Flow Behavior and Displacement Mechanisms of Nanoparticle Stabilized Foam Flooding for Enhanced Heavy Oil Recovery

Abstract

In this study, nanoparticle stabilized foam experiments were performed in bulk tests, micromodels, and sandpacks at elevated temperatures and pressures to investigate the flow behavior and displacement mechanisms for enhanced heavy oil recovery. The results from the bulk tests showed that the stability of the foam and oil in water (O/W) emulsion improved when silica nanoparticles (SiO₂) were added, compared with the anionic surfactant alone. Also, the SiO₂ nanoparticles increased the dilatational viscoelasticity of the gas-water interface, which is an important fluid property and mechanism for improving heavy oil recovery. The micromodel studies demonstrated that several gas bubbles and oil droplets were stably dispersed during the nanoparticle stabilized foam flooding. The gas bubbles and oil droplets plug pores through capture-plugging and bridge-plugging, thereby increasing the sweep efficiency. The trapped residual oil is gradually pushed to the pores by the elastic forces of bubbles. Subsequently, the residual oil is pulled into oil threads by the flowing gas bubbles. Then, a greater improvement in displacement efficiency is obtained. The sandpack tests showed that the tertiary oil recovery of nanoparticle stabilized foam flooding can reach about 27% using 0.5 wt % SiO₂ nanoparticles. The foam slug size of 0.3 pore volume (PV) and the gas liquid ratio (GLR) of 3:1 were found to be the optimum conditions in terms of heavy oil recovery by nanoparticle stabilized foam flooding in this study. A continuous nanoparticle dispersion and N₂ could be more effective compared with the cyclic injection pattern.

1. Introduction

Due to increasing petroleum demand, the exploitation of heavy oil reserves has increased significantly in recent years. The global heavy oil reserves are abundant and are estimated to be around 6.04 × 10¹¹ m³–9.86 × 10¹¹ m³ in total. However, due to high oil viscosity, the performance of the primary recovery is bad in heavy oil reservoirs [1,2,3].

For the heavy oil reservoirs with oil viscosities lower than 10,000 mPa·s, water flooding can be conducted as a secondary oil recovery method. However, because of the poor mobility ratio between heavy oil and water, water flooding does not have a satisfactory performance [4,5]. Due to this reason, even after water flooding, a significant amount of oil still remains in the reservoir. To recover the remaining oil after water flooding in heavy oil reservoirs, enhanced oil recovery (EOR) methods can be applied such as thermal recovery methods, surfactant flooding, polymer flooding, and foam flooding [6,7,8,9].

Because the viscosity of heavy oil is much higher than that of gas, the recovery performance of gas flooding is quite low. Previous studies have demonstrated that foam flooding can reduce the gas mobility and increase the sweep efficiency, thereby improving the heavy oil recovery [10,11,12]. The stability of foam is very important for the foam flooding [13,14]. Nanoparticles, such as silica, can improve the stability and recovery performance of foam flooding. Sun et al. [15] investigated the stability and the effect of foam stabilized with silica nanoparticles and Sodium Dodecyle Sulfate (SDS) on enhanced light oil recovery. Nguyen et al. [16] studied the flow behaviors of nanoparticle-stabilized CO₂ foam in improving light oil recovery at the pore and micromodel scales. The authors of the study found that the effectiveness of nanoparticle-stabilized CO₂ foam is comparable for representative light, medium, and heavy oils. All three oils showed substantial additional oil recovery and a potentially valuable reservoir homogenization effect. Singh and Mohanty [17] studied foam stabilization by in-situ surface activation of hydrophilic nanoparticles for subsurface applications. The corefloods in the sandstone cores in a light oil reservoir showed that immiscible foams with a surface-modified nanoparticle (SMNP) solution can recover significantly more oil than water flooding. By using computed tomography (X-ray CT) scanning techniques, Aminzadeh et al. [18] found that compared to the foam flooding without nanoparticles, the SiO₂ stabilized N₂ foam can improve sweep efficiency and reduce the gravity override.

While nanoparticle-stabilized foam flooding has extensively been investigated in light oil reservoirs, limited data are available for nanoparticle-stabilized foam flooding in improving the recovery of heavy oil. For the heavy oil reservoir, the mobility of heavy oil is much lower than water and gas. Improving the stability of foam is very important for the application of foam flooding in the heavy oil reservoir. Therefore, the flow behaviors of nanoparticle-stabilized foam flooding in heavy oil reservoirs are more complicated than that in light oil reservoirs. This paper provides the microscopic displacement mechanisms and recovery performance of nanoparticle-stabilized foam flooding for enhancing heavy oil recovery.

