SPE/Petroleum Society of CIM/CHOA 78988

Laboratory Investigation Of Gravity-Stable Waterflooding Using Toe-To-Heel

Displacement: Part L: Hele Shaw Model Results
A.T. Turta, C. Ayasse* J. Najman**, D. Fisher and A. Singhal, Petroleum Recovery Institute, Alberta Research Council
Copyright 2002, SPE/PS-CIM/CHOA International Thermal Operations and Heavy Oil
Symposium and International Horizontal Well Technology Conference.
This paper was prepared for presentation at the 2002 SPE International Thermal Operations
and Heavy Oil Symposium and International Horizontal Well Technology Conference held in
Calgary, Alberta, Canada, 47 November 2002.
This paper was selected for presentation by the ITOHOS/ICHWT 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, the
Petroleum Society of CIM, or CHOA 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, the Petroleum Society of CIM, or CHOA, its officers, or members. Electronic
reproduction, distribution, or storage of any part of this paper for commercial purposes without
the written consent of the Society of Petroleum Engineers, Petroleum Society of CIM, or CHOA
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,
P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Abstract
This paper presents results of the laboratory investigations
leading to development of a novel, radically different
waterflooding process, which is gravity stable. The new
process is called Toe-To-Heel Waterflooding (TTHW) and
uses vertical wells for injection, while horizontal wells are
utilized for production. Normally, the system uses a staggered
line drive configuration in which the toe of the horizontal
producer is close to the line of vertical injectors. The
horizontal section of producer is located at the top of oil
formation while the vertical injectors are perforated on the
lower section of the layer. By using the proposed
configuration, a short-distance oil displacement process is
achieved. The water/oil segregation causes the injected water
to slump, while the horizontal well as a sink causes water to
flow upwards; the composite of these two tendencies leads to
an advancement of water through almost all the entire pay
section, with the water break-through at the toe, followed by
an advancement of the displacement front along and under the
horizontal leg.
A 2-D Hele-Shaw model was used for investigations. The
Hele-Shaw model was composed of two parallel plexiglass
plates, which were held together at 0.1 mm; these vertical
plates form between them a simulated porous medium with a
permeability of about 833 D. The rectangular chamber
forming the simulated porous medium has the dimensions of
52cm* 21.4cm*0.01cm.
The Hele-Shaw model was intended to mimic the vertical
section of an oil reservoir and the investigations were designed

specifically for a comparative study of vertical sweep

efficiency for conventional waterflooding and toe-to-heel
waterflooding. These experiments do not permit an evaluation
of areal sweep efficiency.
In fact, two toe-to-heel processes were explored: one-phase
TTHW and two-phase TTHW; in the last case a blanket of
water is intentionally generated at the lower part of layer (by
keeping opened the vertical pilot hole of the horizontal
producer, while the horizontal leg is closed) and then only the
horizontal leg is kept opened (while the vertical pilot hole
is closed).
More than 20 low pressure (40-80psi) tests were carried out at
room temperature. First, some preliminary experiments were
conducted in order to establish a base line performance, to
provide calibration, or to study the main mechanisms of this
process. In these tests, colored water miscibly displaced plain
water in a toe-to-heel configuration; no gravity segregation
effect between displacing and displaced liquid existed. Also,
no mobility effects and two-phase flow along the horizontal
leg existed. Then, detailed experiments with five different oils
(with viscosity in the range 10 mPa.s to 12,000 mPa.s), and
water injection rates of 2.5 ml/hr to 320 ml/hr were
carried out.
A comparison between conventional waterflooding and
TTHW showed that in general for the heavy oils with
moderate viscosities, oil recoveries at water-break-through are
similar, but at the water-out, at the end of the test, the ultimate
oil recoveries were much higher for TTHW; this was a direct
consequence of higher vertical sweep efficiencies obtained.
Later on, 3-D tests conducted in a real porous medium reinforced that the concept was sound; the 3-D tests results will
be reported in part ll of this series.
Introduction
Conventional waterfloods in light-heavy oil reservoirs are
limited by three main factors, namely:
Rock heterogeneity, leading to the water channelling
Gravity segregation (due to oil/water density contrast),
leading to under-riding of the injected water

