This paper was prepared for presentation at the SPE/DOE Thirteenth Symposium on Improved

Oil Recovery held in Tulsa, Oklahoma, 1317 April 2002.

This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300
words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
Abstract
When considering cold production from heavy oil fields, one
of the main issues is to define a reliable recovery factor. Many
field simulations are performed with black oil simulators, and
predictions are very often below the observed production
levels. One explanation is the difficulty of such simulators in
representing the physical behavior of this type of "bubbling"
oil : as depletion reaches the bubble point, small bubbles of
gas appear; they grow according to the solution gas drive
process, then connect to create a free gas phase.
A series of experiments has been performed under reservoir
conditions, by applying different depletion rates, on two long
cores of consolidated sandstone with similar characteristics :
Berea (0.8 bar/d and 8 bar/d) and Clashach -Scotland-
(0.5 bar/d and 2.5 bar/d). Two recombined oils from the Zuata
field -Orinoco- were tested. Reservoir oil and gas production,
in-situ gas saturation profiles, critical gas saturations were
measured and compared.
A black oil simulator had been used to match these
experiments whereby a separate gas relative permeability (krg)
curve was derived for each experiment.
Simulations were also run with a chemical-like simulator. It
considers three forms of gas : the solution gas, the dispersed
gas (bubble gas) and the free gas. This simulator applies one
krg for free gas and one for dispersed gas. It applies specific
kinetic factors for each transformation from dissolved gas to
dispersed gas, then from dispersed gas to free gas.
The simulations give a set of krg curves and kinetic factors
that match all experiments. A methodology is described giving
a set of parameters which are independent of the depletion
rate.
The simulations provide production profiles for each type of
gas. As, during experiments, only the total gas is measured,
some observations are made which could help to distinguish
between dispersed and free gas during the experiment.
Introduction
This work aims at studying and improving the understanding
of the solution gas drive mechanism in heavy oils below the
saturation pressure. This phenomenon is often called foamy
oil behavior, likely due to the appearance of the mixture
produced at the wellhead, even though the term of bubbling
oil would seem to be more appropriate.
Recently, studies have been carried out to explain the
advantages - higher recovery and production rate - of this
process in heavy oil reservoirs (mainly in Canada and
Venezuela de Mirabal et al).
Nevertheless, this complex mechanism still remains
controversial according to the different assumptions reported :
- high critical gas saturation (Maini et al, 1993).
- - lubrication effect caused by nucleation of gas bubbles on the
pore wall (Shen and Batycky, 1996);
- low gas mobility (Pooladi-Darvish and Firoozabadi, 1999;
Tang and Firoozabadi, 1999; Kumar et al, 2000).
As regards numerical simulation, in order to tackle the original
behavior of bubbling oils, the first modeling attempts have
tried to modify the Darcys law, implemented in the existing
simulators to describe the two phase flow of oil and gas :
- Kraus et al (1993) introduced the pseudo-bubble point
model assuming that all the gas produced remains entrained
in the oil phase as long as the pressure does not reach the
pseudo-bubble point, lower than the actual saturation
pressure measured in a PVT cell.
- Lebel (1994) developed the concept of both modified
fractional flow - to represent the entrainment of part of the
gas by the oil - and reduced foamy oil viscosity - above a
certain gas saturation corresponding to the liberation of the
gas.
- Claridge and Prats (1995) proposed the reduced viscosity
model, assuming that the oil viscosity decreases when the
asphaltenes content in oil is reduced, while the asphaltenes
would contribute to stabilize the smaller gas bubbles.
Although these models succeeded in reproducing some of the
characteristics of the heavy-oil flow under depletion, they fail
to properly simulate the non-equilibrium effects and all the
time-dependent phenomena, therefore being non-predictive.
Thus, some researchers developed non-equilibrium models :
SPE 75133
A New Methodology To Match Heavy-Oil Long-Core Primary Depletion Experiments
