Experimental Study of Hot Water Injection into Low-Permeability

Energy & Fuels 2008, 22, 2353–2361

2353

Experimental Study of Hot Water Injection into Low-Permeability Carbonate Rocks Behnam Sedaee Sola*,† and Fariborz Rashidi‡ Department of Petroleum Engineering and Department of Chemical Engineering, AmirKabir UniVersity of Technology, 15875-4413, Tehran, Iran ReceiVed January 6, 2008. ReVised Manuscript ReceiVed March 28, 2008

Hot water drive involves the flow of only two heated phases. The major mechanisms on hot water injection are thermal expansion, viscosity reduction, wettability alteration, and oil/water IFT reduction. In this study, hot water injection experiments were carried out using unpreserved limestone and dolomite core samples obtained from the oil zones of heavy oil low-permeability reservoirs. These experiments were conducted at reservoir pressure but in various temperature ranges up to 500 °F using a wide variety of oils. The final oil recovery, residual oil saturation, irreducible water saturation, and pressure drop were compared in each experiment. The results of dynamic isothermal displacements were interpreted using numerical simulation method to obtain reliable relative permeabilities. Hence, the effects of temperature on oil/water relative permeabilities were obtained for low-permeability carbonate rocks. Results show that it is possible to recover a high percent of oil using high-pressure/high-temperature injection even from heavy oils in low-permeability carbonate reservoirs. In the heavy oil system, the oil production to hot water injection ratio is higher than in the medium and extra heavy oil, but the values are less than the reported values for conventional heavy oil reservoirs. Also, it was found that the relative permeabilities of oil/water depend on temperature and the residual oil saturation decreases and irreducible water saturation increases when the rock is heated.

1. Introduction Many believe that the era of conventional oil will soon come to an end and heavy and nonconventional oil will be replaced by easy producing oils. There is more than 1600 billion barrels of heavy oil in carbonate rocks and most of these kinds of reservoirs require thermal enhanced oil recovery methods for production.1 Also, low-permeability fractured reservoirs contain a large volume of the worldwide oil resources.2 Thermal enhanced oil recovery methods that have been applied in the field include hot water drive, steam injection, and in situ combustion. Because of the presence of water in all petroleum reservoirs, flow of hot water will occur to some extent in all thermal recovery processes. Oil viscosity decreases manyfold upon heavy oil heating and therefore the mobility ratio of the fluids in the heated zones is more favorable than the cold water injection. This results in better macroscopic displacement efficiency and would improve the ultimate recovery. The improvement in recovery of viscous crude oils by hot fluid injection is primarily due to the improved oil mobility and reduction in residual oil saturation. Figure 1 shows schematically the effect of oil gravity on the most significant mechanisms in hot fluid injection. As it seems, in heavy oil reservoirs, the * To whom correspondence should be addressed. E-mail: [email protected]. Fax: +98-21-88946426. † Department of Petroleum Engineering. ‡ Department of Chemical Engineering. (1) Briggs, P. J.; Beck, D. L.; Black, C. J. J.; Bissell, R. Heavy oil from fractured carbonate reservoirs. SPE ReserVoir Eng. 1992, 173–179. (2) Saidi, A. M. Presented at the SPE Reservoir Simulation Symposium, San Francisco, Nov 1983; Paper SPE 12270; pp 15-18. (3) Butler, R. Thermal recovery of oil and Bitumen; Prentice-Hall: Englewood Cliffs, NJ, 1991.

Figure 1. Relative contributions of various mechanisms on the displacement efficiency of oil by hot water injection.4

objective of hot water is to increase production by reduction in oil viscosity. Hot waterflooding is usually less effective than steamflooding because of the lower heat content of hot water than steam and the steam-distillation effect of steam. Also, it is found that the residual oil saturation level that can be achieved with a hot (4) Prats, M. Thermal Recovery; Society of Petroleum Engineers: Dallas, TX, 1982; SPE Monograph Volume 7.

