SPE 84868
Evaluation of CO2 Based Vapex Process for the Recovery of Bitumen
from Tar Sand Reservoirs
Khelifa Talbi and Brij. B. Maini, University of Calgary, Canada
Copyright 2003, Society of Petroleum Engineers Inc.
This paper was prepared for presentation at the SPE International Improved Oil Recovery
Conference in Asia Pacific held in Kuala Lumpur, Malaysia, 20–21 October 2003.
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300
words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, and
P.O. BOX 833836, RICHARDSON, TX 75083-3836 U.S.A., fax 01-972-952-9435.
Abstract
The vapour extraction (Vapex) process has recently
gained considerable attention for its promising application for
the recovery of heavy oil and bitumen reservoirs that are
deemed unsuitable for thermal methods. In conventional
Vapex process, a mixture of vaporized solvent (propane and/or
butane) and a commercially available non-condensable gas
(methane, natural gas) is injected into the reservoir to reduce
the oil viscosity. Therefore, the cost of these injected gases per
barrel of produced oil is critical for the economic viability of
the process. With predicted increase in the gas prices, the
process has lost some of its initial attraction.
For reducing the solvent cost, the injection of CO2 as a
non-condensable gas appears to be an attractive alternative.
The advantage of this low cost alternative is that the solubility
of CO2 in heavy oils is significantly higher than that of
methane. Consequently, the mixture of propane and CO2 can
provide greater reduction in oil viscosity compared to the
equivalent mixture of propane and methane. In addition, this
CO2 injection can play a
significant role in sequestering CO2 emissions. However,
the CO2-propane-heavy oil system shows complex phase
behaviour, forming multiple liquid phases. Such multiple
liquid phases can reduce the effectiveness of gravity drainage
process by introducing complex relative permeability effects.
This paper presents an experimental study aimed at
developing a new CO2 based Vapex process for in situ
recovery of bitumen from tar sand reservoirs. The experiments
have been performed in a partially scaled physical model at
different operating conditions. The main conclusion from this
study is that the CO2 based Vapex process is more cost
effective and environmentally friendly than the conventional
Vapex process.
Introduction
Heavy oil and bitumen represent a huge natural resource;
the word’s total estimated oil in place (OIP) in these forms is
about six trillion barrels, which is almost six times the total
conventional reserves1. The majority of these resources are
located in Venezuela, Canada and the united states2. Heavy oil
and bitumen are characterized by their high viscosities, and
low-degree API gravities5. In some Canadian reservoirs, such
as Athabasca, the oil viscosity is in the millions of mPa.s at
reservoir conditions. Consequently, this high viscosity rules
out the primary production and even in lower viscosity
reservoirs the primary recoveryis less than 10% of the original
oil in place (OOIP) 3,4. Many non-thermal improved oil
recovery (IOR) methods, have been tried such as
waterflooding, water-soluble polymers, etc., unfortunately, the
overall incremental recovery that can be achieved is very low.
In thermal recovery processes, namely, cyclic steam
stimulation (CSS), in situ combustion (ISC), steam-assisted
gravity drainage (SAGD), etc., the viscosity is reduced by
heating the reservoir. With CSS, the maximum recovery rarely
exceeds 20%. Currently, Steam-assisted gravity (SAGD) 6,7
has become a popular method for the recovery of heavy oil
and bitumen. There are several field projects currently in
operation using this principle. In this process the heat is
induced into the reservoir by injecting steam through a
horizontal well; steam condenses at the oil interface and heats
the oil. Consequently the viscosity is lowered and the hot oil
drains to another horizontal well by gravity.
Although thermal methods, especially SAGD, are
successful in exploiting these resources, the large heat
requirement can make them inefficient and uneconomic in
many reservoirs particularly those with thin pay zones, low
porosity, high water saturation, an overlying gas cap, vertical
fractures and/or fissures, low rock thermal conductivity and
reservoirs with aquifers. As an alternative, the Vapex process
is one of the few options for the recovery of huge resources
available in the form of highly viscous heavy oil and bitumen.
