Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Gas/water foams stabilized with a newly developed anionic surfactant for gas mobility control applications

Gas/water foams stabilized with a newly developed anionic surfactant for gas mobility control... Carbon dioxide (CO ) flooding is one of the most globally used EOR processes to enhance oil recovery. However, the low gas viscosity and density result in gas channeling and gravity override which lead to poor sweep ec ffi iency. Foam application for mobility control is a promising technology to increase the gas viscosity, lower the mobility and improve the sweep effi- ciency in the reservoir. Foam is generated in the reservoir by co-injection of surfactant solutions and gas. Although there are many surfactants that can be used for such purpose, their performance with supercritical CO (ScCO ) is weak causing poor 2 2 or loss of mobility control. This experimental study evaluates a newly developed surfactant (CNF) that was introduced for ScCO mobility control in comparison with a common foaming agent, anionic alpha olefin sulfonate (AOS) surfactant. Experimental work was divided into three stages: foam static tests, interfacial tension measurements, and foam dynamic tests. Both surfactants were investigated at different conditions. In general, results show that both surfactants are good foaming agents to reduce the mobility of ScCO with better performance of CNF surfactant. Shaking tests in the presence of crude oil show that the foam life for CNF extends to more than 24 h but less than that for AOS. Moreover, CNF features lower critical micelle concentration (CMC), higher adsorption, and smaller area/molecule at the liquid–air interface. Furthermore, entering, spreading, and bridging coefficients indicate that CNF surfactant produces very stable foam with light crude oil in both deionized and saline water, whereas AOS was stable only in deionized water. At all conditions for mobility reduction evaluation, CNF exhibits stronger flow resistance, higher foam viscosity, and higher mobility reduction factor than that of AOS surfactant. In addition, CNF and ScCO simultaneous injection produced 8.83% higher oil recovery than that of the baseline experiment and 7.87% higher than that of AOS. Pressure drop profiles for foam flooding using CNF was slightly higher than that of AOS indicating that CNF is better in terms of foam–oil tolerance which resulted in higher oil recovery. Keywords Supercritical CO foam · Foam mobility control · Foam flooding · Enhanced oil recovery (EOR) · Foam assisting CO EOR 1 Introduction It is estimated that two-thirds of the original oil in place (OOIP) are left underground after the primary and second- Edited by Yan-Hua Sun ary oil recovery processes (Green and Willhite 1998). For tertiary recovery, many enhanced oil recovery (EOR) meth- * Mohammed A. Almobarky ods can be used to extract more oil from reservoirs. Among mmobarky@ksu.edu.sa these EOR methods, CO injection is one the most used pro- * Zuhair AlYousif cesses globally (Taber et al. 1997). However, gas injection Zuhair.Yousif@aramco.com processes face many challenges such as gas channeling and Department of Petroleum and Natural Gas Engineering, gravity override that lead to poor sweep efficiency (Healy College of Engineering, King Saud University, Riyadh, et al. 1994). Many techniques have been applied to enhance Kingdom of Saudi Arabia the sweep efficiency such as water alternating gas (WAG), Saudi Aramco, Dhahran, Kingdom of Saudi Arabia polymer, and foam. Foam is a promising technology that Harold Vance Petroleum Engineering Department, Texas can be used to reduce the mobility of the injected gas by A&M University, College Station TX, USA Vol.:(0123456789) 1 3 1026 Petroleum Science (2020) 17:1025–1036 increasing its viscosity (Enick et al. 2012) and diverting the provided better displacement efficiency, and resulted in flow toward lower permeability zones where the remaining higher oil recovery. oil exists (Fried 1961). The aim of this study is to evaluate the ability of a newly Surfactants are the main component in a foam sys- developed anionic surfactant (CNF) to control the mobil- tem. They facilitate the foam generation by reducing the ity of Sc CO . Moreover, the performance of CNF is com- gas–water interfacial tension and adsorb at the interfaces pared with C AOS (anionic surfactant) which is one of 14–16 to make the foam with the required stability by stabilizing the most widely used in literature with C O in gaseous and the thin films between bubbles (Schramm 2000). Thus, the supercritical states. This newly developed surfactant and the surfactant screening is the first step toward a successful foam major challenges for its utilization with ScCO may provide project (Boeije et al. 2017). In foam applications, in general, more opportunities for foam applications in foam-assisting and particularly in C O EOR, the surfactant structure is a miscible CO EOR projects. 2 2 significant factor that affects the efficiency in every aspect of the process: gas viscosity, mobility control, and EOR (Adkins et al. 2010). These effects are related to different 2 Experimental materials interactions of surfactants and C O (Adkins et al. 2010). Moreover, the presence of supercritical C O (ScCO ) results 2 2 Table  1 shows the general properties of both surfactants in low pH acidic environment where some types of sur- used in experimental work. Surfactants were diluted with factants hydrolyze and lose their interfacial activity such as deionized (DI) water (ASTM, type II) provided by LabChem sulfates (Talley 1988). Inc. Moreover, tests were conducted at 0.5 wt% surfactant Alpha olefin sulfonate (AOS) is hydrolytically and ther - concentration. Besides deionized water, the salinity effect mally stable, and soluble at low to medium hard water (Por- was investigated using brine solutions (NaCl solutions) at ter 1994). Farajzadeh et al. (2010) experimentally inves- 10,000, 20,000, and 30,000 ppm (NaCl was purchased from tigated the use of AOS for mobility control and EOR in Cole-Parmer). The crude oil used in this study was from miscible and immiscible flooding with the aid of CT scanner North Burbank Unit, OK, USA (NBU). It is light crude oil for simultaneous monitoring of the flooding process. They with API gravity of 33.7° and viscosity of 8 cP at room reported 19% more oil recovery with ScC O than that of temperature of 23 °C, and 39.5° API and 3.27 cP at 50 °C the immiscible CO flooding. However, no sharp front was which is the reservoir temperature. The glass-bead pack was observed with the use of ScC O . They attributed this to the made using glass beads with a specific gravity of 2.5 and poor foam stability with oil. Haugen et al. (2012) experimen- a diameter of 100 µm, which were purchased from Potters tally used AOS for mobility control and EOR in fractured oil Industries LLC. wet and water wet cores and reported that the pre-generated foam is better than in situ foam generation in terms of mobil- ity reduction and oil recovery. They attributed the results to the poor foam–oil tolerance. Li et al. (2012) used AOS in 3 Methodology surfactant-alternating-gas (SAG) injection mode for foam generation using N . Their experiments were conducted on The experimental work was divided into three stages: static a two-dimensional sand pack with 19–1 permeability con- foam tests, interfacial tension measurements, and dynamic trast. They attributed the poor sweep efficiency to the weak foam tests. The dynamic foam tests were divided into three foam stability in the presence of crude oil. They suggested sections: mobility reduction evaluation in the high-permea- that enhancing the foam–oil tolerance could provide higher bility glass-bead pack, mobility reduction evaluation in low- oil recovery because this may enhance the sweep efficiency. permeability Bentheimer sandstone cores, and core flooding Indeed, mixing the surfactant with a foam booster CTAB experiments. The surfactant concentrations were constant zwitterionic surfactant improved the foam–oil tolerance, at 0.5  wt%, diluted with deionized water, and prepared Table 1 Properties of the surfactants Surfactant Form Chemical family pH Density, g/mL Charge Flash point, °C Carbon chain length CNF Liquid Alpha olefin sulfonate, isopropyl 7.73 1.07 Anionic > 93.3 – alcohol, and citrus terpenes AOS Liquid Alpha olefin sulfonate 8.20 1.06 Anionic > 94.0 14–16 1 3 Petroleum Science (2020) 17:1025–1036 1027 with NaCl solutions of three salinities: 10,000, 20,000, and S =  −( +  ) a/w o/w o/g (2) 30,000 ppm. 2 2 2 B =  +  − (3) a/w o/w o/g 3.1 Static foam tests Foam was generated by shaking 3 mL of surfactant solu- a/w L = (4) tions in 13 × 100-mm (9-mL) Pyrex glass test tubes. Care has o/w been taken to perform 10–15 gentle and uniform shakings where  ,  ,  are the air–water, oil–water, and oil–gas for all samples. Samples were prepared at 0.5 wt% concen- a/w o/w o/g interfacial tensions, respectively. tration in deionized water, 10,000, 20,000, and 30,000 NaCl solutions. After the foam has been generated with shaking inside the test tube, the foam columns were monitored by 3.3 Dynamic foam tests taking images at different times. Then, the foam column lengths were measured from images using ImageJ software. These experiments were designed for mobility reduction The foaming ability was investigated using the initial foam evaluation and oil recovery investigation by conducting core column length (h ), and the foam stability was measured by flood experiments. fi the foam half-life (FHL), t , which is the time at which the 1/2 foam column loses half of the initial foam column length 3.3.1 Mobility reduction evaluation h . The samples were prepared for static tests without oil fi in the high‑permeability glass‑bead pack and stirred for about 12 h before testing. For static tests with crude oil, the samples were prepared at 0.5 wt% surfactant These experiments were conducted at a high shear rate of −1 −1 concentration and stirred for 12 h, and the surfactant solution 317 s and a low shear rate of 9.51 s . Furthermore, three was placed in 9-mL test tubes above which the crude oil was injection qualities 50%, 70%, and 90% were used. All experi- simply poured. Then, the sample was shaken immediately. ments were conducted at 1800 psi and 50 °C to ensure the supercritical conditions of C O . The foam was generated 3.2 Interfacial tension measurements by simultaneously injecting the surfactant solution and supercritical CO (ScCO ) through the glass-bead pack. 2 2 Air–water surface tension measurements were conducted at The pressure drop was measured using two sets of pressure different surfactant concentrations in DI water using a Data- transducers: 500-psi for high range and 50-psi for low range. physics OCA 15 Pro IFT instrument, pendant drop method. The pressure drop data were collected using a data acqui- The surface measurements were used for critical micelle sition system. The onset of a strong foam generation was concentration (CMC) determination and interfacial activity recognized as a rapid increase in pressure drop according predictions for both surfactants. to Dicksen et al. (2002). The flow continued with monitor - The air–water surface tension versus the logarithmic val- ing the