SPE 149052

Cyclic Pressure Pumping (CPP): A Potential Recovery Method for Fractured

Carbonate Reservoirs
A. Ikwumonu, K. Rawnsley and S. Bettembourg, Petroleum Development Oman

Copyright 2011, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE/DGS Saudi Arabia Section Technical Symposium and Exhibition held in Al-Khobar, Saudi Arabia, 15–18 May 2011.

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 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 the SPE meetings are subject to publication review by Editorial Committee of Society of Petroleum Engineers. Electronic reproduction,
distribution, or storage of any part of this paper 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 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

The large permeability contrast between fractures and matrix in highly fractured carbonate reservoirs has been a
hindrance to the efficient recovery of the oil from the matrix. Gas oil gravity drainage (GOGD) has been the most
appealing option to date. However it is a slow process. Cyclic Pressure Pumping (CPP) is proposed which has the
potential of increasing the rate of recovery over GOGD under certain conditions. CPP utilises the fact that due to
the much lower permeability in the matrix, the matrix pressure lags behind the fracture pressure during a rapid
pressure change. A sudden drop in pressure in the fractures results in expansion of fluids in the adjoining matrix
which are discharged into the fracture system. The fluids in the fracture system can then be produced. The
reservoir can be re-pressurised by gas injection and depressurised by producing the fluids in an alternate manner.

This process originally called batch gas cycling was suggested in the 50s for improving recovery in non-
homogeneous reservoirs over conventional gas flooding. It was tried successfully in a small field in the late 50s.
Laboratory experiments in the 70s on low-permeability matrix in contact with high-permeability fractures indicated
the effectiveness of the process.

The process was re-discovered during reservoir production optimisation simulation studies in a heavy oil steam
injection project in Oman and will be applied to minimise oil loss during steam plant shut downs. Early field
observations support the process. The paper presents a study of the process using both dual permeability and
single porosity simulation modeling. It also highlights the conditions which are favorable to the process as well as
certain practical problems that could be encountered in implementing it.

Introduction
Fractured carbonate reservoirs contain a significant amount of oil world wide. In many cases most of the oil in
these reservoirs is held in the matrix blocks of very low permeabilities usually less than 20 mD. These matrix
blocks are surrounded by relatively very permeable fractures (5 – 100D) that contain very little oil. This permeability
contrast makes flooding these reservoirs very inefficient as the displacing fluid (water or gas) by passes the oil that
is held in the matrix.
One current method of recovering the oil from such reservoirs is the gas oil gravity drainage (GOGD) process. In
this process the fractures are filled with gas (or steam) normally injected near the top of the gas column in the
fractures. The density difference between the oil in the matrix and the gas in the surrounding fractures causes the
oil in the matrix to drain downwards by gravity through the matrix. At the bottom of the fracture gas column the oil
flows from the matrix into the fractures to form a fracture oil rim from where it is produced. Because the oil drains
through the low permeability matrix, and is resisted by capillary forces, the process is usually very slow. Usually the
only way of accelerating the process is to use steam in the fractures rather than gas that heats the matrix and
reduces the oil viscosity in the matrix thereby increasing the drainage rate. The driving force being gravity cannot
be changed.
Cyclic pressure pumping or pulsing (with water and gas) has been used in heterogeneous reservoirs from the
1950’s [1], [2]. A laboratory study was carried out in 1970 by S.H. Raza [3] which showed the effectiveness of the
2 [Paper Number]

process in a fractured tight reservoir condition as well as the superiority of using gas over water. More recently
published papers on cyclic pressure pulsing in fractured reservoirs have focused more on the single well process
of huff ‘n’ puff with CO2 or N2 injection [4], [5]. This paper focuses on a reservoir process where by the entire
reservoir pressure is increased by injecting any gas or steam into the fractures and the pressure is allowed to drop
by shutting down the injectors and producing the oil and gas spiked into the fracture system from the matrix as a
result of fluid expansion. This process depends heavily on the expansion of the oil and gas in the matrix as a result
of the pressure drop. If in addition the injected gas is able to dissolve in the oil during the period of increasing
pressure and released during the de-pressurisation then the recovery will be greatly enhanced. This study shows
that even in the event of no dissolution of gas and for more viscous crude (greater than 20 cP) the process can be
faster than the more conventional field wide process of GOGD.

