- Email: [email protected]
Experiments on CO2 foam seepage characteristics in porous media
PETROLEUM EXPLORATION AND DEVELOPMENT Volume 43, Issue 3, June 2016 Online English edition of the Chinese language journal Cite this article as: PETROL. EXPLOR. DEVELOP., 2016, 43(3): 499–505.
RESEARCH PAPER
Experiments on CO2 foam seepage characteristics in porous media DU Dongxing1,*, WANG Dexi1, JIA Ninghong2, LYU Weifeng2, QIN Jishun2, WANG Chengcheng1, SUN Shengbin1, LI Yingge3 1. College of Electromechanical Engineering, Qingdao University of Science and Technology, Qingdao 266061, China; 2. PetroChina Research Institute of Petroleum Exploration & Development, Beijing 100083, China; 3. College of Automation and Electronic Engineering, Qingdao University of Science and Technology, Qingdao 266061, China
Abstract: Simulative experiments were carried out for CO2 foam flooding process in homogeneous porous media prepared with the sand packing method, and the CO2 foam seepage characteristics in porous media were studied with CT technology. CO2 foam flooding experiments were carried out under different packing sand sizes, different surfactant concentrations and different gas-liquid ratios. CT technology was employed to visualize the displacement process and to obtain the water saturation data along the sample, and the pressure distribution in the sample during the foam seepage process was measured at the same time. Experimental results show that, CO2 foam flooding has higher pressure drop and lower water saturation entrance effect in the porous media with lower average grain sizes; when surfactant concentrations are higher than CMC (Critical Micelle Concentration), the generated foam is stable, without showing obvious difference for the foam displacement efficiency in the sample, and water saturation entrance effect gradually decreases with increase of surfactant concentrations; improving gas-liquid ratio can lead to higher foam seepage pressure drop, but has little effect on residue water saturation after foam displacement. Key words: CO2 flooding; CO2 foam; flooding experiment; seepage characteristics; CT technology; EOR
Introduction In recent years, extensive research works and application studies concerning CO2 Enhanced Oil Recovery (CO2 EOR) technology have been carried out. Producing oil by injecting CO2 can not only enhance oil recovery, but also reduce greenhouse gas emission, benefiting both economy and environment[12]. However, enhanced oil recovery with CO2 has some problems, one major problem is that CO2, low in viscosity, could make early breakthrough in the high permeability channels, resulting in failure of a project[34]. Foam technology can significantly improve the apparent viscosity of the gas phase, therefore can be used to solve gas fingering and channeling, to enhance oil recovery[59]. In order to apply foam technology successfully to industrial practice of CO2 EOR, the characteristics and mechanism of CO2 foam flow in porous media have to be investigated. Up to now, quite a few works have been done concerning CO2 foam flow characteristics in porous media[1018]. Due to the unique advantage of obtaining liquid phase saturation in a multiphase system, CT scan technology has been widely used
in studying CO2 foam flow characteristics in porous media[1218]. Although there have been fairly a lot literature reports on various disciplines involved in the foam system, the CO2 foam flow hasn’t been systematically studied in terms of flow characteristics and flow mechanism. For example, in most foam flow characteristic tests, the pressure measurement points are set at inlet and outlet of the sand core only[11,1418], few research papers reported the pressure distribution along the sample in the transient foam displacement process, which is essential for understanding foam flow mechanism in porous media. In this paper, an organic glass pressure vessel was designed and manufactured, a homogeneous porous medium sample was made by packing sand method, and the experimental platform was designed and constructed for studying foam seepage characteristics in porous media. CT technology was employed to visually monitor the CO2 foam displacement process, and test the pressure variation history along the sample to investigate the CO2 foam flow characteristics in porous media.
Received date: 31 Jul. 2015; Revised date: 28 Mar. 2016. * Corresponding author. E-mail: [email protected] Foundation item: Supported by the National Natural Science Foundation of China (51476081); PetroChina Innovation Foundation (2014D50060210). Copyright © 2016, Research Institute of Petroleum Exploration and Development, PetroChina. Published by Elsevier BV. All rights reserved.
DU Dongxing et al. / Petroleum Exploration and Development, 2016, 43(3): 499–505
1.
Introduction of experiments
Sodium dodecyl sulfate (SDS, with molecular formula of C12H25SO4Na and molecular weight of 288.38) was used as surfactant, and CO2 gas was used as the internal phase of the foam system. The porous medium sample was prepared by packing sand in an organic glass pressure vessel (30 mm in internal diameter, 80 mm in external diameter and 200 mm in length). Fig. 1 is the schematic diagram of the experimental system. The organic glass pressure vessel was placed on the test couch of the CT device. The gas and the surfactant solution regulated through gas mass flow controller and piston pump respectively, were injected into the bubble generator at a constant flow rate ratio to generate stable foam fluid, and the generated foam fluid was injected into the water saturated porous medium. CT measurement was carried out at a regular time interval and dozens of cross section CT images were taken along the flow direction to study the dynamic distribution of water phase saturation in the foam displacement process. At the same time, the pressure distribution along the sample was obtained by multiple pressure sensors uniformly installed along the sample holder, and the pressure signals were collected in real time with data acquisition system. Through comprehensive analysis of phase saturation and pressure distribution data, the CO2 foam flow characteristics under different packing sand sizes, different surfactant concentrations and different gas-liquid ratios have been found out.
2. 2.1.
