Petroleum Technology 7

PPaappeerr IIDD 3388

The Effect of Foam Stability in CO2-Foam Flooding

K. Teerakijpaiboon1*, F. Srisuriyachai1 1Department of Mining and Petroleum Engineering, Chulalongkorn University, Thailand

*E-mail: kunwadee.t@student.chula.ac.th

ABSTRACT CO2-foam flooding is implemented to minimize drawbacks of CO2-flooding by reducing high mobility

of CO2. Performance of CO2-foam flooding depends on several factors including foam stability. In this study, results shows that varying foam stability slightly affects flooding performance. CO2-foam application is favorable when reservoir wettability is water-wet condition. For oil-wet reservoir, CO2 flooding shows similar or slightly better result compared to CO2-foam flooding. Oil composition also affects CO2-foam. Benefit from CO2-foam over CO2 flooding is greater when hydrocarbon contains low intermediate component. The best strategy for CO2-foam flooding is injecting one slug of CO2-foam followed by chasing water. KEY WORDS: CO2-Foam flooding /Foam stability 1. INTRODUCTION Carbon dioxide or CO2 flooding is one of the most widely used and the most well-known EOR techniques because CO2 can perform both miscible and immiscible conditions in wide range of hydrocarbon properties to improve oil recovery. However, CO2 viscosity is much lower than crude oil. This results in high mobility ratio and consecutively leaves oil by-passed behind. Viscous fingering and gravity overriding effects are commonly results from unfavorable mobility ratio, leading to an early breakthrough of injected CO2. In order to minimize drawbacks of CO2 flooding, foam is generated to reduce mobility of gas phase and hence, sweep efficiency is improved by decreasing mobility ratio. Foam has been also used as controlling and blocking agent to prevent rapid gas invasion in high permeability streaks[1].

When foam gets in contact with oil, foam disintegrates and turns into CO2 gaseous and surfactant liquid forms[2]. Oil recovery mechanism obtained by foam flooding is a combination between surfactant and CO2 characteristics which are: 1) lowering interfacial tension (IFT) to proper value that oil can be liberated and consecutively stabilized as small droplets in aqueous phase and; 2) CO2 can be miscible with oil, reduce oil viscosity and hence, make oil ready to flow.

Efficiency of foam flooding depends on many factors. One of the most important parameters is foam stability. Foam stability is defined as half-life of foam or time required for half of foam volume to decay or collapse[3]. In this study, the reservoir simulator called STARS® commercialized by Computer Modeling Group Ltd. (CMG) is used as investigation tool. A homogeneous reservoir model is constructed.

Properties of hydrocarbon and reservoir are obtained from Sirikit oil field provided by PTT Exploration and Production Public Company Limited (PTTEP). Appropriate values of foam stability are applied for foam slug. Lithology of rock is the first parameter in sensitivity analysis. A range of wettability from moderately water-wet to strongly oil-wet is chosen. Percentage of intermediate in volatile oil is also considered. As CO2 is known as a potential vaporizer, amount of intermediate plays a major role in this process. Last, foam slug sizes are chosen for operational parameter study. Foam slugs are divided into many slugs and injected alternately with chasing water. Total foam volume is equal in all cases and ratio of foam to alternating water slug is kept constant.

2. RESERVOIR SIMULATION INFORMATION

2.1 Reservoir section

Reservoir model is created as Cartesian grid and represents homogeneous reservoir. Properties are shown in Tab. 1 Tab. 1 General reservoir model properties

Property Value Top reservoir depth, feet 6,000 Grid block number 30 × 15 × 20 Grid size, feet 100 × 100 × 10 Thickness, feet 200 Porosity 0.25 Initial water saturation 0.28 Horizontal permeability, mD 220 Vertical permeability, mD 22

Petroleum Technology 8

2.2 PVT properties section

Oil composition of Sirikit oil field shown in Tab.2 was analyzed to determine phase behavior and to create properties of components by using Winprop application provided by CMG[4]. Tab. 2 Hydrocarbon compounds from Sirikit oil field

