broad patent coverage
from the original Ensyn's RTP patents. These patents
presently provide coverage in key markets, including Canada
and USA. In addition, Ivanhoe Energy has a number of patents
and patents pending applications for petroleum HTL
applications of RTP in the USA and Canada and other
countries as well. Once granted, they will provide patent
protection beyond 2022.

HTL ADVANTAGES
Viscosity Reduction
Significant viscosity reduction and gravity increases are a
favorable result of HTL upgrading as asphaltenes, which
contribute to the very high viscosity, are removed or cracked.
The fraction of the feed that does not boil under the atmospheric
pressure conditions of the process stays in close contact with the
sand, which facilitates its rapid decomposition into smaller
molecules and coke. Even in a once through or High Yield
configuration the HTL process will destroy in excess of 90% of
the asphaltines.
The relationship between viscosity and API for four heavy
crudes and the resulting products using the HTL technology is
presented in Figure 6 and Table 1. It is apparent that there is a
dramatic reduction in the viscosity using the High Yield
configuration. Further reductions in viscosity are achievable by
processing in the High Quality mode of operation which uses a
recycle stream to reduce the +1000 F cut in the final product.
Since the products are low viscosity (typically below 100 cSt at
40 C) the need for diluents required for transportation is
eliminated since the products meet or exceed pipeline
specifications with respect to viscosity. This protects the heavy
oil or bitumen producer from exposure to market conditions
related to the purchase of condensate or synthetic crude oils for
diluent.

Yield
The HTL process has the ability to achieve higher yields than
delayed coking. It is believed the reason for this is the kinetics
of the process. A unique combination of very short contact time
with sand, short residence time of the flashed distillable oil
fraction in the high temperature zone (helps to minimize
secondary cracking reactions), complete lack of porosity and
surface catalytic activity of sand, and a high ratio of sand-to-oil
helps to achieve higher liquid yields and lower by-product gas
make as compared to delayed coking.
Yields (on a C5+basis) are shown in Figure 7 and Table 1
for several heavy oils and bitumen as a function of processing
configuration. As residue content increases there is a trade off in
volumetric yield in both the High Yield and High Quality
processing configurations. The higher residue content
feedstocks have a tendency to have a higher propensity to coke
resulting in more carbon being rejected thereby resulting in a
higher yield loss.
Reactions in thin films of Athabasca vacuum residue have
been shown experimentally to give lower coke yield and a
lighter product
2
. Grey et al. found that as film thickness
approached 20 microns, the transport through the reacting film
of lighter cracked products is more easily facilitated as a result
of the reduced diffusional resistance and resulted in an overall
increase in liquid yield.
The HTL upgrading system has an estimated feedstock film
thickness of less than 20 microns. The operation of the HTL
process results in a thin film of coke being deposited on the
sand and, as discussed above, has a beneficial impact on the
product quality and yields. Another important aspect of the
HTL process is that there are no coke by-products to dispose of,
as is typical in conventional coking technologies, as all of the
coke is oxidized in the reheater vessel.

4
By-Product Heat
Another important benefit is the utilization of the by-
product heat that becomes available due to the rapid oxidation
of the thin film of coke coating the sand. The HTL process has
the unique characteristics of being able to efficiently oxidize
coke in the reheater or regenerator vessel and to utilize that heat
in the production of steam for enhanced oil recovery operation
or for the generation of power. There is a direct relationship
between the residue content of the crude and the amount of by-
product energy available. See Figure 8.
This benefit tilts the economics for HTL upgrading
favorably, e.g., as the price of natural gas increases, the wider
the advantage of HTL upgrading. HTL may be the best
upgrading option in such situations where the excess energy of
the process can be utilized for enhanced recovery of heavy oil.
For Athabasca bitumen (with a residue content of 52 wt%), the
process can support a steam-oil-ratio (SOR) of 3.0, thereby
effectively eliminating or reducing the exposure to market
conditions related to the purchase of fuel gas.