Experimental results and discussion Analysis of porous media porosity
Various porous medium models were prepared by packing quartz sand of different sizes, namely, 0.1800.250 mm (6080 mesh), 0.1500.180 mm (80100 mesh) and 0.1060.150 mm (100150 mesh) respectively. With the average particle size of 0.21 mm, 0.17 mm and 0.13 mm, the porous medium models had a measured porosity of 43.29%, 43.62% and 43.18% respectively. It can be seen the porous
Fig. 1. Schematic diagram of experimental system. 1—Pressure relief valve; 2—Pipe valve; 3—Gas mass flow controller; 4—Oneway valve; 5—Foam generator; 6—Precise pressure gauge; 7—Pressure transmitter; 8—Data acquisition device; 9—CO2 gas bottle; 10— Piston pump; 11—Organic glass pressure vessel (with packing sand medium inside); 12—Back pressure system; 13—Vacuum pump.
medium models packed with different sizes of sand have little difference in porosity, indicating nearly the same pore volume of the packing sand samples in the same vessel. 2.2. Effect of packing sand size on foam displacement process Under the fixed conditions of surfactant concentration of 10.4 mmol/L, gas liquid ratio of 13.5:1 and atmospheric back pressure, the pressure and water saturation during CO2 foam flow process along the packing samples made of sand with average particle size of 0.21 mm, 0.17 mm and 0.13 mm were tested, as shown in Fig. 2 to Fig. 4 respectively. The packing-sand models were 200 mm long, with the entrance at 0 mm and the exit at 200 mm. It can be seen from Figs. 2a, 3a and 4a that: (1) the pressure at various points gradually increased with experiments going on, and became steady after a period of time, forming a stable pressure drop between inlet and outlet of the samples. As indicated in Figs. 2a, 3a and 4a, pressure became stable at 10, 12 and 16 pore volume (PV), when the pressure difference between the inlet and the outlet were around 0.30 MPa, 0.53 MPa and 1.12 MPa respectively. It can be concluded that the smaller the particle size of sand used to pack the porous medium, the higher the pressure difference will be needed to push the foam from inlet to the outlet, and the higher the PV will be needed to reach stable pressure distribution. (2) For the porous medium made of larger particle sand (such as 0.21 mm and 0.17 mm in average diameter), pressure gradient in the entrance part was lower, became steeper clearly near the exit of the sample; for the porous medium made of smaller size sand (0.13 mm in average diameter), however, the pressure gradient was nearly the same in the whole length range of the sample. The experimental observations can be well explained based on the apparent viscosity model of CO2 foam. In the rheology model based on two phase assumption, flow resistance for one bubble and the bubble number are the most important parameters. According to the bubble population balance model based on statistical theory[1920], the generation and development of bubble numbers in homogeneous porous medium can be worked out by Eq. (1). At the entrance of the porous medium where the foam is in the stage of generation and development, the foam destruction rate coefficient Kd can be ignored and the bubble number is only related closely to the foam generation rate coefficient, Kg.. In the porous medium with larger sand particles, the CO2 foam bubble grow slower in number and can only reach full development stage after certain distance, therefore the bubble number near the entrance of the porous medium is smaller. In addition, the pressure drop for a single flowing bubble is related to the surface tension between the gas and the liquid phase[2123], the lower the surface tension, the lower the pressure drop will be. Due to the water-solubility of CO2, the surface tension between the gas phase CO2 and the liquid phase of surfactant solution de-
500
DU Dongxing et al. / Petroleum Exploration and Development, 2016, 43(3): 499–505
Fig. 2.
Foam fluid seepage characteristics in the porous medium with average packing sand size of 0.21 mm.
Fig. 3.
Foam fluid seepage characteristics in the porous medium with average packing sand size of 0.17 mm.
Fig. 4.
Foam fluid seepage characteristics in the porous medium with average packing sand size of 0.13 mm.
creases with the increase of pressure[2425], therefore the foam apparent viscosity is lower under higher inlet pressure, which leads to lower pressure drop and obvious entrance effect.
Whereas near the exit of the porous medium, the apparent viscosity of the foam fluid no longer changes after bubble number reaching balance and the foam flow process still
501
DU Dongxing et al. / Petroleum Exploration and Development, 2016, 43(3): 499–505
obeys Darcy’s law. In the porous medium with smaller particle size, on the other hand, foam generation rate is higher and the maximum bubble number is also larger due to the increased pore throat structures per unit volume. At the entrance side of the porous medium, the foam bubble increases rapidly and reaches maximum value in a shorter distance. No obvious entrance effect is observed because the effect of water solubility on CO2 foam seepage resistance is partially compensated by the larger bubble numbers. nSf (1) nuf Sf K g n n K d n t where Sf—foam saturation, f; t—time, s; Kd—foam destruction rate coefficient, f; Kg—foam generation rate coefficient, f; n—bubble number, bubbles/m3; n∞—maximum bubble number in porous medium after foam seepage flow process reaching stable, bubbles/m3; uf—velocity vector of gas phase, m/s; —porosity of porous medium, f. It can be seen from Figs. 2b, 3b and 4b that the porous media were pre-saturated with water at the beginning of the experiments. Water saturation started to decrease firstly near the entrance and gradually decreases from inlet to the outlet of the sample as foam displacement process went on, and finally maintained at around 10% after foam flooding, indicating a stable foam propagation process with satisfactory sweeping efficiency. In the porous media with larger particle sizes, obvious entrance effect indicates the foam is still in the development stage and the bubble number does not reach the steady state at the entrance of the sample. Whereas in the porous medium with smaller particle size, no obvious entrance effect has been observed due to the higher foam generation and development rate. 2.3. Effect of surfactant concentration on foam displacement process Under the fixed conditions of average packing sand size of 0.17 mm, gas liquid ratio of 13.5:1 and atmospheric back-
Fig. 5.
