Note: Descriptions are shown in the official language in which they were submitted.
e~~.~~~:~'~
1
APPLICATION FOR PATENT
INVENTORS: TAPANTOSH CHAKRABARTY and JOSEPH S, TANG
TITLE: IMPROVING SWEEP IN THERMAL FOR USING
EMULSIONS
SPECIFICATION
BackcLround of the Invention
1. Field of the Invention
The present invention relates to improved recovery
of viscous oils from formations by steam processes.
More specifically, a method of diverting steam from
flooded-out portions of the formations to oil-
containing portions of the formations by use of a
water-in-oil emulsion is provided.
2. Descri~tian of Related Art
A substantial portion of the world's known
reserves of hydrocarbons is in the form of oil having a
very high viscosity under existing subterranean
conditions. The viscosity is often such that the flow
rate of the oil through wells is uneconomic using
common oil industry techniques. The subterranean
formation containing the oil is often too far below the
surface of the earth to allow economic recovery by
mining techniques. The only alternatives available far
recovery of such oils employ wells and a technique to
decrease the viscosity of the oil. The most common
technique for decreasing viscosity of the oil is to
increase its temperature, by processes referred to
generally as thermal recovery. The most common thermal
recovery processes are cyclic steam stimulation, steam
22658/27/1-I-1/900
~~~~~~r:~
2
injection or hot water injection. In cyclic steam
injection, steam in injected for a tune into a well,
then the well is converted to a production well and
fluids are produced from that same well. Cyclic steam
stimulation of wells in a field may be employed before
a steam or hot water. flood is used to drive fluids from
injection wells toward spaced-apart production wells.
When steam or hot water is injected into the
formation, it by-passes much of the oil-saturated rock
because the injected fluid has much lower viscosity
than the cold oil. A variety of efforts have been
made to find fluids which can be economically injected
to plug the swept-out region of the formation, where
steam has displaced most of the viscous oil, and to
divert steam which is injected thereafter to the cold
or unswept portion of the formation.
Two types of fluids have been suggested for
diverting the flow of injected steam from the portion
of the formation which has been largely swept of its
oil and into more oil-saturated portions. These fluids
are foams and oil-in-water emulsions. Foams are
disclosed, for example, in U.S. Patent 4,607,695. A
mixture of steam, a non--condensable gas and a
surfactant is injected into the formation. A foam
which has a higher apparent viscosity than steam is
formed in the pore spaces of the rock to block the flow
of steam and divert steam from the swept zones. U.S.
Patent 4,609,044 discloses a process for recovering
acidic viscous oil by injecting steam along with
dissolved alkaline salt and surfactants for foaming the
steam. The use of emulsions for diverting steam flow
in thermal recovery processes has been discussed in SPF
Paper No. 15052, "Use of Emulsions for Mobility Control
During Steamflooding," by T.R. French et al. These
studies were related to oil-in-water emulsions, both
those produced in situ and those injected into wells.
U. S. Patent 4,161,218 discloses the use of a coarse
22658/27/1-1-1/900
~~r~~'~~'~~
3
oil-in-water emulsion formed in the pore spaces by a
surfactant injected in an injection fluid. The oil
droplets in an oil-in-water emulsion plug pore throats,
thereby diverting steam into unswept portions of the
formation.
U.S. Patent 4,444,261 discloses a method of
diverting steam flow in an oil recovery process by
injecting into a formation a slug of high molecular
weight hydrocarbon which has been heated to a high
temperature. The diverting hydrocarbon then cools and
increases in viscosity to divert the follawing steam.
While many suggestions have been proposed for
diverting fluids, there remains the problem of
achieving greater sweep of formations containing high
viscosity oil in thermal recovery processes employing
cyclic steam stimulation, hot water flooding or steam
flooding. A diverting fluid is needed which is low in
cost, stable in the formation at high temperatures, and
which is compatible with formation fluids.
Summary of the Invention
According to one embodiment, there is provided a
method of increasing recovery of viscous oils in a
thermal recovery process by following injection of
steam with an emulsion of water in oil. In another
embodiment, the water-in-oil emulsion is heated on the
surface before it is injected. In yet another
embodiment, the oil and water are injected separately
or in the form of an oil-in-water emulsion and
converted into a water-in-oil emulsion in the wellbore.
The emulsion is preferably stabilized by adding an
alkaline material to the water which reacts with acidic
components in the oil. Alternatively, the emulsion is
stabilized by an added effective surfactant. The
emulsion is preferably formed from the viscous oil
recovered from the farmation. The emulsion diverts
steam or hot water injected after the emulsion is
22658/27/1-1-1/900
~~~~'~i3
4
injected to unswept portions of the formation
containing viscous oil and increases oil recovery in a
more economical means than hitherto available.