Product Quality
The reduction in the +1000 F cut is seen more explicitly on
Figure 9. High Temperature Simulated Distillation curves are
presented for the case of using Athabasca bitumen as the feed.
The bitumen was processed in both a High Yield (once through)
and High Quality (recycle) configuration. It is apparent from the
curves that the bitumen had a 1000+F fraction of 52 wt% and
a 1300+F fraction of approximately 22 wt%. The processing
of this material resulted in the complete conversion of the
1300+F material in both the High Yield and High Quality
products and the overall reduction in the 1000+F material
from approximately 52 wt% to 30 wt% in the High Yield
Product and approximately 2 wt% for the High Quality product.
Also apparent from the curves is that the VGO cut (650-1050
F) increases from 43 wt% to 57 wt % in the High Quality
product. There is a similar dramatic increase in the 650- F
material from approximately 10 wt% to 43 wt%.
In addition to the dramatic viscosity reduction and
conversion of high boiling point hydrocarbons (+1000 F) to
more valuable lighter hydrocarbons and by-product energy, a
number of other characteristics of the HTL process add value.
Among those discussed below is the reduction in metals content
and acid number of the upgraded product compared to the raw
feed. Also discussed is the relative stability of the HTL
products.

Total Acid Number (TAN) Reduction,
Sulphur Capture and Demetallation

A novel approach has also been developed for reducing the
acid content of crudes that are rich in naphthenic acid, a
corrosive component found in many heavy oil crudes (see Table
1). A calcium-based additive can be introduced to the reaction
system to serve as a sorbent. As the sorbent interacts with the
oil in the upgrading process, it reduces the total acid number
(TAN), absorbs sulphur in flue gas, facilitates metals removal
and effectively removes sulphur from by-product gas.
The addition of a calcium compound as sorbent to the
feedstock neutralizes acids within the oil as determined by total
acid number (TAN), and reduces sulphur oxides (SO
x
) in the
flue gases. The reduction of the TAN value of the oil at an early
stage of its processing can lead to improved performance and
lifetime of the equipment used in the HTL system as well as in
downstream refineries that would receive the upgraded product
from the field. Furthermore, addition of calcium-based sorbent
to the reheater effectively scrubs the flue gases of SO
x
as well
as providing a site for the nickel and vanadium to report to in
the regeneration step in the reheater.
Sorbent addition to the feedstock also enables significant
reduction of H
2
S in the by-product gas streamthat is generated
as high sulphur vacuum tower bottoms are cracked in the
reactor. This can allow for the direct use of the by-product gas
in gas turbines or exporting it for other uses.
High metals (vanadium and nickel) reduction is also
possible. Metals are easily recovered with the spent sorbent at
the exit of the reheater cyclone. Metals react with the calcium
based sorbent to form high melting point compounds. Since
most of the impurity metals are concentrated in the asphaltene
portion of the crude, >90% asphaltene destruction under the
process conditions lets the metals break away from the organic
structures, report to the calcium based sorbent and leads to very
high demetallation. See Table 1.

Stability
The stability of the product may be one of the most
important benefits of HTL upgrading over delayed coking.
Hydroprocessing the delayed coker product is costly. Though
5
HTL upgrading is a thermal cracking process, just like delayed
coking and fluid coking, the HTL products are stable, whereas
the delayed coking and fluid coking products are unstable and
need further stabilization via hydrotreating.
The HTL products have been tested by third parties for
stability and compatibility. The cracked products formed from
the conversion of the heaviest fraction of the feedstock through
contact with the hot sand can be unstable at reaction
temperature, if allowed to proceed to secondary reactions.
However, these products are quenched immediately after
separation of the product vapor from the solid heat carrier, so
that the entire liquid product is stabilized.
Unlike delayed coker products, light distillate products
using HTL upgrading do not appear to be prone to oxidation or
slow polymerization reactions that result in gum formation. The
products are low in unstable olefins with conjugated double
bonds, such as butadiene and diolefins. Compared to delayed
coker distillate, the light distillate of HTL product contains
mostly aliphatic olefins formed by the cracking of waxy
molecules and side chains of aromatic molecules. This fraction
also has an increased content of aromatic molecules that
improve its stability and compatibility with other crudes.
Stability testing was conducted through the Shell Hot
Filtration test, Total Sediment Analysis (D4870-99), Potential
Total Sedimentation (IP390), NCUT Stability testing and
viscosity testing over several weeks. Similarly, product
compatibility testing was performed through the Weihe method,
which tested the blendability of the HTL products against
paraffinic and aromatic stocks. As a result, the light distillates
using the HTL technology are not expected to require
hydrotreating for stability and compatibility.