pressure, the pressure and water saturation variations during the foam displacement process were obtained at different surfactant concentrations of 5.2 mmol/L, 10.4 mmol/L and 17.4 mmol/L (Figs. 5-7) respectively. Comparison of Figs. 5a, 6a and 7a shows the foam generated at surfactant concentration of 5.2 mmol/L, (which is lower than the Critical Micelle Concentration (CMC) of 8 mmol/L), was unstable, so the pressure rise at each point in the medium fluctuated, and larger injection PV number was needed to reach steady pressure at the inlet; while in the cases of surfactant concentration of 10.4 mmol/L and 17.4 mmol/L, the generated foam was stable because the employed surfactant concentrations were higher than CMC. In both cases, pressure increased gradually point by point in the foam displacement process, the time at all points needed to reach stable pressure was almost the same, and the pressure drop finally maintained at around 0.53-0.55 MPa when the flow process reached stable. Careful observation of Figs. 6a and 7a shows in the case of 10.4 mmol/L surfactant concentration, there was a lower pressure gradient at the entrance side, while in the case of 17.4 mmol/L surfactant concentration, the pressure gradient at the inlet was almost the same as other points, because the higher surfactant concentration can meet the requirement of generating sufficient bubbles to satisfy sand adsorption and form foam fluid in-time. It can be seen from Figs. 5b, 6b and 7b also that in the case of 5.2 mmol/L, the frontal interface in the foam displacement process inclined more and fluctuated, and the average water saturation was higher at the end of the displacement; in the case of surfactant concentration of 10.4 mmol/L, obvious entrance effect still could be observed but the water saturation was lower after the foam flooding process; while in the case of surfactant concentration of 17.4 mmol/L, the propagation process behaved in a piston-like way, with negligible entrance effect and the lowest average water saturation at the end of foam displacement.
Foam fluid seepage characteristics at surfactant concentration of 5.2 mmol/L.
502
DU Dongxing et al. / Petroleum Exploration and Development, 2016, 43(3): 499–505
Fig. 6.
Foam fluid seepage characteristics at surfactant concentration of 10.4 mmol/L.
Fig. 7.
Foam fluid seepage characteristics at surfactant concentration of 17.4 mmol/L.
2.4. Effect of gas-liquid ratio on foam displacement process Under the conditions of average packing sand size of 0.17 mm, surfactant concentration of 10.4 mmol/L and atmospheric back pressure, the foam fluid flow characteristics in porous medium were measured at different gas-liquid ratios of 7.0:1, 13.5:1, 20.0:1 and 30.0:1 respectively (Fig. 8). During the experiment, after the flow process of a foam fluid with a certain gas-liquid ratio reached stable, that is the pressure drop didn’t change anymore, the gas-liquid ratio was changed deliberately for another set of experiment. It can be seen from Fig. 8a and 8b that the pressure at Point 1, 2, 3, 4 and 5 (the distance to the sample inlet of them are 0, 50 mm, 100 mm, 150 mm, 200 mm respectively) increased in order with experiment going on, and the pressure drop between Point 1 and Point 5 finally stablized at around 0.55 MPa. With the increase of gas-liquid ratio, the measured
pressure at each point became higher accordingly and the pressure difference between inlet and outlet of the sample also increased after reaching steady state. This is because the foam apparent viscosity is higher at higher gas-liquid ratio, which leads to increased pressure drop in the foam seepage process. Fig. 8c shows the corresponding water saturation curves under different gas-liquid ratios in the foam flooding process. Among all the curves, the eleven curves with PV numbers ranging from 0.13 to 6.56 are the water saturation variation curves for the foam fluid generated at gas-liquid ratio of 7.0:1; while the four curves at PV numbers of 13.80, 30.20, 46.00 and 65.30 show the water saturation distribution of flooding process at the gas-liquid ratio of 7: 1, 13.5: 1, 20: 1, 30: 1 respectively. It is also found from Fig. 8c that although the pressure drop along the sample increased significantly with the increase of gas-liquid ratio, the residue water saturation after foam flooding was nearly the same.
503
DU Dongxing et al. / Petroleum Exploration and Development, 2016, 43(3): 499–505
Fig. 8.
3.
Seepage characteristics of CO2 foam fluid with different gas-liquid ratios in porous medium. [3]
Conclusions
HOLM L W. Evolution of the carbon-dioxide flooding processes. Journal of Petroleum Technology, 1987, 39(11): 1337–1342.
