Method and apparatus for stimulating heavy oil production

CA 02299790 2005-09-01

- 1 Title: METHOD AND APPARATUS FOR STIMULATING HEAVY

OIL PRODUCTION

FIELD OF THE INVENTION

This invention relates to the extraction of hydrocarbons such as heavy oil and bitumen. In particular this invention relates to reducing the viscosity of hydrocarbons such as heavy oil in situ to permit the heavy oil to flow more readily and thus to improve the recovery thereof.

BACKGROUND OF THE INVENTION

Heavy oils refer to crude oils which have high specific gravity and viscosity and are therefore difficult to extract commercially because they do not readily flow. Heavy oils are found, for example, in the tar sand deposits in Alberta, Canada. Typically these heavy oils will have viscosities greater than 1000 centiPoise or specific gravities greater than .934 at 60°F (i.e. less than 20° API). There has long been sought a means to accelerate the heavy oil production process by permitting the oil to flow more readily thereby increasing the rate of return on capital and decreasing the financial risk of such heavy oil production projects.

One thermal extraction technique, called fireflood, is generally uneconomic due to very severe operating problems including corrosion, scale precipitation and explosion hazards after breakthrough, not to mention the difficulty in controlling the process and the production of plugging deposits such as coke.

Another prior approach that has had some merit is to use steam in a thermal stimulation for improving heavy oil extraction. Steam raises the temperature of the oil and thereby reduces its viscosity and allows it to flow more easily. Steam stimulation is subject to a number of problems, including heat losses during injection, clay swelling problems, thief zones, emulsions, capillary surface tension effects and lack of confinement for

CA 02299790 2005-09-01

-2shallower zones. Further, injecting steam creates water (condensate) in the formation which is much less viscous than oil and which will therefore be preferentially produced due to relative permeability effects. Preferential production of water perversely makes the oil production or recovery more difficult.

An additional problem, which has become more important recently, is that most thermal recovery processes such as steam require large amounts of methane gas to be burned to provide the energy to vaporize the water above grade. This can lead to the emission of enormous amounts of greenhouse gases such as carbon dioxide. For example a 100,000 bbl oil/day heavy oil facility requires 200,000 - 300,000 bbl water /day to be converted into steam at 200°C. Thus, for a methane gas burner system, to recover 100,000bbl oil/day requires producing more than 12 million pounds per day of carbon dioxide emissions. The two main traditional approaches used in steam recovery systems have been huff and puff (i.e., cyclic steaming) and steam floods. Recently, however, steam assisted gravity drainage (SAGD) has become popular.

SAGD begins with the formation of a steam chamber in the formation. The steam is injected into the chamber and transfers heat to the surface of the chamber thereby mobilizing oil at the chamber surface. The heated oil flows down the walls of the chamber under the influence of gravity and drains into the producing well, thereby increasing the size of the chamber. The advantage of SAGD is that the countercurrent flow of steam upwards into the reservoir and oil down and out of the reservoir is relatively efficient, thus the heavy oil production rates are high enough to provide favourable economics in some situations.

There are many possible SAGD geometries including single well (injection and production from the same well) and dual or multiple well. The wells may be either horizontal or vertical. Generally horizontal wells are favoured by producers because they offer a longer exposure to the pay zone and thereby offer increased production rates for highly viscous oils.

CA 02299790 2005-09-01

-3Single well SAGD offers the least capital cost, but heat losses due to countercurrent flow of steam into and oil out of the wellbore are severe. Quite simply, as the hot steam going into the well passes the cold oil coming out of the well and the steam loses heat to the oil. For example, at an injection pressure of 1000 psig and 285°C, the enthalpy of the steam is 1192 btu/lb and the enthalpy of the water is 542 btu/lb. Due to countercurrent heat exchange the produced fluids (water and oil) are at the same temperature as the injected steam. For typical injection conditions, the steam quality is 80% (i.e., 80% vapour and 20% liquid). Thus, the maximum heat delivered to the formation is only the latent heat of vaporization (i.e. about 50% of the total heat input). With additional heat losses through the well casing, the net heat delivery to the formation is quite low and thus this process is inefficient.

There have also been in the past suggestions to use cold solvent vapour to lower the viscosity of the heavy oil in situ. This was first proposed by Nenniger¹ (1979). This idea has shown much promise for production of heavy oil with minimal environmental impact, primarily because such a process does not require heating large volumes of steam nor huge amounts of fresh water suitable for steam generation. Energy requirements for solvent extraction are expected to be less than 4% of those required for steam extraction. Insitu recovery has minimal environmental impact compared to surface mining techniques.