Highly unfavourable water/oil mobility ratio, which

aggravates the negative effects of the first two factors.
Usually, rock heterogeneity is reflected in a pronounced
vertical stratification, due to a relatively large contrast in
horizontal permeability of different layers. On the other hand,
the negative effect of gravity segregation is mainly felt when
the stratification is not very pronounced, and pay thickness
and permeability of the pay zone are relatively large.
The effect of gravity segregation on waterflood performance
was fully recognized in 1953 when the first mathematical model
of water tonguing (under-riding) was published by Dietz1.
Initially, Dietz theory was believed to be applicable mostly to
thick formations. However, subsequently, Outmans showed
that this theory is equally valid for thin oil formations2.
Recently, for such horizontal waterfloods, the effect of gravity
segregation on waterflood performance in light oil reservoirs
was analytically expressed by Ekrann3, who showed that it
becomes important for the high permeability formations, for
moderate ratios between vertical and horizontal permeability.
The recovery decreases with increasing water/oil mobility ratio,
and increases with increasing ratio between viscous and gravity
forces. The last condition means that from a practical point of
view, high water injection rates discourage under-ride; therefore
they lead to higher oil recoveries.
For a stratified two-layer system, Ekrann found that the most
unfavourable stratification occurs when the high permeability
interval is located at the bottom, while the most favourable one
is when the high permeability interval is located at the top.
Ekranns theoretical results were close to the laboratory results
reported by Craig4. All his calculations were for a case of
water/oil density difference of 200 kg/m3. From Ekranns
paper, one can conclude the following:
- Vertical sweep efficiency is less than 40% for
homogeneous systems, at water/oil mobility ratios higher
than 2.
- Vertical sweep efficiency is less than 20% when a
high permeability interval is located at the bottom; this is valid
for a large range of water/oil mobility ratios (higher than 0.5).
- The under-ride phenomenon is not a concern in
achieving good vertical sweep for the case of the high
permeability interval located at the top.
To combat under-riding phenomenon in the conventional
waterflooding, in order to achieve a gravity stable waterflood,
the first alternative is to inject something similar to the water,
but with a density very close to the oil. However, in this
project, a second alternative was adopted; in order to achieve a
gravity stable waterflood, a new approach to re-engineering of
the gravity effect during displacement, using horizontal wells,
was developed5. This approach focussed on the novel
technology of Toe-To-Heel Waterflooding (TTHW) process.
The TTHW process is a Toe-To-Heel (TTH) displacement
process. It can be classified as a short-distance oil

SPE/PS-CIM/CHOA 78988

displacement (SDOD) process, which is important mainly for

heavy oil reservoirs6.
Conventional oil reservoir flooding methods (pattern or line
drive configurations), as applied during the past 60 years, are
long-distance oil displacement (LDOD) processes; mobilized
oil typically needs to travel several hundred metres to the
production wells. Actually, the oil is pushed by the injectant
from its original location to the outlet and, most of the time oil
flows along the stratification. In spite of being relatively
inefficient, these are still economically acceptable in situations
with low oil viscosity (<10 mPa.s), and the need to switch to
SDOD was not felt so acutely. However, for flooding heavy
oil pools SDOD has now become feasible, especially in view
of the widespread use of horizontal well technology in recent
years. The main disadvantages of the long-distance
displacement processes are outlined elsewhere5,6. The most
important disadvantage of LDOD processes is the dependence
of the displacement front advance on the distribution of
properties (mainly permeability and viscosity) along the flow
path, from the injection to the production well. This crucial
negative effect is eliminated in the SDOD processes, where
displacement front advance depends only on the distribution
of the same properties (mainly viscosity and permeability) in a
region immediately ahead of the displacement front. This
region can be narrower or larger, depending upon
transmissibility, which is the ratio between permeability and
the product of oil viscosity, oil compressibility and porosity;
generally, the higher the oil viscosity, the narrower is
this region.
The SDOD processes are defined as a broad category of
displacement types in which the mobility (viscosity) of
injectant is important, but it no longer dominates the process.
A more significant feature is the short distance travel for any
oil particle, before it is produced. The injectant could still
flow long distances, as SDOD are specifically designed for
high mobility injection fluids, generally with unfavourable
mobility ratios between injectant and oil. So, instead of
looking for solutions for making mobility ratio more
favourable like in LDOD, the SDOD processes tend to reduce
its importance. This approach is by far more practical, since
for most heavy oil pools even if a mobility ratio of one is
attained, injection pressures required to sustain an
economically acceptable oil rate would be impractical and
may lead to fracturing, which is always undesirable in
displacement processes.
Steam Assisted Gravity Drainage (SAGD) and Vapor
Extraction (Vapex) are also SDOD processes. However, the
essential difference between TTH displacement processes, and
SAGD and Vapex is that the last two processes are integrally
gravity controlled processes, while TTH displacement
processes are pressure gradient and gravity controlled, in
other words, TTH displacement processes include an active
drive component.