A homogeneous porous medium model was prepared with sand packing method, and the experimental platform was constructed to study foam seepage characteristics. CT technology was employed to monitor the displacement process and obtain water saturation profiles during the process. Together with the pressure data measured with uniformly distributed pressure transducers along the samples, the CO2 foam flow characteristics have been investigated in this study. The experimental results show: (1) In the porous medium with larger average particle size (0.170.21 mm), the pressure gradient declined gently near the entrance, but became much steeper near the outlet. While in the porous medium with smaller average particle size (0.13 mm), the pressure gradient of the foam fluid near the entrance had no obvious difference with other locations of the porous medium. (2) When the surfactant concentration was lower than the CMC, the generated foam had lower stability and therefore lower displacement efficiency; when the surfactant concentration was slightly higher than the CMC, obvious entrance effect was observed although the water saturation after displacement was lower; at further higher surfactant concentration, the entrance effect diminished and the water saturation was the lowest at the end of the foam flooding. (3) With the increase of gas-liquid ratio, the pressure drop during the displacement process became larger, but the residue water saturation after foam flooding process, did not show obvious difference at different gas-liquid ratios.
[4]
SMITH D H. Promise and problems of miscible-flood enhanced oil recovery: The need for surfactant-based sweep and mobility control. In: ACS Symposium Series 373. Washington: American Chemical Society, 1988: 2–37.
[5]
WANG Gaofeng, ZHENG Xiongjie, ZHANG Yu, et al. A new screening method of low permeability reservoirs suitable for CO2 flooding. Petroleum Exploration and Development, 2015, 42(3): 358–363.
[6]
LAKE L W. Enhanced oil recovery. New Jersey: Prentice Hall, 1989.
[7]
HOLM L W, JOSENDAL V A. Mechanism of oil displacement by carbon dioxide. Journal of Petroleum Technology, 1974, 26(12): 1427–1438.
[8]
ALI J, BURLEY R W, NUTT C W. Foam enhanced oil recovery from sand packs. Chemical Engineering Research and Design, 1985, 63: 101–111.
[9]
PATZEK T W. Field application of foam for mobility improvement and profile control. SPE Reservoir Engineering, 1996, 11(2): 79–86.
[10] CHOU S I. Conditions for generating foam in porous media. SPE 22628, 1991. [11] LI Chun, YI Xiangyi, LU Yuan. Experimental study on CO2 foam profile modification. Drilling & Production Technology, 2008, 31(1): 107–108, 142. [12] WELLINGTON S L, VINEGAR H J. Surfactant-induced mobility control for carbon dioxide studied with computerized tomography. In: ACS Symposium Series 373. Washington: American Chemical Society, 1988. [13] LEE H O, HELLER J P. Carbon dioxide foam mobility meas-
References
urement at high pressure. In: ACS Symposium Series 373. Washington: American Chemical Society, 1988.
[1]
QIN Jishun, HAN Haishui, LIU Xiaolei. Application and
[14] DU D X, ZITHA P L J, UIJTTENHOUT M G H. Carbon di-
enlightenment of carbon dioxide flooding in the United States
oxide foam rheology in porous media: A CT scan study. SPE J,
of America. Petroleum Exploration and Development, 2015, 42(2): 209–216. [2]
2007, 12(2): 245–252. [15] DU D X, NADERI B A, FARAJZADEH R, et al. Effect of
SHEN Pingping, JIANG Huaiyou, CHEN Yongwu, et al.
water solubility on carbon dioxide foam flow in porous media:
Study on CO2 injection technique in EOR. Special Oil & Gas
An X-ray computed tomography study. Industrial & Engi-
Reservoir, 2007, 14(3): 1–4, 11.
neering Chemistry Research, 2008, 47(16): 6298–6306.
504
DU Dongxing et al. / Petroleum Exploration and Development, 2016, 43(3): 499–505
[16] FARAJZADEH R, ANDRIANOV A, BRUINING J, et al. Comparative study of CO2 and N2 foams in porous media at low and high pressure-temperatures. Industrial & Engineering
461–474. [21] BRETHERTON F P. The motion of long bubbles in tubes. J Fluid Mech, 1961, 10(2): 166–188. [22] HIRASAKI G J, LAWSON J B. Mechanisms of foam flow in
Chemistry Research, 2009, 48(9): 4542–4552. [17] FARAJZADEH R, ANDRIANOV A, ZITHA P L J. Investigation of immiscible and miscible foam for enhancing oil recovery. Industrial & Engineering Chemistry Research, 2010,
porous media: Apparent viscosity in smooth capillaries. SPE J, 1985, 25(2): 176–190. [23] DU D X, ZHANG J, SUN S B, et al. Mechanistic study on foam lamellae flow characteristics in tubes. Material Research
49(4): 1910–1919. [18] ANDRIANOV A, FARAJZADEH R, MAHMOODI N M, et al. Immiscible foam for enhancing oil recovery: Bulk and po-
Innovations, 2015, 19(S5): 526–529. [24] AGGELOPOULOS C A, ROBIN M, VIZIKA O. Interfacial
rous media experiments. Industrial & Engineering Chemistry
tension between CO2 and brine (NaCl+CaCl2) at elevated
Research, 2012, 51(1): 2214–2226.
pressures and temperatures: The additive effect of different
[19] ZITHA P L J, DU D X. A new stochastic bubble population model for foam flow in porous media. Transport in Porous Media, 2010, 83(3): 603–621.
salts. Advances in Water Resources, 2011, 34(4): 505–511. [25] BACHU S, BRANT B D. Interfacial tension between CO2, freshwater, and brine in the range of pressure from (2 to 27)
[20] DU D X, ZITHA P L J, VERMOLEN F J. Numerical analysis
MPa, temperature from (20 to 125) C, and water salinity from
of foam motion in porous media using a new stochastic bubble
(0 to 334 000) mg·L−1. J Chem Eng Data, 2009, 54(3):
population model. Transport in Porous Media, 2011, 86(2):
765–775.