pressure drop until the steady-state pressure drop ues of the surfactant concentrations below the CMC is a across the glass-bead pack was reached. Then, the steady- linear relationship. The slope of this straight line can be used state pressure data were averaged and used to calculate the to interpret the interfacial activities: adsorption and area/ mobility, foam effective viscosity, and mobility reduction molecule at the interface. According to the Gibbs adsorption factor (MRF) using the following equations. equation, the higher the slope is, the higher the adsorption at ql the air–water interface. Furthermore, the higher the adsorp- = = (5) AΔP tion of surfactant molecules at the interface results in smaller area/molecule and stronger packing that induces higher foam stability (Rosen and Kunjappu 2004). (6) eff The interfacial tension measurements, also, can be also used to investigate the foam–oil tolerance by calculating the entering coefficient (E ) (Robinson and Woods 1948), spread- ΔP foam f = (7) ing coefficient (S ) (Harkins 1941), bridging coefficient (B ) mr ΔP baseline (Denkov 2004), and the lamellae number (L) (Schramm and Novosad 1990) using Eqs. (1), (2), (3), and (4). where  is the mobility, k is the permeability,  is the vis- cosity, q is the flow rate, l is the length of the porous media, E =  +  − a/w o/w o/g (1) A is the cross-sectional area of the glass-bead pack, ∆P is the pressure drop across the porous media, μ is the foam eff effective viscosity, and f is the mobility reduction factor. mr 1 3 1028 Petroleum Science (2020) 17:1025–1036 Table 2 shows the dimensions and petrophysical proper- to ensure 100% core saturation. Although the XRD tests ties of the glass-bead pack which was filled with 100 µm for these rocks show that their composition is 100% quartz, glass beads. Figure 1 shows the experimental setup for the 1 PV of the surfactant solution was injected into the core at −1 mobility reduction evaluation. 5 ft/day (~ 9 s shear rate) to mitigate the effect of surfactant The baseline experiment, in which ScC O was injected adsorption on the rock surfaces. After that, the foam was without surfactant, was also conducted for comparison pur- applied by simultaneously injecting the surfactant solution poses. Detailed experimental conditions are described in and ScCO or N gas at 5 ft/day. The foam injection was 2 2 Sect. 4.4. continued until the steady pressure drop across the core was reached. The recorded steady-state pressure drop data were 3.3.2 Mobility reduction evaluation in low‑permeability averaged and used to calculate the mobility, effective viscos- Bentheimer sandstone cores ity of foam, and MRF using Eqs. 5, 6, and 7. Figure 1 above shows a schematic diagram of the exper- All tests in this section were conducted on homogeneous imental setup for mobility evaluation in sandstone cores. Bentheimer sandstone cores (diameter, 1 in.; length, 12 in.). Properties of Bentheimer sandstone cores and experimental The core was left in an oven overnight for drying. Then, conditions for all runs are described in Sect. 4.5. Moreo- it was mounted in the core holder and 500 psi overburden ver, the last two runs listed (runs 9 and 10) are baseline pressure was applied. After that, the air was removed from experiments conducted using N and ScCO injection for 2 2 the core using a vacuum pump followed by saturating the comparison purposes, respectively. In all runs, the salinity core with 10,000 ppm NaCl brine at which its pore volume of the surfactant solution or water injected is 10,000 ppm. (PV) and porosity can be measured. The system was then pressurized using the back-pressure regulator, as shown in 3.3.3 Core flooding experiments Fig. 1. After the system was pressurized and the tempera- ture was maintained at 50 °C, the absolute permeability was The core flooding experiments were also conducted on measured by obtaining the pressure drop at different flow homogeneous Bentheimer sandstone cores with a diameter rates using Darcy’s law. Then, 5–6 PV of brine were injected of 1″ and a length of 12″. The core was left in an oven over- night for drying. Then, it was vacuumed and saturated with 10,000 ppm brine at which the pore volume and porosity were measured. After that, 5–6 PV of 10,000 ppm brine Table 2 Dimensions and petrophysical properties of the glass-bead pack were injected into the core at 5 ft/day to ensure that the core sample was 100% water-saturated. Then, the absolute per- Length, in. Diameter, Pore vol- Porosity, % Permeabil- meability was calculated using Darcy’s law. The setup was in. ume, mL ity, D pressurized to 1450 psi, keeping the overburden pressure 13 0.18 1.625 30 17.1 500 psi higher than the test pressure, and at test temperature Overburden pressure Oven Core holder Glass bead pack BPR Graduated Pressure transducers Pressure transducers cylinder Data acquisition system Pumps Fig. 1 Schematic of an experimental setup for mobility reduction evaluation and core flooding 1 3 Water Surfactant Crude oil Gas Petroleum Science (2020) 17:1025–1036 1029 of 50 °C. Once the pressure and temperature were stable, CNF AOS crude oil was injected at a flow rate of 5 ft/day until no more 6.4 water was observed in the eu ffl ent. The water produced from 5.6 the core sample was collected in a graduated cylinder for the 4.4 4.3 4.0 3.9 original oil in place (OOIP) estimation. Then, water flooding 4 3.5 2.8 was applied by injecting 5 PV of 10,000 ppm brine at a flow rate of 5 ft/day until no more oil was observed in the efflu- 2 ent. The high amount of water injection was to ensure that the optimum oil recovery by waterflooding was achieved, 0 (DI water) 10,000 20,000 30,000 no more oil was produced by water injection, and the end NaCl salinity, ppm effects would be diminished. Then, the second stage was to inject 1–1.5 PV of the surfactant solution (surfactant pre- Fig. 2 Foam half-lives for both surfactants (0.5  wt%) in deionized flush) to mitigate the surfactant adsorption on the rock. After water, 10,000, 20,000, and 30,000 ppm NaCl solutions that, 5 PV of simultaneous injection of the surfactant solu- tion and ScCO was applied for 24 h at a flow rate of 5 ft/day. Moreover, CNF surfactant is slightly better than AOS at The pressure drop was recorded for the three oil recovery stages. Figure 1 shows the schematic diagram of the setup any salinity (0–30,000 ppm). For both surfactants, the sta- bility of foam prepared with deionized water is the best. In for the core flooding experiments. One baseline experiment was conducted by injecting ScCO only for comparison addition, both surfactants exhibit an enhancement of foam stability at the salinity of 30,000 ppm. Liu et al. (2005) purposes with oil recovery of CNF and AOS foam floods. Table 3 shows the properties for the Bentheimer sandstone reported that the foam stability with C O decreased with increasing salinity up to 2 wt%. Then, a further increase samples used to conduct the core flooding experiments and the experimental conditions. in salinity enhanced the foam stability. However, the foam stability plateaued shortly with further increase in salinity. The foam stability with crude oil was impressive for both surfactants. The foam half-life for CNF lasted for 4 Results and discussion more than 24 h and 12–18 h for AOS. Figure  3a shows images for CNF and Fig.  3b for AOS. The images were 4.1 Static foam test results taken after 24 h for CNF foams and after 18 h for AOS foams. Each image shows samples in DI water, 10,000 and Foaming ability (or foamability) was investigated by meas- uring the initial foam heights h for both surfactants from 20,000 ppm salinity left to right. Clearly, both surfactants fi produced stable foam with crude oil, but CNF foam was the shaking tests. Both surfactants gave almost the same ini- tial foam column length h . Therefore, both CNF and AOS significantly stronger than that of AOS in the presence of fi crude oil. are good foaming agents in terms of foaming ability. This also indicates the efficiency of both surfactants to reduce the However, shaking involves eventually very high shear rates which provide high energy for any surfactant to give air–liquid surface tension. The surface tension at the air–liq- uid interface will be discussed shortly. its optimum performance as foaming agent regardless of how the shaking was performed. Therefore, it is difficult Figure 2 shows the foam half-life for both surfactants at different salinities of 0 (deionized water), 10,000, 20,000, to recognize the differences in foaming ability and foam stability as well. Therefore, combining the shaking test and 30,000  ppm. Both surfactants provided good foam stability. As observed, the foam stability decreases as the observations with the interfacial tension measurements is next. salinity increases at the salinity of 0–20,000 ppm, which is attributed to a reduction in repulsive forces between the surfactant molecules due to the addition of salts. Table 3 Petrophysical properties of the sandstone cores and the experimental conditions in core flooding experiments Run No. Surfactant Core length, Core diameter, Core pore vol- Core porosity, % Core permeabil- Experiment type in. in. ume, mL ity, D 1 – 12 1 33.52 21.71 1.87 Baseline 2 AOS 12 1 34.74 22.50 1.71 Foam flood 3 CNF 12 1 33.74 21.85 1.91 Foam flood 1 3 Foam half-life, t 1/2 1030 Petroleum Science (2020) 17:1025–1036 with a decrease in the surfactant concentration and no foam was generated at surfactant concentration below its CMC. Therefore, the lower the CMC of the surfactant is the bet- ter from different perspectives such as lowering the cost of the project due to the use of low surfactant concentrations without compromising the foaming efficiency. According to Gibbs surface adsorption equation (Eq. 8), the higher the slope is, the higher the adsorption of the sur- factant at the liquid–air interface, and consequently, the better the foamability and foam stability (Rosen and Kun- jappu 2004). This is because the higher the adsorption at the air–liquid interface is, the smaller the area/molecule and the stronger the packing of the molecules at the interface (Rosen and Kunjappu 2004). =− (8) RT d ln C Fig. 3 a Images of CNF foam after 24 h and b images of AOS foam where  is the surfactant adsorption at the air–liquid inter- after 18 h, both foam samples at 0.5 wt% surfactant concentration and at a salinity of 0 (deionized water), 10,000 and 20,000 ppm from left face, R is the gas constant, T is temperature, γ is the surface to right tension, and C is the surfactant concentration. Furthermore, for concentrations above 0.007  wt% in Fig. 4, CNF is able to reduce the interfacial tension lower 4.2 Interfacial tensions than that of AOS which also indicates that CNF is predicted to perform better than AOS in foam generation. Table  4 As mentioned earlier, surface tensions were measured for shows the air–water and oil–water interfacial tensions for different concentration samples prepared with deionized both 0.5 wt% surfactant solutions at different salinities. water and the measured results are shown in Fig.  4. The critical micelle concentrations (CMCs) for CNF and AOS 4.3 Foam stability in the presence of crude oil are 0.011 wt% and 0.028 wt%, respectively. The surfactant concentration for foam applications