Product Value
In general, HTL products generated with High Quality
processing configuration are distinguished by the low
concentration of vacuum residual, compared to the
corresponding native crude. In the High Quality configuration
the vacuum residue is either converted to coke or vacuum gas
oil (VGO) and lighter material. Given the high percentage of
VGO in the overall products, the VGO fractions qualities will
have the largest effect on the value of the RTP product for a
particular refinery
3
.
Refineries with excess VGO processing and hydrotreating
capacity will find the most value in HTL product since these
processing capabilities allow them to process the additional
volumes and reduce the contaminants impact on overall
refinery yields. Although there is very little sulphur removal
occurring in the upgrading process, some key vacuum residue
properties are typically improved such as the viscosity and
Conradson Carbon (CCR).
Many of the VGO quality characteristics of interest to
refiners measure the materials crackability or the expected
conversion to lighter products in the FCC process. Some of the
important quality characteristics of VGO that would impact the
VGO value are sulphur and nitrogen content and aniline point,
refractive index and K-factor. The value of any feedstock to a
refinery is typically also dependent on the amount processed
relative to its other feedstocks.
A number of marketing studies have been completed by
third parties that investigate the value of HTL products in
specific markets
3
. HTL processing of California heavy oils such
as Midway Sunset, Belridge and San Ardo were evaluated in
the California marketplace and Athabasca bitumen was
evaluated in the Midwestern refineries. Typically, value uplifts
of 50-80% of the price differential between the raw feed and
light oil prices were indicated to be achievable. These uplifts
depended on the raw feed and upgraded product qualities and
the configuration and throughputs of the target refineries.

Field Upgrading: Impact on Greenhouse
Gas Reduction

In order to determine if partial field upgrading of Athabasca
bitumen could potentially have a net positive impact on carbon
dioxide emissions; Enbridge, Inc. conducted a "well to wheels
analysis that compared generic Steam Assisted Gravity
Drainage (SAGD)

operations with and without generic partial
field upgrading. This analysis was published in a paper
4
that
was presented at the Petroleum Societys Canadian
International Petroleum Conference 2002, Calgary, Alberta,
Canada, J une 11 13, 2002.
The base case of the analysis was a stand-alone SAGD
operation located in the Athabasca oil sands with diluted
bitumen transported to a coking refinery located in the United
States to produce transportation fuels. The coke generated at the
refinery was eventually sold and used as fuel. The majority of
6
the bitumen produced from the oil sands today is processed in
this manner. See Table 2.
One of the comparison cases had partial upgrading with by-
product carbon removal integrated as fuel in the field to
generate steam for SAGD. In this analysis it was assumed that
the generic partial upgrader would supply 65% of the energy
required for steam; a conservative assumption with regard to
HTL, which is able to support 100% of the assumed SOR of 3.0
when starting with Athabasca 8 API bitumen.
It was assumed that by-products from the partial upgrading
process (gas and coke) were used to reduce the volume of
natural gas required to make steam. The steam generated from
the combustion of by-product coke and gas supplied 65% of the
required steam load (SOR=1.95). The partially upgraded
product was stabilized in a mid-stream polisher and transported
without diluent to a US refinery that produced an equal amount
of transportation fuels for all cases. The stabilization process
consisted of mild hydrotreating (hydrogen addition) to stabilize
olefins in the cracked product. This is a rather typical approach
and is considered to be a value-add process as it generates a
more desirable product for refiners. As shown in Table 2, using
the carbon removed by partial upgrading as fuel for SAGD and
avoiding the burning of coke in low efficiency and
environmentally substandard coke-fired furnaces reduces the
well to wheels greenhouse gas emissions by 22%.
It is apparent that on a field or point source basis, burning
coke off of the sand using generic field upgrading is a net
negative with respect to CO
2
emissions (500 vs. 900 CO
2e
kg/m
3