Vapex process is an emerging technology that has
recently gained considerable attention for its promising
application for the recovery of heavy oils and bitumen
reservoirs (Das, 1995; Jiang, 1997). This process involves a
drastic reduction of oil viscosity by diluting the oil with
vaporized hydrocarbon solvents (HCS). It was conceived by
Butler and Mokrys8, 9,10 as a solvent analogue of the steam
assisted gravity drainage (SAGD) process. The process
2 SPE 84868
normally involves horizontal well pairs, with an injection well
located directly above the production well as schematically
shown in Figure-1. The vaporized solvents, injected through
the injection well, dissolve into the bitumen and dramatically
reduce its viscosity. The diluted oil is then mobile enough to
drain down under the influence of gravity into the production
well located near the bottom of the formation, from where it is
pumped out to the surface. The pore space around the injection
well from where the oil has been drained out remains filled
with the solvent vapor and is referred to as “vapor chamber” in
Figure-1. This vapor chamber grows laterally as more oil is
drained out of the reservoir. The dissolution of solvent vapors
into the oil occurs at the walls of this growing vapor chamber
and the diluted oil flows downwards in a thin layer (drainage
edge) adjoining the vapor chamber.
Swept Zone
Solvent Vapour chamber
Mixing Zone Drainage
Edge
Inj
Undiluted Bitumen
Producer
Figure-1: Schematic Illustration of the Vapex Process
(Solvent Analogue to SAGD).
Many Experimental studies have been carried out with various
solvent types and using different conditions11, 12,13,14,15,16. The
experimental work done before indicates that; propane works
better than butane for the vapex process because of its higher
diffusivity; mixture of propane and butane perform better than
butane alone and work as almost well as pure propane11. Even
though ethane has high vapour pressure; it is not a good
solvent for the Vapex process due to its phase behavior15. Up
to now the focus of work on this process has been on the
application of light alkanes as the solvent. The most frequently
considered would be a mixture of methane and propane. The
economic viability of the process depends on maintaining a
low solvent to oil ratio and is sensitive to the cost of natural
gas and propane relative to the selling price of the produced
oil. The recent increase in the price of natural gas and propane
has somewhat diminished the attractiveness of the process.
Therefore it is desirable to find ways of reducing the solvent
cost in vapex process and one attractive option appears to be
to use CO2 as a major component of the injected solvent.
The main objectives of this paper involved designing and
performing laboratory experiments with mixtures of CO2 and
Propane in sand packed model. The same experiments were
repeated under the same conditions with a mixture of
methane-propane and the performances of both solvents were
compared.
Why CO2:
There is a general agreement that the increase in
atmospheric concentration of greenhouse gases (GHG’s), such
as CO2, methane, nitrous oxide, ect., over the last hundred
years presents a real threat to the environment. Therefore,
alternative solutions for coping with the increased GHG
emissions must be sought. Obviously, utilization of
greenhouse gases for improved oil recovery has significant
impact in reduction of these emissions. Having a high
absorption capacity and a very long atmospheric lifetime, CO2
has gained considerable importance over other greenhouse
gases and it is believed to have the greatest contribution to
global warming. One of the enhanced oil recovery applications
of CO2 that appears to be good in both economical and
environmental sides is the use of CO2 based Vapex for the
recovery of heavy oils and bitumen. CO2 is considerably more
soluble in heavy oil than methane and is known to provide
greater viscosity reduction on its own17. Therefore a mixture
of CO2 and propane could work even better than a similar
mixture of methane and propane. In the view of desirability of
reducing CO2 emissions into the atmosphere and the current
focus on sequestering CO2 in underground formations, it is
likely that low-cost CO2 will be available for oil recovery
applications. Hence a mixture of CO2 and propane would be
expected to be more cost-effective than the methane-propane
mixture for application in Vapex. Such use of CO2, if
successful, will not only reduce the operation cost of the
Vapex process but will also provide substantial environmental
benefits.
Experimental Set-up and Procedures
A simplified process flow diagram of the experimental
set-up is schematically shown in Figure-2. The major
components of this set-up are a cylindrical stainless packed
model with production and injection wells, a high pressure cell
with a transparent window along the length was used as gas-
oil separator, a positive displacement liquid propane injection
pump, a high pressure methane gas and /or carbon dioxide
cylinder, mass flow meter (MFM) for methane and /or carbon
dioxide vapor injection, a gathering facility for free and
solution gases, and collection bottles for oil samples.
In order to perform experiments at a high operating
pressure, the physical model (pressure cell) used in the
experiments described in this paper was cylindrical in shape
and fabricated with steel. The model as shown in Figure-3,
consisted of two coaxial cylinders, closed at both top and
bottom ends with circular stainless plates. The cavity of the
model is represented by the annular space between the two-
cylinders. The model was packed uniformly with technical
quality (12-16) U.S mesh glass beads, which gave a typical
permeability of about 640 Darcies and porosity of about 35 %.