The physics of cold solvent stimulation are not fully understood. The measured solvent diffusion rates are typically 100 -1000 times higher than predicted by theory²,³. A key economic requirement is

Nenniger, E.H., Hydrocarbon Recovery, Canadian Patent 1,059,432 ² Dunn, S.G.; E.H. Nenniger, V.S.V. Rajan, A Study of Bitumen Recovery by Gravity Drainage Using LowTemperature Soluble Gas Injection, The Canadian Journal of Chemical Engineering, Vol 67, December 1989.

³ Lim, et al, Three dimensional Scaled Physical Modelling of Solvent Vapour

Extraction of Cold Lake Bitumen, JCPT, April 1996, Page 37

CA 02299790 2005-09-01

-4efficient recovery of the solvent, so light gases such as ethane and propane which can be recovered by pressure blowdown are generally preferred. A recent study has reported the ratio of ethane solvent loss to bitumen produced, was as low as seven percent (wt/wt). However, the calculated production rates for solvent extraction are marginal for commercial application and to date there has never been a successful commercial pilot.

In a bench test⁴ warm solvent (propane) was injected into a sample of warmed heavy oil. This experiment showed that if the solvent temperature was raised and the heavy oil temperature was also raised to the same temperature (ie. Isothermal conditions) production rates could be increased about 20 fold simply by increasing the temperature from 20°C to 90°C.

This observation led to the development of the Vapex process⁴which proposes to combine solvent with steam or hot water heated above grade to provide downhole heat. Because of the water/steam this process suffers from all the problems mentioned above (countercurrent heat exchange, formation damage problems with clays, emulsions, capillary pressure, water treatment, water supply, reduced oil relative permeability due to high water saturations and the like).

A key requirement for both steam assisted gravity drainage and solvent assisted gravity drainage is the formation of a steam or solvent chamber in the reservoir. The chamber allows efficient countercurrent flow of solvent vapour (or steam) upwards . and flow of the heavy crude downwards along the walls of the chamber. The predicted oil drainage rate is proportional to the square root of the height of the chamber (reference 4). Thus the oil production rates are predicted to be very small initially and then grow with time until the roof of the chamber encounters a boundary such as an impermeable shale.

This has been confirmed by lab tests which have shown that ⁴ See Table 1 and Figure 7 of Buder et al, A New Process lor Recovering Heavy Oils using Hot Water and Hydrocarbon Vapours, JCPT Jan 1991, pg 100

CA 02299790 2005-09-01

-5the maximum oil production rates will not occur until a large solvent chamber is formed. Unfortunately, in the field this means that peak oil production rates do not occur until 3-4 years after the well is placed on production.

Thus, for solvent vapour extraction the peak oil production rates are not typically achieved until perhaps three years after the capital costs of the well and the production facilities are incurred. The delayed production response decreases the rate of return and increases the risk to the operator. For example thief zones, etc, may not be identified until substantial costs have been incurred (i.e. until after three years of solvent injection).

Thus, there is a need for the solvent chamber to be quickly established. For example, the capital cost of drilling and completing a horizontal well pair might be typically 1,800,000 dollars. The minimum internal rate of return for a oil project is typically about 15%. Thus, the opportunity cost of a one year delay in the peak production rate is 275K$. If peak production is accelerated, so it occurs in the first year rather than the third, then the value added by early development of the solvent chamber would be about 800K$ per well pair.

Thus, while the cold solvent vapour extraction process has great advantages due to energy efficiency and minimal environmental damage, it has never been successfully used. The primary reason is the cold solvent vapour production rates are too low to be economic, particularly with a 3 - 4 year delay in achieving peak production rates. Another way of looking at this issue, is to apply a discount to value of the produced oil if the production is delayed. At 15% rate of return, the 3 year delay gives a discount of 33%, so the value of the oil production is reduced by 1/3. In other words, if the market price of oil is 20$/bbl, the effective price the producer receives is only 14$/bbl, due to the three year delay. Obviously this delayed startup has a huge negative impact on the commercial feasibility of this environmentally friendly technology.

What is desired is a way of stimulating production of heavy oil

CA 02299790 2005-09-01

-6which is energy efficient and yet is effective. In this respect it should not require the use of very high temperatures or high energy use rates as is the case presently. Further, it would be preferable to avoid introduction of steam or water into the formation which has negative effects on the production rates.