505
RESEARCH PAPER
Experiments on CO2 foam seepage characteristics in porous media DU Dongxing1,*, WANG Dexi1, JIA Ninghong2, LYU Weifeng2, QIN Jishun2, WANG Chengcheng1, SUN Shengbin1, LI Yingge3 1. College of Electromechanical Engineering, Qingdao University of Science and Technology, Qingdao 266061, China; 2. PetroChina Research Institute of Petroleum Exploration & Development, Beijing 100083, China; 3. College of Automation and Electronic Engineering, Qingdao University of Science and Technology, Qingdao 266061, China
Abstract: Simulative experiments were carried out for CO2 foam flooding process in homogeneous porous media prepared with the sand packing method, and the CO2 foam seepage characteristics in porous media were studied with CT technology. CO2 foam flooding experiments were carried out under different packing sand sizes, different surfactant concentrations and different gas-liquid ratios. CT technology was employed to visualize the displacement process and to obtain the water saturation data along the sample, and the pressure distribution in the sample during the foam seepage process was measured at the same time. Experimental results show that, CO2 foam flooding has higher pressure drop and lower water saturation entrance effect in the porous media with lower average grain sizes; when surfactant concentrations are higher than CMC (Critical Micelle Concentration), the generated foam is stable, without showing obvious difference for the foam displacement efficiency in the sample, and water saturation entrance effect gradually decreases with increase of surfactant concentrations; improving gas-liquid ratio can lead to higher foam seepage pressure drop, but has little effect on residue water saturation after foam displacement. Key words: CO2 flooding; CO2 foam; flooding experiment; seepage characteristics; CT technology; EOR
Introduction In recent years, extensive research works and application studies concerning CO2 Enhanced Oil Recovery (CO2 EOR) technology have been carried out. Producing oil by injecting CO2 can not only enhance oil recovery, but also reduce greenhouse gas emission, benefiting both economy and environment[12]. However, enhanced oil recovery with CO2 has some problems, one major problem is that CO2, low in viscosity, could make early breakthrough in the high permeability channels, resulting in failure of a project[34]. Foam technology can significantly improve the apparent viscosity of the gas phase, therefore can be used to solve gas fingering and channeling, to enhance oil recovery[59]. In order to apply foam technology successfully to industrial practice of CO2 EOR, the characteristics and mechanism of CO2 foam flow in porous media have to be investigated. Up to now, quite a few works have been done concerning CO2 foam flow characteristics in porous media[1018]. Due to the unique advantage of obtaining liquid phase saturation in a multiphase system, CT scan technology has been widely used
in studying CO2 foam flow characteristics in porous media[1218]. Although there have been fairly a lot literature reports on various disciplines involved in the foam system, the CO2 foam flow hasn’t been systematically studied in terms of flow characteristics and flow mechanism. For example, in most foam flow characteristic tests, the pressure measurement points are set at inlet and outlet of the sand core only[11,1418], few research papers reported the pressure distribution along the sample in the transient foam displacement process, which is essential for understanding foam flow mechanism in porous media. In this paper, an organic glass pressure vessel was designed and manufactured, a homogeneous porous medium sample was made by packing sand method, and the experimental platform was designed and constructed for studying foam seepage characteristics in porous media. CT technology was employed to visually monitor the CO2 foam displacement process, and test the pressure variation history along the sample to investigate the CO2 foam flow characteristics in porous media.
Received date: 31 Jul. 2015; Revised date: 28 Mar. 2016. * Corresponding author. E-mail: [email protected] Foundation item: Supported by the National Natural Science Foundation of China (51476081); PetroChina Innovation Foundation (2014D50060210). Copyright © 2016, Research Institute of Petroleum Exploration and Development, PetroChina. Published by Elsevier BV. All rights reserved.
DU Dongxing et al. / Petroleum Exploration and Development, 2016, 43(3): 499–505
1.
Introduction of experiments
Sodium dodecyl sulfate (SDS, with molecular formula of C12H25SO4Na and molecular weight of 288.38) was used as surfactant, and CO2 gas was used as the internal phase of the foam system. The porous medium sample was prepared by packing sand in an organic glass pressure vessel (30 mm in internal diameter, 80 mm in external diameter and 200 mm in length). Fig. 1 is the schematic diagram of the experimental system. The organic glass pressure vessel was placed on the test couch of the CT device. The gas and the surfactant solution regulated through gas mass flow controller and piston pump respectively, were injected into the bubble generator at a constant flow rate ratio to generate stable foam fluid, and the generated foam fluid was injected into the water saturated porous medium. CT measurement was carried out at a regular time interval and dozens of cross section CT images were taken along the flow direction to study the dynamic distribution of water phase saturation in the foam displacement process. At the same time, the pressure distribution along the sample was obtained by multiple pressure sensors uniformly installed along the sample holder, and the pressure signals were collected in real time with data acquisition system. Through comprehensive analysis of phase saturation and pressure distribution data, the CO2 foam flow characteristics under different packing sand sizes, different surfactant concentrations and different gas-liquid ratios have been found out.
2. 2.1.