Observations during dual permeability study

A dual permeability model for a fractured field in Oman was set up for steam assisted GOGD as shown in Fig.1.
Steam is injected into the crest by injectors such as I1 and production is taken from the flank wells such as P1.

The production profile for the continuous steam injection and another case of a seven day steam shut down are
shown in Fig.2. After an initial drop of oil production due to the sudden movement of the oil column upwards, the oil
production from the shut down case exceeds the case of continuous injection. The reason for this is that the
pressure drop in the fracture associated with the shut down pulls oil horizontally over a relatively short distance
from the low permeability matrix block into the high permeability fracture from where it can travel quickly
downwards in the fracture to the oil rim. In the continuous case the GOGD process requires that oil must travel
vertically slowly through the low permeability matrix until it reaches the oil rim in the fracture. The resulting profiles
are presented in Fig. 3 that shows the oil volume in the matrix and fractures against time. The shut down results in
a large amount of oil desaturation at a rate of 26000 m3/d of oil compared to about 3500 m3/d GOGD rate for the
no shut down case (c.f. green solid & dashed lines). Correspondingly the oil volume in the fractures rises very
quickly (red solid line) for the shut down case compared to the no shut down case (dashed red line).

Single porosity fine grid model (single stack)

A black oil single porosity fine grid model was set up to validate the process. The matrix permeability of 10 mD and
fracture (high permeability cells surrounding the matrix) permeability of 100 D were used. The column thickness
was 100 m and the matrix block was 20m. The PVT, relative permeability and capillary pressure data are shown in
Figures 4 to 6. The oil compressibility above the bubble point pressure of 3300 kPa is 0.74 x 10-6/kPa. The
simulator used the Carlson method [6], [7] to model hysteresis effects. In view of the extreme saturation reversals it
has been considered to incorporate the effects of hysteresis. Gas was injected into the reservoir at the top blocks
in the high permeability (fracture) cells by wells I1-I4, while oil and gas is produced from wells P1 – P4 from the
lower blocks at the high permeability (fracture) cells as shown in Figure 7. In the GOGD case both gas injection
and fluid production was continuous while in the CPP case the gas injection was shut down intermittently with the
oil production still continuous. The oil production from the two cases are shown in Fig.8. The much higher oil rates
from the intermittent injection case (solid green line) compared to the continuous case (solid red line) can be seen.
The cumulative oil production from the intermittent case (dashed green line) is almost twice that of the continuous
case (dashed red line) for the 10 years production. Figure 9. shows the detailed initial production profiles.

A similar simulation was made by reducing the oil viscosity by a factor of 10. The results shown in Figure 10. show
only a marginal benefit over GOGD predominantly only in the early years. The end production saturations are
shown in Figures11 and 12.
As the oil viscosity is reduced the duration of the pressure difference between fracture and matrix becomes small
resulting in very short cycle times and relatively low benefit over GOGD.
Further sensitivities were explored. A similar simulation set up was made for 50 m matrix block and 20 cP oil. The
results are shown in Figure 13. Also the case of pressure cycling below the bubble point was simulated to illustrate
the benefit of dissolved gas is shown (Figure 13. blue lines). As can be seen cycling below the bubble point when
some of the injected gas can be re-dissolved and later released during depressurisation significantly increases the
recovery. However how much of the gas would dissolve in practice will depend on the type of gas and the oil.

Table 1. Summaries the results which are discussed in the following section.

Factors affecting the process

Two critical factors affect the recovery efficiency of the process:
• The pressure pulse amplitude
• The length of the pressure cycle
[Paper Number] 3

The pressure amplitude determines how much the fluids in the matrix will expand. The time interval over which a
pressure difference can be maintained between the fracture and the matrix controls how much oil is discharged
into the fracture system.