SUMMARY OF THE INVENTION

What is desired therefore, is a means to accelerate the oil production rate by encouraging the rapid extraction of heavy oil or bitumen. According to the present invention it is possible to accelerate the extraction process by the injection of heated solvent vapor into the reservoir in the absence of a water/steam phase under certain predetermined conditions. As the solvent condenses on the cold bitumen surface it supplies heat to the bitumen interface, by releasing the latent heat of condensation, and greatly accelerates the extraction without the problems associated with a liquid water phase. Furthermore, by using solvent condensation as a heat transfer mechanism, it is possible to significantly increase the proportion of solvent in the bitumen solvent blend, thereby reducing blend viscosity, improving drainage rates (production) and also achieving enhanced insitu upgrading of the oil. Further according to the present invention countercurrent heat exchange losses can be avoided by injecting the heated solvent from an injection well and removing the produced fluid from an adjacent well which is communication with the injection well. Thus, the present invention contemplates establishing such a connection between the production and injections wells prior to injecting a surface heated solvent vapor.

The present invention also takes into consideration various additional factors such as the kinetics of extraction, hydraulics and heat transfer for hot gas delivery to the reservoir and recovery and recycle of solvent from the produced fluid.

Accordingly, in the present invention, there is a provided a

CA 02299790 2005-09-01

-7method of enhanced heavy oil recovery, said method comprising the steps of:

a) establishing a flow path between an injection well and a production well through an oil bearing formation;

b) vapourizing a solvent to inject into said formation;

c) pressurizing said formation sufficiently to raise a condensation temperature of said solvent above an original formation temperature;

d) delivering said vapourized solvent to said formation to permit said solvent to condense within said formation, and to release the latent heat of condensation to said formation at said condensation temperature; and

e) extracting heavy oil from said production well.

According to another aspect of the present invention, there is provided a method of enhanced heavy oil recover, said method comprising:

a) selecting a solvent having a predetermined latent heat of condensation;

b) selecting a predetermined amount of heat to deliver to a formation;

c) vapourizing said selected solvent to permit said solvent to be delivered to said formation in a vapour state;

d) delivering said solvent to said formation at a rate sufficient to deliver the predetermined amount of heat;

e) condensing said solvent onto said formation to deliver said predetermined amount of heat to said formation; and

f) recovering, from said formation, a solvent/heavy oil blend.

According to another aspect of the present invention, there is provided a method of enhanced hydrocarbon recovery comprising the steps of:

a) injecting a vapourized condensing solvent into a formation;

b) condensing said solvent in said formation to deliver a latent heat of condensation to said formation to heat said hydrocarbon to reduce a viscosity of said hydrocarbon;

CA 02299790 2005-09-01

c) dissolving said solvent into said hydrocarbon to form a solvent/hydrocarbon blend having a further reduced viscosity; and

d) recovering from said formation said reduced viscosity hydrocarbon blend.

According to another aspect of the present invention, there is provided a method of improved hydrocarbon recovering comprising:

a) heating said hydrocarbon to be recovered insitu in the presence of a solvent to increase the solvent penetration rate into said hydrocarbon.

According to another aspect of the present invention, there is provided a method of increasing the value of a recovered hydrocarbon comprising the steps of:

a) injecting a condensing solvent vapour into a formation;

b) said solvent vapour at a pressure above its condensation pressure (at original reservoir temperature);

c) condensing said solvent into said heated hydrocarbon thereby providing a mobile hydrocarbon fraction; and

d) extracting said mobile hydrocarbon fraction from said formation; whereby asphaltenes remain in said formation in a substantially immobilized hydrocarbon fraction.

According to another aspect of the present invention, there is provided a method of enhanced recovery of hydrocarbon from a hydrocarbon bearing formation said method comprising the steps of:

a) establishing a flow path between an injection well and a production well;

b) injecting a heated solvent into said formation under sufficient pressure to raise a condensation temperature of said solvent above a formation temperature;

c) heating said formation by condensing said solvent onto said formation to help remove hydrocarbons from said formation and to create a chamber;

d) increasing the size of the chamber by continuing to inject heated

CA 02299790 2005-09-01

-9solvent;

e) reducing the injection pressure as said chamber expands; and

f) reducing the energy delivered to said formation as said pressure is reduced.

According to another aspect of the present invention, there is provided a method of enhanced hydrocarbon recovery comprising the steps of:

a) injecting a heated solvent into a formation under sufficient initial injection pressure to raise a condensation temperature of said heated solvent above a naturally occurring formation temperature; and

b) reducing, over time, said injection pressure from said initial pressure.