10.1021/ef800009r CCC: $40.75  2008 American Chemical Society Published on Web 05/20/2008

2354 Energy & Fuels, Vol. 22, No. 4, 2008

Sedaee Sola and Rashidi

Table 1. Different Properties of the Used Oils in the Experiments

a

name

density of oil at standard condition, g/cm3

dead oil viscosity at 100 °F, cP

C7+, %

EHOa HOb MOc

0.99 0.95 0.90

16000 2094 10

85.60 84.81 92.00

thermal expansion factor, °F-1, 10-4

compressibility, psi-1, 10-6

4.00 4.24 -

4.67 5.90 -

Extra heavy oil. b Heavy oil. c Medium oil.

waterflood is markedly higher than that found with steam even at the same temperature.3 Many experimental and theoretical works have been carried out regarding steam injection into carbonate reservoirs. However, hot water injection has received relatively less attention, probably due to its limited application in carbonate reservoirs as an enhanced oil recovery method. In most cases, hot water injection is used as a stimulation method. Although there are published works on hot water injection into sandstone rocks and carbonate reservoirs,5–11,17 it seems there is no published study on hot water injection into lowpermeability carbonate reservoirs. Therefore, the objective of this study was to investigate hot water injection into lowpermeability carbonate rocks in heavy and medium oils. 2. Experiments 2.1. Material. 2.1.1. Fluid Properties. Various types of crude oils having a wide range of densities and viscosities (medium to extra heavy oil) from different Iranian fields were used in this study. Density, dead oil viscosity at 100 °F, mole fraction of C+7 fraction, thermal expansion factor, and compressibility of these samples are given in Table 1. One of the most important parameters that affects any thermal enhanced oil recovery is the oil viscosity variations with temperature. Therefore, it is important to use measured viscosity versus temperature rather than correlations. For this purpose, the viscosity of crude oils in experimental conditions was measured using the rolling ball method and is shown in Figure 2. Initially all of the oil samples were at stock tank condition. Fortunately, none of the samples had any water content at this condition. 2.1.2. Rock Properties. Two types of low-permeability carbonate rocks were used in this study (Table 1). The cores are dolomite or limestone or combination of these. The absolute permeability of all samples was less than 3 mD (millidarcy). Due to the carbonate nature of the cores, special care was needed to cut them as small plugs from the whole core. After cutting, the samples were washed completely with toluene over three days and dried in an oven at more than 300 °F for at least 24 h to remove all possible oil and fines from rocks. The core plug diameter and length were 3.83 and 8.2 cm in all of experiments, respectively. Table 2 shows the measured porosity (by helium porosimeter) and absolute permeability (by air permeameter) of each sample. The absolute permeability of these cores varied from 0.1 to 3 mD which is an indication of no fractures and low-permeability core samples. (5) Ditz, D. N. Presented at 7th World Petroleum Congress, Mexico City, 1967. (6) Dietrich, W. K.; Willhite, G. P. Presented at the Petroleum Industry Conference on Thermal Oil Recovery, Los Angeles, June 6, 1966. (7) Bursell, C. G.; Taggart, H. J.; Demirjian, H. A. Producers Monthly 1966, (Sept), 18. (8) Socorro, J. B.; Reid, T. B. Presented at the Terceras Jordans Tencicas the Petroleum, Sociedad Venezolana de Petroleum Maracaibo, Venezuela, 1971. (9) Sahuquet, B. C.; Ferrier, J. J. J. Pet. Technol. 1982, 873–880. (10) Chierici, G. L.; Delle Canne A. ; Properzi. Presented at the 1985 European Meeting on Improved Oil Recovery, Rome, Italy, April 1985. (11) Dreher, K. D.; Kenyon, D. E. Presented at the 1986 SPE/DOE Symposium on Enhanced Oil Recovery, Tulsa, OK, April 1986; Paper SPE 14902, pp 20-23.