Experimental results and discussion Analysis of porous media porosity
Various porous medium models were prepared by packing quartz sand of different sizes, namely, 0.1800.250 mm (6080 mesh), 0.1500.180 mm (80100 mesh) and 0.1060.150 mm (100150 mesh) respectively. With the average particle size of 0.21 mm, 0.17 mm and 0.13 mm, the porous medium models had a measured porosity of 43.29%, 43.62% and 43.18% respectively. It can be seen the porous
Fig. 1. Schematic diagram of experimental system. 1—Pressure relief valve; 2—Pipe valve; 3—Gas mass flow controller; 4—Oneway valve; 5—Foam generator; 6—Precise pressure gauge; 7—Pressure transmitter; 8—Data acquisition device; 9—CO2 gas bottle; 10— Piston pump; 11—Organic glass pressure vessel (with packing sand medium inside); 12—Back pressure system; 13—Vacuum pump.
medium models packed with different sizes of sand have little difference in porosity, indicating nearly the same pore volume of the packing sand samples in the same vessel. 2.2. Effect of packing sand size on foam displacement process Under the fixed conditions of surfactant concentration of 10.4 mmol/L, gas liquid ratio of 13.5:1 and atmospheric back pressure, the pressure and water saturation during CO2 foam flow process along the packing samples made of sand with average particle size of 0.21 mm, 0.17 mm and 0.13 mm were tested, as shown in Fig. 2 to Fig. 4 respectively. The packing-sand models were 200 mm long, with the entrance at 0 mm and the exit at 200 mm. It can be seen from Figs. 2a, 3a and 4a that: (1) the pressure at various points gradually increased with experiments going on, and became steady after a period of time, forming a stable pressure drop between inlet and outlet of the samples. As indicated in Figs. 2a, 3a and 4a, pressure became stable at 10, 12 and 16 pore volume (PV), when the pressure difference between the inlet and the outlet were around 0.30 MPa, 0.53 MPa and 1.12 MPa respectively. It can be concluded that the smaller the particle size of sand used to pack the porous medium, the higher the pressure difference will be needed to push the foam from inlet to the outlet, and the higher the PV will be needed to reach stable pressure distribution. (2) For the porous medium made of larger particle sand (such as 0.21 mm and 0.17 mm in average diameter), pressure gradient in the entrance part was lower, became steeper clearly near the exit of the sample; for the porous medium made of smaller size sand (0.13 mm in average diameter), however, the pressure gradient was nearly the same in the whole length range of the sample. The experimental observations can be well explained based on the apparent viscosity model of CO2 foam. In the rheology model based on two phase assumption, flow resistance for one bubble and the bubble number are the most important parameters. According to the bubble population balance model based on statistical theory[1920], the generation and development of bubble numbers in homogeneous porous medium can be worked out by Eq. (1). At the entrance of the porous medium where the foam is in the stage of generation and development, the foam destruction rate coefficient Kd can be ignored and the bubble number is only related closely to the foam generation rate coefficient, Kg.. In the porous medium with larger sand particles, the CO2 foam bubble grow slower in number and can only reach full development stage after certain distance, therefore the bubble number near the entrance of the porous medium is smaller. In addition, the pressure drop for a single flowing bubble is related to the surface tension between the gas and the liquid phase[2123], the lower the surface tension, the lower the pressure drop will be. Due to the water-solubility of CO2, the surface tension between the gas phase CO2 and the liquid phase of surfactant solution de-
500
DU Dongxing et al. / Petroleum Exploration and Development, 2016, 43(3): 499–505
Fig. 2.
Foam fluid seepage characteristics in the porous medium with average packing sand size of 0.21 mm.
Fig. 3.
Foam fluid seepage characteristics in the porous medium with average packing sand size of 0.17 mm.
Fig. 4.
Foam fluid seepage characteristics in the porous medium with average packing sand size of 0.13 mm.
creases with the increase of pressure[2425], therefore the foam apparent viscosity is lower under higher inlet pressure, which leads to lower pressure drop and obvious entrance effect.
Whereas near the exit of the porous medium, the apparent viscosity of the foam fluid no longer changes after bubble number reaching balance and the foam flow process still
501
DU Dongxing et al. / Petroleum Exploration and Development, 2016, 43(3): 499–505
obeys Darcy’s law. In the porous medium with smaller particle size, on the other hand, foam generation rate is higher and the maximum bubble number is also larger due to the increased pore throat structures per unit volume. At the entrance side of the porous medium, the foam bubble increases rapidly and reaches maximum value in a shorter distance. No obvious entrance effect is observed because the effect of water solubility on CO2 foam seepage resistance is partially compensated by the larger bubble numbers. nSf (1) nuf Sf K g n n K d n t where Sf—foam saturation, f; t—time, s; Kd—foam destruction rate coefficient, f; Kg—foam generation rate coefficient, f; n—bubble number, bubbles/m3; n∞—maximum bubble number in porous medium after foam seepage flow process reaching stable, bubbles/m3; uf—velocity vector of gas phase, m/s; —porosity of porous medium, f. It can be seen from Figs. 2b, 3b and 4b that the porous media were pre-saturated with water at the beginning of the experiments. Water saturation started to decrease firstly near the entrance and gradually decreases from inlet to the outlet of the sample as foam displacement process went on, and finally maintained at around 10% after foam flooding, indicating a stable foam propagation process with satisfactory sweeping efficiency. In the porous media with larger particle sizes, obvious entrance effect indicates the foam is still in the development stage and the bubble number does not reach the steady state at the entrance of the sample. Whereas in the porous medium with smaller particle size, no obvious entrance effect has been observed due to the higher foam generation and development rate. 2.3. Effect of surfactant concentration on foam displacement process Under the fixed conditions of average packing sand size of 0.17 mm, gas liquid ratio of 13.5:1 and atmospheric back-
Fig. 5.