Component Mole fraction Carbon dioxide (CO2) 0.0091 Nitrogen (N2) 0.0006 Methane (C1) 0.3383 Ethane (C2) 0.0904 Propane (C3) 0.0799 Isobutane (i-C4) 0.0197 Normal butane (n-C4) 0.0469 Isopentane (i-C5) 0.036 Normal pentane (n-C5) 0.0178 Hexane (C6) 0.0501 Heptane plus (C7+) 0.3112

It is noted that specific gravity of hepthane plus (C7+) is 0.8615, whereas its average molecular weight is 267. Calculated Minimum Miscibility Pressure (MMP) of this oil is about 2,800 psi.

2.3 SCAL section

Relative permeabilities to oil and water as a function of water saturation and relative permeabilities to gas and liquid as a function of liquid saturation that are used for base case simulation are shown in Fig.1 and Fig.2, respectively. From both figures, it can be seen formation wettability is a strongly water-wet condition.

Fig. 1 Relative permeabilities to oil and water as a function of water saturation.

Fig. 2 Relative permeabilities to gas and liquid as a function of liquid saturation.

2.4 Well and recurrent section

Production well is located at the edge of reservoir model, while injection well is at the opposite edge on another side. Constraints and economic limits of both wells are listed in Tab. 3 Tab. 3 Constraints and economic limits of production well and injection well

Constraints and economic limites value Production well Maximum oil rate, STB/D 2,000 Maximum water rate, STB/D 2,000 Maximum gas rate, MMSCF/D 10 Minimum bottomhole pressure, psi 800 Cut-off oil production rate, STB/D 100 Water cut, % 95 Injection well Maximum bottom hole pressure, psi 4,100 Injection pressure, psi 3,000

3. METHODOLOGY 1. Create a homogeneous reservoir model. 2. Simulate CO2 flooding base case. 3. Perform CO2-foam flooding on the same

reservoir mode of CO2 flooding base case. Five foam stability values are selected which are 20, 40, 80, 160 and 320 days.

4. Study effect of wetting condition of reservoir formation. The study is performed on formation wettability varied from original value in a direction to a more oil-wet condition. The wetting condition is varied to moderately water-wet, neutral-wet, moderately oil-wet, and strongly oil-wet conditions.

5. Study effect of intermediate percentages of hydrocarbon in volatile oil which is adjusted by increasing and decreasing percentage of intermediate compounds approximately 10% and 20% compared to base case.

6. Study effect of slug size by dividing 0.4 pore volume into two slugs of 0.2 pore volume and

Petroleum Technology 9

three slugs of 0.133 pore volume, each slug is alternated by chasing water.

7. All simulation results are compared among cases of each study parameter to determine effectiveness of foam flooding.

8. Compare, analyze and summarize the most suitable foam stability in each circumstance which yields the highest oil recovery.

4. RESULTS AND DISCUSSION After reservoir model has been constructed, CO2-foam models are simulated. The varied five foam stabilities are applied in these foam models. Alteration of oil compositions, rock wettability, and foam slug size are performed. In this study, CO2 flooding is selected as reference case used for comparison with other foam flooding cases. Fluid injection schedule can be divided in two periods. First, CO2 is injected for 0.4 hydrocarbon pore volume, afterwards water is injected after CO2 slug to chase all slugs until cumulative amount of water reaches 0.4 pore volume for both CO2 flooding and CO2-foam flooding. The simulation results are considered at the time where cumulative water is 0.4 pore volume.

4.1 CO2 flooding

Injection rate is initially kept at 9.5 MMSCF/D. Injection pressure is fixed at 3,000 psi in order to ensure that miscibility of CO2 can be achieved throughout the process.