BRIEF DESCRIPTION OF THE DRAWINGS

Reference will now be made, by way of example only, to preferred embodiments of the invention as illustrated in the accompanying drawings and in which:

Figure 1 illustrates a process schematic of the present invention showing formation of a solvent chamber;

Figure 2 illustrates the solvent chamber along section A-A of Figure 1 in more detail;

Figure 3 is a graph which shows a relationship between viscosity and temperature for Athabasca bitumen, and the predicted relationship between diffusion rate and temperature based on the StokesEinstein equation;

Figure 4 is a graph which illustrates a relationship between temperature rise and volume of a theoretical reservoir heated at a constant power rate of 1 megawatt;

Figure 5 is a graph which illustrates the vapour pressure of propane solvent as a function of temperature;

Figure 6 is a graph which shows the latent heat of vaporization

CA 02299790 2005-09-01

-10for propane solvent as a function of temperature and the mass of propane solvent vapour required to deliver one megawatt of heat (via latent heat of condensation);

Figure 7 is a graph which shows the volumetric heat capacity of vapour (via latent heat of condensation) as a function of temperature for several solvents compared to steam;

Figure 8 is a graph which shows volume fraction of propane solvent in produced fluid vs chamber temperature;

Figure 9 illustrates the bitumen- propane blend viscosity at 8C as a function of propane solvent volume fraction and the favorable reduction in viscosity at higher solvent ratios;

Figure 10 illustrates the propane solvent/bitumen blend viscosity as a function of temperature; and

Figure 11 illustrates the extraction rate for the heated propane solvent vapour as a function of temperature and how the rate is limited by mass transfer at temperatures below 40C and limited by heat transfer at temperatures above 40C.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

Figure 1 shows a schematic of a process of stimulating heavy oil or bitumen recovery according to the present invention. Generally, a hot solvent 10 is injected down an injection well 12 into a reservoir 14. The hot solvent 10 is most preferably a vapour, enters a solvent chamber 16 through a perforated or slotted casing 18 or the like and flows out to condense on the cold bitumen interface 20 to form a solvent/bitumen blend. The terms “bitumen” and “heavy oil” are used interchangably in this specification and for the purposes of this invention means hydrocarbons which are recovered from naturally occurring formations and which in their natural state are generally too viscous to readily flow into a production well. It will be appreciated that the present invention is most suitable for such formations as tar sands, but may also be used in other formations.

CA 02299790 2005-09-01

- 11 The solvent bitumen blend 15 formed at the interface drains to the bottom of the chamber 16 (shown at 22 in Figure 2), where it is removed via a production well 24 and produced to surface 26. Valves 17 are located at each well head. The bitumen is separated from the solvent at the surface 26 and the bitumen is sold 27. The separation at 29 of a solvent such as propane from the bitumen, might involve a flash at a temperature above the critical temperature of the solvent. The present invention comprehends that there may be several stages of separation to maximize solvent recovery, which of course will minimize solvent losses in the sold bitumen. It will be appreciated by those skilled in the art that some factors to consider in establishing solvent recovery are energy efficiency, reliability, and potential for fouling problems (i.e. deposition of asphaltenes). The recovered solvent 33 is then compressed, and/or heated at 32 and then reinjected into the injection well 18. Additional make up solvent is added as needed to replace the void volume created by the extracted bitumen at 30. It may also be necessary to remove light gases from solvent/bitumen blend which may have been co-produced from the reservoir. These may also be used as fuel in re-heating the solvent.

The present invention comprehends a process in which a flow path has already been established between an injection well and a production well. This flow path could be established by any of a number of means including downhole heaters or the like. The establishment of a flow connection is desirable because this avoids countercurrent heat losses which might otherwise occur. However, this step is not deemed essential if such countercurrent heat losses can be mitigated through use of other strategies such as insulated injection tubing or the like. It will be appreciated though that the most preferred form of the present invention is a flow through from an injection well to a production or recovery well.

Figure 2 shows the solvent chamber 16 formed in this formation in more detail. Also shown is a pressure containment layer, such as shale barrier layer 21. The heated solvent vapour rises within the chamber 16 to condense on the walls and roof 19 of the chamber 16. As the solvent

CA 02299790 2005-09-01

-12condenses it releases its latent heat of condensation thereby heating the bitumen interface at the chamber surface. As the solvent dissolves and is mixed into the bitumen the bitumen is upgraded by the precipitation out of asphaltenes. At this stage the bitumen begins to flow as its viscosity has been lowered by two effects namely, the heating effect from the latent heat of condensation and the dilution effect from being blended with the now liquid solvent. The bitumen- solvent liquid blend 25 drains along the wall or down off the ceiling into the sump 22. The liquid is then drained into the production well 24. As will be more fully understood from the description below, the production of bitumen solvent blend is preferably restricted to avoid solvent gas bypassing. This is accomplished via a steam trap type control as currently practiced in SAGD technology.