Y.M. Bayon, Ph.R. Cordelier, SPE, A. Nectoux
2 Y.M. BAYON, PH.R. CORDELIER, A.NECTOUX SPE 75133
- Coombe and Maini (1994) described a non-equilibrium
reaction model (implemented in CMG Stars simulator)
defining chemical-like transformations to mimic processes
which deviate from equilibrium : (a) the gas dissolved in the
oil into micro-bubbles - called the dispersed gas- (b) which
then evolves to the free gas.
- Sheng et al. (1998) introduced the dynamic nucleation
model, after Maini et al. researches, to better represent the
transient phenomena occurring during a depletion; it takes
into account the rates associated with the nucleation, the
growth and the coalescence of bubbles. The formation rate
of the dispersed gas is function of both the super-saturation
and time, the coalescence rate is assumed to be proportional
to the volumetric fraction of the dispersed gas in the oil. One
limitation of this model lies in its predictive ability because
the main parameters are dependent on the depletion rate and
require a history match.
Further to the earlier paper of Urgelli et al, our current
approach is to test and, if necessary, improve the predictive
flow simulators. Based on representative experimental
depletions, the respective capabilities of a black-oil simulator
and a non-equilibrium model are compared. The methodology
derived is detailed on the basis of the model developed by
Coombe et al, which accounts for the non-equilibrium nature
of the heavy oil flow, coupled with a reduced mobility of gas.
Description of the experiments
Two sets of depletion experiments were performed with
heavy oils on outcrop sandstones at the nearer reservoir
conditions. Consolidated cores were selected to avoid any
compaction effects within the range of pressure drop.
Reservoir oils were recombined from stock tank oils from a
heavy-oil field and adequate quantities of required natural gas
components. All experiments were initiated at a pressure
above the saturation pressure in order to obtain data in the
single phase conditions.
The lower depletion rates were applied to be closer to actual
field conditions while the higher could represent nearer
wellbore conditions. In table 1 the experimental conditions
and main results are summarized.
The first set uses an homogeneous long core of Berea
sandstone (Kw=630 mD), saturated with a recombined heavy
oil, at initial water saturation (Swi), and set at reservoir
temperature, above the saturation pressure. The temperature
and confining pressure are kept constant during the
experiment. The effect of the temperature and the presence of
connate formation water has been reported to increase the gas
relative permeability and thus to reduce the oil recovery (Tang
& Firoozabadi, 2001).
The core holder is vertical, the withdrawal being applied at the
top, at two depletion rates : 0.8 bar/d and 8 bar/d.
During these two experiments, the pressure, oil and gas
produced are measured both at reservoir and surface
conditions. An original production manifold permits the
visualization and the differentiation of the liquid and gaseous
effluents at reservoir conditions. It is then possible to fully
characterize each of the oil or gas phases produced in terms of
volumes, GOR, volume factors. Such an apparatus allows the
differentiation of the liquid and gas, at ambient conditions,
from each of the phases (oil and/or gas) identified under
reservoir conditions. The gas breakthrough is estimated from
these measurements. A scheme is given (Fig. 1).
These experiments were performed under X-ray scan; they
record the gas saturation profiles all along the core during the
depressurization, as illustrated in Figure 4.
The second set of experiments was performed with a
composite core of Clashach sandstone, homogeneous, slightly
more permeable (Kw=850 mD) than the Berea sample. The
depletion rates 0.5 bar/d and 2.5 bar/d were applied to the
sample. In these experiments, a more viscous oil was selected
in order to establish comparisons with the first series; the
production was began at the top outlet until the free gas
mobility was observed, then switched to the bottom outlet.
Experiment
depletion rate (bar/d)
first set
0.8 / 8.0
second set
0.5 / 2.5
Core sample Berea Clashach
Length of core (cm) 91.2 66.4
Diameter (cm) 5.0 6.8
Mean porosity 0.216 0.16
Mean permeability (mD) 630 850
Oil viscosity @ Pb (cPo) 308 1746
Saturation pressure (Pb) 52 bar @ 49C 50 bar@ 49C
GOR (Sm3/m3) 15.5 11
Oil density (g/cm3) 0.984 1.011
Swi 0.18 0.15