The annular cavity of the model was 12 in. (30.48 cm) high
and 12.1 in. (30.73 cm) OD and 10.71 in. (27.2 cm) ID. A
digital photograph of the experimental set-up with all the
equipment used to perform the runs is shown in Figure-4. The
crude oil used in experiments had a viscosity of 3300 mPa.s at
the room temperature of 24oC and 0.9826 as specific gravity.
SPE 84868
The glass beads-pack cavity was first deaerated by
applying a vacuum, and then it was fully saturated with
degased water. The amount of water imbibed provided a
measure of the pore volume. To establish the initial oil
saturation and the irreducible water saturation in the model,
the water was displaced vertically downward with the oil. The
typical oil saturation in the model ranged from 94 to 95 % of
pore volume.
Before starting solvent injection, the model was
pressurized to the operating pressure by injecting either
mixture of solvent and non-condensable gas or liquid solvent.
To maintain the desired pressure drop between the injection
well and production well when production starts, and to ensure
that the driving force for the oil production was only the
gravity, the separator was pressurized to the operating pressure
through the bypass. Liquid propane was injected with a
positive displacement pump at a constant and pre-determined
rate (40 ml/h). The propane was mixed with a non-
condensable gas (carbon dioxide or propane) before injection
into the model. The pressure of the non-condensable gas was
adjusted and controlled by a regulator mounted on the high-
pressure cylinder. The amount of non-condensable gas
injected into the model was measured using a mass flow meter
(MFM) that was calibrated earlier to the experiment
conditions. During the run, the diluted oil with dissolved
solvent was produced from the well located at the bottom of
the model and accumulated in a 700-cm3 gas-oil separator.
The volume of live oil accumulated was monitored through
the glass window of the separator. A metering valve was used
to control the rate of the free gas and the volume produced was
monitored by collection in gas-metering cylinder via the
displacement of water.
Once the separator was almost full, the oil was
transferred to 700-cm3-collection bottle (sampling bottle) at
atmospheric pressure and the volume of solution gas released
from the collected sample was measured by another gas-
metering cylinder. All the experiments were run continuously
for 9 hours. At the end of each experiment, the model was
blown down to atmospheric pressure to liberate the dissolved
solvent and gas mixture remaining in the pressure cell. To
determine the amount of dead oil produced, all the oil samples
collected were allowed to degas at atmospheric pressure and
weighed five days after each experiment.
Experiments Performed
This paper presents the results of four experiments. The
first run was performed at an operating pressure of 250 psig
(263.1 psia), using carbon dioxide as the non-condensable gas,
which was mixed with the propane to form the injected solvent
into the model. The same experiment was later repeated under
the same conditions except for the non-condensable gas,
which was replaced with methane (run # 2). Expt.4 and
Expt.#1, as well as Expt. #3 and Expt. #2 are similar except
for the operating pressure, which was increased from 250 to
600 psig (613.1 psia). All experiments were conducted at
room temperature. A description of the experiments done is
given in Table 1.
3
Results and discussions
The effect of the non-condensable gas/carbon dioxide
mixed with propane on the performance of the vapex process
was investigated and compared to methane-propane mixture.
Four different runs were conducted in a physical sand-packed
model at a two different operating pressures where two cases
were studied. The data set recorded in each experiment
included, solvent and non-condensable gas (carbon dioxide or
methane) injection rates, free and solution gas production
rates, and oil production rates.
Case1:
Case1 includes run #1 and run #2 where the operating
pressure was 250 psig (263.1 psia) for both runs. In run#1,
carbon dioxide from the cylinder was combined with the
liquid propane from the pump discharge and the vapor mixture
was injected into the model. Run #2 was conducted using a
mixture of methane and liquid propane to form the injected
solvent under the same condition of pressure and temperature
(room temperature). The results of these run are summarized
in Table-2.
Figure-5 shows a comparison of the cumulative injected
amount of non-condensable gases between run#1 and run #2.
It shows that the amount of CO2 injected in the first run is
higher than the amount of methane injected in second run.