SPE/PS-CIM/CHOA 78988

A TTH displacement process can be applied as a non-thermal

recovery technology, such as Toe-To-Heel Waterflooding
(TTHW) or as a thermal technology such as Toe-To-Heel Air
Injection (THAI), with its variant, catalytic THAI (CAPRI),
for in situ upgrading of the oil. The THAI process was
conceived and confirmed as an effective displacement process
for in situ combustion, early in 19925,6,7.
A schematic a THAI is shown in Figure 1. As can be seen, the
oil flows only in a region immediately ahead of the
displacement front, within the so-called mobile oil zone
(MOZ). A viscosity gradient, due to high temperature in
MOZ, does exist. More than 70 laboratory tests and some
simulation work have confirmed the concept of thermal TTH
displacement is a sound concept8,9,10. The laboratory work so
far has focussed mostly on THAI and catalytic THAI
(CAPRI).
The potential for the application of waterflooding in a TTH
configuration was explored only very recently8. During
1996/1998 period, more than 20 Hele-Shaw tests were carried
out. These tests showed that the TTHW concept has merit. In
this paper these initial semi-quantitative tests results are
presented. However, later on, 3-D tests conducted in a real
porous medium proved that the concept was sound; the 3-D
tests results will form part ll of this series, while the results of
detailed simulations will be included in part lll.

the horizontal leg (this tendency was strongly influenced by

oil-rock properties and horizontal well design.

TTH W aterflooding Process

V ertical injector

H orizontal producer

Brine: Sp.G rav. >1.0

Fig. 2: Schematics of Toe-to-Heel Waterflooding

(TTHW) process, as resulted from Hele Shaw
model tests
Experimental Set-up, Procedure and Materials
The experimental set-up is shown in Figure 3a. Two positive
displacement pumps regulate injection rates for oil and water.
The Hele-Shaw cell is made of transparent plexiglass. A set
of lights enables a video camera to continuously record images
of water invaded zone. At any time, the ratio of this water
invaded area to the total model area yields the vertical sweep
efficiency, which is equivalent to the oil recovery for this
simplified system, representing a vertical cross section.
Figure 3b shows details for the Hele-Shaw cell itself. It is
composed of two 2.5 inch thick plexiglass plates 21.4 cm by
52 cm, which are held together by a series of long bolts with a
sealing gasket setting the plates 0.1 mm apart. In this way a
rectangular chamber (52cm*21.4cm*0.01cm) is formed. The
0.1 mm distance between plates is simulating a porous
medium having a (calculated) permeability of 833 D. There
are auxiliary outlets, which are used for vacuum, upper
injection, or cleaning of the model.

Fig. 1: Schematics of Toe-To-Heel Air Injection Process

(after CMG)
A schematic of TTHW process is provided in Figure 2. As far
as gravity segregation is concerned, its negative effect is
largely mitigated as a consequence of two opposing effects:
downwards flow due to gravity segregation and upwards flow
due to the horizontal producer being a pressure sink. Another
important feature is the abatement of heterogeneity effect due
to the horizontal well acting as a line-sink, covering a
considerable length; there is a pressure drop along this linesink. So, due to a slightly lower pressure drop towards the
heel, oil has the tendency to flow within the reservoir along

SPE/PS-CIM/CHOA 78988

WATER

OIL

vertical sweep efficiency for this simplified system.