Initial Pressure (bar) 73 / 102 60
Final Pressure (bar) 1.0 / 1.0 1.2 / 16
Outlet side top top then bottom
Critical gas saturation (%) 5 / 17 4.2 / 4.5
Gas mobility pressure (bar) 42 / 26 40 / 39
Pb (gas mobility pressure) 10 / 26 10 / 11
Final recovery 0.25 / 0.35 0.22 / 0.21
Table 1 : Experiment data
Results
The variations of the outlet pressure against time during the
experiments are plotted in Figure 2. Figure 3 shows the
evolution of the oil recovery versus pressure. In spite of the
low gas content, these recoveries are significantly higher than
expected for such a depletion mechanism with a conventional
oil. A late gas breakthrough is observed, the gas remaining in
the porous medium. As reported by many authors, the critical
gas saturation, Sgc, is a function of the depletion rate : the
higher the pressure decline rate, the higher the Sgc. This
notion will be discussed later on.
In the first set of depletions (outlet on the top side), a major oil
production is observed after breakthrough, despite adverse
gravity effects. For the 0.8 bar/d depletion rate, an increase in
the recovery is noticed for the lower pressures. This is due to
the evolution of the GOR, the gas volume factor, and the
SPE 75133 A NEW METHODOLOGY TO MATCH HEAVY-OIL LONG-CORE PRIMARY DEPLETION EXPERIMENTS 3
gas/oil relative permeabilities. Indeed, the phase of expansion
below 5-8 bars is characterized by a decrease of the GOR (the
downhole volume of gas produced carries on increasing but it
corresponds to a decreasing mass flow rate of gas). Moreover,
the gas dilation effect is dominant at the lower pressures, thus
resulting in a relative pressure stabilization.
Based on the saturation profiles (see Fig. 4), once the pressure
declines below the bubble point, the gas saturation does not
increase uniformly, confirming observations from other
researchers (Sahni et al, Egermann et al). Each saturation
profile presents two slopes : the greater one, which
corresponds to the higher gas saturation growth, is located
close to the bottom end of the core model; the smaller one is
close to the top.
Simulation of the experiments
As a matter of fact, the mobility of gas appears to be a key
factor of the behavior of heavy oils. In the present paper, the
mobility is addressed in two ways. Firstly, some simulations
were performed only taking the relative permeability into
account. Secondly, simulations were run considering the
availability of gas according to the non-equilibrium reaction
model.
From the four experiments described above, two were used to
define the methodology that is later described : 8 bar/day and
0.8 bar/day. The two others were used to check it.
Using a black-oil simulator.
In the first stage, a black-oil simulator was used to match the
experiments. The data set was built with the characteristics of
the fluid and the core of the experiment (table 1). The core is
represented by a series of thin cells about one centimeter -
whose section has the same area as the core. The aim of these
matches was to find the gas-to-oil relative permeability krg.
The results are presented in Fig 5 a-b-c. The match of the
experiment was satisfactory. Production and pressure curves
superimpose almost perfectly.
Fig 5 c presents the krg curves. One can see a big step at gas
saturation around 35%. Only the part of the curve below 35%
has a major impact on the simulation.
Two different krg curves were derived from the simulations :
one for the 8 bar/d experiment and one for the 0.8 bar/d
experiment. The relative permeability of gas is lower for the
8 bar/d experiment, suggesting that the gas is less mobile in
this experiment.
The main conclusion from these black-oil simulations, is that
the krg are dependant on the depletion rate. Therefore, when
studying the mobility of gas during the heavy foamy oil flow,
we cannot only consider the relative permeability of gas. The
results confirm that the depletion in heavy oils is a time-
dependent process regarding the pressure gradients induced in
the porous medium.
Using a chemical-like simulator
Furthermore, the availability of the gas must be addressed.
Due to the high viscosity of the oil, and to the high interfacial
tension, the exsolution of the solution gas is not instantaneous,
as it is for lighter oils at our macroscopic scale of analysis.
The kinetics of this transformation is a major concern in the
solution gas drive process and will be later studied with the
help of the kinetics model implemented in Stars