Figure 2. Oil viscosity of different oil samples used in this study as a function of temperature. Table 2. Porosity, Permeability, and Lithology of the Cores Used in Experiments core name

preferred lithology

porosity, %

abs permeability, mD

KM-B KM-C MN-X

limestone limestone dolomite

20 21 21

0.3 0.2 3.0

2.2. Equipment. Figure 3 shows a schematic representation of the displacement apparatus used in this study. This apparatus includes three sections: Injection, coreholder and production. Water was injected using two ISCO-LC-5000 positive displacement pumps which were working in parallel. These pumps can inject with high accuracy from 0.10 to 400 cm3/h at pressures higher than 4000 psi. The injected water is heated in the steam generator at the desired temperature. A Fenwal temperature controller was used to inject the hot fluid at constant temperatures. For oil injection, a mercury Ruska pump which can inject at pressures higher than 10 000 psi was used. This pump also has two parallel limbs which allow continuous injection. It was possible to inject oil at constant rate and pressure with this pump. The coreholder was built from stainless steel which was capable of loading overburden pressure over 10 000 psi and operating temperatures up to 650 °F. Coreholder heads were sealed with rubber O-rings and Teflon pads. Nitrogen was used as overburden fluid. Overburden pressure was always 200 psi more than the core pressure. Back pressure regulator was used to control the core pressure and maintain the system in the two-phase mode (oil-hot water) and prevent steam formation. The coreholder and measurement tools were placed in an oven to control its temperature as shown by dotted line in Figure 3. A fan was blowing air inside the oven to maintain constant temperature inside the oven. The effluent was collected in graduated cylinders for analysis and separation of oil and water. Production fluid was cooled in a condenser before collection. Graduated cylinders were used to measure produced fluid. Due to the formation of emulsion, the weighting method of the produced fluid would not yield correct results. Pressure and temperature transducers and pump calibrations were checked before each experiment. The accuracies of the measurement system of the production fluids, thermocouples, and pressure transducers were 0.05 cm3, 2 °F, and 10 psi, respectively. The dead volumes of all of the flow lines were measured and considered in all material balance calculations.

Hot Water Injection into Carbonate Rocks

Energy & Fuels, Vol. 22, No. 4, 2008 2355

Figure 3. Schematic of experimental apparatus.

2.3. Experimental Procedure. After the cores were washed and dried, they were placed into the core holder and vacuum was applied for about 1 h to all flow lines and the core. After enough vacuum was established, the core was saturated with water and then absolute permeability was determined at the experiment temperature. No significant difference in absolute permeability was observed. To reach ion exchange equilibrium between rock and water components, 24 h were allowed prior to each experiment. Then, the irreducible water saturation can be established by displacing water by oil at the experimental temperature and pressure. For high-temperature cases, oil was heated up to the experiment temperature during injection to attain injectivity and thermal equilibrium. Oil injection continued until the water fraction of the produced fluid decreased below 1%. After the irreducible water saturation was established, the oven temperature was fixed at the desired level and the system was allowed to reach thermal equilibrium. The volumes of oil and water produced during this process were monitored. In most runs, more than 5 days passed before the oil reached equilibrium with the water at high temperature and pressure. During this period, oil was produced due to thermal expansion. This ensured that no oil was produced during thermal injection due to expansion. This period was assumed as restore or aging time. Next, the oil permeability was measured at irreducible water saturation. All produced fluids were collected in graduated calibrated cylindrical tubes and then a number of drops of demulsifier were added to them and they were centrifuged at high speed for about 1 h. For extra heavy crude oil tests, a small amount of solvent (toluene) was added using high accuracy syringe to make the separation of the oil and water easy for the next step. After thermal equilibrium was reached, as indicated by cessation of oil production, constant rate water flooding was started in vertical direction. The waterflood was continued until oil production ceased.

Table 3. Operational Condition of the Experiments expt name

core name

crude oil name

temp, °F

av press., psi

KM-B-MO

KM-B

MO

MN-X-HO

MN-X

HO

KM-C-EHO

KM-C

EHO

100 200 200 300 450 150 200 250 300 400 500

1200 800 1800 1400 1000 2000 1800 1500 1400 900 700

At the end of runs, the cores were washed using hot toluene and, subsequently, water permeability was measured to investigate the effect of high-temperature process on absolute permeability.

3. Results and Discussion In this section, the results of three experiment sets for three different oil samples, i.e., medium oil, heavy oil, and extra heavy oil will be analyzed. Table 3 summarizes those experiments and operational conditions. It should be mentioned that it was impossible to inject at a minimum rate to overcome capillary end effect problem. Rapoport and Leas12 suggested that the following criterion should be satisfied to avoid any end effect: luµw > 3.5, and this number was less than 1 in all of the experiments. However, the pressure gradients were high enough to ignore the capillary pressure effects and the fluid velocity was close to reservoir velocity.