2. Experimental

2.1. Materials

The oil sample was collected from the Shengli Oilfield in China. The viscosity, density, saturates, aromates, resins, and asphaltenes (SARA) analyses results, and the acid number of the heavy oil are listed in

Table 1. Properties of the heavy oil sample.

Density at 60 °C (kg/m3)	Viscosity at 60 °C (mPa·s)	Saturate (wt %)	Aromatic (wt %)	Resin (wt %)	Asphaltenes (wt %)	Acid Number (mg KOH/g oil)
986.8	2240	39.19	38.49	19.58	2.35	2.71

.

N₂ gas used was provided by Tianyuan Co., Ltd., Qingdao, China and had a purity of 99.99%. A commercial surfactant Hengye-2 (HY-2) (anionic, provided by Shengli Oilfield, Dongying, China, with a purity of 33.5%) was used as the foaming agent. The brine was prepared by a NaCl solution with a salinity of 7500 mg/L. The partially hydrophobic SiO₂ nanoparticles (HDK, H15) were supplied by Wacker Chemie AG, Burghausen, Germany. The average diameter and surface area of the hydrophobic SiO₂ nanoparticles were about 14 nm and 200 m²/g, respectively. Distilled water was used to make all the necessary solutions.

2.2. Nanoparticle-Surfactant Dispersion

A nanoparticle-surfactant dispersion was prepared by mixing 1.0 wt % nanoparticles and 0.5 wt % HY-2 surfactant. The mixture was stirred for 24 h and was then ultrasonicated for 1 h.

2.3. Preparation and Characterization of Foam

The foam was generated according to the Waring Blender method in a thermostat under different temperatures [19]. Using a blender, a 100 mL dispersion was stirred for 3 min at 7000 rpm to generate the foam. The foam was poured into a cylinder. The time taken to drain 50 mL liquid from the foam (half-life) was recorded.

The surface dilational viscoelasticity of the dispersion was measured by using a bubble/drop profile analysis tensiometer (Tracker-H, Teclis Instruments, Longessaigne, France). The principle of surface dilational viscoelasticity has been described in one of our previous studies [19]. The same setup was used to determine the surface rheological properties. After 3000 s of bubble formation, oscillations were performed with bubble area sinusoidal oscillations of relative amplitude δS/S₀ = 15%. The dilatational modulus versus the oscillation frequency ranged from 0.05 to 0.2 Hz. The temperature of the measuring cell was kept constant at 30 °C.

2.4. Emulsification Tests

Two kinds of emulsification tests were conducted in the present study. The first one was the bottle test, which was conducted in 50 mL bottles. Two bottles were used, which contained brine and the nanoparticle-surfactant dispersion. First, 40 mL of aqueous phase (brine or nanoparticle-surfactant dispersion) was placed in a test bottle. Then, using a syringe, 10 mL of oil was poured into the top of the brine or nanoparticle-surfactant dispersion slowly. The bottles were turned upside down to mix the oil and aqueous phase.

The second emulsification test was conducted to investigate the effect of nanoparticles on the emulsion stability. For these tests, 100 mL of aqueous phase (0.5 wt % HY-2 solution or nanoparticle-surfactant dispersion) was placed in the blender. Then, 20 mL of oil was added to the aqueous phase in the blender. To form the oil in water (O/W) emulsion, the aqueous phase with oil was stirred for 3 min at 500 rpm. Then the emulsion was poured into a sealed cylinder. The time taken to drain 50 mL of the aqueous phase from the emulsion (half-life) was recorded. The emulsification tests were conducted at room temperature.