The pressure amplitude is a matter of choice based on the required gas volume to inject. The time which a
pressure difference can be maintained depends on the reservoir/fluid properties. The higher the matrix fluid
mobility (k/μ) the shorter this time will be and hence the less the benefit over GOGD. Also the smaller the matrix
block size (or spacing between fractures) the shorter this time will be. It should be noted that as the process
continues and the gas saturation in the matrix increases the overall matrix fluid mobility increases thereby reducing
the time of pressure difference. This in addition to the lower relative permeability to oil further reduces the benefit of
the process over GOGD. Hence the maximum benefit is at the beginning of the process. This can be seen in the
huge benefit of CPP in the first 10 years in test 3 compared to the moderate benefit over a 30 years production life
in test 4 of Table 1. When the average rate from CPP becomes similar to GOGD rate later in the field life, CPP can
be stopped and GOGD continued. The following can be deduced:

a) The lower the oil viscosity (higher matrix fluid mobility) the less favourable CPP will be over GOGD as
explained above and seen in the results of test 1 and test 2 in Table.1
b) The larger the matrix block size the more favourable CPP will be in comparison to GOGD as explained
above and seen in the results of test 1 and test 3 in Table. 1.
c) The lower the matrix permeability the more favourable CPP will be in comparison to GOGD due to lower
fluid mobility allowing a longer period of pressure difference between matrix and fracture
d) Horizontal baffles in the matrix do not have a negative impact on CPP recovery as flow is horizontal unlike
in GOGD where baffles can greatly reduce recovery.
e) Large oil column height is required in GOGD but is not required in CPP as oil flows from everywhere
directly into the fractures.

Test 5 of Table.1 shows the extra benefit that can be derived by gas going back into solution during the re-
pressurisation and released during de-pressurisation.

Field observations
Figure 14 shows the result of a recent field observation. In this fractured reservoir the matrix has a significant oil
column while the fracture system has a very limited oil column whose thickness is monitored using gradio pressure
surveys in a well. The oil column and hence the amount of oil in the fracture is indicated by the levels of the gas oil
contact (GOC) and the oil water contact (OWC) in the fractures.
In figure 14a, it can be seen that upon start of production the oil rim started to move down. When the production
was increased it can be seen that not only did the oil rim further move down but its thickness was significantly
increased. The corresponding pressure drop during this increased production is shown in Figure 14b. Thus
indicating that the pressure drop has resulted in oil being pulled from the matrix into the fracture system.

Practical issues
In reservoirs with large gas saturations, it may be difficult to have any significant pressure amplitude due to the
very high compressibility of the system. The gas volume required for any desired pressure amplitude can be
calculated using the method proposed in reference [1]. Also reservoirs with strong aquifer connection could have
similar difficulties. An additional problem with reservoirs with strong aquifers is the difficulty in targeting the oil
spiked into the fractures as the aquifer influx would make the location of the oil rim variable not being necessarily at
the base of the reservoir.
It should be noted that pressure changes are much easier to achieve during steam injection/shut down as
condensation of the steam during shut down helps in lowering the pressure.
As suggested in reference [1] the process may be much easier to manage if multiple reservoirs are targeted so that
the injection shut down period of one is the injection period of the other.

Cycle times can be determined from simple modelling with the reservoir/fluid properties and modified as the
process progresses.

Conclusions
1) During studies of a steam assisted GOGD project in Oman the effect of a shut down in steam injection was
modelled. It was observed that more oil was produced into the fracture oil rim. This observation led to
further study of this phenomenon and the rediscovery of Cyclic Pressure Pumping.
2) Cyclic Pressure Pumping can potentially recover the matrix oil faster than GOGD.
4 [Paper Number]

3) A number of the conditions such as small oil columns, lower matrix permeability, higher oil viscosity and
horizontal baffles are more favourable to CPP than to GOGD.
4) A number of factors affecting the recovery of CPP have been studied.
5) For specific applications further studies would be required to quantify the benefit, if any, after taking into
account the impact of factors such as the economics of repressurisation to create the required pressure
amplitude, matrix permeability and matrix block size on the recovery.
6) Increasing oil rim thickness associated with a sudden drop in fracture pressure has already been observed
in the case reservoir and CPP is the most likely explanation for this.