Figure 4. Oil production and pressure drop from thermal expansion (temperature increased 200 °F).

In some of the high-temperature tests, after irreducible water saturation was established, the temperature of the oven was fixed at the designed level and the system was allowed to reach thermal equilibrium. The volume of oil and water produced in this process were monitored. In one case, extra heavy crude oil was injected at reservoir temperature and then the oven temperature increased to 400 °F for monitoring the thermal expansion effect. The oil production and the pressure drop of this test are shown in Figure 4. As is seen, 11% of oil which is extra heavy oil can be recovered by heating the system about 200 °F above its reservoir temperature. This is a valuable result for thermal EOR process. As mentioned in the previous sections at the end of runs, several pore volumes of toluene were injected to wash the cores and subsequently water permeability was measured to investigate the effect of process on the absolute permeability. One of these effects is formation damage due to heating and oil/water injection. The formation damage process during thermal processes would include mineral transformation, mineral solubilization or dissolution, wettability alterations, and in situ emulsion formation.13 For investigation of this effect, two SEM pictures were taken before and after experiment. Figure 5 shows these images in two different magnifications. It seems hot fluid injection process has changed the shape and structure of the rock. This phenomenon maybe due to transformation and dissolution of the carbonate minerals. Also wettability alterations of medium were concluded in the next experiments. Therefore, the rest of the experiments were conducted on stable rocks (no shale). Another phenomenon that could affect the displacement is plugging problem. An experiment was conducted to investigate for blockage by shale swelling. For this purpose, a shale carbonate sample with absolute permeability of 0.5 mD and porosity of 21% was used. After irreducible water saturation was established, hot water injection started at 350 °F. After 25 pore volumes of hot water were injected, about 70% oil recovery was obtained, but the pressure drop had not yet been stabilized. As Figure 6 shows after 20 PV hot water injection, the pressure drop is still fluctuating. This can be attributed to plugging due to shale rock swelling. Therefore, in the next experiments clean core was used to prevent this problem. In the following sections, the results of three sets of tests are presented. (12) Rapoport, L. A.; Leas, W. J. Trans. AIME 1953, 198, 139–148. (13) Bennion, D. B.; Thomas, F. B.; Sheppard, D. A. Paper SPE 23783 presented at the SPE Symposium, 1992.


3.1. Medium Oil Experiments. In this set of experiments, medium oil and limestone rock were used for investigation of hot water flooding into low-permeability carbonate rocks. The experiments were run at 100 °F (reservoir temperature) and 200 °F. After finishing the first experiment at 100 °F, the core was washed and cleaned following the same process as explained in the Experiments section. Next, the absolute permeability to water was measured and no significant difference between the absolute permeability at two temperatures was observed. In the first experiment (100 °F), the test terminated after injecting 7 PV of hot water and was stabilized at 48% oil recovery. The pressure drop across the core decreased from 1200 to 400 psi during this test. However, in the next experiment, hot water injection continued up to 25 pore volume and the recovery increased to 54% and pressure drop decreased from 740 to 240 psi. It was noticed that the oil production rate at the 100 °F test is higher than at the 200 °F test. However, the final recoveries are not similar (Figure 7). The reduction in the oil production rate may be explained in this way that with increasing temperature the limestone rocks become more oil wet and therefore the oil recovery decreases. Also, it is possible that in the drainage process at the lower temperature, the small pores have not been filled with the viscous oil but at higher temperature, the lower viscous oil has more chance to pass small pore throats. Therefore, oil is producing from the large pores more quickly than the small ones in the imbibition process. In Figure 8 the produced oil per pore volume of injected water is compared for early times of the experiments. It seems in the earlier time the O/HW at 100 °F water injection is higher than at the 200 °F experiment but in later times both tests show similar amount of O/HW. The cumulative O/HW ratio at 5 PV hot water injections is 0.067 and 0.023 for 100 °F and 200 °F, respectively. 3.2. Heavy Oil Experiment. This set of experiments was carried out to determine the effect of injection temperature on the heavy oil recovery. The experiments were conducted at three different temperatures, namely 200, 300, and 450 °F. As a result of the preliminary tests, in this set of experiments the system is allowed to reach thermal equilibrium and hot waterflood was started at the desired temperature. After the cores were under vacuum and were saturated with water, the temperature of the oven was set to the experimental temperature. Oil flooding was then conducted until the irreducible water saturation was reached. Eventually, hot waterflooding was performed. At the end of each experiment, the core was washed with toluene and acetone and then the same process was repeated for the next two higher temperatures. Due to relatively medium injectivity of the system (3 mD), it was possible to inject hot water at higher constant rates. Hot water was injected at 12, 15, and 20 cm3/h in the 200, 300 and 450 °F runs, respectively. Figures 9 and 10 show the oil recovery and pressure drop of these experiments, respectively. As is seen, in the first two temperatures, initially pressure drop jumps up and then it decreases until finally it becomes stable. This phenomenon is due to the fluid compressibility.14 For part of the fluid displaced by the constant rate of pump was absorbed by the fluid compressibility effects. Therefore, the actual rate which occurs in the core should be less than the amount that is flowing through the pump. So pressure in the inlet goes up and subsequently pressure drop also goes up. Also as seen from Figure 10, when the temperature increases, pressure drop fluctuation damps out and at 450 °F it ceases. This could be attributed to the increase of oil mobility due to viscosity reduction at higher temperatures. (14) Maini, B. B.; Batycky, J. P. J. Pet. Technol. 1985, 1500–1510.