bitumen). From a global perspective though, the net reduction
of greenhouse gas emissions becomes more clearly understood.
Generating coke at a refinery, transporting the coke to the
power generation site, typically offshore, and then combusting
the coke in coke-fired furnaces leads to a significant level of
emissions (650 CO
2e
kg/m
3
bitumen).
A deeper look at the environmental impact would suggest
that in terms of NO
x
(nitrous oxides) and SO
x
(sulphur oxides)
emissions the net reduction will be significant with HTL
technology due to the application of our novel sulphur capture
process using calcium hydroxide, known flue gas de-
sulphurization techniques and nitrogen capture technologies that
will allow us to meet California air quality requirements the
Gold Standard. On the other hand, coke is normally shipped
offshore and burned in furnaces in locations where the
environmental regulations are much more lenient. The impact
on the planet, with respect to sulphur and nitrogen oxides, is not
accounted for in this high level analysis.

Commercialization of HTL Upgrading

Pilot Plant
Since 1998, Ensyn has been developing the technology for
heavy oil upgrading on a nominal 20 bpd (~5 bpd vacuum
residue) pilot plant, at their research facility in Ottawa, Canada.
Ensyn operated the unit from 1998 to 2002, testing a variety of
heavy oil and bitumen feeds, with over 90 pilot plant runs
completed. During this time, process scheme developments
took place, which now form the basis of the CDF demonstration
unit and commercial unit designs. A number of patents have
been issued and are pending for use of the process to upgrade
heavy oils.

Commercial Demonstration Facility (CDF)
During 2003 and 2004 a nominal 300 bpd vacuum residue
feed commercial demonstration facility, which demonstrates an
overall processing capacity of 1,000 bpd of California heavy
crude, was designed and built at Aera Energys (owned by
affiliates of ExxonMobil and Shell) Belridge field, near
Bakersfield, California. See Figure 10.
The CDF is designed to demonstrate upgrading performance
under a variety of operating parameters and process
configurations. The CDF was commissioned in late 2005 and
operation with heavy crude oil feed commenced. Test runs with
whole crudes and vacuum tower bottoms in once-through and
recycle operation confirmed pilot plant performance using
similar crudes and processing schemes. Further test runs are
underway using California heavy oil and Athabasca VTB and
ATB (atmospheric tower bottoms) feedstocks.

Commercial Plant Design
A number of third party engineering studies have been
performed to establish a commercial plant design, together with
estimated capital and operating costs.
In 2000, SNC Lavalin developed a conceptual design for a
30,000 bpd commercial HTL upgrading plant. The design
consisted of a number of reaction and reheater sections. In
7
2005, based on improved pilot plant performance data,
improved design parameters and target process performance,
Colt Engineering of Calgary, Alberta was contracted by Ivanhoe
Energy to prepare a preliminary process design package (PDP)
for a single train 5,000 bpd vacuum residue feed HTL plant.
This would be the core residue processing unit for an integrated
plant size of 15,000 bpd of raw California heavy crude oil
feedstock. As part of their design package, Colt prepared a
preliminary capital cost estimate and overall schedule for
project execution. This improved design basis and target
process performance form the basis of the current commercial
HTL plant designs.
AMEC Group Ltd. has completed a Preliminary Design
Package (PDP) that builds on the Colt PDP work by expanding
the focus to include all of the systems and facilities required to
enable total operations outside the core processing unit
previously defined and estimated by Colt. In particular, this
work included the feed fractionator and associated equipment,
all offsites and utilities, including boiler feed water, nitrogen,
flare, fuels, feed and product tankage, lime and ash handling
systems and instrument air. Also addressed was site
infrastructure, including all buildings, control system, site civils
and fire protection. This work was completed in J une 2006.
Preliminary engineering has been completed and will be the
basis for front-end engineering and design (FEED).