References:
1. Alton B. Cook: “Alternate Producing and Gas Repressuring for Greater Oil Recovery” SPE 778-G (May,
1957)
2. George E. Crosby, Robert J. Cochran: “Performance of An Alternate Repressuring and Producing Project”
SPE 1395-G (September, 1960)
3. S.H. Raza: “Water and Gas Cyclic Pulsing Method for Improved Oil Recovery” SPE 3005 (October, 1970)
4. E. Artun, T. Ertekin, R. Watson, B. Miller: “Optimized Design of Cyclic Pressure Pulsing in a Depleted,
Naturally Fractured Reservoir” SPE 117762 (October, 2008)
5. E. Artun, T. Ertekin, R. Watson, M. Al-Wadhahi, B. Miller: “Performance and Economic Evaluation of Cyclic
Pressure Pulsing in Naturally Fractured Reservoirs” SPE 129599 (April, 2010)
6. Carlson, F.M.: “Simulation of Relative Permeability Hysteresis to the Non-Wetting Phase” SPE 10157
(1981)
7. C.A. Kossack: “Comparison of Reservoir Simulation Hysteresis Options” SPE 63147 (October, 2000)
8. D.G. Longeron, François Kalaydjian, Charles Bardon, L.M. Desremaux: “Gas/Oil Capillary Pressure:
Measurements at Reservoir Conditions and Effects on Gas-Gravity Drainage” SPE 28612 (September,
1994)
9. D.J. Element, J.H.K Masters, N.C. Sargent, A.J. Jayasekera, S.G. Goodyear: “Assessment of Three-Phase
Relative Permeability Models Using Laboratory Hysteresis Data” SPE 84903 (October, 2003)

Table. 1 Summary of simulation results

Test Matrix blck size oil viscosity Simulation years GOGD recovery CPP recovery % increase of CPP over GOGD
m cP % of STOIIP % of STOIIP
1 20 27 10 3.4 5.9 73.5
2 20 2.7 10 26.6 26.6 0.0
3 50 20 10 4.6 10.0 117.4
4 50 20 30 13.4 20.4 52.2
5 50 20 23 34.2 CPP below bubble point pressure

Figures
[Paper Number] 5

P1 S1 P2

Figure 1. Fluid saturation distribution in the fracture system in a fractured reservoir during steam injection. A cross
section through the center of the reservoir is shown with the gas (green), oil rim (red) and water (blue) in the
fractures displayed.

Figure 2. Production forecast for steam shut down and no shut down cases. Red solid – Oil rate from steam
injector shut down case. Red dashed – Oil rate from no shut down case. Green solid – Cumulative oil production -
steam shut down case. Green dashed – Cumulative oil - no shut down case
6 [Paper Number]

Red solid – Oil volume in fracture-steam injector shut down case

Red dashed – Oil volume in fractures-no shut down case
Green solid – Oil volume in matrix - steam shut down case
Green dashed – Oil volume in matrix - no shut down case

Figure 3 Oil volume in matrix and fractures for steam shut down and no shut down cases.