Figure 5. SEM image of the core MN-X (dolomite).

Figure 6. Oil recovery and pressure drop when plugging happened.

The final oil recovery at the 200 °F test is 47% while when temperature increases to 300 and 450 °F the recovery increases up to 71% and 83%, respectively. Also, the stabilized cumulative oil/hot water ratios are 0.07, 0.13, and 0.12 for 200, 300, and 450 °F, respectively. As is seen, the cumulative O/HW ratios in the heavy oil experiments are much higher than in the medium oil ones. This is a conclusive result that hot water injection in heavy oils is more efficient than in lighter oils which can be explained according to the oil properties such as viscosity. The viscosity reduction in the heavy oils is much higher than light oils; therefore, the heating process will be more efficient than lighter ones. 3.2.1. Effects of Temperature on Sor and Swirr. The effects of temperature on irreducible water saturation and residual oil saturation in sandstone systems have been studied previously. However, there is little published literature regarding work on carbonate systems. Therefore, in this study temperature effects were taken into consideration at some temperatures. Figure 11

Figure 7. Comparison of the oil recovery and pressure drop at 100 and 200 °F in medium oil experiments.

shows that the irreducible water saturation increased linearly with increasing temperature, while the residual oil saturation decreased nonlinearly with increasing temperature. It should be mentioned that due to the high viscosity of the oils used in these experiments, it was a very time-consuming task to reach the real end points. Therefore, the end points were found from extrapolation. Due to this problem, some researchers believe that the residual oil saturation obtained in this way is the practical residual oil saturation and not the real residual oil saturation.14 3.3. Extra Heavy Oil Experiments. In this set of experiments, a constant temperature displacement was conducted at reservoir conditions (150 °F and 2000 psi) followed by incremental temperature process. It was desired to measure the recovery potential of thermal technique at 200, 250, 300, 400,



Figure 8. Comparison of oil/hot water injection ratio in medium oil experiments. Figure 11. Irreducible water saturations and residual oil saturation versus temperature (heavy oil case).

Figure 9. Comparison of the oil recovery at 200, 300, and 450 °F (heavy oil case).

Figure 10. Comparison of the pressure drop at 200, 300, and 450 °F (heavy oil case).

and 500 °F. After the residual oil saturation was reached, the whole system temperatures were increased to the temperature of the next experiment. Therefore, some oil and water were produced due to thermal expansion in each run except the first experiment in which the flow lines, injection/production fluids, and the oven were set to thermal equilibrium. The viscosity of heavy crude was more than 2000 cP in the first temperature run. With this high viscosity and low permeability (0.20 mD), the injection with constant rate was very difficult; therefore, stepwise constant rate injection was applied. The range of water injection rates were 0.19 and 33.42 cm3/h. The overall performance of this set of experiments is shown in Figure 12.

Figure 12. Oil production and pressure drop data versus PV of water injected (extra heavy oil case).