is recommended above As shown in Table 4, the σ for both surfactants dropped o/w the CMC (Nikolov et al. 1986). Using the surfactant con- with the addition of salts (from deionized water to centration above the CMC provides the best foam stability, 10,000 ppm salinity), but no influence was observed on σ o/w whereas foam has less opportunity to be of good stability with increasing the salinity. This shows that both surfactants when the surfactant concentration is below the CMC (Rafati would probably perform efficiently in the presence of oil, but et al. 2012). The CMC defines the foaming efficiency of the CNF had higher σ which suggests that it is better in terms o/w surfactants. The lower the CMC is, the higher the foaming of foam–oil tolerance. efficiency (Rosen and Kunjappu 2004). Moreover, Man- Table  5 shows the entering, spreading, bridging coef- nhardt et al. (2000) found that the foaming ability decreased ficients, and lamellae number, respectively. The entering coefficient gives positive values for all surfactant solutions, CNF AOS Table 4 Interfacial tensions σ and σ for AOS and CNF (0.5 wt%) a/w o/w at 23 °C Salinity of surfactant solu- IFT between gas IFT between gas tion (water phase), ppm (or oil) and AOS (or oil) and CNF solution, mN/m solution, mN/m σ σ σ σ a/w o/w a/w o/w 0.001 0.010 0.100 1.000 0 (deionized water) 32.50 1.40 30.70 5.88 logC, wt% 10,000 32.30 0.52 31.00 3.94 20,000 32.10 0.44 31.25 3.51 Fig. 4 Interfacial measurements for AOS and CNF (C is the sur- 30,000 32.15 0.38 31.11 3.11 factant concentration, wt%) 1 3 Interfacial tension, mN/m Petroleum Science (2020) 17:1025–1036 1031 Table 5 Entering, spreading, bridging coefficients, and lamellae number at 0.5 wt% surfactant concentration at 23 °C Salinity of surfactant solution (water Entering coefficient Spreading coefficient Bridging coefficient Lamellae number phase), ppm AOS CNF AOS CNF AOS CNF AOS CNF 0 (Prepared with deionized water) 2.40 5.08 − 0.40 − 6.68 65.96 − 15.19 3.48 0.78 10,000 1.32 3.44 0.28 − 4.44 51.31 − 15.73 9.32 1.18 20,000 1.04 3.26 0.16 − 3.76 38.35 − 3.37 10.94 1.34 30,000 1.03 2.72 0.27 − 3.50 41.52 − 14.75 12.69 1.50 which clarifies that oil will enter the lamellae. However, 4.4 Mobility reduction evaluation the spreading coefficient implies that oil will spread at the in the high‑permeability glass‑bead pack air–water interface to destabilize AOS foams in saline solu- tions only. The spreading coefficient for CNF is negative for Table 6 shows the results for 16 runs on the high-permea- all samples. Moreover, the bridging coefficient gives nega- bility glass-bead pack at high and low shear rates for both tive values for CNF at all conditions. These observations surfactants with ScC O foam. Moreover, one baseline exper- imply that CNF is going produce very stable foam with oil. iment was conducted with ScC O injection only and resulted However, AOS produces stable foam with oil when the sur- in a very low steady-state pressure drop (0.04 psi) which factant prepared with deionized water. Furthermore, lamel- corresponds with 0.03 cP effective viscosity. The average lae number values confirm the same conclusions. The CNF of the recorded steady-state pressure drop data were used to foam is stable in deionized water and semi-stable in saline calculate the mobility, foam effective viscosity, and mobil- solutions, whereas AOS is semi-stable in deionized water ity reduction factor. Experimental conditions and results for and unstable in saline solutions. these tests are listed in Table 6. Figure 5 shows the effect of three injection qualities on the viscosities of both surfactants at the high shear rate in deionized water. The AOS foam viscosity decreased but the CNF foam viscosity increased as the injection quality decreased. The opposite behavior of these foaming agents Table 6 Experimental conditions and results for AOS and CNF with CO in glass-bead packs a −1 Run Surfactant Salinity, ppm Injection Flow rate Shear rate, s Steady-state pres- Mobility Effective Mobility No. quality, % q, mL/min sure drop ΔP , λ, mD/cP viscosity μ , reduction fac- ss eff psi cP tor f mr 1 AOS 0 90 0.5 317 208 118 145 – 2 AOS 10,000 90 0.5 317 164 150 114 4100 3 AOS 20,000 90 0.5 317 161 156 109 – 4 AOS 30,000 90 0.5 317 216 459 37 – 5 AOS 0 70 0.5 317 175 142 121 – 6 AOS 10,000 70 0.5 317 99 249 69 2475 7 AOS 20,000 70 0.5 317 93 265 64 – 8 AOS 30,000 70 0.5 317 91 273 63 – 9 AOS 0 50 0.5 317 122 202 85 – 10 AOS 10,000 90 0.015 9.51 9.97 74 230 250 11 CNF 0 90 0.5 317 162 153 112 – 12 CNF 0 70 0.5 317 207 119 143 – 13 CNF 0 50 0.5 317 225 109 154 – 14 CNF 10,000 90 0.5 317 178 138 123 4450 15 CNF 10,000 90 0.015 9.51 16 46 371 400 16 – 10,000 – 0.5 317 0.04 – 0.03 – The surfactant concentrations were 0.5 wt% in runs 1–15. No surfactant was used in the baseline experiment (run 16) 1 3 1032 Petroleum Science (2020) 17:1025–1036 CNF AOS CNF AOS 9.97 ΔPss, psi λ, mD/cP μeff, cP fmr 90 70 50 Quality, % Fig. 7 Comparison of experimental results at low shear rate between AOS and CNF foams (0.5 wt%, 10,000 ppm salinity) with ScCO at Fig. 5 Effect of injection quality on the effective viscosities of AOS 90% injection quality and CNF (0.5 wt% in deionized water) foams in the glass-bead pack −1 at a high shear rate of 317 s lower mobility, higher foam viscosity, and higher mobility can be attributed to many reasons that were not covered in reduction factor. this study. In fact, the relationship of the foam viscosity with injection quality is still controversial. In an experimental study conducted by Marsden and Khan (1966), they found 4.5 Mobility reduction evaluation that the higher the injection quality is, the higher the foam in low‑permeability Bentheimer sandstone viscosity. Lee and Heller (1990) experimentally concluded cores that the increase in injection quality decreased the viscosity. The foam viscosity depends on flow rate, permeability of Two baseline runs were conducted using N injection at the porous media, and foam texture (Hirasaki and Lawson 850 psi and ScCO injection at 1800 psi for comparison. 1985). All experimental conditions and results are listed in Table 7. −1 The effect of shear rate (317 and 9.51 s ) on the effective The effect of permeability is shown in Fig.  8 which com- viscosities of both AOS and CNF foams at 90% injection pares the foam viscosities for AOS and CNF with ScCO at −1 quality and 10,000 ppm salinity are shown in Fig. 6. The 90% injection quality at a shear rate of 9 s . The perme- higher the shear rate resulted in the lower viscosity because abilities of the cores and the glass-bead pack are 1.62–1.80 of the shear thinning nature of the foam. As shown in Fig. 6, Darcy (Table 7) and 17.1 Darcy (Table 2), respectively. CNF the viscosities of CNF foam at both shear rates are higher is repeatedly proved to be better than AOS by generating than that of the AOS foam. higher foam viscosity in both cases at high and low-perme- Figure 7 also shows all results for CNF and AOS foams ability porous media. The results in Fig. 8 are in agreement −1 (0.5 wt%, 10,000 ppm salinity) at a low shear rate of 9.51 s with the fact that foam favors the higher permeability porous and 90% injection quality. Again, the CNF foam proves pow- media. The viscosities of surfactant foams in the glass-bead erful performance with higher steady-state pressure drop, pack are extremely higher than that in sandstone cores. The effect of injection quality at low-permeability sand- stone cores was investigated using two injection qualities 90% and 70% for both surfactants with ScC O at a shear 371 −1 CNF AOS rate of 9.51 s . The pressure drop curves for both surfactant foams are shown in Fig.  9 for 90% injection quality and Fig. 10 for 70% injection quality. According to the pressure profiles, the CNF foam is stronger at 90% injection quality, whereas the AOS foam is stronger at 70% injection quality. Therefore, the results suggest that CNF is better at more realistic conditions (i.e., low shear rate and sandstone reser- voirs), CNF co-injection with ScC O provides higher foam 9.51 317 viscosity at higher injection qualities, while AOS requires -1 low injection qualities for better performance. This also indi- Shear rate, s cates that using CNF with ScC O for foam mobility con- trol will eventually reduce the cost as the amount of liquid Fig. 6 Effect of shear rate on the effective viscosities of AOS and decreases as the injection quality increases. CNF foams at 0.5  wt% surfactant concentration and 90% injection quality 1 3 Effective viscosity, cP Effective viscosity, cP Petroleum Science (2020) 17:1025–1036 1033 CNF AOS 5.65 4.13 17.1 1.7 Permeability, D Fig. 8 Effect of permeability on effective viscosity of AOS and CNF foams (0.5  wt%, 10,000  ppm salinity) with ScC O at 90% injection quality 1.0 90% quality AOS CNF 0.8 0.6 0.4 0.2 0 123456 7 Injection volume, PV Fig. 9 Pressure drop curves for AOS and CNF foams (with ScCO ) at 90% injection quality in sandstone cores 70% quality AOS CNF 0246 810 Injection volume, PV Fig. 10 Pressure drop curves for AOS and CNF foams (with ScCO ) at 70% injection quality in sandstone cores The mobility control with N gas was also tested for both surfactants. The pressure profiles for AOS and CNF are shown in Fig. 11. These experiments were conducted on sandstone core samples at two velocities of 5 and 10 ft/ −1 day (9 and 18 s ) with an injection quality of 90%. In these experiments, the surfactant solutions used had a concentra- tion of 0.5 wt%, which were prepared with 10,000 ppm NaCl 1 3 Table 7 Experimental conditions and results for AOS and CNF in sandstone cores −1 Run No. Surfactant Gas Pressure P, psi Velocity, Shear rate, s Injection Pore vol- Permeabil- Steady-state pres- Mobility λ, Foam effective Mobility ft/day quality, % ume, mL ity k, D sure drop ΔP , psi mD/cP viscosity, cP reduction fac- ss tor f mr 1 AOS ScCO 1800 5 8.77 90 31.24 1.70 0.40 411 4.13 1.67 2 AOS ScCO 1800 5 9.30 70 33.74 1.58 25.42 6.79 232.8 105.9 3 CNF ScCO 1800 5 9.10 90 34.74 1.70 0.60 301 5.65 2.5 4 CNF ScCO 1800 5 8.92 70 34.74 1.78 2.00 89 20 8.3 5 AOS N 850 5 9.15 90 34.24 1.65 1.35 130 12.7 3.4 6 AOS N 850 10 18.31 90 34.24 1.65 3.46 100 18 8.6 7 CNF N 850 5 8.76 90 34.24 1.80 12.89 13.61 121.23 32.2 8 CNF N 850 10 17.53 90 34.24 1.80 34.4 10.20 176.5 86 9 – N 850 5 9.05 – 33.74 1.62 0.40 – 10 – ScCO 1800 5 8.83 – 32.74 1.70 0.24 680 0.04 – Effective viscosity, cP Pressure dop, psi Pressure dop, psi 1034 Petroleum Science (2020) 17:1025–1036 50 100 CNF 5 ft/day CNF 10 ft/day AOS 5 ft/day AOS 10 ft/day 40 80 ScCO2 flooding 30 60 Water flooding 20 40 10 20 0 0 05 10 15 20 0123456789 10 Injection volume, PV Injection volume, PV Fig. 11 Pressure drop curves for AOS and CNF foams with N at dif- Fig. 13 Oil recovery of the baseline experiment ferent injection velocities (90% injection quality, 850  psi injection pressure) 100 10 Surfactant 80 Water flooding AOS foam flooding 8 pre-flush CNF AOS 176.5 Oil recovery 60 6 121.3 40 4 20 2 Pressure drop 0 0 18.0 12.7 0 246 81012 Injection volume, PV 5 10 Velocity, ft/day Fig. 14 Oil recovery and pressure drop curves across the sandstone core for waterflooding, AOS surfactant (0.5  wt% and 10,000  ppm Fig. 12 Foam effective viscosity in sandstone cores for CNF and salinity) pre-flush, and AOS surfactant