simulator.
Three different type of gases, and availability :
This reservoir simulator enables the definitions of different
forms of gas components; in this study, the three following
components are considered :
- the Solution gas (SolG) : dissolved in the oil phase,
- the Dispersed gas (DispG), released from the oil into a
non-connected bubbles form dispersed in the live oil, due
to the pressure drop.
- the Free gas (FreeG) representing the connected gas
phase.
The simulator models the transformations of these gases as
chemical reactions. Two reactions are thus defined with their
own kinetics :
Reaction 1 : Solution Gas Dispersed Gas : X
1
= F
1
*[SolG]
Reaction 2 : Dispersed Gas Free Gas : X
2
= F
2
*[DispG]
Xi: volumetric reaction rate
Fi : frequency factor
[ ] : concentration
where F
1
and F
2
are the parameters to be found to match the
experiments.
Mobility of gas
The "chemical-like" simulator attributes a krg curve - Krg
u
- to
the free gas. Although the dispersed gas appears as a
discontinuous phase, its mobility is associated to a relative
permeability curve. It is the result of the interpolation between
this previous Krg
u
curve and another one, Krg
l
as shown in
Figure 6. For numerical reasons, this curve is not a key
parameter in the matching procedure. Within the range of
influence e.g. for gas saturation below 35% - Krg
l
is at least
one hundred times lower than Krg
u
. Consequently, changing
this Krg
l
in a range of one to two will not bring significant
difference in the interpolation leading to the relative
permeability of the dispersed gas.
On the contrary, Krg
u
is a key parameter for matching the
experiments.
Methodology
The aim of the simulation is to find a set of parameters F
1
, F
2
,
Krg
u
, which match both simulations : 8 bar/d and 0.8 bar/d.
At the start, F
1
and F
2
are set at high values. Transformations
of gas are instantaneous. The Krg curve is the same for both
Free gas and Dispersed gas, and corresponds to a black oil
behavior. Then, as transformations of Solution gas into
Dispersed gas and Dispersed gas into Free gas are not
4 Y.M. BAYON, PH.R. CORDELIER, A.NECTOUX SPE 75133
instantaneous, F
1
and F
2
are decreased. To compensate for this
lack of availability of the gas, the relative Krg
u
is increased.
For each experiment, there is not a unique set of parameters
that will match : for each set of F
1
and F
2
factors, a Krg
u
will
correspond. But, only one set of Krg
u
,