Therefore, conclusions from this section refer mainly to the
vertical sweep efficiency. No information on horizontal sweep
efficiency was obtained from these tests. Given the ratio
52/21.4=2.4, between the length and the height of the model
(layer), it can be considered that the tests described mimic a
thick layer.
Oil
collector
Videocamera

Vertical pilot hole

VCR

RUSKA PUMPS

Fig. 3a: Laboratory set-up for TTHW investigation

Five different oils, with viscosities in a broad range, 10 mPa.s

to 12,000 mPa.s, were used in these investigations. The
density contrast was in the range 0.05 g/cc to 0.184 g/cc, while
the injection rate was in the range 2.5 cc/hr to 320 cc/hr. For
each oil, the TTHW process result was compared to a
conventional, vertical injector-vertical producer test,
conducted at the same injection rate.
The tests were performed at room temperature and with an
injection pressure in the range of 40-80 psi; the exit pressure
was very close to the injection pressure and in some cases it
was the atmospheric pressure.
Preliminary Experiments
These preliminary tests were designed to explore whether the
TTH displacement is feasible when no viscosity gradient (due
to temperature distribution) exists. Also, they served as a
basis of comparison for further tests and allowed us to
understand the fundamental mechanisms.

Fig. 3b: Hele Shaw plexiglass cell

The wells, which are 5/16 inch in diameter, are drilled directly
into the plexiglass walls, being connected with the chamber of
the cell by a series of ports placed every 1 cm (Figure 3b).
During TTHW tests, the cell shown in Figure 3b is used
upside-down, such that the horizontal well (HW) is located at
upper part of layer; HW has a length of 34 cm, and between
its toe and the shoe of the vertical well there is a space of
approx. 10 cm. The vertical well is located at approx. one
inch from the edge of the model.
The volume of the cell is 10.8 cc, that of horizontal well is
40.8 cc and that of the vertical well is 16.5 cc. Given the low
volume of the cell compared to that of wells, the water breakthrough may be masked and the production of oil can be
delayed. Because of this, tests are considered semiquantitative and oil recovery is calculated based on water
invaded area. Given the configuration of the system the
porous medium is mimicked by the space confined by the two
closely spaced vertical plates the oil recovery is equal to the

In Figure 1, the TTH concept was demonstrated for thermal

processes; the horizontal section of horizontal well is located
at the bottom layer, while the injection is through the vertical
well, using the upper perforations. Using the same lay out, a
non-thermal TTH process was tried; a heavy liquid (glycerol)
with a density of 1220 kg/m3 and viscosity of 1500 mPa.s was
displaced by a light liquid, plain water with a higher mobility.
The water was dyed with green food color to improve the
visualization of invaded zone. An injection rate of 320 ml/hr
was used.
As seen in Figure 4, a direct channeling of the
water did not occur. Instead, a partial anchoring of the water
body took place in the toe region, and a high recovery of
glycerol was achieved. Therfore, it was demonstrated that
when using a pair of liquids with low-density contrast, it was
possible to achieve a stable TTH displacement. Afterwards, a
TTH displacement for a high-density contrast pair of liquids
was investigated by displacing water by gas in the same
model, the same configuration and the same injection rate,
used previously. This time, the displacement process was not
stable. As seen in Figure 5, the gas segregated and formed a
gas cap at the upper part. Later on, downward displacement of
water occurred; when gas injection rate was higher than the
critical rate, gas coning somewhere along the horizontal well
took place.
Finally, a basic displacement of water by water (same density
and viscosity) was performed this time in the full set up for
TTHW (Figure 6), i.e. with the horizontal section of horizontal
well located at the top of the layer and the injection through

SPE/PS-CIM/CHOA 78988

the vertical well (all thickness perforated). The above

conditions of unity mobility ratio and equal density are ideal
from the point of view of longdistance oil displacement
(LDOD) processes, and it may be viewed as a calibration test.
There are only two forces dominating the advance of the
water; the drive force horizontally, and the action of the
horizontal section as a line-sink - vertically upwards. As a
result, displacement front is curved as shown in Figure 6.
Given the ideal conditions of same density and viscosity,

faster in the upper part of the layer, while some advancement

also occurred at the bottom of layer.

Vertical injector

Horizontal producer

Figure 6: Displacement of water by green water in a

TTH configuration.