Brief Description of the Drawincts
Fig. 1 is a graph of the ratio of oil recovered to
the oil in place and of the oil-to-steam ratio in a
parallel pack model of steam injection with and without
injection of a slug of water-in-oil emulsion.
Fig. 2 is a graph of oil-to-steam ratio prior to
and following injection of a slug of water-in-oil
emulsion in a three-dimensional packed model of a steam
injection process.
Description of Preferred Embodiments
In steam recovery processes, it is common to
inject a predetermined amount of steam into a formation
containing viscous oil through an injection well
penetrating the formation and then to produce steam,
hot water and oil back through the same well in a
process called cyclic steam stimulation. This process
is commonly practiced befare a steam flood operation.
In a steam flood operation, steam is injected into the
injection well or wells and oil, water and sometimes
steam are produced from a production well or wells
which are spaced apart a selected distance from the
injection well or wells. The purpose of the cyclic
steam stimulation process is to establish flow
communication between the wells such that the formation
can be flooded with steam to recover the oil in the
formation between the wells. When steam breaks through
into the production wells, however, the steam zone
often occupies only a small portion of the vertical
extent of the formation, the steam having moved along
the top of the formation or through more permeable
streaks in the formation. A slug of viscous fluid can
then be injected into the zone where steam has swept
22658/27/1-1-1/900
5
oil from the formation to partially block flow of steam
into this zone and cause the steam to be diverted. The
efficacy of such a slug injected into the steam-swept
zone will be manifest by an increase in the oil
production rate at production wells.
Steam flooding processes are normally studied in
the laboratory by constructing scaled models of a
subterranean formation. In many formations, a fracture
in the formation is formed by injection of steam.
Steam can then channel through the fracture and
establish more rapid flow communication between
injection and production wells. In other formations, a
highly permeable streak in the formation has an effect
on fluid flow during steam injection similar to a
horizontal fracture. These effects of fractures and
permeable streaks occur in cyclic steam stimulation and
in flooding processes. These situations can be modeled
in the laboratory to study the benefits of a viscous
slug of water-in-oil emulsion in flooding process.
Many crude oils in their natural state contain
organic acidic components. The amount of acid present
is measured as an "acid number," which is defined as
the number of milligrams of potassium hydroxide
required to neutralize all the acidic components in 1
gram of oil. These organic acids will react at an oil-
water interface with alkaline components in the water
phase to produce soaps, which are surface active at the
oil-water interface and which can serve to stabilize
emulsions. We have found that with the proper ionic
composition of the water phase and with high shear
energy applied, a water-in-oil emulsion can be formed
with crude oil which is stable even at the high
temperatures which exist in steam floods. Water-in-oil
emulsions made with viscous oil are several times more
viscous than the oil, and several orders of magnitude
more viscous than steam. Surprisingly, this type of
emulsion was shown to be effective in models of steam
22658/27/1-1-1/900
~~~~~'~~~
6
flooding processes to substantially increase the net
oil recovered in a steam flood. Alternatively,
emulsifiers for stabilizing water-in-oil emulsions at
high temperature are added to the oil. Examples of
such emulsifiers are alkyl aryl sulfonates and alpha-
olefin sulfonates.
For example, a water-in-oil emulsion for diverting
steam can be created using viscous crude oil which has
a viscosity in the range from about 900 cp to about
1100 cp at a temperature of 60° C. and an acid number
of 1.1 mgm KOH per gm oil, and using water that
contains from about 100 ppm to about 2000 ppm sodium
hydroxide, 10,000 ppm sodium chloride, 80 ppm calcium
ion and 24 ppm magnesium ion. Chlorides of other
monovalent ions such as potassium and ammonium can be
substituted for the sodium chloride. The range of
concentration of sodium chloride in the water phase is
preferably from about 500 ppm to about 100,000 ppm.
The optimum concentration will depend on the
composition of the oil phase, the acidic components
naturally present in the oil phase, the amount of other
ions present and the relative amounts of oil and water
in the emulsion. The water phase preferably contains a
sufficient amount of divalent ions such as calcium and
magnesium to further stabilize the emulsion. The
divalent calcium and magnesium ions can be supplied by
any soluble salt containing these elements. The
optimum concentration of divalent ions will depend on
the amount of sodium or other monovalent ion present,
the composition of the oil phase, temperature and the
relative amount of oil and water present in the
emulsion. The concentration of divalent ions will
preferably be in the range from about 10 ppm to about
1,000 ppm. Other alkaline materials would be suitable
substitutes in equivalent amounts for the sodium
hydroxide, such as hydroxides of ammonium, potassium,
22658/27/1-1-1/900
7
calcium and magnesium. Alkaline silicates would also
be suitable.