2.5. Micromodel Setup

The micromodel test experimental setup is shown in Figure 1. An ISCO pump (model 100 DX) delivered crude oil, brine, and the nanoparticle-surfactant dispersion to the micromodel at a preset and constant flow rate. A diaphragm back pressure regulator (BPR) with a pressure accuracy of 0.001 MPa was used to control the back pressure. The high-pressure visual cell allowed an overburden pressure of up to 10.0 MPa, and a temperature up to 180 °C. The confinement fluid in the visual cell was water. The visual cell had two transparent windows: the upper one for viewing the micromodel, and the lower one for illuminating it. A heating muff outside of the visual cell was used to heat the cell to a certain high temperature. As shown in Figure 2, a quarter 5-spot glass-etched micromodel was placed in the high-pressure visual cell horizontally. The micromodel was made of two high pressure-resisting glasses. The lower plate had two holes for injecting and retrieving the fluids. The micromodel was constructed by etching a two-dimensional network of pores and throats using a photochemical method. The pore network used in this study was patterned based on the pore structure of a core obtained from the reservoir. The size of the network was 5 cm × 5 cm, and the depth and width of the microscopic channel were approximately 40 µm each. The porosity and permeability of the micromodel were about 0.45 and 2.5 μm², respectively. A Nikon Model L110 digital camera was used to record images within the micromodel. As shown in Figure 1b, a foam generator, which was a porous medium sealed in the flow channel, was used upstream of the micromodel. The foam generator was filled with the silica sand, with diameters of about 70 μm–100 μm. Then the foam was generated by passing the gas and surfactant-containing liquid simultaneously through the foam generator.

After the micromodel was vacuumed, it was saturated by the oil. The back pressure during the micromodel test was set to be 2.0 MPa. Then, the brine was introduced into the micromodel at 0.004 mL/min. After 1.0 pore volume (PV) of brine was injected, nanoparticle-stabilized foam flooding was conducted. The nanoparticle-surfactant dispersion and N₂ were injected simultaneously. The injection rate of the nanoparticle-surfactant dispersion and N₂ were both set to be 0.002 mL/min. Therefore, the gas liquid ratio (GLR) was 1:1 during the micromodel test. To distinguish the water phase from the gas phase, 0.05 wt % eosin-Y was added to the nanoparticle-surfactant dispersion [2]. The experiment was conducted at 60 °C.

2.6. Sandpack Flood Studies

The sanpack test experimental setup is shown in Figure 3. All the sandpacks had a diameter and length of 2.54 and 60.00 cm, respectively. Firstly, the sandpack was packed as follows: a coreholder filled with brine was positioned vertically, and fresh quartz sands with size fractions of 80–100 meshes were added in several increments to fill the coreholder. In each step, the sand was shaken slightly after being poured in. During this process, the water surface was kept above the top of the sand to ensure that air was not introduced into the sample. The sandpack tests were conducted horizontally. The brine was injected into the sandpack to obtain the porosity and permeability. The sandpack was flooded with the crude oil to establish the initial oil saturation with the injection rate of 0.05 mL/min at 60 °C. The water flooding was conducted in the sandpack until the water cut was up to 98%. To study the recovery performance of foam flooding, N₂ and chemical slugs were injected. The foam flooding was followed by a subsequent water flooding until the water cut was up to 98%. The back pressure and temperature of the sandpack tests were set to be 3.0 MPa and 60 °C, respectively.

3. Results and Discussion

3.1. Characterization of the Foam in Bulk Tests

Figure 4 shows the half-life of different foam systems at different temperatures. It is observed that the half time of the SiO₂/HY-2 foam system was much higher than that of the HY-2 foam system at different temperatures, indicating that SiO₂ nanoparticles can improve the stability of the foam. Figure 5 shows the dilational viscoelasticity of the different systems. The dilational viscoelasticity is an important property for gas bubbles. The dilational viscoelasticity of the surface layer determines its ability to resist external disturbances and to avoid the bubble coarsening and rupture [20,21]. It can be seen that the dilational viscoelasticity of the SiO₂ nanoparticles with the HY-2 dispersion is higher than that of the HY-2 solution within the experimental frequency range. This is because the SiO₂ nanoparticles can attach to the gas-liquid interface and form stronger viscoelastic layers.

3.2. Emulsification Tests

Emulsification tests were conducted to study the emulsification performance of the nanoparticle-surfactant dispersion. Emulsions of oil in two aqueous phases (brine and nanoparticle-surfactant dispersion) are shown in Figure 6. Figure 6a,b show the bottle test for the oil and brine, respectively. It can be seen that the O/W emulsion could not be formed in the brine. Figure 6c,d show the emulsification test for the oil and nanoparticle-surfactant dispersion. The results show that oil could be well dispersed in the nanoparticle-surfactant dispersion.

To study the effectiveness of nanoparticles on the O/W emulsion stability, the half-life of the emulsion at different temperatures was investigated. As shown in Figure 7, the half-life of the emulsion was higher with a lower temperature. At 30 °C, the half-life of the nanoparticle-surfactant dispersion emulsion was about eight times that of the surfactant emulsion. At 80 °C, the half-life was about 13 times that of the surfactant emulsion. These results indicate that nanoparticles can significantly improve the stability of the O/W emulsion. Nanoparticles could form monolayer bridges among the oil droplets (see Figure 8). Nanoparticle bridging plays an important role in the stability of the O/W emulsion and inhibits the individual droplet-droplet coalescence [17].