Figure
4:
Lowdensity
contrast
displacement of glycerol by green
TTH configuration.

displacement;
water in a
GAS

GAS

WATER

Figure 5: Highdensity contrast displacement; displacement of

water by gas in a TTH configuration.
the displacing liquid front profile is related only to the
pressure drop along the horizontal leg of the horizontal
producer. As seen in Figure 6, contrary to our expectations, a
direct channeling of the injected water through the horizontal
leg did not take place, and more than 60% of the resident
water was displaced. The advance of the injected water was

In Figure 2, water invaded zone for a real TTHW process was

illustrated. Although water viscosity is slightly lower and
density of water is slightly higher than those for oil, which
from the point of view of LDOD processes is very
unfavorable, in the TTH system, in general, it led to a
recovery higher than that obtained for a more favorable case
of same density and viscosity fluids. This shows that gravity
segregation plays an important role in these processes. There
are three forces dominating the advance of the high mobility
injected water: the drive force horizontally, the action of the
horizontal section as a line-sink - vertically upwards, and
oil/water gravity segregation - vertically downwards. When
the water first enters the porous medium the oil flow
dominates the whole process and the water tends to flow
directly towards the toe; afterwards, as more water has been
injected, the oil/water segregation - vertically downwards - is
more and more active and causes the formation of a nose of
the water body directed downwards. The shape of the water
body, as shown in Figure 2, is typical for a TTHW process.
The displacement front is loosely anchored to the horizontal
section of horizontal well, and although the front is not
perfectly vertical, the water body advance is still relatively
efficient from the point of view of sweep, which increases
in time.
Results of the Main Tests
The results are presented in Tables 1 to 5 in order of
increasing oil viscosity. In Figures 7 to 10 the shapes of
water invaded zone are shown. In Figures 7-a to 7-b for a 10
mPa.s oil, the shape of water invaded zone at the water breakthrough for the conventional waterflood (Figure 7-a) is
compared to the shape of water invaded zone at the water
break-through for the TTHW process (Figure 7-b).

SPE/PS-CIM/CHOA 78988

In Figures 8-a to 8-b, for the 780 mPa.s oil, these shapes are
shown for TTHW at the beginning (Figure 8-a), and at
watering out (Figure 8-b).
As seen in Figure 8a, at the
beginning of injection the fingering is quite extensive, but it is
still possible for the water body to have horizontal and
downwards advancement, so that the final vertical sweep is
relatively good, as seen in Figures 8-b.

Oil viscosity, mPa.s

851.7
3

Brine-oil density contrast, kg/m

168

Brine injection rate, ml/h

20

Conventional

Well

%Recovery

%Recovery

Types

at brine

at watering

break-through

out

22

46

VI-VP

TTHW

Oil density, kg/m

VI-HP

93

96

780

945
3

Brine-oil density contrast, kg/m

75

Brine injection rate, ml/h

10

Well

%Recovery

%Recovery

Types

at brine

at watering

break-through

out

Conventional

VI-VP

14

23

TTHW

VI-HP

12

54

10

Flood Type

Oil viscosity, mPa.s

Flood Type

Table 1: Pembina Oil

Oil density, kg/m

Table 3: Court Oil

Table 4: Lindberg Oil

Oilviscosity, mPa.s
Oil density, kg/m

1200

971
3

Legend:
VI = Vertical injector
VP = Vertical producer
HP = Horizontal producer

Brine-oil density contrast, kg/m

49

Brine injection rate, ml/h

2.5

Flood Type

Well

%Recovery

%Recovery

Types

at brine

at watering

break-through

out

Conventional

VI-VP

22

40

TTHW

VI-HP

21

67

Table 2: Dunsmore Oil

Oil viscosity, mPa.s
Oil density, kg/m

Table 5: Bodo Oil

112

918
3

Brine-oil density contrast, kg/m

101

Brine injection rate, ml/h

20

Oil viscosity, mPa.s

Oildensity,kg/m

12,000

988.1
3

Flood Type

Well

%Recovery

%Recovery

Types

at brine

at watering

break-through

out

Conventional

VI-VP

23

27

TTHW

VI-HP

24

58

Brine-oil density contrast, kg/m

184

Brine injection rate, ml/h

2.5

Flood Type

Well

%Recovery

%Recovery

Types

at brine

at watering

break-through

out

Conventional

VI-VP

TTHW

VI-HP

14

32

SPE/PS-CIM/CHOA 78988

Table 6: TTHW Process Effect of Brine

Injection Rate on Oil Recovery
Light oil

Pembina oil

Oil viscosity, mPa.s

Oil density, kg/m

10

851.7
3

168

Brine-oil density contrast, kg/m

Injection

%Recovery

%Recovery

Rate

at brine

at watering

ml/hr

break-through

out

60

78

84

20

93

96

Fig. 8a: Court oil: TTH waterflooding.