Figure 3 shows the viscosity of a typical Athabasca bitumen as a function of temperature by way of example. The Stokes-Einstein law states that the diffusion coefficient for any solvent is inversely related to the solute viscosity. Using this relationship an estimate can be made of the improvement in the diffusion coefficent as the temperature is increased and the bitumen viscosity decreases. For example, at 40C the diffusion coefficient is increased by 100 fold above that at 8C (i.e. original reservoir temperature).

The thermal diffusivity in the Athabasca tar sands is typically about 100 times larger than the molecular diffusivity at 8C. Thus, Figure 3 shows that the heat transfer process becomes the rate limiting process step at temperatures above 40C, while the molecular diffusion will be the rate limiting process step at temperatures below 40C.

Figure 4 shows the volume of reservoir heated per day with a power delivery rate of 1 megawatt. This figure illustrates a simple heat balance and does not reflect any heat transfer limitations. As a point of reference 1 megawatt will heat 600m3/day of reservoir from 8C to 70C. Assuming a recovery rate of about 80% recovery of the original bitumen in place (assuming 35% porosity and 85% bitumen saturation) 1 megawatt will provide 140 m3/day of bitumen production at 70C.

CA 02299790 2005-09-01

-13Figure 5 shows the vapour pressure of one preferred solvent, propane, as a function of temperature. As can now be appreciated, the present invention comprehends enhancing the delivery of heat to the heavy oil insitu by increasing the pressure of the solvent vapour which in turn 5 increases the dew point temperature. As the vapour pressure is increased the dew point temperature increases. Above the critical temperature a separate liquid phase ceases to exist, so the vapour pressure concept no longer applies. By way of example, assuming that the solvent used is propane, at 70C the vapour pressure of propane is about 375 psia. This means that if the 10 solvent chamber is pressurized to 375 psia, then liquid propane will condense on any surface which is at a temperature below 70C. This condensation will eventually heat the surface (via the latent heat of condensation), to a temperature approaching 70C. Conversely if the target temperature was 40C, then the pressure of propane in the chamber would have to be held only at 15 about 200 psia. Thus, according to the present invention by pre-heating a solvent under predetermined pressure and injecting the same into a formation, a predetermined amount of heat can be delivered to a formation by controlling the injection rate of heated solvent vapour.

Figure 6 shows the latent heat of condensation for propane as 20 a function of temperature. As the temperature approaches the critical temperature the latent heat of vaporization drops to zero. Figure 6 also shows the metric tons of propane required per day to supply 1 megawatt via the latent heat of condensation. At 70C about 350 metric tons of propane vapour per day are required to supply one megawatt of heat. Thus, according to the 25 present invention heat can be delivered at a predetermined rate to the hydrocarbon bearing formation by latent heat of condensation. As will now be appreciated with appropriate pressure maintenance such a heat delivery mechanism avoids many of the problems of the prior art.

Figure 7 compares the latent heat of condensation as a function 30 of temperature for several solvents and water. The latent heat is presented on a volumetric basis (i.e. per m3 of saturated vapour at temperature and

CA 02299790 2005-09-01

-14pressure). The saturation pressure is the same thing as the vapour pressure and can be obtained from Figure 5. On this basis, propane at 70C has a latent heat content comparable to steam at 180C. Ethane has an even higher heat content but is not as useful due to its low critical temperature. Figure 7 5 also shows that butane would be useful if one wanted to achieve a reservoir temperature between 85 and 115C. While ethane, butane and propane are all possible solvents, many other solvents could also be used without departing from the present invention. Essentially for the purposes of this invention, the term solvent means any material which mixes with oil in a liquid 10 phase and which can be injected into a formation as a gas to deliver a latent heat of condensation to the formation. Solvents which are substantially miscible with the hydrocarbon or bitumen are preferred. By way of example, light volatile hydrocarbons such as propane, propylene, butane, ethylene, ethane and pentane are most preferred. While many solvents are available, 15 the most preferred ones will have a dew point temperature above a formation temperature at reasonable operating pressures (i.e. below formation orcasing fracture pressures).