Fig.3 Oil, water and gas production rates of CO2 flooding base case. From Fig. 3, produced oil is constantly maintained for while. After that, oil rate declines as a result of gas breakthrough. Gas production rate is controlled not to exceed 10 MMSCF/D throughout production period. The first slight increment of oil rate starts at 2,950 days or approximately two years after injecting of chasing water. This incident is a result from an increase of relative permeability of water. Hence, influence of gas on oil flowing is minimized. High rate of oil production is maintained for while before a drastic decline due to reduction of bottmehole pressure. The second increment of oil rate

starts at around 4,300 days. This incremental recovery comes from chasing water sweeps oil from bottom zone. This arrival of oil is in coincidence with high water production rate that is water breakthrough period. Oil rate rises up for while and then declines again because water is a dominant phase in flowing. Water rate escalates rapidly, and remains constant at the maximum rate throughout production period. Oil recovery factor of CO2 flooding base case is around 42.62%

4.2 CO2–foam flooding

Due to complexity of co-injecting two fluids in the same well by simulation program, CO2 and water are injected from separated imaginary wells but they are both located at the same coordinate. Injection rates in this study are 8.78 MMSCF/D for CO2 and 500 STB/D for surfactant solution. Summation of CO2 injection rate and surfactant solution injection rate in foam application is approximately equal to CO2 injection rate in CO2 flooding. Foam stabilities are varied from 20 days, 40 days, 80 days, 160 days, and 320 days in order to study their effect on foam flooding performance. For all foam simulations in this study, surfactant concentration is kept constant at 0.5% w/v.

Fig. 4 Oil production rates of CO2–foam flooding base case.

Fig. 5 Water production rates of CO2–foam flooding base case. Oil and water production rates are displayed in Fig. 4 and Fig.5, respectively. Oil can be produced at constant rate for 4,350 days. Afterwards, water

Petroleum Technology 10

breakthrough occurs and this diminishes oil production rate, consequently oil rate drops drastically. Although water production rate reaches the maximum limit and oil rate decreases abruptly, oil can still be produced at small rate at approximately 100-200 STB/D until oil rate approaches the production constraints which are 95% of water cut and 100 STB/D of oil production rate.

Fig. 6 Gas production rates of CO2–foam flooding base case. From Fig. 6 there are obvious peaks of gas rate in the period of 440-1,000 days of production. These peaks result from part of CO2 which cannot be captured by lamella. Foam therefore cannot be formed immediately at initial stage of fluid injection. CO2 that is not accounted for foam generation in the initiation stage then flows deep in reservoir. After that CO2 undergoes miscibility with reservoir oil. Methane and intermediates are vaporized and move forward to production well. Because there is only small amount of gas that cannot be encapsulated by foam, produced gas peaks are not high enough to interrupt oil production rate. At about 4,320 days of production, gas production rates drop as almost the same time of where oil production rates suddenly decline. This is a result from water breakthrough. Around 700 days after that, all gas production rates rise up again. This situation occurs because CO2 and surfactant solution injections are stopped since 4,950 days, accordingly. When existing foam in reservoir coalescences and breaks into free CO2 and no new foam are generated, free CO2 is miscible with reservoir oil and so vaporized methane and intermediates in oil substantially increase. In late period, all produced fluid rates are dropped due to declining of reservoir pressure. It is found that variation of foam stabilities does not affect significantly production performances. This result could be from injecting too big slug of CO2-foam. When exist foam lamella breaks, there are new foams generated continuously so an effect of foam stability cannot be seen obviously. The ultimate oil recovery factors of CO2–foam flooding are in range of 55.12-55.63%. Foam stability of 80 days provides the best oil recovery.

Base on oil recovery factor, it can be seen that foam flooding can recovery oil much better that CO2 flooding since foam can reduce gas mobility. This yields smoother flood front compared to the case of CO2 flooding. Since there is no direct way to visualize mobility of gas phase in STARS simulator, mobility is however an ability to move and this can be seen from flowing speed. Fig. 7 shows an evolution of flood front of CO2 flooding compared to CO2-foam flooding captured at the same time. It is observed that in CO2 flooding, flood front moves very fast but CO2-foam flood front flows is much slower.

Fig. 7 Evolution of flood front of CO2 flooding compared to CO2-foam flooding.