pressure, the pressure and water saturation variations during the foam displacement process were obtained at different surfactant concentrations of 5.2 mmol/L, 10.4 mmol/L and 17.4 mmol/L (Figs. 5-7) respectively. Comparison of Figs. 5a, 6a and 7a shows the foam generated at surfactant concentration of 5.2 mmol/L, (which is lower than the Critical Micelle Concentration (CMC) of 8 mmol/L), was unstable, so the pressure rise at each point in the medium fluctuated, and larger injection PV number was needed to reach steady pressure at the inlet; while in the cases of surfactant concentration of 10.4 mmol/L and 17.4 mmol/L, the generated foam was stable because the employed surfactant concentrations were higher than CMC. In both cases, pressure increased gradually point by point in the foam displacement process, the time at all points needed to reach stable pressure was almost the same, and the pressure drop finally maintained at around 0.53-0.55 MPa when the flow process reached stable. Careful observation of Figs. 6a and 7a shows in the case of 10.4 mmol/L surfactant concentration, there was a lower pressure gradient at the entrance side, while in the case of 17.4 mmol/L surfactant concentration, the pressure gradient at the inlet was almost the same as other points, because the higher surfactant concentration can meet the requirement of generating sufficient bubbles to satisfy sand adsorption and form foam fluid in-time. It can be seen from Figs. 5b, 6b and 7b also that in the case of 5.2 mmol/L, the frontal interface in the foam displacement process inclined more and fluctuated, and the average water saturation was higher at the end of the displacement; in the case of surfactant concentration of 10.4 mmol/L, obvious entrance effect still could be observed but the water saturation was lower after the foam flooding process; while in the case of surfactant concentration of 17.4 mmol/L, the propagation process behaved in a piston-like way, with negligible entrance effect and the lowest average water saturation at the end of foam displacement.
Foam fluid seepage characteristics at surfactant concentration of 5.2 mmol/L.
502
DU Dongxing et al. / Petroleum Exploration and Development, 2016, 43(3): 499–505
Fig. 6.
Foam fluid seepage characteristics at surfactant concentration of 10.4 mmol/L.
Fig. 7.
Foam fluid seepage characteristics at surfactant concentration of 17.4 mmol/L.
2.4. Effect of gas-liquid ratio on foam displacement process Under the conditions of average packing sand size of 0.17 mm, surfactant concentration of 10.4 mmol/L and atmospheric back pressure, the foam fluid flow characteristics in porous medium were measured at different gas-liquid ratios of 7.0:1, 13.5:1, 20.0:1 and 30.0:1 respectively (Fig. 8). During the experiment, after the flow process of a foam fluid with a certain gas-liquid ratio reached stable, that is the pressure drop didn’t change anymore, the gas-liquid ratio was changed deliberately for another set of experiment. It can be seen from Fig. 8a and 8b that the pressure at Point 1, 2, 3, 4 and 5 (the distance to the sample inlet of them are 0, 50 mm, 100 mm, 150 mm, 200 mm respectively) increased in order with experiment going on, and the pressure drop between Point 1 and Point 5 finally stablized at around 0.55 MPa. With the increase of gas-liquid ratio, the measured
pressure at each point became higher accordingly and the pressure difference between inlet and outlet of the sample also increased after reaching steady state. This is because the foam apparent viscosity is higher at higher gas-liquid ratio, which leads to increased pressure drop in the foam seepage process. Fig. 8c shows the corresponding water saturation curves under different gas-liquid ratios in the foam flooding process. Among all the curves, the eleven curves with PV numbers ranging from 0.13 to 6.56 are the water saturation variation curves for the foam fluid generated at gas-liquid ratio of 7.0:1; while the four curves at PV numbers of 13.80, 30.20, 46.00 and 65.30 show the water saturation distribution of flooding process at the gas-liquid ratio of 7: 1, 13.5: 1, 20: 1, 30: 1 respectively. It is also found from Fig. 8c that although the pressure drop along the sample increased significantly with the increase of gas-liquid ratio, the residue water saturation after foam flooding was nearly the same.
503
DU Dongxing et al. / Petroleum Exploration and Development, 2016, 43(3): 499–505
Fig. 8.
3.
Seepage characteristics of CO2 foam fluid with different gas-liquid ratios in porous medium. [3]
Conclusions
HOLM L W. Evolution of the carbon-dioxide flooding processes. Journal of Petroleum Technology, 1987, 39(11): 1337–1342.
A homogeneous porous medium model was prepared with sand packing method, and the experimental platform was constructed to study foam seepage characteristics. CT technology was employed to monitor the displacement process and obtain water saturation profiles during the process. Together with the pressure data measured with uniformly distributed pressure transducers along the samples, the CO2 foam flow characteristics have been investigated in this study. The experimental results show: (1) In the porous medium with larger average particle size (0.170.21 mm), the pressure gradient declined gently near the entrance, but became much steeper near the outlet. While in the porous medium with smaller average particle size (0.13 mm), the pressure gradient of the foam fluid near the entrance had no obvious difference with other locations of the porous medium. (2) When the surfactant concentration was lower than the CMC, the generated foam had lower stability and therefore lower displacement efficiency; when the surfactant concentration was slightly higher than the CMC, obvious entrance effect was observed although the water saturation after displacement was lower; at further higher surfactant concentration, the entrance effect diminished and the water saturation was the lowest at the end of the foam flooding. (3) With the increase of gas-liquid ratio, the pressure drop during the displacement process became larger, but the residue water saturation after foam flooding process, did not show obvious difference at different gas-liquid ratios.