Returning again to propane, Figure 8 shows the volume fraction of propane in the bitumen propane blend as a function of temperature. This 20 graph was derived from Figures 4 and 6, which show bitumen production and solvent injection rate at 1 megawatt of heat delivery. Figure 8 shows a great advantage of the present invention, namely that solvent proportion in the blend can be increased by operating at higher temperatures. This increased solvent proportion at high temperatures is made possible because the solvent 25 circulation rate is determined by heat transfer requirements rather than solubility in the bitumen under those conditions. In other words, to deliver the desired rate of heat transfer involves injecting enough solvent under pressure to provide the predetermined heat. This higher injection rate leads to a higher solvent fraction in the produced blend, with a beneficially lowered blend 30 viscosity.

Figure 9 shows the blend viscosity at 8C as a function of

CA 02299790 2005-09-01

-15propane volume fraction. It is clear that higher solvent proportions in the blend are very advantageous in terms of reducing viscosity. As the solvent proportion increases the viscosity of the blend decreases quite rapidly. This low blend viscosity provides rapid drainage of the bitumen from the chamber interface and exposes fresh cold bitumen to fresh hot condensing solvent vapour. Figure 9 also shows the approximate viscosity range expected for a typical VAPEX solvent/oil ratio at bracket 100 and the preferred much lower approximate viscosity range preferred for the present invention at higher solvent oil ratios at range 102.

Figure 10 shows the blend viscosity as a function of temperature. At 70C the blend viscosity is reduced by at least 10 fold over blend viscosity at original reservoir temperature. This again increases extraction relative to an unheated or ambient process.

Consider the rate of bitumen extraction with warm solvent vapour according to the present invention. Making a determination of this rate in advance is complicated because factors to simultaneously consider include heat, mass and momentum transfer in a porous medium. Furthermore, the measured mass transfer rates (diffusion coefficients) for cold solvent vapour extraction are higher than predicted by theory. Therefore the calculation which follows is an approximation only.

Consider temperatures above 40°C where the molecular diffusivity is higher than the thermal diffusivity. Assuming the process is limited by thermal diffusivity it is possible to model the process as a solvent SAGD with appropriate adjustments to the viscosity and permeability. Butler (Canadian Patent 1,130,201, pg. 19) gives a formula which states that the rate is proportional to (k/u)^%. =(k*p/p)^1/2. O’Rourke, J.C. (Canadian Journal of Petroleum Technology, Sept. 1999, pg, 50, Fig. 5.1) reports that the SAGD extraction rate at 200°C is about 5cm/day.

A condensing propane flood will increase permeability k by 4-5, because there is no relative permeability reduction due to high water saturations from steam (see Table I on pg. 14 of Butler Canadian Patent

CA 02299790 2005-09-01

-161,130,201 ). Δρ is reduced by ¹Z> due to the lower density difference between condensed and vaporized propane relative to water and steam (i.e. 0.5 for propane vs 1 for water). The blend viscosity μ at 40°C is 0.3cP vs 10cP for steam at 200°C. Therefore, the production rate using solvent vapours at 40°C is predicted to increase by (4*0.5*10/0.3)¹⁷² = 8 above the rate for SAGD at 200°C.

With the SAGD extraction rate of 5 cm/day at 200°C (where cm/day equals the distance the steam chamber expands), one can predict a hot pressurized propane solvent vapour extraction rate according to the present invention of about 8x5 = 40cm/day. Thus, the present invention, with condensing propane in gravity drainage solvent extraction process can give bitumen production rates about 8 times larger than a SAGD, with about 1/6 of the energy requirement of a SAGD (due to the lower reservoir temperature 40°C vs 200°C) and 1/6 of the greenhouse gas emissions. Furthermore, the produced oil will more valuable due to the insitu upgrading (i.e. loss of undesirable asphaltenes).

Figure 11 shows the extraction rate as a function of temperature for a heat transfer limited case for propane. The formation extraction temperature is shown ranging from 10 degrees C to 80 degrees C in ten degree increments. As noted earlier, the mass transfer rate via molecular diffusion will be limiting at lower temperatures, so a different mechanism occurs at temperatures below 40°C. Dunn et al. (Canadian Journal of Chemical Engineering, Vol. 67, December 1989, pg. 979) present an analogous equation for the case where mass transfer is limiting. In this case the rate is proportional to (D/u)^1/2.=(D*p/p)’^/2.