with ScCO foam flood at 90% AOS foams with N gas at different injection velocities (90% injec- injection quality tion quality, 850 psi injection pressure) solution. Each surfactant solution was injected simultane- 4.6 Core flooding experiments ously with N gas at 5 ft/day until the steady-state pressure drop was reached, then the velocity was raised to 10 ft/day The oil recovery from the baseline experiment is shown as until the steady-state pressure drop was reached for the new a function of pore volume injected in Fig. 13. The ultimate velocity. From Fig. 11, the CNF foam appeared always better oil recovery of water flooding is 39% of OOIP. The water in terms of flow resistance at both velocities than the AOS flooding was followed by 5–6 PV of continuous ScCO foam. Moreover, Fig. 12 compares the foam viscosities for injection which resulted in 27.54% OOIP more oil recov- AOS and CNF at both velocities. Surprisingly, the viscosity ery. The total oil recovery from the baseline experiment increases as the velocity (i.e., shear rate) increases for both is 66.54% of the OOIP. foams. Although this is shear thickening behavior, foam is Figure 14 shows the results for AOS as a foaming agent known of its non-Newtonian shear thinning nature (Sch- after water flooding. The same procedure was conducted. ramm and Wassmuth 1994). This behavior could be related This run started with injecting 4.56 PV water flooding, to the procedure of the experimental work. At 5 ft/day, the which resulted in an oil recovery of 35.42% of the OOIP. foam was very efficient to provide high flow resistance due Then, 1.62 PV of the AOS surfactant solution was injected to the gas blockage effect. Therefore, the gas relative perme- to reduce the surfactant adsorption on the rock surfaces. ability is already low, and increasing the velocity to 10 ft/ The surfactant pre-flush stage resulted in an oil recovery day at such conditions would promote the foam generation of 4.75% of the OOIP. The third stage was the foam flood because of the high shear rate in a blocked porous media. As with simultaneous injection of 5 PV of AOS-ScCO . The a result, foam contradicted its shear thinning nature by pro- AOS foam flood resulted in an additional oil recovery of viding higher viscosity at higher shear rates, lesson learned. 28.5%. Then, the total oil recovery was 68.67% of the 1 3 Effective viscosity, cP Pressure drop, psi Oil recovery, % Oil recovery, % Pressure drop, psi Petroleum Science (2020) 17:1025–1036 1035 of salts in terms of foam–oil tolerance. However, the lamellae number for AOS shows semi-stable in deion- Surfactant 80 8 Water flooding CNF foam flooding pre-flush ized water and unstable at all salinities. These results Oil recovery show that CNF foam is very stable with oil more than 60 6 AOS foam. (4) At 90% injection quality, the mobility reduction with 40 4 Pressure drop ScCO , and with N at lower pressures, CNF shows 2 2 20 2 higher foam viscosity and better mobility reduction than AOS. 0 0 (5) For oil recovery, AOS foam produced 1% more than 0246 810 12 the baseline experiment, whereas CNF foam produced Injection volume, PV almost 7.87% more than AOS foam, and 8.38% more than the baseline experiment. Fig. 15 Oil recovery and pressure drop curves across the sandstone (6) It is not recommended to investigate foam viscosity at core for waterflooding, CNF surfactant (0.5  wt% and 10,000  ppm more than one shear rate during one experiment. Test- salinity) pre-flush, and CNF surfactant with ScCO foam flood at 90% injection quality ing two shear rates in series in one experiment would provide erroneous results. In this study, measuring the foam viscosity at two shear rates in one experiment OOIP. The additional recovery by AOS foam accounted resulted in shear thickening foam behavior, while foam for 1% higher than that of baseline experiment. is non-Newtonian shear thinning in nature. CNF foam core flood results are shown in Fig.  15. Water flooding stage produced an oil recovery of 39.66% of the Open Access This article is licensed under a Creative Commons Attri- OOIP. Moreover, no oil production was observed during the bution 4.0 International License, which permits use, sharing, adapta- surfactant pre-flush stage in which 1.5 PV of CNF surfactant tion, distribution and reproduction in any medium or format, as long solution was injected to reduce the adsorption effect. The as you give appropriate credit to the original author(s) and the source, final stage, CNF-ScCO simultaneous injection or foam provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are flooding resulted in a recovery of 36.3% of the OOIP after included in the article’s Creative Commons licence, unless indicated water flooding. The amount of oil produced by the CNF otherwise in a credit line to the material. If material is not included in foam was 7.87% higher than that by the AOS foam and the article’s Creative Commons licence and your intended use is not 8.83% higher than the baseline experiment. The total recov- permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a ery for this core flood was 75.96%. copy of this licence, visit http://creativ ecommons .or g/licenses/b y/4.0/. 5 Conclusions References (1) Both AOS and CNF surfactants show good foaming ability and foam stability without oil in shaking tests. Adkins S, Chen X, Nguyen Q, Sanders A, Johnston K. Effect of Such behavior is not representative to the actual foam- branching on the interfacial properties of nonionic hydrocarbon ing ability of both surfactants because the shaking tests surfactants at the air–water and carbon dioxide–water inter- faces. J Colloid Interface Sci. 2010;346(2):455–63. https ://doi. are naturally involved high shear rate which enforces org/10.1016/j.jcis.2009.12.059. the surfactants to perform at their optimum abilities as Boeije C, Bennetzen M, Rossen W. A methodology for screening foaming agents. However, CNF was able to reduce the surfactants for foam enhanced oil recovery in an oil-wet res- air–water interfacial tension lower than that of AOS. ervoir. SPE Res Eval Eng. 2017;20(04):795–808. https ://doi. org/10.2118/18518 2-PA. (2) The newly developed surfactant, CNF, shows impres- Denkov N. Mechanisms of foam destruction by oil-based antifoams. sive foam oil tolerance than AOS. It does not reduce the Langmuir. 2004;20(22):9463–505. https ://doi.org/10.1021/la049 oil–water interfacial tension to low value as with AOS 676o. in saline solutions. This makes it a good surfactant for Dicksen T, Hirasaki G, Miller C. Conditions for foam generation in homogeneous porous media. In: SPE/DOE improved oil recov- the applications of foam for mobility control. ery symposium, 13–17 April, Tulsa, Oklahoma; 2002. https: //doi. (3) CNF provides negative values of entering, spread- org/10.2118/75176 -MS. ing, and bridging coefficients, whereas AOS provides Enick R, Olsen D, Ammer J, Schuller W. Mobility and conformance negative values in deionized water only. Moreover, the control for C O EOR via thickeners, foams, and gels—a literature review of 40 years of research and pilot tests. In: SPE improved oil lamellae number indicates that the CNF foam is stable in deionized water and semi-stable with the addition 1 3 Oil recovery, % Pressure drop, psi 1036 Petroleum Science (2020) 17:1025–1036 recovery symposium, 14–18 April, Tulsa, Oklahoma, USA; 2012. Marsden S, Khan S. The flow of foam through short porous media and https ://doi.org/10.2118/15412 2-MS. apparent viscosity measurements. SPE J. 1966;6(01):17–25. https Farajzadeh R, Andrianov A, Zitha P. Investigation of immiscible ://doi.org/10.2118/1319-PA. and miscible foam for enhancing oil recovery. Ind Eng Chem. Nikolov A, Wasan D, Huang D, Edwards D. The effect of oil on foam 2010;49(4):1910–9. https ://doi.org/10.1021/ie901 109d. stability: mechanisms and implications for oil displacement by Fried A. The foam-drive process for increasing the recovery of oil. foam in porous media. In: SPE annual technical conference and Washington, DC: U.S. Bureau of Mixes, Rep. Inv., 5866; 1961. exhibition, 5–8 October, New Orleans, Louisiana; 1986. https :// Green D, Willhite G. Enhanced oil recovery. Volume 6, SPE Textbook doi.org/10.2118/15443 -MS. Series. Richardson, Texas: SPE; 1998. Porter MR. Handbook of surfactants. London: Blackie Academic & Harkins W. A general thermodynamic theory of the spreading of liquids Professional; 1994. to form duplex films and of liquids or solids to form monolayers. Rafati R, Hamidi H, Idris A, Manan M. Application of sustainable J Chem Phys. 1941;9(7):552–68. https://doi.or g/10.1063/1.17509 foaming agents to control the mobility of carbon dioxide in 53. enhanced oil recovery. Egypt J Pet. 2012;21(2):155–63. https :// Haugen Å, Fernø M, Graue A, Bertin H. Experimental study of foam doi.org/10.1016/j.ejpe.2012.11.010. flow in fractured oil-wet limestone for enhanced oil recovery. SPE Robinson J, Woods W. A method of selecting foam inhibitors. J Soc Res Eval Eng. 2012;15(02):218–28. https ://doi.or g/10.2118/12976 Chem Ind. 1948;67(9):361–5. https: //doi.org/10.1002/jctb.50006 3-PA. 70908 . Healy R, Holstein E, Batycky J. Status of miscible flooding technology. Rosen M, Kunjappu J. Foaming and antifoaming by aqueous solutions In: 14th world petroleum congress, 29 May-1 June, Stavanger, of surfactants and interfacial phenomena. Hoboken: Wiley; 2004. Norway; 1994. p. 308–35. Hirasaki G, Lawson J. Mechanisms of foam flow in porous media: Schramm L. Surfactants: fundamentals and applications in the petro- apparent viscosity in smooth capillaries. SPE J. 1985;25(02):176– leum industry. Cambridge: Cambridge University Press; 2000. 90. https ://doi.org/10.2118/12129 -PA. p. 3–50. Lee H, Heller J. Laboratory measurements of CO -foam mobility. SPE Schramm L, Novosad J. Micro-visualization of foam interactions Res Eng. 1990;5(02):193–7. https ://doi.org/10.2118/17363 -PA. with a crude oil. Colloids Surf. 1990;46(1):21–43. https ://doi. Li R, Hirasaki G, Miller C, Masalmeh S. Wettability alteration and org/10.1016/0166-6622(90)80046 -7. foam mobility control in a layered, 2D heterogeneous sandpack. Schramm L, Wassmuth F. Foams: basic principles foams: fundamentals SPE J. 2012;7(04):1207–20. https ://doi.org/10.2118/14146 2-PA. and applications in the petroleum industry, vol. 242. Washington: Liu Y, Grigg R, Bai B. Salinity, pH, and surfactant concentration American Chemical Society; 1994. p. 3–45. effects on CO -foam. In: SPE international symposium on oilfield Taber J, Martin F, Seright R. EOR screening criteria revisited—Part chemistry, 2–4 February, The Woodlands, Texas; 2005. https :// 1: introduction to screening criteria and enhanced recovery doi.org/10.2118/93095 -MS. field projects. SPE Res Eng. 1997;12(03):189–98. https ://doi. Mannhardt K, Novosad J, Schramm L. Comparative evaluation of foam org/10.2118/35385 -PA. stability to oil. SPE Res Eval Eng. 2000;3(01):23–34. https://doi. Talley L. Hydrolytic stability of alkylethoxy sulfates. SPE Res Eng. org/10.2118/60686 -PA. 1988;3(01):235–42. https ://doi.org/10.2118/14912 -PA. 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Petroleum Science Springer Journals

Gas/water foams stabilized with a newly developed anionic surfactant for gas mobility control applications

Download PDF
12 pages

Loading next page...