The initial water content of the emulsion is
preferably in the range from about 5 per cent to about
70 per cent by volume. More preferably, the initial
water content is from about 20 per cent to about 50 per
cent by volume. Emulsion stability experiments can be
performed to determine the amount of water remaining in
the water-in-oil emulsion after different times the
emulsion is maintained at the steam temperatures of
interest, with different compositions of the water
phase and using the oil to be injected in the emulsion.
The composition of the water phase is preferably varied
until at least about 50 per cent of the initial water
present remains in the emulsion after one week at steam
temperatures.
An important feature of our invention is that the
emulsion can be repeatedly injected in slugs during the
cyclic steam stimulation or steam flooding process.
Thus, while the emulsions farmed are not completely
stable with time at steam temperatures and lose a part
of the water by demulsification, they are inexpensive
to prepare and axe sufficiently stable to allow
diversion of the steam for an adequate time to achieve
substantial benefits in oil recovery.
The emulsions are preferably heated before
injection, by heating the component fluids on the
surface of the earth either before or after the
emulsification step. After heating, the emulsion then
has low enough viscosity to be injected into the
injection well or wells. The emulsion is sufficiently
stable and the water content is high enough to
significantly increase the viscosity of the emulsion
over that of the oil phase. For the same slug size,
emulsion is less expensive than oil since a lower
volume of oil is injected.
22658/27/1-1-1/900
s
Emulsification is preferably achieved by imparting
high shear conditions to a mixture of oil and water.
Satisfactory emulsification can be achieved at the
surface by centrifugal blade devices, by flowing the
fluids at high pressure through jets, or by other
emulsification methods commonly used in industry.
Alternatively, the emulsions can be formed by imparting
high shear to the oil and water after the fluids have
been pumped down an injection well. In another
alternative method, part or all of the oil and water
are formed into an unstable oil-in-water emulsion at
the surface by adding the oil to an excess amount of
water phase and the water-in-oil emulsion is then
formed by high shear imparted in the wellbore or as the
unstable emulsion flows through perforations in the
casing of the well and into the formation. The amount
of shear imparted to the fluid should preferably be
such that the water droplets in the emulsion are
smaller than the pore spaces of the formation swept by
the steam so as to achieve sufficient stability of the
emulsion and to allow the emulsion to flow through the
formation as a viscous fluid.
In some fields, water-in-oil emulsion will be
produced from at least some wells. This emulsion may
be suitable for injecting as a slug in the method of
this invention. Alternatively, the produced emulsion
can be mixed with additional water and a suitable
emulsion can be formed as described before.
Example 1
An emulsion was prepared using crude oil from the
Cold Lake field, the crude oil having a viscosity of
1000 cp at a temperature of 60° C. The emulsion
contained 39 per cent aqueous phase, the aqueous phase
containing 750 ppm sodium hydroxide, 10,000 ppm sodium
chloride, 80 ppm calcium ion and 24 ppm magnesium ion.
The emulsion was formed by heating the fluids to 60° C.
22658/27/1-1-1/900
9
and forming the emulsion in a blaring blender, a
centrifugal blade device well-known for use in a
laboratory. The average size of the water droplets in
the emulsion was 1 micrometre. The viscosity of the
emulsion was 7,000 cp at a temperature of 60° C.
When heated to 85° C., 'the emulsion could easily be
injected into a sand-pack having a permeability of 2
darcies, which is about the permeability of the Cold
Lake formation sand in many areas. The pore size of
such sand is about 20 micrometres, so water droplets of
the size produced will flow through the pore spaces of
such a sand as a fluid. Because the water is not in
contact with the sand surface when in the form of such
fine droplets, the chemical reaction between the alkali
in the water and components of the sand is expected to
be greatly reduced, thus improving the stability of the
emulsion in the formation. In stability tests of the
emulsion in a pressure vessel, the water content of the
emulsion was still over 35 per cent after one week at
250° C.
Two sand packs were prepared having a permeability
of about 2 darcies. The two packs ware insulated and
heated to 95° C. Both packs were saturated with Cold
Lake crude oil and connate water. One pack was
individually flooded with steam to simulate a steam-
swept zone. A steamflood in the combined parallel pack
provided the base case. Then steam floods at 140° C.
in the parallel pack were conducted. An emulsion slug
equal in volume to 10 per cent of the pore volume of
the steam-swept pack was injected. The emulsion
contained 40 per cent brine and 60 per cent Cold Lake
crude oil or bitumen.
Referring to FIG. 1, curve 2 shows the oil
recovery ratio for the flood with the emulsion slug.