Assuming that the extraction rate at 40°C is the same for the heat transfer limited case above (i.e. 40cm/day) as the mass transfer rate limited case (since the rates must converge to the same value at some temperature). The variation in D (diffusion rate) is known from Figure 3. The blend viscosity is known from Figure 10. Figure 11 also shows the predicted extraction rates for temperatures less than 40°C where the mass transfer is

CA 02299790 2005-09-01

-17the rate limiting step. This low temperature part of the curve is very steep, due to the relationship between viscosity and diffusion coefficient. Thus a relatively small increase in the solvent vapour chamber temperature can increase the extraction rates significantly.

It will now be understood that as the chamber grows in size, the requirements for solvent, (such as propane vapour) delivery will rapidly increase due to the increased surface area if the temperature is to be maintained. The ability to deliver hot vaporized propane to the injection well may become rate limiting. To some extent, the solvent vapour delivery can 10 be improved by injecting at higher pressures and temperatures. However, this will require very high bitumen-propane separation rates in the surface facilities.

For example, consider a heat delivery of 1 megawatt at 70°C. This requires 350 metric tons per day of propane vapour delivered to the 15 reservoir. At saturation pressure of 375 psia at 70°C, the propane vapour requirement is about 8800m3/day. This gives a velocity of 5m/s in 7” casing and a pressure drop of about 1 psi/100m. Over 700 meters of horizontal injection well the total pressure drop is less than 3 psi, which corresponds to a hydrostatic head variation of about 3 meters of propane bitumen blend, (n.b.

the pressure drop along the horizontal section is less than 7 psi due to leakoff into the formation). If the injection and production wells are separated by 5 meters, the liquid interface can be kept between the injector and the producer.

Consider the case where SAGD production is 3000 bopd so the predicted production according to the present invention will be 24000 bopd at 25 40°C. This yields a propane volume fraction of 0.67, so the propane injection rate will be 48000 bbl/day of liquid solvent equivalent. This corresponds to a volumetric flowrate of about 220,000 m3/day of vapour at 200psia and 40°C. In 9” casing the velocity is 65m/s which gives a pressure gradient of 100 psi/100m. It is desirable to minimize the pressure gradient along the injector 30 and to this end flow control means 40 (see Figures 1 and 2) can be used. For example, the pressure gradient can be mitigated by using larger casing or a

CA 02299790 2005-09-01

-18tubing string with orifices or the like to help distribute the solvent more evenly. The orifices can be metered to deliver a constant flow over different pressures, or can be designed to yield a variable flow at different pressures. Further, the flow control means can be varied along the length of the well to yield a more constant injection pressure in spite of line losses. Of course, at such high volumes, an additional challenge will be to separate the solvent from the bitumen at surface.

At some point increasing the injection/separation rates probably won’t be practical. When the supply of propane vapour to the reservoir becomes rate limiting, the pressure in the solvent chamber will begin to drop. This will lead to a reduction in the dew point or saturation temperature and a reduction in the solvent penetration rate as the bitumen surface viscosity is increased and the molecular diffusivity the solvent is reduced. Thus, it is anticipated that the pressure in the solvent chamber will gradually decrease with time and the process will eventually trend towards a process at the original ambient temperature of the reservoir. Thus, the present invention comprehends an extraction process which begins hot and pressurized and in which overtime both heat and pressure are reduced as the production volume increases. The supply bottleneck for solvent vapour could also be mitigated, by using shorter horizontal wells, but this may not be economically desirable. It can now be appreciated that a cold or ambient process may be used once the solvent chamber has been made large enough by the hot process first to give reasonable production rates.

Thus the proposed hot vapour extraction technique will be most useful for providing high initial production rates by rapidly forming a'chamber of size and quickly recovering the upfront capital costs. By growing a chamber quickly, the hot vapour extraction technique described here will allow the operator to have a large chamber much more quickly and thereby allow subsequent energy efficient, cold extraction to proceed economically.

For example, one can now estimate the minimum chamber size at 40°C and 200 psi for 1 megawatt of heat via condensing vapour. At

CA 02299790 2005-09-01

-1940cm/day x 750 m long x .35 porosity x .85 oil saturation x .8 recovery factor, the production rate is 71 m3 of solvent per meter of chamber circumference. Therefore for 270m3/day of bitumen production, the circumference of the solvent chamber must be greater than 4m, or the solvent chamber diameter should be larger than about 2 m. Since this is small relative to the distance between the wells (5 m), high rates of bitumen extraction should be feasible immediately after breakthrough between the wells.

The advantages of the present invention can now be understood. The prior art, a cold (unheated) solvent vapour extraction process the solvent- bitumen ratio is largely determined by the solubility of propane in the bitumen (it also depends somewhat on the mobility of the blend).