This difference can be explained by the high solubility of CO2
in bitumen compared to the solubility of methane and higher
free gas production in the CO2 run. Comparisons of oil rate
production and cumulative oil produced between CO2
experiment (run #1) and the methane one (run #2) are
presented in Figure-6 and Figure-7 respectively. With the
same operating pressure (250 psig) and the same average room
temperature (24.55 0C), the average oil production rate is
about 80 ml/h (± 5) for both experiments. The final oil
recovery for methane experiment was slightly higher (2%)
than CO2 experiment. At this operating pressure, CO2 did not
perform better than methane, but the use of CO2 can be
justified by its cost effectiveness and environmental purposes.
Case2:
To investigate the effect of high pressure for both
solvents (CO2-Propane mixture and Methane-Propane
mixture) on the performance of the process, the operating
pressure was increased from 250 psig to 600 psig (613.1 psia).
Under the same experimental conditions as in run#1 and
run#2, Run#3 and run#4 were conducted using the new
operating pressure. Methane-Propane mixture was used in
run#3; whereas CO2-Propane mixture was used in run#4.
Figure-8, Figure-9 and Figure-10 present successively
comparisons of the cumulative injected amount of non-
condensable gases, oil production rate and the cumulative oil
produced between run#3 (Methane) and run#4 (CO2). Even
though the room temperature in CO2 run was 2oC less than
methane run, the cumulative oil production was higher by
about 25% in run #4, i.e. the final oil recovery was 11% less in
methane experiment (run #3) compared to CO2 one (run #4).
This promising result can be explained by the high solubility
4 SPE 84868
of CO2 in bitumen at the high pressure, which leads to more
reduction of the crude viscosity.
It is important to note that the final oil recovery for high
operating pressure (600 psig) was less than the final oil
recovery for lower operating pressure (250 psig) for both
methane-propane and CO2-propane mixtures. This
phenomenon can be explained by the high injection rate of the
non-condensable gas (in high operating pressure), which
decreases the concentration of propane in the solvent mixture.
As a result, the concentration of solvent at vapor-oil interface
and the mass transfer rate by diffusion are reduced. Because of
the higher fraction of non-condensable gas dissolving in the
oil, the propane solubility, at the same partial pressure of
propane in the gas, is reduced.
Conclusions
1. At high operation pressure, CO2-propane mixture
shows better performance in vapex process compared
to methane-propane mixture.
2. At lower operating pressures, the cost effectiveness
and environmental purposes can justify the use of
CO2 in the vapex process rather than methane.
References
1. Janisch, A.: “Oil Sands and Heavy Oil: Can They Ease the Energy
Shortage,” First United Nations Inst. for Training and Research
(UNITAR) Conference, Edmonton, Alberta (4–12 June 1979), The
Future of Heavy Oils and Tar Sands, McGraw Hill, New York
(1981), 33–41.
2. Stosur, G.J.: “Heavy Oil Recovery in The Low Oil Price Regime,”
Proc., Sixth UNITAR Intl. Conference on Heavy Crude and Tar
Sands, Houston (12–17 February 1995) 3–9.
3. Borregales, C.J.: “Production Characteristics and Oil Recovery in
Orinoco Oil Belt”, Proceedings of The 1rst UNITAR Conference on
Heavy Crude and Tar Sands, Edmonton, Alberta, Canada, June 4-
12, 1979 pp498-50.
4. Christopher, J.E. and Knudsen, R.H.: “Heavy crude Oil Potential
of Saskatchewan”, Proceedings of The 1rst UNITAR Conference on
Heavy Crude and Tar Sands, Edmonton, Alberta, Canada, June 4-
12, 1979 pp61-68
5. Speight, J.G., 1991 Chemistry and Technology of Petroleum, 2nd
edn. Marcel Dekker, New York, pp.316-325.
6. Butler, R.M., Stephens, D.J., 1981. “The Gravity Drainage of
Steam-Heated Heavy Oil to Parallel Horizontal Wells” J. Can. Pet.
Techno. (April-June), 90-91.
7. Butler, R.M., McNab, G.S., Lo, H.Y., 1981 “ Theoretical Studies
on the Gravity Drainage of Heavy Oil during Steam Heating”. Can.
J. Chem. Eng. 59, 455-460.
8. Butler, R.M., and Mokrys. I.J., “Solvent Analog Model of Steam-
Assisted Gravity Drainage”; AOSTRA Journal of Research. Vol. 5
pp. 17-32, 1989.