invaded zone at the start of water injection

Shape of water

Fig. 8b: Court oil: TTH waterflooding.

invaded zone at the watering out

Shape of water

Fig. 7a: Pembina oil: Conventional waterflooding. Shape of

water invaded zone at water break-through

Fig. 9a: Lindberg oil: TTH waterflooding. Shape of water

invaded zone at water break-through

Fig. 7b: Pembina oil: TTH waterflooding. Shape of water

invaded zone at water break-through

SPE/PS-CIM/CHOA 78988

these zones; also, the fingering is more accentuated for the

heavier oil. However, it is seen that for both oils, the fingering
is more pronounced close to the injection well compared to the
region far away from the injection well, where the water zone
becomes more compact. At this time, the mechanisms leading
to this phenomenon are not understood.
While interpreting these results it should be noted that the
injection rate was not the same for all the tests: it was 20 ml/hr
for the 10 mPa.s and 112 mPa.s oils; 10 ml/hr for the 780
mPa.s oil and 2.5 ml/hr for the 1200 mPa.s and 12,000 mPa.s
oils. Watering out was deemed when a water cut of 95%
was attained.
Fig. 9b: Lindberg oil: TTH waterflooding. Shape of water
invaded zone at watering-out

From Tables 1 to 5 and Figures 6 to 9, the following

conclusions can be drawn:
At water break-through oil recovery of the TTHW process
was close to that of conventional waterflooding. An
exception to this was observed during light oil tests (10
mPas viscosity oil), for which break-through oil recovery
was much higher for TTHW process. This conclusion
seems to indicate that for relatively light-heavy oils once
onset of water fingering occurs, the main finger advances
extremely quickly, such that the value of distance from
the injection point to the outlet becomes unimportant. In
fact, the distance between the vertical injector and the
vertical producer was at least 4 times higher than the
distance between the shoe of vertical injector and the toe
of horizontal producer.

Fig. 10a: Bodo oil: TTH waterflooding.

invaded zone at water break-through

Shape of water

Oil recovery at watering out, compared to conventional

waterflood, increased as follows: 46% to 96% for the 10
mPa.s viscosity oil, 27% to 58% for the 112 mPa.s
viscosity oil, 23% to 54% for the 780 mPa.s viscosity oil,
and 14% to 32% for the 12,000 mPa.s viscosity oil.
The unfavorable effect of water/oil mobility ratio was
reduced, but not totally eliminated as witnessed by the
difference in the vertical sweep efficiency for the most
viscous oils (Figures 9 and 10)
The main unswept zone is located above the horizontal
section of horizontal well; smaller unswept zones are
found at upper corner, between the vertical well shoe and
the horizontal well toe and close to the vertical well shoe
in the lower section, mainly for more viscous oils. This
last finding reveals that TTHW, unlike conventional
waterflooding, would leave some undisplaced oil close to
the injection point.

Fig. 10b: Bodo oil: TTH waterflooding.

invaded zone at watering-out

Shape of water

The water invaded zones for the heaviest oils with viscosities
of 1200 mPa.s and 14,000 mPa.s, at water break-through and
watering out are shown in Figures 9-a to 9-b, and 10-a to 10-b,
respectively. There are important differences in the size of

The direct effect of injection rate on oil recovery for the 10

mPa.s oil is shown in Table 6. There is a slight increase in oil
recovery (both at water break-through and watering out) when
a low injection rate is used. Also, for the same viscosity ratio,
higher density contrasts will lead to high oil recovery. This
was verified for the case of 1200 mPa.s oil; higher sweep