However, with a heated pressurized solvent vapour the solvent injection rate is determined by the heat balance. In other words, the amount of liquid solvent condensed within the reservoir depends on the volumetric heating requirements required to heat the reservoir to the dewpoint of the solvent, (i.e. the temperature difference between the solvent vapour at its dewpoint temperature and the ambient reservoir temperature, the heat capacity of the reservoir and the latent heat of vaporization of the solvent.). Thus the first advantage is that the solvent - bitumen ratio is uncoupled so that higher solvent proportions can be achieved in the blend.

A second advantage is that higher propane ratios provide a higher degree of deasphalting and thereby enhance the value of the produced oil (i.e. add up to 30% of incremental value to the oil). For a 100,000 bopd facility each dollar of incremental value/bbl adds 36 million dollars per year to the cash flow, so a higher degree of insitu upgrading could add up to 100 million dollars of cashflow to a project annually.

A third advantage is that the solvent penetration rate into the bitumen increases as the bitumen temperature is raised, because the diffusion rate increases as the viscosity is decreased, and thermal diffusivity is 100x faster than molecular diffusion at ambient reservoir temperature.

CA 02299790 2005-09-01

-20A fourth advantage of higher solvent ratios is that the bitumen solvent blend will have significantly lower viscosities than a cold or ambient process and therefore will drain faster and thereby speed up the extraction process. This is important because the production rate is minimal for the first three years of a cold start Vapex due to the small size of the solvent chamber. At 15% rate of return, the three year delay in the cash flow reduces the value of the oil production by 30%. For example if the oil is sold for 20$/bbl, the 3 year delay means that the producer is effectively paid only 14$/bbl. Thus, on a 100,000 bopd facility, the fast start up will add $600,000/day of value to the production or 220million$ of value to the cash flow per year.

As will be appreciated with higher production rates fewer wells are required to produce the same cash flow which is more efficient economically.

A further advantage of the present invention is that the elevated reservoir pressure can enormously simplify production of the fluids. For example, at elevated reservoir pressure it may not be necessary to supply a recovery pump on the production well side, because the reservoir pressure may be sufficient to overcome the hydrostatic head. In this case the production well would be choked back to maintain the pressure in the horizontal portion of the production well above the bubble point, in a manner analogous to the steam trap technique used for SAGD. This could save 3M$.

A further advantage of the present invention is that the energy requirements are quite modest compared to SAGD. For example, if the entire reservoir is heated to 40°C, instead of the 200°C for SAGD, then the greenhouse gas emissions are reduced by about 80%. This is particularly significant, since greenhouse gas emissions from heavy oil, bitumen and tar sands account for 25% of the excess above Canada’s obligation under the Kyoto Accord.

As will be appreciated by those skilled in the art, off setting these benefits are the requirement to recover and recycle higher volumes of solvent per bbl of bitumen production. It is expected that in the end stages of

CA 02299790 2005-09-01

-21 the extraction process, the solvent recovery may become a bottleneck, so solvent pressure (i.e. dewpoint temperature) in the solvent chamber will be reduced. However, this will help to offset higher heat losses to the overburden as the chamber spreads along the top of the oil bearing zone. Thus, the final 5 stages of extraction may occur at ambient reservoir temperature as previously described.

Thus we can see that the advantages of hot solvent gas injection include accelerated cash flow (fast start up), increased cash flow (upgrading) delayed capital expenditures, reduced solvent inventory and lifting 10 costs, reduced energy costs (relative to steam) and reduced greenhouse gas emissions (relative to steam). The hot solvent extraction process described here has the potential to add about 1 million$/day of incremental value to a 100,000 bopd cold vapex project.

As will be appreciated, the example reference conditions 15 discussed in this patent have been injection of propane solvent vapour at 40°C and 200psia. This particular choice of solvent, temperature and pressure was intended to teach by way of preferred example only. The optimum choice of temperature and solvent for a particular reservoir will depend on both cost factors (i.e., solvent separation rates) and bitumen 20 production rates.

While the foregoing description of the present invention includes various alternatives and variations, it will be apparent to those skilled in the art that various additional modifications are possible without departing from the broad spirit of the invention as noted in the appended claims. Some 25 of the variations are discussed above, such as the various pressures and temperatures which are suitable for the different solvents which are suitable according to the present invention. Others will be apparent to those skilled in the art. What is considered important in this invention is the selection of a suitable solvent which can effectively deliver heat to the formation by a latent 30 heat of condensation to decrease the viscosity of the hydrocarbons being recovered.