Curve 4 shows the oil recovery ratio for the base case
flood. Curve 12 and 14 show the oil steam ratio for
the flood with the emulsion slug and the base case,
22658/27/1-1-1/900
i
respectively. Oil recovery ratio is defined as the
fraction of the initial oil-in-place recovered. Oil
steam ratio is defined as the volume of oil recovered
divided by the volume of water converted to steam and
5 injected. It is apparent from FIG. 1 that the emulsion
slug resulted in significantly higher oil recovery
ratio and a higher oil steam ratio over most of the
process. Detailed examination of FIG. 1 shows that the
emulsion slug resulted in a 35 per cent improvement
10 (after subtracting the volume of oil in the emulsion
slug) in oil recovery in 30 per cent less time over the
base case. The cumulative oil steam ratio of the
emulsion slug flood in the first 90 minutes was 0.25
after deducting the amount of oil injected in the
emulsion slug, compared with 0.14 in the first 120
minutes for the base case. The emulsion slug thus
resulted in more efficient utilization of the injected
steam.
Example 2
A scaled three-dimensional model of a subterranean
oil-productive formation was used. The model was 56 cm
in diameter and 38 cm in thickness. It was packed with
sand having a permeability of 180 darcies and a
porosity of 42 per cent, to scale a formation with a
permeability of 2 darcies. The model contained two
wells on opposite sides of the center to simulate an
injector and producer well-pair. Wire mesh was placed
at each well to simulate a horizontal fracture
extending about 40 per cemt of the distance to the
other wall. The model was saturated with 79 per cent
pore volume crude oil, 11 per cent pore volume connate
water and 10 per cent pore volume gas. The inclusion
of a gas phase provided the high compressibility
required for early cycle steam injection. Just prior
to the beginning of steaming, a slug of water was
injected.
22658/27/1-1-1/900
11
Referring to FIG. 2, the oil steam ratio is shown
at different times. First, two cycles of cyclic steam
stimulation were conducted at each well. The oil steam
ratios for the first and second cycles are shown at
point 1 and point 2, respectively, The cyclic steam
stimulation cycles were conducted to establish thermal
communication between wells. The simulated horizontal
fracture helped distribute the steam and set up a
thermal communication channel between the wells, a
scenario expected in many subterranean formations.
Steam at 500 psig was injected into each well until the
model reached injection pressure, then each well was
produced until pressure and fluid rate were low. After
two cycles of steam stimulation, thermocouple
measurements in the model showed the wells were in
thermal communication. The wells were then shut in for
8 minutes and a steam flood was then initiated in one
well. The oil steam ratio for this flood is shown at
curve 3 in FIG. 2. The model was 'then shut-in again
for about 7 minutes and a second steamflood was
initiated which lasted for 38 minutes, about the same
as the first steamflood. The oil steam ratio for this
flood is shown at curve 4 in FIG.2. Then a slug of
water-in-oil emulsion was injected, the emulsion being
heated to 80° C. before injection. The time of
injection of emulsion is shown by area 10 of FIG. 2, A
third steamflood was then initiated, this flood lasting
abaut 100 minutes. The oil steam ratio for this flood
is shown at curve 12 in FIG. 2.
The emulsion contained 40 per cent by volume
brine, the brine containing 750 ppm sodium bydroxide,
10,000 ppm sodium chloride, 80 ppm calcium and 24 ppm
magnesium. The emulsion was formed by blending the
crude oil and water for 20 minutes until the average
water droplet size was 1.5 micrometres. Emulsion
viscosity was 5100 cp at 60° C.
22658/27/1-1-1/900
12
Referring to FIG. 2, it is apparent that the oil
steam ratio decreased during the cyclic steam
stimulations and continued to decrease with time during
the steamfloods. Injection o~ the emulsion slug at 10
is seen to have a dramatic effect in increasing the oil
steam ratio in Curve 12.
The extrapolated decline curve of oil steam ratio
before injection of the emulsion slug is shown at curve
16. The extrapolated decline curve of oil steam ratio
after injection of the emulsion slug is shown at curve
14. Curve 18 is the value in the model of the oil
steam ratio corresponding to the economic cutoff value,
or the minimum oil steam ratio that will allow
continued economic production in the formation. The
cutoff value in the model of 0.26 corresponds to an oil
steam ratio in the formation of 0.15. Using these
three curves, incremental oil recovered by injection of
the emulsion slug (subtracting oil injected in the
emulsion slug) was estimated. The result was a total
recovery of 23.3 per cent of the original oil in place
for the emulsion slug process, compared with only 15.7
'per cent without the emulsion slug. This represents an
increase in oil recovery of about 25 per cent over the
recovery for steamflood alone.
The invention has been described with reference to
its preferred embodiments. Those of ordinary skill in
the art may, upon reading this disclosure, appreciate
changes or modifications which do not depart from the
scope and spirit of the invention as described above or
claimed hereafter.
22658/27/1-1-1/900