4.3 Effect of varied parameters on CO2-foam flooding

Several parameters are applied in the CO2-foam model to evaluate their effects and sensitivity on the performance of CO2-foam. All cases are simulated under the same production constraints as well as injection rates of base case CO2-foam model consisting of CO2 injection rate of 8.78 MMSCF/D and surfactant solution injection rate of 500 STB/D. Each parameter is independent to others. The comparison of CO2-foam flooding performance is performed by comparing with result obtained from CO2 flooding.

4.3.1 Effect of wetting condition of reservoir rock

This study is performed to evaluate effects of wetting conditions that are varied from an original value in a direction to more oil-wet. Sensitivity analysis of wetting condition on effectiveness of foam flooding is carried out by adjusting SCAL data. From the relative permeability rule of thumb, reducing irreducible water saturation, increasing relative permeability to water at residual oil saturation, and

Petroleum Technology 11

decreasing of crossover saturation is a sign of direction to a more oil-wet condition. Four types of wettability are investigated in this study. Wettability of reservoir in the base case is considered as strongly water-wet, therefore wetting conditions which are further investigated are varied to moderately water-wet, neutral-wet, moderately oil-wet, and strongly oil-wet. Base on simulation results, performance obtained by any foam stability does not show much different, only one of five foam stabilities is a representative for all cases.

Fig. 8 Oil production rates of CO2–foam flooding cases with variation of wettability conditions.

Fig. 9 Water production rates of CO2–foam flooding cases with variation of wettability conditions. From Fig. 8 and Fig. 9, illustrating produced oil rates and produced water rates in each wettability condition, it can be summarized that the more oil wet condition, the earlier water breakthrough at production well and hence, the lower ultimately oil recovery factor. This is because stronger oil-wet tends to attach less water onto surface than water-wet condition. Therefore water moves quicker and reaches earlier to production well. Late water breakthrough causes longer production period and results in greater oil recover factor as shown in Fig. 10

Fig. 10 Oil recovery factors of CO2–foam flooding cases with variation of wettability conditions. It can be concluded that CO2-foam flooding is suitable for a reservoir that its wettability is in range of neutral-wet to strongly water-wet. For oil-wet formation, solely CO2 flooding is preferable due to no effect of water breakthrough during gas injection period as shown in case of strongly oil-wet condition in Fig.11

Fig. 11 Oil production rates of CO2–foam and CO2 flooding cases in strongly oil-wet reservoir. For strongly oil-wet condition, period that oil is produced with constant rate of CO2-foam is not much longer than solely CO2 injection. So an advantage of CO2-foam is not obviously seen because in late production period water breakthroughs at production well drastically reduces oil rate of CO2-foam flooding.

4.3.2 Effect of intermediate percentages of hydrocarbon in volatile oil

Percentage of intermediate component is varied from oil composition data obtained from S1 oil field by increasing percentage of intermediate compounds about 10% and 20% compared to base case. When intermediate content is increased, part of heavy compounds (C7+) is decreased proportionally. Other two additional cases of lower percentage of intermediate component of about 10% and 20% are also investigated. Similarly, part of heavy compounds is increased in these cases.

Petroleum Technology 12

Fig. 12 Oil production rates of CO2–foam and CO2 flooding cases when increasing intermediate component in oil approximately 20%.

Fig.13 Water production rates of CO2–foam ad CO2 flooding cases when increasing intermediate component in oil approximately 20%.