F
1
and F
2
factors match
both experiments. These values should be representative of the
availability and mobility of gas in this heavy foamy oil.
Results
Simulated and experimental productions and pressures are
compared in Figures 7a-b-c, showing a good match for both
experiments.
Final Krg curves presented in Figure 8 are compared to the
ones obtained with a black-oil simulator. The upper curve is
much higher than the one from the black-oil simulator for the
slow depletion experiment at 0.8 bar/d, and do not exhibit such
a distinct step. The gas shows a continuous mobility as gas
saturation increases, with a much higher relative permeability,
when gas saturation is below 35%. The lower curve is very
similar to the one derived from the black-oil simulator for the
faster depletion experiment 8 bar/d.
The chemical-like simulator provides the amount of simulated
production for all forms of gas : the solution gas, the dispersed
gas and the free gas. (Figure 7a) shows that hardly any free
gas is produced. Gas production comes mainly from the
dispersed and solution gas. This agrees with that which is
observed during the experiment : the oil coming off the core
still contains bubbles, and continues to release its gas, a long
time after its production.
Application to 2.5 bar/d and 0.5 bar/d experiments
The above results were then applied to the second set of
measurements to test the influence of a difference in the
viscosity of the fluid, and therefore, check the robustness of
the methodology described above, the properties of the
Clashach core being close to the Berea sandstone (see table 1).
Nevertheless, due to the fact that some of the PVT data was
not available for these recombined oils, it was not possible to
obtain Kr curves and frequency factors in the same way as in
the first set. Therefore, the frequency factors and relative gas
permeability, derived from the former simulated experiments,
were applied based on the former detailed PVT data.
The simulated results are presented in Figures 9a-b, showing a
fairly good match with the experiment data. For the 2.5 bar/d
experiment, simulated productions of oil and gas are very
close to the measurements. For the 0.5 bar/d experiment, the
simulation gives slightly too much cumulated gas. The oil
curves do not superimpose perfectly, although the total
amount is very close.
One explanation of these differences is related to the higher
viscosity of the actual fluid compared to the one used in the
simulations. One could expect that the transformations from
solution gas to dispersed gas and then to free gas are
slower when the viscosity is higher. A lower mobility of gas
could be also suspected. Nevertheless, despite some minor
differences, the application of the previous matching
parameters (F1, F2, Krgu, Krgl) to these two experiments does
not counter the methodology.
Critical Gas saturation
The definition of the critical gas saturation, Sgc, appears to be
too loose as it considers the gas saturation at which the gas is
seen to be macroscopically in motion (i.e. gas breakthrough,
first krg non null).
In fact, regarding the experiments, gas is observed at the outlet
of the core sample with some delays which can be more
significant for heavy oils than lighter oils.
As regards the simulations, most of the relative permeability
krg derived from experiments are in the range of 10
-2
,