Figure 13. Residual oil saturation versus temperature (extra heavy oil case).

Figure 13 shows the residual oil saturation decreases nonlinearly when temperature increases. The cumulative O/HW ratio for this set of experiments was about 0.03, which is low. This maybe due to the high viscosity of oil and low permeability of the limestone core used in this set of experiments. 4. Numerical Simulation Numerical simulation was used to obtain flow functions and compare the results. There are two practical techniques for relative permeability calculation from dynamic displacement. These techniques can be classified as explicit and implicit


Energy & Fuels, Vol. 22, No. 4, 2008 2359 Table 4. Oil and Water Relative Permeability Exponents and End Point Relative Permeabilities Obtained by History Matching (Medium Oil Case) relative permeability end point of

Figure 14. Constructed grid for simulation of the experiments [r - z (left) and r-angle (right)].

methods. The JBN method15 is the widely used explicit technique. This method applies the Buckley-Leveret theory to obtain flow functions. The second technique is history matching method using numerical simulation of the core flooding tests. Production and pressure drop across the core length are matched by adjusting flow functions. According the experimental conditions, the implicit method was used to obtain the relative permeabilities. To obtain the relative permeability functions, power law equations were used: ° (1 - S)no kro ) kro

(1)

° (S)nw krw ) krw

(2)

S)

Sw - Swi 1 - Sor - Swi

(3)

where S is normalized saturation. A commercial simulator was used for history matching data. A radial grid system of 10 × 8 × 8 blocks (Figure 14) was selected after grid sensitivity studies.16 To consider the viscous fingering in the model, the central grids were refined. Only one grid was used for the injection and one grid used for the production while the fluid was injected from the top of the core into the first radial grid. Subsequently, the fluid filled the first vertical grid quickly and started moving downward. Three-parameter Peng-Robinson equation of state showed the best match between the measured parameters in the laboratory and the thermodynamics model for the fluid properties data. After tuning the PR-3 equation of state using the laboratory fluid properties, dead oil and gas properties were generated at experimental conditions for each test, individually. For the constant temperature experiment, initial experimental data were used in the simulation runs as initial conditions. But, in the incremental temperature experiments, final conditions for (15) Johnson, E. F.; Bossler, D. P.; Naumann, V. O. Trans. AIME 1959, 216, 370–372. (16) Sedaee Sola, B. Investigation of temperature effect on carbonate reservoir water/oil relative permeabilities thermal EOR. Ph.D. Thesis, AmirKabir University, Tehran, Iran, 2006. (17) Esfahania, M. R.; Haghighi, M. J. Pet. Sci. Eng. 2004, 42, 257– 265.

exponents for

temp, °F

oil

water

oil phase

water phase

100 200

1.0 0.2

0.25 0.40

1.8 0.8

2.0 0.4

the previous temperature experiment were used as the initial condition for the next temperature run. After model initialization using the experimental data, the transmissibility of the first and the last vertical direction layers were tuned by matching experimental injectivity and productivity data. The remaining parameters for matching pressure and injection/production data were the relative permeabilities. The curves obtained were compared to check whether the error between the estimated and observed data was minimized. History matching parameters were the injection/production rates, injection/production pressures, and water/oil production data. In the following section, the results of simulation to obtain relative permeability will be presented. 4.1. Medium Oil. Table 4 and Figure 15 show the relative permeability exponents and shapes that were obtained from numerical simulations. It is obvious that the curvature of the relative permeabilities change with temperature which is indicated by the variation of power law equation exponents for oil and water phases. Also, it seems that rock wettability is more oil wet when the temperature increases. This is inferred from the idea that if the crossover saturation where the oil and water relative permeabilities are equal is less than 50%, the rock is pronounced oil wet. Previously, similar results have been reported for light oil and limestone cores.17 4.2. Heavy Oil. Figure 16 shows the experimental and simulation results of oil recovery and pressure drop for one of the heavy oil experiments. As it seems, the matches are reasonable and the same was observed for other temperatures and experiments. Figure 17 shows the relative permeability obtained from the history matching method and Table 5 presents the relative permeability exponents for these sets of experiments. It is obvious that the curvature of the relative permeabilities change with temperature that is indicated by variations in the power law equation exponents for oil and water phases. Figure 18 shows the obtained oil phase relative permeability exponent for heavy oil systems. The water relative permeabilities at 200 and 300 °F were extremely low and this could be due to the high water wettability of the system, according to the rule of thumb which says the

Figure 15. Relative permeabilities obtained from the history matching method (medium oil at 100 and 200 °F).