Fig.14 Gas production rates of CO2–foam and CO2 flooding cases when increasing intermediate component in oil approximately 20%. Fig. 12-14 show oil production rate, water production rate, gas production rate, respectively. For CO2-foam flooding, it is found that initial produced gas rates are already close to limit gas production rate. Oil rates starts to decline due to gas and water breakthroughs which both of them come from CO2-foam breakthrough (observed from arrivals almost the same time of water and gas at the production well). Foam moves fast in this case because flow property of foam is controlled by relative permeability to water. When intermediate portion in oil is increased about

20%, components of liquid oil after emerging of miscibility are reduced rapidly. Consider water and oil relative permeabilities, when oil saturation reduces relative permeability to water increases and hence, foam can flow much faster. Both gas and water liberated from foam compete against oil, flowing to production well. Therefore, oil rates fall rapidly and reach economic limit. It is obvious that CO2-foam can prolong constant oil rate longer than solely CO2 flooding case. For CO2 flooding in increasing intermediate component 20% cases, oil production rate rapidly fall as a result of gas breakthrough. But this drop is not as high as base case because initial produced gas rate is not much different from limit gas production rate. This can be inferred that relative permeability to oil is not reduced much. Nevertheless, oil rate is continuously decreased and finally production is terminated. For decreasing intermediate component 20% cases, oil production rates, water production rates, and gas production rates are illustrated in Fig. 15-17.

Fig. 15 Oil production rates of CO2–foam and CO2 flooding cases when decreasing intermediate component in oil approximately 20%.

Fig.16 Water production rates of CO2–foam ad CO2 flooding cases when decreasing intermediate component in oil approximately 20%.

Petroleum Technology 13

Fig.17 Gas production rates of CO2–foam ad CO2 flooding cases when decreasing intermediate component in oil approximately 20%. Oil production rate of all case can be kept operated constantly for a while. After that, rates fall roughly due to arrival of CO2-foam. It is observed that CO2-foam can travel in reservoir for longer time than other cases (foam breaks before 4,000 days in base case). The reason that most of foam does not break during travelling is that foam tends to be more stable in heavier oil. Lighter oil composing of short chain alkanes has ability to enter into CO2 and surfactant interfaces of lamellae. This leads to weakening of foam bubble and eventually, foam ruptures[5]. Therefore, more heavy oil component causes higher stability of foam [6]. Regarding CO2 flooding in cases where reduction of intermediate component 20% is applied, results show similar trend as seen in results of CO2 base case. But, period of gas breakthrough is shorter than other cases. Moreover, period that water sweeps oil in lower zone is longer than other cases. This can be described that, saturation of intermediate component which can be vaporized is lower and heavy hydrocarbon remained after vaporization is higher. It can be concluded that higher intermediate component in oil composition induces faster flow of CO2-foam slug. Another thing to be noticed is that low intermediate oil does not cause rupture of CO2-foam as much as in high intermediate component oil. In cases of decreasing intermediate component, foam hardly rupture and be produced simultaneously with oil. An arrival of foam at production well can be noticed from coincidence of gas and water production rate.

4.3.3 Effect of slug injection

Effect of slug size is studied by dividing foam injection from 0.4 pore volume into two slugs of 0.2 pore volume and three slugs of 0.133 pore volume. Each slug is alternated with chasing water slugs. Total foam volume must be equal in all cases and also ratio between foam and alternating water slug is kept constant. Simulation results show that division of CO2-foam into smaller slugs, alternating with chasing

water slightly impacts production characteristic of CO2-foam flooding. Small size of CO2-foam reduces capability of pressure maintenance by foam. Injecting of one slug of CO2-foam followed by one slug of water, results in maintaining high pressure at first period and sudden drop of pressure in latter period. On the other hand, splitting CO2-foam into small slugs causes lower reservoir pressure at first and higher pressure is followed. Fig. 18 displays bottomhole pressures at production well which is previously mentioned.

Fig. 18 Bottomhole pressures at production well with variation of CO2-foam slugs. Lower pressure in first stage of double and triple slug cases result in slightly retarding water breakthrough and late declining of oil rate as shown in Fig. 19. Pressure in latter stage is more important because high pressure leads to maintaining of produced water after water breakthrough. Therefore, production is terminated due to water cut reaches preset value of 95% for both double-slug and triple-slug cases. On the other hand, lower pressure of single-slug case causes drop of water production and hence, water cut does not reach preset limit and oil can be produced until it approaches economic limit of 100 STB/D. In summary, in order to obtain the best result from CO2–foam flooding, single-slug mode is recommended.