while
simulators can manage krg down to 10
-6
. Sgc usually refers to
the free gas mobility but the dispersed gas must also be
considered : the dispersed gas starts to move at lower gas
saturation than the free gas.
Due to the use of a "chemical-like" simulator, and the use of
very low krg values, this notion of "critical gas saturation"
should be adapted.
Conclusion.
A series of depletion experiments has been performed on
heavy oils under reservoir conditions which indicate the
following characteristics of the solution gas drive process :
- A recovery significantly higher than expected for a depletion
mechanism with a conventional oil,
- Gas bubbles appear to remain trapped in the oil, the gas
breakthrough being postponed late after the saturation
pressure,
- A non uniform gas saturation is observed in the core model.
This study outlines an original methodology that simulates
long core primary depletion experiments with heavy foamy
oils. It defines a balanced position which considers both the
mobility and the availability of gas, and proves to be
independent from the depletion rate. Simulations with a
"chemical-reaction-like simulator" provide a set of frequency
factors and relative permeability curves.
Compared to a black-oil simulation, the relative permeability
of the free gas is higher.
Furthermore, the simulation distinguishes free gas, dispersed
gas and solution gas, whereas the experiments only monitor
the total amount of gas. An in-depth analysis of the gas
saturation profile (see Fig. 4) could help differentiate between
these forms of gas.
The variation of both the free gas and the dispersed gas should
be measured or modeled by means of a network modeling
approach, and CT-scan measurements.
The methodology proposed must be validated on well data,
then on field data, before being able to derive relevant
recovery factor at field scale. Additionally, one could expect
to represent the extension of gas, free gas and dispersed gas, in
the field and, therefore optimize the location of the wells, their
spacing and the development scheme of the field.
SPE 75133 A NEW METHODOLOGY TO MATCH HEAVY-OIL LONG-CORE PRIMARY DEPLETION EXPERIMENTS 5
Acknowledgements
The authors wish to thank Mr. Dennis Coombe of CMG, for
the numerous and fruitful discussions they had with him. A
big thank you to all those who conducted the core depletions :
D. Heritier, D. Mller, J.F. Besnier, J.Ph. Chaulet,
Ph. Maquignon. The authors gratefully acknowledge several
discussions with G. Hamon and S. Sakthikumar. The authors
wish to acknowledge TotalFinaElf S.A. for permission to
publish this paper.
References
Mirabal, M.de, Gordillo, R., Fuenmayor, M., Rojas, G., Rodriguez,
H., Sanchez, R., Integrated study for the characterization and
development of the MFB-53 reservoir, North Hamaca - Orinoco
Belt, Venezuela , SPE 36095, Fourth Latin American and
Carribean Petroleum Engineering Conference, April, 1996
Maini B.B. " Foamy oil Flow" Journal of Petroleum Technology,
October 2001 pp 54-63.
Maini, B.B., Sarma, H.K. and George, A.E.: Significance of Foamy-
Oil Behavior in Primary Production of Heavy Oils, Journal of
Canadian Petroleum Technology, Vol. 32, No. 9, pp 50-54,
1993.
Shen, C., and Batycky, J.: Some observations of mobility
enhancement of heavy oils flowing through sand pack under
solution gas drive, proceedings of the 47th Annual Technical
Meeting of the Petroleum Society of CIM in Calgary, Alberta,
Canada, June 10 12, 1996.
Pooladi-Darvish, M., and Firoozabadi, A.: Solution-Gas-Drive in
Heavy Oil Reservoirs, Petroleum Society of CIM, n 97-113,
June 1997
Tang, G.Q. Gas and Liquid-Phase Relative Permeabilities for Cold
Production from Heavy Oil, SPE 56540, presented at the SPE
Annual Technical Conference and Exhibition, Houston (October
3-6, 1999).
Kumar, R., Pooladi-Darvish, M., "An Investigation into Enhanced
Recovery under Solution Gas Drive in Heavy Oil", SPE 59336
presented at the 2002 SPE/DOE improved Oil Recovery
Symposium held in Tulsa, Oklahoma 3-5- April 2000.
Kraus, W. P., McCaffrey, W.J. and Boyd, G.W.: Pseudo-bubble
Point Model for Foamy Oils, proceedings of the CIM 1993
Annual Technical Conference in Calgary, May 9-12, 1993.
Lebel, J.P., Performance implications of various reservoir access
geometries, 11
th
Annual Heavy Oil and Oil sands Technical
Symposium, Calgary, March, 2
nd
, 1994
Claridge, E. L., and Prats, M.: A proposed model and mechanism for
anomalous foamy heavy oil behavior, proceedings of the
International Heavy Oil Symposium, Calgary, Canada, 19-21
June 1995.
Coombe, D., Maini, B., Modeling Foamy Oil Flow, Workshop on
Foamy Oil Flow, Petroleum Recovery Institute, Calgary, April,
27
th
, 1994
Sheng, J.J., Foamy Oil Flow in Porous Media, Ph.D. Thesis,
University of Alberta, Department of Civil and Environmental
Engineering, Edmonton, Alberta, 1997
Sheng, J.J., Maini, B.B., Hayes, R.E. and Tortike, W.S.: Critical
Review of Foamy Oil Flow, Transport in Porous Media, 35,
157-187, 1999.
Urgelli, D., Durandeau, M., Foucault, H. and Besnier, J-F.:
Investigation of Foamy Oil Effect from Laboratory
Experiments, SPE 54083, Proceedings of the 1999 SPE
International Thermal Operations and Heavy Oil Symposium,
Bakersfield, California, California, March 17-19, 1999.
Tang, G. and Firoozabadi, A.: Effect of GOR, Temperature, and
Initial Water Saturation on Solution-Gas Drive in Heavy-Oil
Reservoirs, SPE 71499 presented at the 2001 SPE Annual
Technical Conference and Exhibition held in New Orleans,
Louisiana, 30 September-3 October 2001.
Sahni A., Gadelle F., Kumar M: "Experiments and Analysis of Heavy
Oil Solution Gas Drive", SPE 71498 presented at the SPE
annual Technical Conference and Exhibition held in New
Orleans, Louisana, 30 September 3 October 2001.
Egermann, P. and Vizika, O., Critical gas saturation and relative
permeability during depressurization in the far field and the
near-wellbore region, SPE 63149, SPE Annual Technical
Conference and Exhibition, Dallas, Texas, USA, October 1-4,
2000
Kamp A.M., Henry C., Andarcia L, Lago M., Rodriguez A., :
"Experimental Investigation of Foamy Oil Solution Gas Drive."
SPE 69725 presented at the 2001 SPE Thermal Operations and
Heavy Oil Symposium held in Porlamar, Margarita, Venezuela.
Javadapour F., Pooladi-Darvish M., " Network Modeling of Gas
Relative Permeability in Heavy Oil., presented at the
Petroleum Society's Canadian International Petroleum
Conference 2001, Calgary Alberta Canada, June 12-14 2001
Stars Users Guide, Computer Modeling Group, Calgary, 2000
Top production pump
Pressure maintenance unit
Pressure maintenance unit
Atmospheric separator
X-rays emission
and detection
unit
Bottom production pump
Liquid separator
Production cell (reservoir conditions)
Production cell (reservoir conditions)
Visual cell (reservoir conditions)
Core holder and
porous medium
(reservoir conditions)
Gas separator
Long core holder
Depletion
volume
1
2
3
P out
Pump
1
3
0