 
/lp/springer-journals/gas-water-foams-stabilized-with-a-newly-developed-anionic-surfactant-g3xGc2jqCB

References (31)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2020
ISSN
1672-5107
eISSN
1995-8226
DOI
10.1007/s12182-020-00437-x
Publisher site
See Article on Publisher Site

Abstract

Carbon dioxide (CO ) flooding is one of the most globally used EOR processes to enhance oil recovery. However, the low gas viscosity and density result in gas channeling and gravity override which lead to poor sweep ec ffi iency. Foam application for mobility control is a promising technology to increase the gas viscosity, lower the mobility and improve the sweep effi- ciency in the reservoir. Foam is generated in the reservoir by co-injection of surfactant solutions and gas. Although there are many surfactants that can be used for such purpose, their performance with supercritical CO (ScCO ) is weak causing poor 2 2 or loss of mobility control. This experimental study evaluates a newly developed surfactant (CNF) that was introduced for ScCO mobility control in comparison with a common foaming agent, anionic alpha olefin sulfonate (AOS) surfactant. Experimental work was divided into three stages: foam static tests, interfacial tension measurements, and foam dynamic tests. Both surfactants were investigated at different conditions. In general, results show that both surfactants are good foaming agents to reduce the mobility of ScCO with better performance of CNF surfactant. Shaking tests in the presence of crude oil show that the foam life for CNF extends to more than 24 h but less than that for AOS. Moreover, CNF features lower critical micelle concentration (CMC), higher adsorption, and smaller area/molecule at the liquid–air interface. Furthermore, entering, spreading, and bridging coefficients indicate that CNF surfactant produces very stable foam with light crude oil in both deionized and saline water, whereas AOS was stable only in deionized water. At all conditions for mobility reduction evaluation, CNF exhibits stronger flow resistance, higher foam viscosity, and higher mobility reduction factor than that of AOS surfactant. In addition, CNF and ScCO simultaneous injection produced 8.83% higher oil recovery than that of the baseline experiment and 7.87% higher than that of AOS. Pressure drop profiles for foam flooding using CNF was slightly higher than that of AOS indicating that CNF is better in terms of foam–oil tolerance which resulted in higher oil recovery. Keywords Supercritical CO foam · Foam mobility control · Foam flooding · Enhanced oil recovery (EOR) · Foam assisting CO EOR 1 Introduction It is estimated that two-thirds of the original oil in place (OOIP) are left underground after the primary and second- Edited by Yan-Hua Sun ary oil recovery processes (Green and Willhite 1998). For tertiary recovery, many enhanced oil recovery (EOR) meth- * Mohammed A. Almobarky ods can be used to extract more oil from reservoirs. Among mmobarky@ksu.edu.sa these EOR methods, CO injection is one the most used pro- * Zuhair AlYousif cesses globally (Taber et al. 1997). However, gas injection Zuhair.Yousif@aramco.com processes face many challenges such as gas channeling and Department of Petroleum and Natural Gas Engineering, gravity override that lead to poor sweep efficiency (Healy College of Engineering, King Saud University, Riyadh, et al. 1994). Many techniques have been applied to enhance Kingdom of Saudi Arabia the sweep efficiency such as water alternating gas (WAG), Saudi Aramco, Dhahran, Kingdom of Saudi Arabia polymer, and foam. Foam is a promising technology that Harold Vance Petroleum Engineering Department, Texas can be used to reduce the mobility of the injected gas by A&M University, College Station TX, USA Vol.:(0123456789) 1 3 1026 Petroleum Science (2020) 17:1025–1036 increasing its viscosity (Enick et al. 2012) and diverting the provided better displacement efficiency, and resulted in flow toward lower permeability zones where the remaining higher oil recovery. oil exists (Fried 1961). The aim of this study is to evaluate the ability of a newly Surfactants are the main component in a foam sys- developed anionic surfactant (CNF) to control the mobil- tem. They facilitate the foam generation by reducing the ity of Sc CO . Moreover, the performance of CNF is com- gas–water interfacial tension and adsorb at the interfaces pared with C AOS (anionic surfactant) which is one of 14–16 to make the foam with the required stability by stabilizing the most widely used in literature with C O in gaseous and the thin films between bubbles (Schramm 2000). Thus, the supercritical states. This newly developed surfactant and the surfactant screening is the first step toward a successful foam major challenges for its utilization with ScCO may provide project (Boeije et al. 2017). In foam applications, in general, more opportunities for foam applications in foam-assisting and particularly in C O EOR, the surfactant structure is a miscible CO EOR projects. 2 2 significant factor that affects the efficiency in every aspect of the process: gas viscosity, mobility control, and EOR (Adkins et al. 2010). These effects are related to different 2 Experimental materials interactions of surfactants and C O (Adkins et al. 2010). Moreover, the presence of supercritical C O (ScCO ) results 2 2 Table  1 shows the general properties of both surfactants in low pH acidic environment where some types of sur- used in experimental work. Surfactants were diluted with factants hydrolyze and lose their interfacial activity such as deionized (DI) water (ASTM, type II) provided by LabChem sulfates (Talley 1988). Inc. Moreover, tests were conducted at 0.5 wt% surfactant Alpha olefin sulfonate (AOS) is hydrolytically and ther - concentration. Besides deionized water, the salinity effect mally stable, and soluble at low to medium hard water (Por- was investigated using brine solutions (NaCl solutions) at ter 1994). Farajzadeh et al. (2010) experimentally inves- 10,000, 20,000, and 30,000 ppm (NaCl was purchased from tigated the use of AOS for mobility control and EOR in Cole-Parmer). The crude oil used in this study was from miscible and immiscible flooding with the aid of CT scanner North Burbank Unit, OK, USA (NBU). It is light crude oil for simultaneous monitoring of the flooding process. They with API gravity of 33.7° and viscosity of 8 cP at room reported 19% more oil recovery with ScC O than that of temperature of 23 °C, and 39.5° API and 3.27 cP at 50 °C the immiscible CO flooding. However, no sharp front was which is the reservoir temperature. The glass-bead pack was observed with the use of ScC O . They attributed this to the made using glass beads with a specific gravity of 2.5 and poor foam stability with oil. Haugen et al. (2012) experimen- a diameter of 100 µm, which were purchased from Potters tally used AOS for mobility control and EOR in fractured oil Industries LLC. wet and water wet cores and reported that the pre-generated foam is better than in situ foam generation in terms of mobil- ity reduction and oil recovery. They attributed the results to the poor foam–oil tolerance. Li et al. (2012) used AOS in 3 Methodology surfactant-alternating-gas (SAG) injection mode for foam generation using N . Their experiments were conducted on The experimental work was divided into three stages: static a two-dimensional sand pack with 19–1 permeability con- foam tests, interfacial tension measurements, and dynamic trast. They attributed the poor sweep efficiency to the weak foam tests. The dynamic foam tests were divided into three foam stability in the presence of crude oil. They suggested sections: mobility reduction evaluation in the high-permea- that enhancing the foam–oil tolerance could provide higher bility glass-bead pack, mobility reduction evaluation in low- oil recovery because this may enhance the sweep efficiency. permeability Bentheimer sandstone cores, and core flooding Indeed, mixing the surfactant with a foam booster CTAB experiments. The surfactant concentrations were constant zwitterionic surfactant improved the foam–oil tolerance, at 0.5  wt%, diluted with deionized water, and prepared Table 1 Properties of the surfactants Surfactant Form Chemical family pH Density, g/mL Charge Flash point, °C Carbon chain length CNF Liquid Alpha olefin sulfonate, isopropyl 7.73 1.07 Anionic > 93.3 – alcohol, and citrus terpenes AOS Liquid Alpha olefin sulfonate 8.20 1.06 Anionic > 94.0 14–16 1 3 Petroleum Science (2020) 17:1025–1036 1027 with NaCl solutions of three salinities: 10,000, 20,000, and S =  −( +  ) a/w o/w o/g (2) 30,000 ppm. 2 2 2 B =  +  − (3) a/w o/w o/g 3.1 Static foam tests Foam was generated by shaking 3 mL of surfactant solu- a/w L = (4) tions in 13 × 100-mm (9-mL) Pyrex glass test tubes. Care has o/w been taken to perform 10–15 gentle and uniform shakings where  ,  ,  are the air–water, oil–water, and oil–gas for all samples. Samples were prepared at 0.5 wt% concen- a/w o/w o/g interfacial tensions, respectively. tration in deionized water, 10,000, 20,000, and 30,000 NaCl solutions. After the foam has been generated with shaking inside the test tube, the foam columns were monitored by 3.3 Dynamic foam tests taking images at different times. Then, the foam column lengths were measured from images using ImageJ software. These experiments were designed for mobility reduction The foaming ability was investigated using the initial foam evaluation and oil recovery investigation by conducting core column length (h ), and the foam stability was measured by flood experiments. fi the foam half-life (FHL), t , which is the time at which the 1/2 foam column loses half of the initial foam column length 3.3.1 Mobility reduction evaluation h . The samples were prepared for static tests without oil fi in the high‑permeability glass‑bead pack and stirred for about 12 h before testing. For static tests with crude oil, the samples were prepared at 0.5 wt% surfactant These experiments were conducted at a high shear rate of −1 −1 concentration and stirred for 12 h, and the surfactant solution 317 s and a low shear rate of 9.51 s . Furthermore, three was placed in 9-mL test tubes above which the crude oil was injection qualities 50%, 70%, and 90% were used. All experi- simply poured. Then, the sample was shaken immediately. ments were conducted at 1800 psi and 50 °C to ensure the supercritical conditions of C O . The foam was generated 3.2 Interfacial tension measurements by simultaneously injecting the surfactant solution and supercritical CO (ScCO ) through the glass-bead pack. 2 2 Air–water surface tension measurements were conducted at The pressure drop was measured using two sets of pressure different surfactant concentrations in DI water using a Data- transducers: 500-psi for high range and 50-psi for low range. physics OCA 15 Pro IFT instrument, pendant drop method. The pressure drop data were collected using a data acqui- The surface measurements were used for critical micelle sition system. The onset of a strong foam generation was concentration (CMC) determination and interfacial activity recognized as a rapid increase in pressure drop according predictions for both surfactants. to Dicksen et al. (2002). The flow continued with monitor - The