9. Butler, R.M., and Mokrys. I.J.,”A New Process (VAPEX) for
recovering Heavy Oils Using Hot Water and Hydrocarbon
Vapour” J. of Can. Pet. Tech. Vol.30 No.1 pp.97-106. January-
February 1991.
10. Mokrys. I.J. and Butler, R.M.,”In Situ Upgrading of Heavy Oils
and Bitumen By Propane Deasphalting: the Vapex Process”; SPE
Paper No 25452, SPE Productions Operations Symp. Oklahoma
City, Oklahoma. U.S.A., March 21-23, 1993.
11. JIANG.Q. “Recovery of Heavy Oil and Bitumen using Vapex
Process In Homogeneous and Heterogeneous Reservoirs”; PHD
thesis, University of Calgary, Calgary, AB, March 1997.
12. Butler, R. M. and Jiang, Q.: “Improved Recovery of Heavy Oil by
Vapex with Widely Spaced Horizontal Injectors and Producers”,
Journal of Canadian Petroleum Technology, vol.39, no.1, January
2000.
13. Das.S.K., and Butler.R.M., “ Investigation of Vapex Process in a
Packed Cell Using Butane as a Solvent”; Paper No. HWC 94-47,
presented at the Canadian SPE/CANMAT International
Conference on Recent Advance in Horizontal Well Application,
Calgary, AB, March 20-23, 1994.
14. JIANG.Q. and Butler.R.M., ”Experimental Studies on Effects of
Reservoir Heterogeneity on Vapex Process” Journal of Canadian
Petroleum Technology, volume 35, No. 10 December 1996.
15. Butler, R.M., and Mokrys. I.J., and Das.S.K.,”The solvent
Requirement for Vapex Recovery” SPE30293, presented at the
International Heavy Oil Symposium, Calgary, AB, June 19-21.
1995.
16. Karmaker, K. and Maini, B. B.: “Further Investigation of the
Vapex Process for Heavy Oil and Bitumen Recovery”, presented at
the Petroleum Society’s Canadian International Petroleum
Conference 2003, Calgary, Alberta, Canada, June 10-12, 2003.
17. Sankur,V.,Creek, J.L., DiJulio, S.S and Emanuel, A.S.: “A
Laboratory Study of Wilmington Tar zone CO2 Injection Project”,
SPE Reservoir Engineering, Jan. 1986, 95-104.
18. Das S.K., “VAPEX: an efficient Process for the Recovery of
Heavy Oil and Bitumen”, SPE J., Sept, p.232-237, (1998).
19. JIANG.Q. and Butler.R.M., ”Selection of Well Configurations in
Vapex Process”, SPE No. 37145 , Horizontal Well tech. Calgary,
Canada, 18-20 Nov. 1995.
20. Das S.K., “ In-Situ Recovery of Heavy Oil and Bitumen Using
Vaporized Hydrocarbon Solvents”, PhD. Thesis, the university of
Calgary, Alberta, Canada, (1995).
21. Das S.K., Butler, R.M., “ Mechanism of The vapor Extraction
Process for Heavy oil and bitumen”, J. Pet. Sci. Eng., 21, p.43-59,
(1998).
22. Das S.K., Butler, R.M., “Effect of Asphaltene Deposition on The
Vapex process: A Preliminary Investigation Using a Hele-shaw
Cell”, JCPT, Vol.33, No.6. June, p.39-45, (1994).
23. Butler, R.M., and Mokrys. I.J., ”Recovery of Heavy Oils Using
Vaporized Hydrocarbon Solvents; Further Development of The
VAPEX process”, JCPT, Vol.32, No.6, June, (1993).
24. Bachu S., “Disposal and sequestration of CO2 in Geological
Media”, AEUB-Alberta Geologic Survey, September, (1999).
SPE 84868 5
6 SPE 84868

M.
F.
Mixture
M
Injection

Cylindrical steel
Packed
Model CO2 / Or
CH4
Pump

Bypass Propane
Free Sol.
Gas Gas

Vacuum Separator
Pump

Collection
Bottle

Figure-2: Schematic Illustration of the Experimental Set- Up used.

Figure-3: Photograph Showing the Shape and the Dimension of the Sand-Pack High-Pressure Stainless
Model Used in the Experimental Work.
SPE 84868 7

Figure-4: A Digital Photograph of the Experimental Set-up with all the Equipment Used to Perform the Experiments.
8 SPE 84868
SPE 84868 9