SPE/PS-CIM/CHOA 78988

efficiency was obtained for 23% salinity case, as compared to

the 3% salinity case. For this reason, in a TTHW process,
using injection brines with the highest feasible salinity is
recommended. This is totally opposite to conventional
waterflooding, where higher water density (due to higher
salinity) leads to more water under-ride and harms
performance. This constitutes a significant difference between
TTHW and conventional waterflooding.
The tests showed that the position/tilting of displacement front
is determined by the combined influence of contrast of density
and viscosity for the displacing/displaced fluid, injection rate,
and the relative location of the vertical injector and horizontal
section of producer.
Also, they showed that TTH
displacement process applications lead to high oil rates from
the very beginning (no waiting period). The oil recovery
increases substantially, especially when the density contrast is
relatively high (high brine salinity and/or low oil density), low
injection rates are used and water/oil mobility ratio is not
extremely unfavourable.
Discussion of Results. Limitations
The main limitation of Hele Shaw tests is that they give us
only semi-quantitative results. Moreover, the permeability is
extremely high, and capillary effects do not exist. Also, ratio
of length to height of the model is approx. 2.4, which is very
low and can be associated with only very thick oil formations.
From this point of view the results are generally
very optimistic.
The Hele Shaw tests showed that in heavy oil pools, the
fingering phenomena are pronounced close to the injection
well; this fingering disappears far from the injection well
where water invaded zone becomes compact. At the end of a
flood, the zone around the injection well is less swept,
compared to zones more remote. That this effect would also
occur in a real porous medium is still to be demonstrated. In a
TTHW process, the whole oil reservoir volume situated above
the horizontal section of the producer will remain unswept,
and this may force operators to locate the horizontal section as
close to the overburden as possible. Such application would
also impose a minimum limit on the pay thickness.
Conclusions
1. The Toe-To-Heel Waterflooding (TTHW) process was
investigated in a Hele Shaw laboratory model, which
mimicked a vertical section of a simulated porous medium.
The semi-quantitative results of the investigations showed that
the process achieves a good vertical sweep efficiency, better
than that for conventional waterflooding.
2. Because, on the one hand, the fact that the results were
semi-quantitative, and on the other hand, the horizontal (areal)
sweep efficiency was not determined, it remained to be
confirmed that the process has a sound foundation. 3-D tests in
real porous media were therefore necessary for validation.
Later on, they were conducted and confirmed the soundness of

the process. These results will be presented in Part ll of this

series of papers on TTHW.
References
1.
Dietz, D.N. : "A Theoretical Approach to the
Problem of Encroaching and Bypassing Edge Water." Konikl.
Ned. Akad. Wetenschap (1953) Proc. B56, 83.
2.
Sandrea, R. and Nielsen, R, Dynamics of Petroleum
Reservoirs Under Gas Injection, Gulf Publishing
Company, 1974
3.
Ekrann, S.: "Gravity and Vertical Sweep Efficiency",
In Situ, 17(2), 183-199, (1993)
4.
Craig, F.G.: The Reservoir Engineering Aspects of
Waterflooding, SPE Monograph, 1993.
5.
Turta, A.T.: Toe-To-Heel Oil Displacement Short
Paper Presented at Canadian International Petroleum
Conference, June 4-8, 2000, Calgary.
6.
Turta, A.T., and Singhal, A., K.: Overview of ShortDistance Oil Displacement Processes Paper presented at
SPE/PS - CIM International Conference on Horizontal Well
Technology, November 6-8, 2000, Calgary
7.
Greaves, M, and Turta A T.: Oilfield In-Situ
Combustion Process U.S. Patent No.5,626,191, May 6,1997.
Canadian Patent No. 2,176,639, August 8, 200.
8.
Greaves, M., Saghr. A.M. Xia, T.X., Turta, A. and
Ayasse, C: THAI New Air Injection Technology for Heavy
Oil Recovery and In-Situ Upgrading Journal Of Canadian
Petroleum Technology, March 2001, Vol 40, No.3.
9.
Xia, T.X. and Greaves, M: Upgrading Athabasca Tar
Sand Using Toe-To-Heel Air Injection (THAI). Fourth
International Conference and Exposition on Horizontal Well
Technology, November 6-8, 2000, Calgary.
10.
Coates, R. and Zhao, L.L.: Numerical Simulation of
THAI Process Canadian International Petroleum Conference,
June 12-14, 2001, Calgary, Canada
11.
Ayasse C. and Turta A T.: Toe-To-Heel Oil
Recovery Process USA Patent 6,167,966, January 2, 2001.
Canadian Patent 2,246,461, June 18, 2002.