air–water surface tension versus the logarithmic val- ing the pressure drop until the steady-state pressure drop ues of the surfactant concentrations below the CMC is a across the glass-bead pack was reached. Then, the steady- linear relationship. The slope of this straight line can be used state pressure data were averaged and used to calculate the to interpret the interfacial activities: adsorption and area/ mobility, foam effective viscosity, and mobility reduction molecule at the interface. According to the Gibbs adsorption factor (MRF) using the following equations. equation, the higher the slope is, the higher the adsorption at ql the air–water interface. Furthermore, the higher the adsorp- = = (5) AΔP tion of surfactant molecules at the interface results in smaller area/molecule and stronger packing that induces higher foam stability (Rosen and Kunjappu 2004). (6) eff The interfacial tension measurements, also, can be also used to investigate the foam–oil tolerance by calculating the entering coefficient (E ) (Robinson and Woods 1948), spread- ΔP foam f = (7) ing coefficient (S ) (Harkins 1941), bridging coefficient (B ) mr ΔP baseline (Denkov 2004), and the lamellae number (L) (Schramm and Novosad 1990) using Eqs. (1), (2), (3), and (4). where  is the mobility, k is the permeability,  is the vis- cosity, q is the flow rate, l is the length of the porous media, E =  +  − a/w o/w o/g (1) A is the cross-sectional area of the glass-bead pack, ∆P is the pressure drop across the porous media, μ is the foam eff effective viscosity, and f is the mobility reduction factor. mr 1 3 1028 Petroleum Science (2020) 17:1025–1036 Table 2 shows the dimensions and petrophysical proper- to ensure 100% core saturation. Although the XRD tests ties of the glass-bead pack which was filled with 100 µm for these rocks show that their composition is 100% quartz, glass beads. Figure 1 shows the experimental setup for the 1 PV of the surfactant solution was injected into the core at −1 mobility reduction evaluation. 5 ft/day (~ 9 s shear rate) to mitigate the effect of surfactant The baseline experiment, in which ScC O was injected adsorption on the rock surfaces. After that, the foam was without surfactant, was also conducted for comparison pur- applied by simultaneously injecting the surfactant solution poses. Detailed experimental conditions are described in and ScCO or N gas at 5 ft/day. The foam injection was 2 2 Sect. 4.4. continued until the steady pressure drop across the core was reached. The recorded steady-state pressure drop data were 3.3.2 Mobility reduction evaluation in low‑permeability averaged and used to calculate the mobility, effective viscos- Bentheimer sandstone cores ity of foam, and MRF using Eqs. 5, 6, and 7. Figure 1 above shows a schematic diagram of the exper- All tests in this section were conducted on homogeneous imental setup for mobility evaluation in sandstone cores. Bentheimer sandstone cores (diameter, 1 in.; length, 12 in.). Properties of Bentheimer sandstone cores and experimental The core was left in an oven overnight for drying. Then, conditions for all runs are described in Sect. 4.5. Moreo- it was mounted in the core holder and 500 psi overburden ver, the last two runs listed (runs 9 and 10) are baseline pressure was applied. After that, the air was removed from experiments conducted using N and ScCO injection for 2 2 the core using a vacuum pump followed by saturating the comparison purposes, respectively. In all runs, the salinity core with 10,000 ppm NaCl brine at which its pore volume of the surfactant solution or water injected is 10,000 ppm. (PV) and porosity can be measured. The system was then pressurized using the back-pressure regulator, as shown in 3.3.3 Core flooding experiments Fig. 1. After the system was pressurized and the tempera- ture was maintained at 50 °C, the absolute permeability was The core flooding experiments were also conducted on measured by obtaining the pressure drop at different flow homogeneous Bentheimer sandstone cores with a diameter rates using Darcy’s law. Then, 5–6 PV of brine were injected of 1″ and a length of 12″. The core was left in an oven over- night for drying. Then, it was vacuumed and saturated with 10,000 ppm brine at which the pore volume and porosity were measured. After that, 5–6 PV of 10,000 ppm brine Table 2 Dimensions and petrophysical properties of the glass-bead pack were injected into the core at 5 ft/day to ensure that the core sample was 100% water-saturated. Then, the absolute per- Length, in. Diameter, Pore vol- Porosity, % Permeabil- meability was calculated using Darcy’s law. The setup was in. ume, mL ity, D pressurized to 1450 psi, keeping the overburden pressure 13 0.18 1.625 30 17.1 500 psi higher than the test pressure, and at test temperature Overburden pressure Oven Core holder Glass bead pack BPR Graduated Pressure transducers Pressure transducers cylinder Data acquisition system Pumps Fig. 1 Schematic of an experimental setup for mobility reduction evaluation and core flooding 1 3 Water Surfactant Crude oil Gas Petroleum Science (2020) 17:1025–1036 1029 of 50 °C. Once the pressure and temperature were stable, CNF AOS crude oil was injected at a flow rate of 5 ft/day until no more 6.4 water was observed in the eu ffl ent. The water produced from 5.6 the core sample was collected in a graduated cylinder for the 4.4 4.3 4.0 3.9 original oil in place (OOIP) estimation. Then, water flooding 4 3.5 2.8 was applied by injecting 5 PV of 10,000 ppm brine at a flow rate of 5 ft/day until no more oil was observed in the efflu- 2 ent. The high amount of water injection was to ensure that the optimum oil recovery by waterflooding was achieved, 0 (DI water) 10,000 20,000 30,000 no more oil was produced by water injection, and the end NaCl salinity, ppm effects would be diminished. Then, the second stage was to inject 1–1.5 PV of the surfactant solution (surfactant pre- Fig. 2 Foam half-lives for both surfactants (0.5  wt%) in deionized flush) to mitigate the surfactant adsorption on the rock. After water, 10,000, 20,000, and 30,000 ppm NaCl solutions that, 5 PV of simultaneous injection of the surfactant solu- tion and ScCO was applied for 24 h at a flow rate of 5 ft/day. Moreover, CNF surfactant is slightly better than AOS at The pressure drop was recorded for the three oil recovery stages. Figure 1 shows the schematic diagram of the setup any salinity (0–30,000 ppm). For both surfactants, the sta- bility of foam prepared with deionized water is the best. In for the core flooding experiments. One baseline experiment was conducted by injecting ScCO only for comparison addition, both surfactants exhibit an enhancement of foam stability at the salinity of 30,000 ppm. Liu et al. (2005) purposes with oil recovery of CNF and AOS foam floods. Table 3 shows the properties for the Bentheimer sandstone reported that the foam stability with C O decreased with increasing salinity up to 2 wt%. Then, a further increase samples used to conduct the core flooding experiments and the experimental conditions. in salinity enhanced the foam stability. However, the foam stability plateaued shortly with further increase in salinity. The foam stability with crude oil was impressive for both surfactants. The foam half-life for CNF lasted for 4 Results and discussion more than 24 h and 12–18 h for AOS. Figure  3a shows images for CNF and Fig.  3b for AOS. The images were 4.1 Static foam test results taken after 24 h for CNF foams and after 18 h for AOS foams. Each image shows samples in DI water, 10,000 and Foaming ability (or foamability) was investigated by meas- uring the initial foam heights h for both surfactants from 20,000 ppm salinity left to right. Clearly, both surfactants fi produced stable foam with crude oil, but CNF foam was the shaking tests. Both surfactants gave almost the same ini- tial foam column length h . Therefore, both CNF and AOS significantly stronger than that of AOS in the presence of fi crude oil. are good foaming agents in terms of foaming ability. This also indicates the efficiency of both surfactants to reduce the However, shaking involves eventually very high shear rates which provide high energy for any surfactant to give air–liquid surface tension. The surface tension at the air–liq- uid interface will be discussed shortly. its optimum performance as foaming agent regardless of how the shaking was performed. Therefore, it is difficult Figure 2 shows the foam half-life for both surfactants at different salinities of 0 (deionized water), 10,000, 20,000, to recognize the differences in foaming ability and foam stability as well. Therefore, combining the shaking test and 30,000  ppm. Both surfactants provided good foam stability. As observed, the foam stability decreases as the observations with the interfacial tension measurements is next. salinity increases at the salinity of 0–20,000 ppm, which is attributed to a reduction in repulsive forces between the surfactant molecules due to the addition of salts. Table 3 Petrophysical properties of the sandstone cores and the experimental conditions in core flooding experiments Run No. Surfactant Core length, Core diameter, Core pore vol- Core porosity, % Core permeabil- Experiment type in. in. ume, mL ity, D 1 – 12 1 33.52 21.71 1.87 Baseline 2 AOS 12 1 34.74 22.50 1.71 Foam flood 3 CNF 12 1 33.74 21.85 1.91 Foam flood 1 3 Foam half-life, t 1/2 1030 Petroleum Science (2020) 17:1025–1036 with a decrease in the surfactant concentration and no foam was generated at surfactant concentration below its CMC. Therefore, the lower the CMC of the surfactant is the bet- ter from different perspectives such as lowering the cost of the project due to the use of low surfactant concentrations without compromising the foaming efficiency. According to Gibbs surface adsorption equation (Eq. 8), the higher the slope is, the higher the adsorption of the sur- factant at the liquid–air interface, and consequently, the better the foamability and foam stability (Rosen and Kun- jappu 2004). This is because the higher the adsorption at the air–liquid interface is, the smaller the area/molecule and the stronger the packing of the molecules at the interface (Rosen and Kunjappu 2004). =− (8) RT d ln C Fig. 3 a Images of CNF foam after 24 h and b images of AOS foam where  is the surfactant adsorption at the air–liquid inter- after 18 h, both foam samples at 0.5 wt% surfactant concentration and at a salinity of 0 (deionized water), 10,000 and 20,000 ppm from left face, R is the gas constant, T is temperature, γ is the surface to right tension, and C is the surfactant concentration. Furthermore, for concentrations above 0.007  wt% in Fig. 4, CNF is able to reduce the interfacial tension lower 4.2 Interfacial tensions than that of AOS which also indicates that CNF is predicted to perform better than AOS in foam generation. Table  4 As mentioned earlier, surface tensions were measured for shows the air–water and oil–water interfacial tensions for different concentration samples prepared with deionized both 0.5 wt% surfactant solutions at different salinities. water and the measured results are shown in Fig.  4. The critical micelle concentrations (CMCs) for CNF and AOS 4.3 Foam stability in the presence of crude oil are 0.011 