Figure 19. Water/oil relative permeability ratio obtained from history matching (extra heavy oil case). Figure 16. Oil production and pressure drop data matches (heavy oil at 200 °F).

Figure 20. Oil phase relative permeability exponents values obtained from history matching versus temperature (extra heavy oil case).

Figure 17. Relative permeabilities obtained from the history matching at 200, 300, and 450 °F in log scale. Table 5. Oil and Water Exponents and End Point Relative Permeabilities Values Obtained from History Matching relative permeability end point of

exponents for

temp, °F

oil

water

oil phase

water phase

200 300 450

0.95 0.95 0.95

0.08 0.06 0.06

2.5 2.2 2.1

1.2 2.1 2.0

end-point values of relative permeability and the crossover saturation are conventional indicators of wettability. The results suggest that the system becomes more water-wet and the oil relative permeability increases while the water relative permeability decreases with increasing temperature.

The results obtained in this study were compared with the previous findings17 which show that there is a discrepancy in the effect of temperature on Sor and Swirr. Based on the above observations, the following heavy oil/ water relative permeability functions are suggested for dolomite rocks and heavy oil systems: ° (1 - S)no, kro ) kro ° (S)nw, krw ) krw

2.0 < no < 2.5,

1.0 < no < 2.5,

0 kro ) 0.95

0 0.06 < krw < 0.08

(4) (5)

4.3. Extra Heavy Oil. The best match was obtained for the pressure drop and production data by trial and error for extra heavy oil too. Figure 19 shows the oil/water relative permeability ratio obtained using history matching method. Due to variation of the end point saturation in each temperature, the best way to judge the effect of temperature on flow function is to use the normalized saturation relative permeability curves which is shown in this figure. The normalized saturation is defined by eq 3. This indicates that the oil relative permeability changes with increasing temperature. The effect of temperature on the water relative permeability curves is not remarkable in this kind of system. Another important point is the low values of the end point water relative permeabilities (which is less than 0.10, as shown in Figure 20). The strongly water-wet characteristics could be one of the reasons for these very small water relative permeability values. 5. Conclusions

Figure 18. Oil phase relative permeability exponent values obtained from history matching versus temperature (heavy oil case).

Based on the results of this investigation, the following conclusions could be drawn for the hot water injection into



low-permeability carbonate medium to extra heavy oils systems and the related flow functions under reservoir conditions:

Acknowledgment. All laboratory personnel’s assistance in the Department of Chemical/Petroleum Engineering of AmirKabir University, Tehran, Iran, is greatly acknowledged.

1. The shale content may plug the core during core flooding and change the structure of the medium due to hot water injection.

Nomenclature

2. Thermal expansion of the oil plays an important role in the incremental oil recovery mechanism. In fact, more than 10% of extra heavy oil can be recovered by applying heat into carbonate reservoirs. 3. Hot water injection increases the heavy oil recovery in low-permeability reservoirs as well as conventional oil reservoirs. However, the amount of its potential differs for various oils. In the heavy oil system, the O/HW ratio is higher than in the medium and the extra heavy oil; however, the values are less than the reported value for conventional heavy oil reservoirs. 4. The shapes of oil and water relative permeabilities in these kinds of rocks and oil significantly change with increasing temperature. In dolomite the rock becomes more water-wet with increasing temperature. 5. The residual oil saturation nonlinearly decreases with increasing temperature.

Symbols k ) permeability, mD kr ) relative permeability, fraction L ) length, cm S ) saturation, fraction u ) velocity, cm/s EOR ) enhanced oil recovery PV ) pore volume SEM ) scanning electron microscopic Greek Letters µ ) viscosity, cP Subscripts and Superscripts o ) oil or ) residual oil irr ) irreducible saturation w ) water wi ) initial water n ) exponent EF800009R

Experimental Study of Hot Water Injection into Low-Permeability

Recommend Documents