SUMMARY
The four key advantages that the HTL technology affords to
the heavy oil and bitumen producer are:
1. Ability to capture the majority of the price differential
between heavy and light oil.
2. Upgraded product does not require diluents or
blending agents to move the product through a
pipeline.
3. By-product energy is used to generate steam and/or
power.
4. Small scale is appropriate to grow field-sited
upgrading capacity along with resource development
(minimum scale of 10,000-15,000 bpd).
These features of the HTL technology enable the producer
to proceed with heavy oil and bitumen development knowing
that:
they will not be exposed to high heavy oil to light oil
price differentials,
they will not be exposed to volatile price swings and
availability of blending agents used to transport the
heavy oil to market,
their exposure to natural gas pricing and supply
concerns for steam generation is reduced or
eliminated,
this can be accomplished where the initial capital
outlay is matched to field development and is a
fraction of the cost of alternative upgrading schemes
and,
they can grow their upgrading capacity inline with the
field development.

Ivanhoe Energys HTL upgrading technology is a unique
thermal cracking technology that solves some of the
disadvantages that exist for delayed coking, fluid coking, and
visbreaking processes. Only the heaviest molecules in the
residue are thermally cracked in this selective thermal cracking
process. The technology is simple with known analogues in the
downstream industry (FCC and RFCC) and a track record of
on-stream efficiency established via the biomass application
using the same core technology in seven commercial plants
over 16 years.
Since >90% of the asphaltenes in the heavy oil are
converted to coke , light oils and by-product gas in the process,
and since most of the metal impurities are eliminated in the
process, the upgraded oil product from the process could be
easily upgraded further to transportation fuels using
conventional refining technologies.

In summary, the HTL technology significantly improves the
economics of heavy oil and bitumen projects and provides, in
effect, a hedge on the risk exposure that heavy oil producers are
typically exposed to.

ACKNOWLEDGEMENTS
The author would like to thank the management teams of
Ivanhoe Energy Inc. and Ensyn Corporation for permission to
publish this paper. Many individuals contributed to the
technology development but special mention of J oe Kuhach,
Doug Clarke, J erry Kriz, Geoff Hopkins, Ron Conrow, Steve
8
Young, Bill Grassi, Thomas Fisher, Trevor Hibbs, Dennis Reed
and Ron Russell is warranted.

NOMENCLATURE
bpd = barrels per day
cSt = centistokes
FCC = fluid catalytic cracker
N = nitrogen
NCUT = National Centre for Upgrading
Technology
Ni = nickel
RFCC = residue fluid catalytic cracker
S = sulphur
TAN = total acid number
V = vanadium
VTBs = vacuum tower bottoms

REFERENCES
1. WEI, E. AND SARNA, M. E., Technical Review Update:
Ivanhoe/Ensyn RTP Upgrading Process, internal
Ivanhoe Energy report prepared by Purvin & Gertz Inc.,
J uly 2005.
2. GREY, M.R., LE, T., MCCAFFREY, W. C., BERUTTI,
F., SOUNDARARAJ AN, S., CHAN, E., HUG, I., and
THORNE, C., Coupling of Mass Transfer and Reaction
in Coking of Thin Films of an Athabasca Vacuum
Residue, Ind. Eng. Chem. Res. Vol. 40, No. 15, 2001,
pp3317-3324.
3. CRUTCHER, T., HOULTON, G. A. and SARNA, M. E.,
Synthetic Crude Oil Analysis For the California Refining
Market, internal Ivanhoe Energy report prepared by
Purvin & Gertz Inc., October 2003.
4. LUHNING, R. W., BLACKMORE, T., and LAWSON,
D. S. of Enbridge Inc., Pipeline Transportation of
Emerging Partially Upgraded Bitumen, presented at the
Petroleum Societys Canadian International Petroleum
Conference 2002, Calgary, Alberta, Canada, J une 11-13,
2002.
5. FREEL, B. and GRAHAM, R.G., Internal Reports, Ensyn
Group Ltd., October 2002 April 2005.
























9


FIGURE 1. HTL Core process showing the sand path through the RTP
Reactor and RTP Reheater. Flow is clockwise from the reheater up
through the reactor and back to the reheater. Raw feed is pre-fractionated,
routing the straight run 1000 minus F fractions to a product tank for
blending in the final upgraded product.