3.3. Micromodel Flood Study

The micromodel test was performed to study the microscopic flow behavior of the nanoparticle-surfactant-stabilized foam for improving heavy oil recovery.

3.3.1. Transport of Gas Bubbles in Porous Media

After the initial water flooding, the nanoparticle-surfactant dispersion and N₂ were injected into the micromodel (see Figure 9a). As a result, nanoparticle-surfactant-stabilized gas bubbles are gradually formed near the inlet of the micromodel. The number of gas bubbles increased significantly with the injection of the surfactant solution and N₂ (see Figure 9b). Due to the addition of nanoparticles, the gas bubbles were stable and the bubble-bubble coalescence was not observed in the micromodel.

It is clearly observed in Figure 10 that many oil droplets could be stripped and dispersed in porous media because of the emulsification effect of the nanoparticle-surfactant dispersion. In addition, we observed that the flowing gas bubbles can accelerate the formation of emulsified oil droplets (Figure 11). This acceleration can occur because the gas bubbles prefer to remain spherical due to the adsorption of the nanoparticles and surfactant at the water-gas interface [22]. These bubbles will deform when in contact with an oil droplet, as shown in Figure 11a. As shown in Figure 4, the nanoparticles can improve the dilational viscoelasticity of a bubble film. Because of the high viscoelasticity, bubbles tend to recover their spherical shape, which results in the production of a micro-elastic force (F_e) acting on the oil droplet (see Figure 12). The micro-elastic force is helpful for stripping the oil droplets. Therefore, several oil droplets are formed under the action of gas bubbles (see Figure 11b).

Gas bubbles and oil droplets are generated during the injection of the nanoparticle-surfactant dispersion and N₂. Because the nanoparticles improve the stability of the foam and O/W emulsion, a large number of gas bubbles and oil droplets are stably flowed in the porous media (see Figure 13).

Figure 14 shows the gas bubbles driving an oil droplet in the pores. Because of the viscoelasticity of the gas bubbles, the micro-elastic force (F_e) will act on the oil droplet (see Figure 15). Due to the micro-elastic force, the displacement of oil droplets by gas bubbles can be more effective than by the water.

3.3.2. Mechanisms for Plugging Pores with Gas Bubbles and Oil Droplets

The recovery performance of water flooding in heavy oil reservoirs is bad because of the unfavorable mobility ratio between the water and heavy oil. Based on the micromodel test, it was found that the gas bubbles and oil droplets can plug pores in the porous media during the nanoparticle-surfactant-stabilized foam flooding, which can increase the sweep efficiency. There are two main mechanisms for the plugging of pores: capture-plugging and bridge-plugging.

Capture-plugging occurs when a gas bubble or an oil droplet is stuck in a pore-throat with a smaller diameter than the gas bubble or oil droplet particle size (see Figure 16). Based upon the stress analysis, the main forces acting on the gas bubble or oil droplet include the driving force of the fluid (F_df) and the supportive force of the pore-throat walls (F_sx) (see Figure 17). The supportive force F_s is related to the deformation of the gas bubble or oil droplet due to the viscoelasticity.

As shown in Figure 18, it is possible for the gas bubble to pass through the pore-throat after capture. It depends on the viscoelasticity of the gas bubble and the difference of the fluid pressure. In Figure 18a, gas bubbles 1 and 2 are captured by the pore-throats. Gas bubble 1 deforms and gradually enters the pore-throat. This is because the driving force of the fluid (F_df) on the gas bubble is high enough to overcome the supportive force of the pore-throat walls (F_s) (see Figure 19a). Then, the gas bubble transforms and flows in the pore-throat (see Figure 18c). After the gas bubble passes through the pore-throat, it will quickly regain its original shape and size (see Figure 18d). The forces acting on the gas bubbles while they remigrate in the pore-throats are shown in Figure 19. However, gas bubble 2 cannot remigrate into the pore-throat because the difference in fluid pressure is not high enough to overcome the supportive force.

The results indicate that the viscoelasticity of the gas bubble is very important for capture-plugging. Since the nanoparticles increase the viscoelasticity of the bubble-film (see Figure 4), the supportive force of the nanoparticle-surfactant-stabilized gas bubble is higher than the surfactant stabilized gas bubble under the effect of similar deformation. As a result, the capture-plugging ability of the nanoparticle-surfactant-stabilized gas bubble is better than that of the surfactant stabilized gas bubble.