wt% and 0.028 wt%, respectively. The surfactant concentration for foam applications is recommended above As shown in Table 4, the σ for both surfactants dropped o/w the CMC (Nikolov et al. 1986). Using the surfactant con- with the addition of salts (from deionized water to centration above the CMC provides the best foam stability, 10,000 ppm salinity), but no influence was observed on σ o/w whereas foam has less opportunity to be of good stability with increasing the salinity. This shows that both surfactants when the surfactant concentration is below the CMC (Rafati would probably perform efficiently in the presence of oil, but et al. 2012). The CMC defines the foaming efficiency of the CNF had higher σ which suggests that it is better in terms o/w surfactants. The lower the CMC is, the higher the foaming of foam–oil tolerance. efficiency (Rosen and Kunjappu 2004). Moreover, Man- Table  5 shows the entering, spreading, bridging coef- nhardt et al. (2000) found that the foaming ability decreased ficients, and lamellae number, respectively. The entering coefficient gives positive values for all surfactant solutions, CNF AOS Table 4 Interfacial tensions σ and σ for AOS and CNF (0.5 wt%) a/w o/w at 23 °C Salinity of surfactant solu- IFT between gas IFT between gas tion (water phase), ppm (or oil) and AOS (or oil) and CNF solution, mN/m solution, mN/m σ σ σ σ a/w o/w a/w o/w 0.001 0.010 0.100 1.000 0 (deionized water) 32.50 1.40 30.70 5.88 logC, wt% 10,000 32.30 0.52 31.00 3.94 20,000 32.10 0.44 31.25 3.51 Fig. 4 Interfacial measurements for AOS and CNF (C is the sur- 30,000 32.15 0.38 31.11 3.11 factant concentration, wt%) 1 3 Interfacial tension, mN/m Petroleum Science (2020) 17:1025–1036 1031 Table 5 Entering, spreading, bridging coefficients, and lamellae number at 0.5 wt% surfactant concentration at 23 °C Salinity of surfactant solution (water Entering coefficient Spreading coefficient Bridging coefficient Lamellae number phase), ppm AOS CNF AOS CNF AOS CNF AOS CNF 0 (Prepared with deionized water) 2.40 5.08 − 0.40 − 6.68 65.96 − 15.19 3.48 0.78 10,000 1.32 3.44 0.28 − 4.44 51.31 − 15.73 9.32 1.18 20,000 1.04 3.26 0.16 − 3.76 38.35 − 3.37 10.94 1.34 30,000 1.03 2.72 0.27 − 3.50 41.52 − 14.75 12.69 1.50 which clarifies that oil will enter the lamellae. However, 4.4 Mobility reduction evaluation the spreading coefficient implies that oil will spread at the in the high‑permeability glass‑bead pack air–water interface to destabilize AOS foams in saline solu- tions only. The spreading coefficient for CNF is negative for Table 6 shows the results for 16 runs on the high-permea- all samples. Moreover, the bridging coefficient gives nega- bility glass-bead pack at high and low shear rates for both tive values for CNF at all conditions. These observations surfactants with ScC O foam. Moreover, one baseline exper- imply that CNF is going produce very stable foam with oil. iment was conducted with ScC O injection only and resulted However, AOS produces stable foam with oil when the sur- in a very low steady-state pressure drop (0.04 psi) which factant prepared with deionized water. Furthermore, lamel- corresponds with 0.03 cP effective viscosity. The average lae number values confirm the same conclusions. The CNF of the recorded steady-state pressure drop data were used to foam is stable in deionized water and semi-stable in saline calculate the mobility, foam effective viscosity, and mobil- solutions, whereas AOS is semi-stable in deionized water ity reduction factor. Experimental conditions and results for and unstable in saline solutions. these tests are listed in Table 6. Figure 5 shows the effect of three injection qualities on the viscosities of both surfactants at the high shear rate in deionized water. The AOS foam viscosity decreased but the CNF foam viscosity increased as the injection quality decreased. The opposite behavior of these foaming agents Table 6 Experimental conditions and results for AOS and CNF with CO in glass-bead packs a −1 Run Surfactant Salinity, ppm Injection Flow rate Shear rate, s Steady-state pres- Mobility Effective Mobility No. quality, % q, mL/min sure drop ΔP , λ, mD/cP viscosity μ , reduction fac- ss eff psi cP tor f mr 1 AOS 0 90 0.5 317 208 118 145 – 2 AOS 10,000 90 0.5 317 164 150 114 4100 3 AOS 20,000 90 0.5 317 161 156 109 – 4 AOS 30,000 90 0.5 317 216 459 37 – 5 AOS 0 70 0.5 317 175 142 121 – 6 AOS 10,000 70 0.5 317 99 249 69 2475 7 AOS 20,000 70 0.5 317 93 265 64 – 8 AOS 30,000 70 0.5 317 91 273 63 – 9 AOS 0 50 0.5 317 122 202 85 – 10 AOS 10,000 90 0.015 9.51 9.97 74 230 250 11 CNF 0 90 0.5 317 162 153 112 – 12 CNF 0 70 0.5 317 207 119 143 – 13 CNF 0 50 0.5 317 225 109 154 – 14 CNF 10,000 90 0.5 317 178 138 123 4450 15 CNF 10,000 90 0.015 9.51 16 46 371 400 16 – 10,000 – 0.5 317 0.04 – 0.03 – The surfactant concentrations were 0.5 wt% in runs 1–15. No surfactant was used in the baseline experiment (run 16) 1 3 1032 Petroleum Science (2020) 17:1025–1036 CNF AOS CNF AOS 9.97 ΔPss, psi λ, mD/cP μeff, cP fmr 90 70 50 Quality, % Fig. 7 Comparison of experimental results at low shear rate between AOS and CNF foams (0.5 wt%, 10,000 ppm salinity) with ScCO at Fig. 5 Effect of injection quality on the effective viscosities of AOS 90% injection quality and CNF (0.5 wt% in deionized water) foams in the glass-bead pack −1 at a high shear rate of 317 s lower mobility, higher foam viscosity, and higher mobility can be attributed to many reasons that were not covered in reduction factor. this study. In fact, the relationship of the foam viscosity with injection quality is still controversial. In an experimental study conducted by Marsden and Khan (1966), they found 4.5 Mobility reduction evaluation that the higher the injection quality is, the higher the foam in low‑permeability Bentheimer sandstone viscosity. Lee and Heller (1990) experimentally concluded cores that the increase in injection quality decreased the viscosity. The foam viscosity depends on flow rate, permeability of Two baseline runs were conducted using N injection at the porous media, and foam texture (Hirasaki and Lawson 850 psi and ScCO injection at 1800 psi for comparison. 1985). All experimental conditions and results are listed in Table 7. −1 The effect of shear rate (317 and 9.51 s ) on the effective The effect of permeability is shown in Fig.  8 which com- viscosities of both AOS and CNF foams at 90% injection pares the foam viscosities for AOS and CNF with ScCO at −1 quality and 10,000 ppm salinity are shown in Fig. 6. The 90% injection quality at a shear rate of 9 s . The perme- higher the shear rate resulted in the lower viscosity because abilities of the cores and the glass-bead pack are 1.62–1.80 of the shear thinning nature of the foam. As shown in Fig. 6, Darcy (Table 7) and 17.1 Darcy (Table 2), respectively. CNF the viscosities of CNF foam at both shear rates are higher is repeatedly proved to be better than AOS by generating than that of the AOS foam. higher foam viscosity in both cases at high and low-perme- Figure 7 also shows all results for CNF and AOS foams ability porous media. The results in Fig. 8 are in agreement −1 (0.5 wt%, 10,000 ppm salinity) at a low shear rate of 9.51 s with the fact that foam favors the higher permeability porous and 90% injection quality. Again, the CNF foam proves pow- media. The viscosities of surfactant foams in the glass-bead erful performance with higher steady-state pressure drop, pack are extremely higher than that in sandstone cores. The effect of injection quality at low-permeability sand- stone cores was investigated using two injection qualities 90% and 70% for both surfactants with ScC O at a shear 371 −1 CNF AOS rate of 9.51 s . The pressure drop curves for both surfactant foams are shown in Fig.  9 for 90% injection quality and Fig. 10 for 70% injection quality. According to the pressure profiles, the CNF foam is stronger at 90% injection quality, whereas the AOS foam is stronger at 70% injection quality. Therefore, the results suggest that CNF is better at more realistic conditions (i.e., low shear rate and sandstone reser- voirs), CNF co-injection with ScC O provides higher foam 9.51 317 viscosity at higher injection qualities, while AOS requires -1 low injection qualities for better performance. This also indi- Shear rate, s cates that using CNF with ScC O for foam mobility con- trol will eventually reduce the cost as the amount of liquid Fig. 6 Effect of shear rate on the effective viscosities of AOS and decreases as the injection quality increases. CNF foams at 0.5  wt% surfactant concentration and 90% injection quality 1 3 Effective viscosity, cP Effective viscosity, cP Petroleum Science (2020) 17:1025–1036 1033 CNF AOS 5.65 4.13 17.1 1.7 Permeability, D Fig. 8 Effect of permeability on effective viscosity of AOS and CNF foams (0.5  wt%, 10,000  ppm salinity) with ScC O at 90% injection quality 1.0 90% quality AOS CNF 0.8 0.6 0.4 0.2 0 123456 7 Injection volume, PV Fig. 9 Pressure drop curves for AOS and CNF foams (with ScCO ) at 90% injection quality in sandstone cores 70% quality AOS CNF 0246 810 Injection volume, PV Fig. 10 Pressure drop curves for AOS and CNF foams (with ScCO ) at 70% injection quality in sandstone cores The mobility control with N gas was also tested for both surfactants. The pressure profiles for AOS and CNF are shown in Fig. 11. These experiments were conducted on sandstone core samples at two velocities of 5 and 10 ft/ −1 day (9 and 18 s ) with an injection quality of 90%. In these experiments, the surfactant solutions used had a concentra- tion of 0.5 wt%, which were prepared with 10,000 ppm NaCl 1 3 Table 7 Experimental conditions and results for AOS and CNF in sandstone cores −1 Run No. Surfactant Gas Pressure P, psi Velocity, Shear rate, s Injection Pore vol- Permeabil- Steady-state pres- Mobility λ, Foam effective Mobility ft/day quality, % ume, mL ity k, D sure drop ΔP , psi mD/cP viscosity, cP reduction fac- ss tor f mr 1 AOS ScCO 1800 5 8.77 90 31.24 1.70 0.40 411 4.13 1.67 2 AOS ScCO 1800 5 9.30 70 33.74 1.58 25.42 6.79 232.8 105.9 3 CNF ScCO 1800 5 9.10 90 34.74 1.70 0.60 301 5.65 2.5 4 CNF ScCO 1800 5 8.92 70 34.74 1.78 2.00 89 20 8.3 5 AOS N 850 5 9.15 90 34.24 1.65 1.35 130 12.7 3.4 6 AOS N 850 10 18.31 90 34.24 1.65 3.46 100 18 8.6 7 CNF N 850 5 8.76 90 34.24 1.80 12.89 13.61 121.23 32.2 8 CNF N 850 10 17.53 90 34.24 1.80 34.4 10.20 176.5 86 9 – N 850 5 9.05 – 33.74 1.62 0.40 – 10 – ScCO 1800 5 8.83 – 32.74 1.70 0.24 680 0.04 – Effective viscosity, cP Pressure dop, psi Pressure dop, psi 1034 Petroleum Science (2020) 17:1025–1036 50 100 CNF 5 ft/day CNF 10 ft/day AOS 5 ft/day AOS 10 ft/day 40 80 ScCO2 flooding 30 60 Water flooding 20 40 10 20 0 0 05 10 15 20 0123456789 10 Injection volume, PV Injection volume, PV Fig. 11 Pressure drop curves for AOS and CNF foams with N at dif- Fig. 13 Oil recovery of the baseline experiment ferent injection velocities (90% injection quality, 850  psi injection pressure) 100 10 Surfactant 80 Water flooding AOS foam flooding 8 pre-flush CNF AOS 176.5 Oil recovery 60 6 121.3 40 4 20 2 Pressure drop 0 0 18.0 12.7 0 246 81012 Injection volume, PV 5 10 Velocity, ft/day Fig. 14 Oil recovery and pressure drop curves across the sandstone core for waterflooding, AOS surfactant (0.5  wt% and 10,000  ppm