Figure 4. Oil PVT data

[Paper Number] 7

1.0

0.8 krg_drainage
Relative permeability

kro drainage
kro imbibition
0.6
krg imbibition

0.4

0.2

0.0
0.0 0.2 0.4 0.6 0.8 1.0
Liquid Saturation

Figure 5. Gas-Oil relative permeability data

200
180
160
140
Pcgo drainage
Pc (kPa)

120
100 Pcgo imbibition
80
60
40
20
0
0.0 0.2 0.4 0.6 0.8 1.0
Liquid saturation

Figure 6. Gas-Oil capillary pressure data

8 [Paper Number]

Figure 7. Initial oil saturation in stack for 20m matrix block

Monthy average production profile

1.4 7.0
CPP average oil rate
GOGD average oil rate
1.2 6.0
CPP Cum fraction of STOIIP produced
Fraction of STOIIP produced

GOGD Cum fraction of STOIIP produced

1.0 5.0
Oil rate m3/d

0.8 4.0

0.6 3.0

0.4 2.0

0.2 1.0

0.0 0.0

0 1000 2000 3000 4000

Time (days since start of production)

Figure 8. Comparison of oil production from CPP and GOGD 20m matrix block and 27 cP oil
[Paper Number] 9

First 5 months simulated production profile

3.5 70

3.0 60

Datum pressure Bar

2.5 50
Oil rate m3/d

2.0 CPP Oil rate 40

GOGD Oil rate
1.5 Datum pressure Bar 30

1.0 20

0.5 10

0.0 0
0 30 60 90 120 150
Time (days)

Figure 9. Comparison of first 5 months detailed oil production from CPP and GOGD 20m matrix block and 27
cP oil

Monthly average production profile

4.0 40.0

3.5 35.0

Fraction of STOIIP produced

GOGD monthly rate
3.0 CPP Monthly rate 30.0
GOGD Cum Oil
Oil rate m3/d

2.5 25.0
CPP Cum oil
2.0 20.0

1.5 15.0

1.0 10.0

0.5 5.0

0.0 0.0
0 500 1000 1500 2000 2500 3000 3500 4000
Time (days since start of production)

Figure 10. Comparison of oil production from CPP and GOGD 20m matrix block and 2.7 cP oil
10 [Paper Number]

Figure 11. Oil saturation after 11.7 years for GOGD for 20m matrix block and 2.7 cP oil

Figure 12. Oil saturation after 11.7 years for CPP for 20m matrix block and 2.7 cP oil
[Paper Number] 11

Comparison of CPP and GOGD for 20cp oil and 50m fracture spacing with pressure amplitude of 30 Bar

5.0 0.50
CPP Average oil rate UR GOGD = 13.4 % STOIIP
GOGD average oil rate UR CPP P>Pb = 20.4 % STOIIP
CPP below Pb average oil rate UR CPP P<Pb = 34.2 % STOIIP

Cumulative recovery fraction of STOIIP

4.0 0.40
CPP Cumulative recovery (fraction of STOIIP)
GOGD Cumulative recovery (fraction of STOIIP)
CPP Cumulative recovery below Pb
3.0 0.30
Oil rate m3/d

2.0 0.20

1.0 0.10

0.0 0.00
0 730 1460 2190 2920 3650 4380 5110 5840 6570 7300 8030 8760 9490 10220 10950

Time days

Figure 13. Comparison of oil production of CPP and GOGD for 50 m matrix block and 20 cP oil

Oil rim lowering

12000 -220
11000 -223
10000 -226

Fluid contact depth below

9000 -229
8000 -232

mean sea level

Offtake rate

7000 -235
6000 -238
5000 -241
4000 Offtake rate -244
3000 Fracture Gas-Oil contact -247
2000 Fracture Oil-Water contact -250
1000 -253
0 -256
1-Sep-09 28-Feb-10 27-Aug-10

Figure 14a Fracture Oil rim montoring of a fractured reservoir impact of production on oil volume in the fractures

Oil rim lowering

3160 -220
Fluid contact depth below

3150 -225
Pressure (kPa)

sea mean level

3140 -230

3130 -235

3120 pressures -240

Fracture Gas-Oil contact
3110 -245
Fracture Oil-Water contact
3100 -250
1-Sep-09 28-Feb-10 27-Aug-10

Figure 14b Fracture Oil rim montoring of a fractured reservoir impact of pressure on oil rim volume in fractures