FIGURE 2. Processing in a High Yield mode: Vacuum tower bottoms are
routed to the RTP Reactor where thermal cracking takes place. Upgraded
VTBs (product) are quenched at the exit of the Reactor Cyclone and routed
to the product tank and blended.
10


FIGURE 3. Processing in a High Quality mode: Vacuum tower bottoms are
routed to the RTP Reactor where thermal cracking takes place.
Upgraded VTBs (product) are quenched at the exit of the Reactor Cyclone
and routed back to the distillation column for further quenching, separation
of upgraded products and recycle of VTBs that survived the first pass
through the reactor. In this diagram the distillation process has been
simplified to show only one tower. In a commercial design there would be
an atmospheric and vacuum tower at the front of the plant with an
additional vacuum tower to separate the products and recycle streams
downstream of the product quench.


FIGURE 4. The coke covered sand from the RTP Reactor is separated
from the product vapor stream via a high-efficiency cyclone and directed to
a fluidized bed reheater for carbon removal. Air is used to fluidize the sand
and facilitate combustion in the reheater. Sorbent is added to the reheater
in order to capture SO
2.
Flue gas polishing to meet SO
2
emission
requirements through a flue gas de-sulphurization unit (FGDSU) is also
included as part of the flue gas system. The ash, spent sorbent and sand
fines are routed to a hopper, ready for disposal as a non-hazardous solid
waste.

11
FIGURE 5. The temperature of the regenerated sand is adjusted to the
proper reactor temperature in a sand cooler where excess heat is recovered
prior to circulation back to the reactor. Flue gas from the reheater is directed
through the reheater cyclone to remove ash, spent sorbent, and any sand
fines from the gas prior to injection in a power recovery turbine. High
pressure steam production is possible by energy recovery from the hot flue
gas, plus burning excess by-product gas, in a WHB and from the heat
recovered in the sand cooler.

Viscosity vs. degrees API for Raw Feed and Upgraded Products
10
100
1,000
10,000
100,000
5 7 9 11 13 15 17 19 21
degrees API
V
i
s
c
o
s
i
t
y

(
c
S
t

@

4
0

C
)
Belridge Midway-Sunset San Ardo Athabasca Bitumen
Raw Feed
High Yield Product
High Quality
Product

FIGURE 6. The relationship between viscosity and API for four heavy crudes and the resulting
products using the HTL technology is presented. It is apparent that there is a significant reduction in
the viscosity using the High Yield configuration. Further reductions in viscosity are achievable by
processing in the High Quality mode of operation which uses a recycle stream to reduce the +1000 F
cut in the final product.
12
Product Yields vs. Processing Configuration / Resid Content
80%
82%
84%
86%
88%
90%
92%
94%
96%
98%
100%
Belridge Midway Sunset San Ardo Athabasca Bitumen
V
o
l
u
m
e
t
r
i
c

Y
i
e
l
d
s

(
v
o
l
%
)
20%
25%
30%
35%
40%
45%
50%
55%
60%
65%
70%
R
e
s
i
d

C
o
n
t
e
n
t

(
w
t
%
)
High Yeld (Once Through) High Quality (Recycle) Resid Content

FIGURE 7. Yields (on a C5+ basis) are presented for several heavy oils and bitumen as a function of
processing configuration. As residue content increases there is a trade off in volumetric yield in both
the High Yield and High Quality processing configurations.


Net By-Product Energy vs. Resid Content in Raw Feed
High Quality Mode of Operation
0
200,000
400,000
600,000
800,000
1,000,000
1,200,000
1,400,000
0% 10% 20% 30% 40% 50% 60%
Resdid Content (wt % 1000+ F)
N
e
t

M
M
B
t
u

o
f

s
t
e
a
m

p
e
r

b
b
l

o
f

f
e
e
d

FIGURE 8. There is a direct relationship between the residue content of the crude and the amount of
by-product energy available. There is roughly 360,000 400,000 MMBtus in a barrel of steam. For
Athabasca bitumen a steam-oil-ratio (SOR) of approximately 3.0 can be supported as there are 1.1
MMBtus of high pressure steam available for export per barrel of whole bitumen from the HTL
facility for SAGD operations.