The second plugging mechanism is bridge-plugging. It is involved with gas bubbles and oil droplets smaller than the pore-throats (see Figure 20). As a gas bubble or oil droplet flows through a pore-throat, the bridging of pore-throats may form. Forming a bridge means that it may get attached to a previously deposited gas bubble and oil droplet. When a bridge is formed and consolidated, the newly arriving gas bubbles accumulate upstream from the bridged pores, thus drastically decreasing the fluid flow rate through these pores. Bridge-plugging is the most important plugging mechanism to enhance plugging strength in actual reservoirs.

Pore-bridging consists of several gas bubbles or oil droplets in the pore-throats. Taking as an example the bridge-plugging of two gas bubbles, the main forces acting on the gas bubbles include the driving force of the fluid (F_df), the supportive force of the pore-throat walls (F_s), and the extruding force (F_ex) between the bubbles (see Figure 21). The bridge-plugging is clearly affected by the pore-throat size, gas bubble size, bubble viscoelasticity, the quantity of bubbles in the bridges, and the degree of compaction of the bubbles. The supportive force (F_s) and the extruding force (F_ex) are both related to the bubble viscoelasticity. The results show that the ability of bridge-plugging increases with an increase in the bubble viscoelasticity.

3.3.3. Mechanisms for Displacing Residual Oil with Gas Bubbles

The nanoparticle-surfactant-stabilized foam flooding can improve sweep efficiency by capture-plugging and bridge-plugging. However, some residual oil droplets remain trapped in the pore-throats, as shown in Figure 22. In the following section, the mechanisms of displacing the residual oil by nanoparticle stabilized foam flooding are discussed.

Water is a Newtonian fluid and during the process of water flooding, it can only push the oil along the displacement front, which leaves a large number of residual oil droplets beside the streamline (see Figure 23). For the nanoparticle-surfactant-stabilized foam flooding, the elastic forces of the gas bubbles will act on the residual oil droplets due to the high viscoelasticity of the gas bubbles (see Figure 24). Therefore, the trapped oil droplets will gradually be pushed into the pores by gas bubbles, as shown in Figure 25.

As the trapped oil droplets are pushed into the pores, the flowing gas bubbles can pull the oil droplet into oil threads (see Figure 26). It can be observed in Figure 27 that the residual oil droplet at the interchange of the pores can be pulled into an oil thread. This is also due to the viscoelasticity of the bubbles. The deformation of gas bubbles will occur when in contact with the trapped oil droplets. Later on, bubbles can recover their spherical shape, which will produce an elastic force (F_e) acting on the oil droplet, as shown in Figure 11. The elastic forces can change the shape of the oil droplet and mobilize it (see Figure 28 and Figure 29). As a result, the residual oil droplets could be pulled into oil threads. Therefore, the residual oil can be recovered by nanoparticle-surfactant-stabilized foam flooding, and a greater improvement in the displacement efficiency can be obtained.

Based on the results obtained from the micromodel test, it can be concluded that the nanoparticle-surfactant-stabilized foam can improve sweep efficiency and displace residual oil effectively, thereby improving the heavy oil recovery. The nanoparticles can increase the viscoelasticity of the bubbles, which seems to be very important for foam stability during foam flooding.

3.4. Sandpack Flood Study

To investigate the recovery performance of nanoparticle-stabilized foam flooding, 14 flooding tests were carried out in sandpacks. The effects of the nanoparticle concentration, foam slug size, GLR, and injection pattern were studied.

Table 2. Summary of the sandpack tests.