Fig. 12 Foam effective viscosity in sandstone cores for CNF and salinity) pre-flush, and AOS surfactant with ScCO foam flood at 90% AOS foams with N gas at different injection velocities (90% injec- injection quality tion quality, 850 psi injection pressure) solution. Each surfactant solution was injected simultane- 4.6 Core flooding experiments ously with N gas at 5 ft/day until the steady-state pressure drop was reached, then the velocity was raised to 10 ft/day The oil recovery from the baseline experiment is shown as until the steady-state pressure drop was reached for the new a function of pore volume injected in Fig. 13. The ultimate velocity. From Fig. 11, the CNF foam appeared always better oil recovery of water flooding is 39% of OOIP. The water in terms of flow resistance at both velocities than the AOS flooding was followed by 5–6 PV of continuous ScCO foam. Moreover, Fig. 12 compares the foam viscosities for injection which resulted in 27.54% OOIP more oil recov- AOS and CNF at both velocities. Surprisingly, the viscosity ery. The total oil recovery from the baseline experiment increases as the velocity (i.e., shear rate) increases for both is 66.54% of the OOIP. foams. Although this is shear thickening behavior, foam is Figure 14 shows the results for AOS as a foaming agent known of its non-Newtonian shear thinning nature (Sch- after water flooding. The same procedure was conducted. ramm and Wassmuth 1994). This behavior could be related This run started with injecting 4.56 PV water flooding, to the procedure of the experimental work. At 5 ft/day, the which resulted in an oil recovery of 35.42% of the OOIP. foam was very efficient to provide high flow resistance due Then, 1.62 PV of the AOS surfactant solution was injected to the gas blockage effect. Therefore, the gas relative perme- to reduce the surfactant adsorption on the rock surfaces. ability is already low, and increasing the velocity to 10 ft/ The surfactant pre-flush stage resulted in an oil recovery day at such conditions would promote the foam generation of 4.75% of the OOIP. The third stage was the foam flood because of the high shear rate in a blocked porous media. As with simultaneous injection of 5 PV of AOS-ScCO . The a result, foam contradicted its shear thinning nature by pro- AOS foam flood resulted in an additional oil recovery of viding higher viscosity at higher shear rates, lesson learned. 28.5%. Then, the total oil recovery was 68.67% of the 1 3 Effective viscosity, cP Pressure drop, psi Oil recovery, % Oil recovery, % Pressure drop, psi Petroleum Science (2020) 17:1025–1036 1035 of salts in terms of foam–oil tolerance. However, the lamellae number for AOS shows semi-stable in deion- Surfactant 80 8 Water flooding CNF foam flooding pre-flush ized water and unstable at all salinities. These results Oil recovery show that CNF foam is very stable with oil more than 60 6 AOS foam. (4) At 90% injection quality, the mobility reduction with 40 4 Pressure drop ScCO , and with N at lower pressures, CNF shows 2 2 20 2 higher foam viscosity and better mobility reduction than AOS. 0 0 (5) For oil recovery, AOS foam produced 1% more than 0246 810 12 the baseline experiment, whereas CNF foam produced Injection volume, PV almost 7.87% more than AOS foam, and 8.38% more than the baseline experiment. Fig. 15 Oil recovery and pressure drop curves across the sandstone (6) It is not recommended to investigate foam viscosity at core for waterflooding, CNF surfactant (0.5  wt% and 10,000  ppm more than one shear rate during one experiment. Test- salinity) pre-flush, and CNF surfactant with ScCO foam flood at 90% injection quality ing two shear rates in series in one experiment would provide erroneous results. In this study, measuring the foam viscosity at two shear rates in one experiment OOIP. The additional recovery by AOS foam accounted resulted in shear thickening foam behavior, while foam for 1% higher than that of baseline experiment. is non-Newtonian shear thinning in nature. CNF foam core flood results are shown in Fig.  15. Water flooding stage produced an oil recovery of 39.66% of the Open Access This article is licensed under a Creative Commons Attri- OOIP. Moreover, no oil production was observed during the bution 4.0 International License, which permits use, sharing, adapta- surfactant pre-flush stage in which 1.5 PV of CNF surfactant tion, distribution and reproduction in any medium or format, as long solution was injected to reduce the adsorption effect. The as you give appropriate credit to the original author(s) and the source, final stage, CNF-ScCO simultaneous injection or foam provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are flooding resulted in a recovery of 36.3% of the OOIP after included in the article’s Creative Commons licence, unless indicated water flooding. The amount of oil produced by the CNF otherwise in a credit line to the material. If material is not included in foam was 7.87% higher than that by the AOS foam and the article’s Creative Commons licence and your intended use is not 8.83% higher than the baseline experiment. The total recov- permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a ery for this core flood was 75.96%. copy of this licence, visit http://creativ ecommons .or g/licenses/b y/4.0/. 5 Conclusions References (1) Both AOS and CNF surfactants show good foaming ability and foam stability without oil in shaking tests. Adkins S, Chen X, Nguyen Q, Sanders A, Johnston K. Effect of Such behavior is not representative to the actual foam- branching on the interfacial properties of nonionic hydrocarbon ing ability of both surfactants because the shaking tests surfactants at the air–water and carbon dioxide–water inter- faces. J Colloid Interface Sci. 2010;346(2):455–63. https ://doi. are naturally involved high shear rate which enforces org/10.1016/j.jcis.2009.12.059. the surfactants to perform at their optimum abilities as Boeije C, Bennetzen M, Rossen W. A methodology for screening foaming agents. However, CNF was able to reduce the surfactants for foam enhanced oil recovery in an oil-wet res- air–water interfacial tension lower than that of AOS. ervoir. SPE Res Eval Eng. 2017;20(04):795–808. https ://doi. org/10.2118/18518 2-PA. (2) The newly developed surfactant, CNF, shows impres- Denkov N. Mechanisms of foam destruction by oil-based antifoams. sive foam oil tolerance than AOS. It does not reduce the Langmuir. 2004;20(22):9463–505. https ://doi.org/10.1021/la049 oil–water interfacial tension to low value as with AOS 676o. in saline solutions. This makes it a good surfactant for Dicksen T, Hirasaki G, Miller C. Conditions for foam generation in homogeneous porous media. In: SPE/DOE improved oil recov- the applications of foam for mobility control. ery symposium, 13–17 April, Tulsa, Oklahoma; 2002. https: //doi. (3) CNF provides negative values of entering, spread- org/10.2118/75176 -MS. ing, and bridging coefficients, whereas AOS provides Enick R, Olsen D, Ammer J, Schuller W. Mobility and conformance negative values in deionized water only. Moreover, the control for C O EOR via thickeners, foams, and gels—a literature review of 40 years of research and pilot tests. In: SPE improved oil lamellae number indicates that the CNF foam is stable in deionized water and semi-stable with the addition 1 3 Oil recovery, % Pressure drop, psi 1036 Petroleum Science (2020) 17:1025–1036 recovery symposium, 14–18 April, Tulsa, Oklahoma, USA; 2012. Marsden S, Khan S. The flow of foam through short porous media and https ://doi.org/10.2118/15412 2-MS. apparent viscosity measurements. SPE J. 1966;6(01):17–25. https Farajzadeh R, Andrianov A, Zitha P. Investigation of immiscible ://doi.org/10.2118/1319-PA. and miscible foam for enhancing oil recovery. Ind Eng Chem. Nikolov A, Wasan D, Huang D, Edwards D. The effect of oil on foam 2010;49(4):1910–9. https ://doi.org/10.1021/ie901 109d. stability: mechanisms and implications for oil displacement by Fried A. The foam-drive process for increasing the recovery of oil. foam in porous media. In: SPE annual technical conference and Washington, DC: U.S. Bureau of Mixes, Rep. Inv., 5866; 1961. exhibition, 5–8 October, New Orleans, Louisiana; 1986. https :// Green D, Willhite G. Enhanced oil recovery. Volume 6, SPE Textbook doi.org/10.2118/15443 -MS. Series. Richardson, Texas: SPE; 1998. Porter MR. Handbook of surfactants. London: Blackie Academic & Harkins W. A general thermodynamic theory of the spreading of liquids Professional; 1994. to form duplex films and of liquids or solids to form monolayers. Rafati R, Hamidi H, Idris A, Manan M. Application of sustainable J Chem Phys. 1941;9(7):552–68. https://doi.or g/10.1063/1.17509 foaming agents to control the mobility of carbon dioxide in 53. enhanced oil recovery. Egypt J Pet. 2012;21(2):155–63. https :// Haugen Å, Fernø M, Graue A, Bertin H. Experimental study of foam doi.org/10.1016/j.ejpe.2012.11.010. flow in fractured oil-wet limestone for enhanced oil recovery. SPE Robinson J, Woods W. A method of selecting foam inhibitors. J Soc Res Eval Eng. 2012;15(02):218–28. https ://doi.or g/10.2118/12976 Chem Ind. 1948;67(9):361–5. https: //doi.org/10.1002/jctb.50006 3-PA. 70908 . Healy R, Holstein E, Batycky J. Status of miscible flooding technology. Rosen M, Kunjappu J. Foaming and antifoaming by aqueous solutions In: 14th world petroleum congress, 29 May-1 June, Stavanger, of surfactants and interfacial phenomena. Hoboken: Wiley; 2004. Norway; 1994. p. 308–35. Hirasaki G, Lawson J. Mechanisms of foam flow in porous media: Schramm L. Surfactants: fundamentals and applications in the petro- apparent viscosity in smooth capillaries. SPE J. 1985;25(02):176– leum industry. Cambridge: Cambridge University Press; 2000. 90. https ://doi.org/10.2118/12129 -PA. p. 3–50. Lee H, Heller J. Laboratory measurements of CO -foam mobility. SPE Schramm L, Novosad J. Micro-visualization of foam interactions Res Eng. 1990;5(02):193–7. https ://doi.org/10.2118/17363 -PA. with a crude oil. Colloids Surf. 1990;46(1):21–43. https ://doi. Li R, Hirasaki G, Miller C, Masalmeh S. Wettability alteration and org/10.1016/0166-6622(90)80046 -7. foam mobility control in a layered, 2D heterogeneous sandpack. Schramm L, Wassmuth F. Foams: basic principles foams: fundamentals SPE J. 2012;7(04):1207–20. https ://doi.org/10.2118/14146 2-PA. and applications in the petroleum industry, vol. 242. Washington: Liu Y, Grigg R, Bai B. Salinity, pH, and surfactant concentration American Chemical Society; 1994. p. 3–45. effects on CO -foam. In: SPE international symposium on oilfield Taber J, Martin F, Seright R. EOR screening criteria revisited—Part chemistry, 2–4 February, The Woodlands, Texas; 2005. https :// 1: introduction to screening criteria and enhanced recovery doi.org/10.2118/93095 -MS. field projects. SPE Res Eng. 1997;12(03):189–98. https ://doi. Mannhardt K, Novosad J, Schramm L. Comparative evaluation of foam org/10.2118/35385 -PA. stability to oil. SPE Res Eval Eng. 2000;3(01):23–34. https://doi. Talley L. Hydrolytic stability of alkylethoxy sulfates. SPE Res Eng. org/10.2118/60686 -PA. 1988;3(01):235–42. https ://doi.org/10.2118/14912 -PA. 1 3

Journal

Petroleum Science – Springer Journals

Published: Aug 7, 2020

There are no references for this article.