b
a
r
(
g
)
,

4
9

c
P in
P
res
=60 bar(g), T= 49c
Gasometer
1
0
0

b
a
r
(
g
)
,

4
9

c
Gamma
First set Second set
Figure 1 : Experiment apparatus.
Figure 2 : Pressure vs Time. Figure 3 : Oil Recovery vs Pressure
0
100
200
300
400
500
600
700
800
900
-0.100 0.000 0.100 0.200 0.300 0.400 0.500 0.600
GAS SATURATION
D
i
s
t
a
n
c
e

f
r
o
m

c
l
o
s
e
d

e
n
d
P = 97 bar
P = 43 bar
P = 39 bar
P = 32 bar
P = 28 bar
P = 18 bar
P = 15 bar
P = 4 bar
Top
Bottom
Figure 4 : Gas saturation profile (example)
Depletion experiments - Oil Recovery
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 10 20 30 40 50 60
Pressure (bar)
O
i
l

r
e
c
o
v
e
r
y

(
%
)
0.5 b/d
2.5 b/d
0.8 b/d
8 b/d
Depletion Experiments - Pressure
0
20
40
60
80
100
120
0 500 1000 1500 2000 2500 3000
Time
( )
P
r
e
s
s
u
r
e
(
)
0.5 b/d
2.5 b/d
0.8 b/d
8 b/d
breakthrough
40b
42b
26b
39b
SPE 75133 A NEW METHODOLOGY TO MATCH HEAVY-OIL LONG-CORE PRIMARY DEPLETION EXPERIMENTS 7
0
20
40
60
80
100
120
0 50 100 150 200 250 300 350 400 450
Time( hour)
O
i
l

P
r
o
d
u
c
t
i
o
n

-

P
r
e
s
s
u
r
e
0
500
1000
1500
2000
2500
3000
3500
4000
4500
G
a
s

P
r
o
d
u
c
t
i
o
n
Oil Producton Experiment Oil Producton Simulated
Outlet Pressure Experiment Outlet Pressure Simulated
Gas Producton Experiment Gas Producton Simulated
0
10
20
30
40
50
60
70
80
0 500 1000 1500 2000 2500
Time( hour)
O
i
l

P
r
o
d
u
c
t
i
o
n

-

P
r
e
s
s
u
r
e
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
G
a
s

P
r
o
d
u
c
t
i
o
n
Oil Producton Experiment Oil Producton Simulated
Outlet Pressure Experiment Outlet Pressure Simulated
Gas Producton Experiment Gas Producton Simulated
Figure 5 a : Black-oil simulator Figure 5 c : Black-oil simulator.
Match of 8 bar/d experiment. Match of 0.8 bar/d experiment.
Figure 5 c : Black-oil simulator.
Gas Relative Permeability.

Dispersed Gas
interpolation
between Krgu and Krgl

Free Gas
Krgu
Krgl
Gas Saturation
Krg
Figure 6 : Chemical-like Simulator.
1.00E-06
1.00E-05
1.00E-04
1.00E-03
1.00E-02
1.00E-01
1.00E+00
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Gas Saturation
L
o
g
a
r
i
t
h
m
i
c

s
c
a
l
e
8b/d 0.8b/d
8 Y.M. BAYON, PH.R. CORDELIER, A.NECTOUX SPE 75133
Determination of gas relative permeability.
Figure 7 a : Chemical-like Simulator. Figure 7 b : Chemical-like Simulator
Match of Gas Production Match of Oil Production.
Figure 7 c : Chemical-like Simulator.
Match of Outlet Pressure.
SPE 75133 A NEW METHODOLOGY TO MATCH HEAVY-OIL LONG-CORE PRIMARY DEPLETION EXPERIMENTS 9
Figure 8 : Gas Relative Permeability.
Figure 9 a : Chemical-like Simulator - Match of Gas Production.
Figure 9 b : Chemical-like Simulator - Match of Oil Production.
0.000001
0.00001
0.0001
0.001
0.01
0.1
1
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Gas Saturation
L
o
g
a
r
i
t
h
m
i
c

s
c
a
l
e
Krgl chemical simulator Krgu chemical simulator
8b/d black-oil simulator 0.8b/d blackoil -simulator