13
H T S D B O I L I N G P O I N T D I S T R I B U T I O N
0
1 0
2 0
3 0
4 0
5 0
6 0
7 0
8 0
9 0
1 0 0
0 1 0 0 2 0 0 3 0 0 4 0 0 5 0 0 6 0 0 7 0 0 8 0 0 9 0 0 1 0 0 0 1 1 0 0 1 2 0 0 1 3 0 0
B O I L I N G P O I N T , F
C
U
M
U
L
A
T
I
V
E

w
t

%

O
F
F
R T P - L P , R e c yc le
R T P - L P , O n c e - T h ro u g h
A th a b a s c a B itu m e n , 8 A P I

FIGURE 9. High temperature simulated distillation results for Athabasca bitumen processed in
both the High Yield (Once Through) and High Quality (Recycle) configuration.


FIGURE 10. Ivanhoe Energys Commercial Demonstration Facility (CDF) located in the Belridge
Oil Field in California. The facility can process 300 bpd of VTBs or an equivalent of 1,000 bpd of
typical California heavy oil which has a residue content of approximately 33 vol%.
14
Feed and Product Properties Belridge
Midway
Sunset
San Ardo
Athabasca
Bitumen
Raw Crude or Bitumen
API 13.4 11 10.4 8
Residue Content* (wt%) 33% 36% 38% 52%
Viscosity (cSt @ 40 deg C) 750 2,500 11,000 40,000
TAN (mgKOH/g) 5.8 4.4 4.6 3.7
Impurities
S (wt%) 1.0 1.8 2.2 4.9
N (wt%) 0.61 0.93 0.73 0.30
Ni (wt ppm) 150 275 100 184
V (wt ppm) 200 300 92 471
Once Through Product
API 16 15 16.9 14
Liquid Yield 94% 91% 87% 90%
Residue Content (wt%) 16% 22% 20% 30%
Residue Conversion % 52% 39% 26% 42%
Viscosity (cSt @ 40 deg C) 100 120 92 150
Viscosity Reduction (@ 40 deg C) 86.7% 95.2% 99.2% 99.6%
TAN (mgKOH/g) <.5 <.5 <.5 <.5
Impurities
S (wt%) 0.9 1.7 2.0 4.0
N (wt%) 0.53 0.65 0.61 0.25
Ni (wt ppm) 115 138 38 110
V (wt ppm) 130 150 30 270
Recycle Product
API 19 20 18.5 19
Liquid Yield 90% 87% 84% 81%
Residue Content (wt%) 1% 2% 3% 2%
Residue Conversion % 97% 94% 92% 96%
Viscosity (cSt @ 40 deg C) 35 50 45 60
Viscosity Reduction (@ 40 deg C) 95.3% 98.0% 99.6% 99.9%
TAN (mgKOH/g) <.5 <.5 <.5 <.5
Impurities
S (wt%) 0.9 1.7 5.0 3.6
N (wt%) 0.5 0.5 0.6 0.2
Ni (wt ppm) 15 28 10 18
V (wt ppm) 20 30 9 47
Note: Residue is the cut that boils above 1000 deg F (570 deg C - a vacuum tower bottom or "VTB" cut)

Table 1. Raw heavy oil or bitumen feed, High Yield and High Quality product properties. Source:
internal Ensyn and Ivanhoe reports
5
.

15
A " Well to Wheels" Analysis
Generic Partial Field Upgrading Impact on Greenhouse Gas Reduction
(values reported as CO2e kg/m3 bitumen at wellhead)
Bitumen Production (Steam/Oil Ratio = 3.0) SAGD
SAGD with Partial
Field Upgrading
Natural Gas for Steam Boilers 500 170
Coke and by-product gas 730
Pipeline Diluted Bitumen or Non Diluted
Partiall y Upgraded to USA

Mid-Stream Polisher (CO
2
from H
2

production)
30 30
Transportation 40 30
Refinery Upgrading (Identical Products)
Refinery Coke & Gas Disposed for Fuel 650
Transportation Fuel Products equal transportation fuels
CO
2e
kg/m3 Bitumen at Wellhead 1,190 930
Reduction in Greenhouse Gas 22%

Table 2. A comparison of the GHG emissions from a generic Steam Assisted Gravity Drainage (SAGD)
operation with and without generic partial field upgrading. This analysis was published in a paper
4

that was presented at the Petroleum Societys Canadian International Petroleum Conference 2002,
Calgary, Alberta, Canada, June 11 13, 2002.


16