[4]
SMITH D H. Promise and problems of miscible-flood enhanced oil recovery: The need for surfactant-based sweep and mobility control. In: ACS Symposium Series 373. Washington: American Chemical Society, 1988: 2–37.
[5]
WANG Gaofeng, ZHENG Xiongjie, ZHANG Yu, et al. A new screening method of low permeability reservoirs suitable for CO2 flooding. Petroleum Exploration and Development, 2015, 42(3): 358–363.
[6]
LAKE L W. Enhanced oil recovery. New Jersey: Prentice Hall, 1989.
[7]
HOLM L W, JOSENDAL V A. Mechanism of oil displacement by carbon dioxide. Journal of Petroleum Technology, 1974, 26(12): 1427–1438.
[8]
ALI J, BURLEY R W, NUTT C W. Foam enhanced oil recovery from sand packs. Chemical Engineering Research and Design, 1985, 63: 101–111.
[9]
PATZEK T W. Field application of foam for mobility improvement and profile control. SPE Reservoir Engineering, 1996, 11(2): 79–86.
[10] CHOU S I. Conditions for generating foam in porous media. SPE 22628, 1991. [11] LI Chun, YI Xiangyi, LU Yuan. Experimental study on CO2 foam profile modification. Drilling & Production Technology, 2008, 31(1): 107–108, 142. [12] WELLINGTON S L, VINEGAR H J. Surfactant-induced mobility control for carbon dioxide studied with computerized tomography. In: ACS Symposium Series 373. Washington: American Chemical Society, 1988. [13] LEE H O, HELLER J P. Carbon dioxide foam mobility meas-
References
urement at high pressure. In: ACS Symposium Series 373. Washington: American Chemical Society, 1988.
[1]
QIN Jishun, HAN Haishui, LIU Xiaolei. Application and
[14] DU D X, ZITHA P L J, UIJTTENHOUT M G H. Carbon di-
enlightenment of carbon dioxide flooding in the United States
oxide foam rheology in porous media: A CT scan study. SPE J,
of America. Petroleum Exploration and Development, 2015, 42(2): 209–216. [2]
2007, 12(2): 245–252. [15] DU D X, NADERI B A, FARAJZADEH R, et al. Effect of
SHEN Pingping, JIANG Huaiyou, CHEN Yongwu, et al.
water solubility on carbon dioxide foam flow in porous media:
Study on CO2 injection technique in EOR. Special Oil & Gas
An X-ray computed tomography study. Industrial & Engi-
Reservoir, 2007, 14(3): 1–4, 11.
neering Chemistry Research, 2008, 47(16): 6298–6306.
504
DU Dongxing et al. / Petroleum Exploration and Development, 2016, 43(3): 499–505
[16] FARAJZADEH R, ANDRIANOV A, BRUINING J, et al. Comparative study of CO2 and N2 foams in porous media at low and high pressure-temperatures. Industrial & Engineering
461–474. [21] BRETHERTON F P. The motion of long bubbles in tubes. J Fluid Mech, 1961, 10(2): 166–188. [22] HIRASAKI G J, LAWSON J B. Mechanisms of foam flow in
Chemistry Research, 2009, 48(9): 4542–4552. [17] FARAJZADEH R, ANDRIANOV A, ZITHA P L J. Investigation of immiscible and miscible foam for enhancing oil recovery. Industrial & Engineering Chemistry Research, 2010,
porous media: Apparent viscosity in smooth capillaries. SPE J, 1985, 25(2): 176–190. [23] DU D X, ZHANG J, SUN S B, et al. Mechanistic study on foam lamellae flow characteristics in tubes. Material Research
49(4): 1910–1919. [18] ANDRIANOV A, FARAJZADEH R, MAHMOODI N M, et al. Immiscible foam for enhancing oil recovery: Bulk and po-
Innovations, 2015, 19(S5): 526–529. [24] AGGELOPOULOS C A, ROBIN M, VIZIKA O. Interfacial
rous media experiments. Industrial & Engineering Chemistry
tension between CO2 and brine (NaCl+CaCl2) at elevated
Research, 2012, 51(1): 2214–2226.
pressures and temperatures: The additive effect of different
[19] ZITHA P L J, DU D X. A new stochastic bubble population model for foam flow in porous media. Transport in Porous Media, 2010, 83(3): 603–621.
salts. Advances in Water Resources, 2011, 34(4): 505–511. [25] BACHU S, BRANT B D. Interfacial tension between CO2, freshwater, and brine in the range of pressure from (2 to 27)
[20] DU D X, ZITHA P L J, VERMOLEN F J. Numerical analysis
MPa, temperature from (20 to 125) C, and water salinity from
of foam motion in porous media using a new stochastic bubble
(0 to 334 000) mg·L−1. J Chem Eng Data, 2009, 54(3):
population model. Transport in Porous Media, 2011, 86(2):
765–775.
505