Fig. 19 Oil production rate of CO2–foam flooding with variation of CO2-foam slug.

Petroleum Technology 14

5. CONCLUSIONS Most of CO2-foam flooding cases have higher potential to enhance hydrocarbon recovery in comparison to solely CO2 flooding. This is because foam reduces mobility of gaseous CO2 and provides smoother flood front. Variation of foam stability does not significantly impact on production performance of CO2-foam flooding. It could be possible that this is a result from injecting continuously big slug of CO2-foam and there are other parameters involve such as ability of foam regeneration. Nevertheless, difference of oil recovery factors by varying foam stability is smaller than 2%. Hence, foam stability seems to be insensitive to production performance of CO2-foam flooding in this study. Influences of study parameters on effectiveness and performance of CO2-foam flooding are summarized as follows:

5.1 Effect of wettability

1) In order to achieve good performance of CO2-foam flooding, the best suit formation wettability should be in a range of neutral-wet to strongly water-wet.

2) CO2 flooding yields better performance compared to CO2-foam flooding when performs in reservoir having rock wettability ranging from oil-wet or strongly oil-wet condition.

5.2 Effect of intermediate percentages of hydrocarbon in volatile oil

1) Implementing CO2-foam flooding with light oil containing high component of intermediate (C2-C6) results in high velocity of injected foam as well as aqueous phase in reservoir.

2) Intermediate compound tends to destabilize foam more than oil containing higher component of heavy compound (C7+). Rupture of foam is caused by smaller molecules of intermediate that can access into the interface of CO2 and surfactant and results in breaking of the film of lamella.

3) Advantage of CO2-foam over CO2 flooding is higher when hydrocarbon in reservoir contains low intermediate component.

5.3 Effect of slug injection

1) Dividing CO2-foam slug into smaller slugs such as double-slug or triple-slug modes can maintain pressure after water breakthrough slightly better than injection a single-slug. Therefore, when water reaches production well, oil production is terminated by reason that water cut reaches its production limitation faster than injecting single-slug.

2) In this study, single-slug mode of CO2-foam provides the most satisfied outcomes.

3) For solely CO2 flooding, double-slug and triple-slug modes yield better production performance due to reduction of gas breakthrough.

ACKNOWLEDGEMENTS

The author would like to express thankfulness to PTT Exploration and Production Public Company Limited (PTTEP) for providing financial support as well as reservoir simulation data.

REFERENCES [1] Y.Zhang, X.Yue, J.Dong, and L.Yu(2008), New

and Effective Foam Flooding To Recover Oil in Heterogeneous Reservoir. Presented at the 2000 SPE/DOE on Improved Oil Recovery Symposium, Tulsa, Oklahoma, USA, April 3-5, 2000, Paper No. SPE 59367

[2] G.C.Wang(1984), A Laboratory Study of CO2 Foam Properties and Displacement Mechanism. Presented at the SPE/DOE Fourth Symposium on Enhanced Oil Recovery, Tulsa, Oklahoma, USA, April 15-18, 1984, Paper No. SPE/DOE 12645

[3] E.C.Donaldson, G.V.Chilingarian, and T.F.Yen(1985), Enhanced Oil Recovery, I: Fundamentals and Analysis. Elsevier Science Publishers B.V., 1985.

[4] Computer Modelling Group, User’s Guide STARS : Advanced Process and Thermanl Reservoir Simulator (2009), Alberta, Canada, 2009.

[5]R.Aveyard, B. P.Binks, P.D.I.Fletcher ,T.G.Peck, and P. R.Garrett (1993), Entry and spreading of alkane drops at the air/surfactant solution interface in relation to foam and soap film stability, Journal of the Chemical Society, Faraday Transactions, Issue 24, 1993.

[6] L.L.Schramm, and J.J.Novosad (1992), The

destabilization of foams for improved oil recovery by crude oil: Effect of nature of oil, Journal of Petroleum Science and Engineering 7, 77-90 Elsevier Science Publishers B.B., Amsterdam, September, 1992.