Test	Properties of Sandpacks			Waterflood Recovery (%)	Chemical Formula	Vl (mL/min)	Vg (mL/min)	GLR	Foam Slug Size (PV)	Injection Pattern	Tertiary Recovery (%)	Final Recovery (%)
	Φ (%)	K (mD)	Soi (%)
1	35.21	1024	0.78	38.14	0.5 wt % HY-2	0.5	0.5	1:1	0.3	co-injection	5.51	43.65
2	32.14	1089	0.75	36.24	0.5 wt % HY-2 + 0.1 wt % SiO2	0.5	0.5	1:1	0.3	co-injection	13.37	49.61
3	35.36	1152	0.71	36.95	0.5 wt % HY-2 + 0.3 wt % SiO2	0.5	0.5	1:1	0.3	co-injection	22.2	59.15
4	36.95	1247	0.76	37.15	0.5 wt % HY-2 + 0.5 wt % SiO2	0.5	0.5	1:1	0.3	co-injection	27.37	64.52
5	34.21	1026	0.81	38.06	0.5 wt % HY-2 + 1.0 wt % SiO2	0.5	0.5	1:1	0.3	co-injection	30.51	68.57
6	36.98	1125	0.79	38.36	0.5 wt % HY-2 + 2.0 wt % SiO2	0.5	0.5	1:1	0.3	co-injection	32.85	71.21
7	35.68	1201	0.81	36.95	0.5 wt % HY-2 + 0.5 wt % SiO2	0.5	0.5	1:1	0.1	co-injection	13.7	50.65
8	39.32	1036	0.71	36.32	0.5 wt % HY-2 + 0.5 wt % SiO2	0.5	0.5	1:1	0.4	co-injection	32.21	68.53
9	35.14	1158	0.74	38.14	0.5 wt % HY-2 + 0.5 wt % SiO2	0.5	0.5	1:1	0.5	co-injection	31.22	69.36
10	36.32	1135	0.72	37.15	0.5 wt % HY-2 + 0.5 wt % SiO2	0.34	0.66	2:1	0.4	co-injection	34.2	71.35
11	34.58	1058	0.73	38.06	0.5 wt % HY-2 + 0.5 wt % SiO2	0.25	0.75	3:1	0.4	co-injection	34.89	72.95
12	38.65	1063	0.76	36.96	0.5 wt % HY-2 + 0.5 wt % SiO2	0.2	0.8	4:1	0.4	co-injection	33.4	70.36
13	36.25	1114	0.75	38.35	0.5 wt % HY-2 + 0.5 wt % SiO2	0.25	0.75	3:1	0.1 + 0.3	nanoparticle dispersion alternating N2	26.33	64.68
14	39.63	1031	0.69	39.65	0.5 wt % HY-2 + 0.5 wt % SiO2	0.25	0.75	3:1	0.05 + 0.15 + 0.05 + 0.15	nanoparticle dispersion alternating N2	28.52	68.17

Φ, porosity; K_w, permeability; V_l, liquid injection rate; V_g, gas injection rate.

lists the experimental conditions and flood results for the sandpack tests.

3.4.1. Effect of the Nanoparticle Concentration

To investigate the effect of the nanoparticle concentration on tertiary oil recovery by nanoparticle-surfactant-stabilized foam flooding, several flooding tests (Runs 1–5) were carried out with nanoparticle concentrations ranging from 0 to 1.0%. The concentration of surfactant was kept constant at 0.5 wt %. For these tests, the injection rates of the nanoparticle-surfactant dispersion and N₂ were both set to be 0.5 mL/min. The total foam slug was set to be 0.3 PV. Figure 30 shows the oil recovery as a function of the nanoparticle concentration. It can be seen that there was a significant increase in recovery with concentrations between 0.1 wt % and 0.5 wt %. Afterwards, the increase in oil recovery with the nanoparticle concentration became very slight. It can be concluded that the 0.5 wt % nanoparticle concentration is the optimum nanoparticle concentration in nanoparticle-stabilized foam flooding.

Figure 31 shows the differential pressure changes as a function of the fluid injected for Runs 1 and 4. It can be seen that the differential pressure of the nanoparticle-stabilized foam flooding (0.5 wt % HY-2 + 0.5 wt % SiO₂) was much higher than the HY-2 foam flooding (0.5 wt % HY-2). The higher built-up differential pressure suggests that the nanoparticle-stabilized foam flooding can block the high-permeability channels and reduce the mobility of the gas phase more effectively. As a result, the subsequent injected water is diverted to the unswept areas, thereby resulting in improved sweep efficiency and increased heavy oil recovery.

3.4.2. Effect of the Foam Slug Size

To investigate the effect of the foam slug size on recovery performance, four sandpack flood tests (Runs 4, 7–9) were conducted by using different foam slug sizes: 0.3, 0.1, 0.4, and 0.5 PV, respectively. The SiO₂ and surfactant concentrations were both set to be 0.5 wt %. Figure 32 shows the oil recovery as a function of the foam slug size. When the foam slug size was less than 0.4 PV, the oil recovery increased obviously with the foam slug size. However, as the foam slug size increased higher the 0.4 PV, the increment of the oil recovery was not significant. Base on the above results, the optimized foam slug size was 0.4 PV.

3.4.3. Effect of Gas Liquid Ratio

The gas liquid ratio is the ratio between the gas flow rate and N₂ flow rate under the experimental conditions during the foam flooding. To study the effect of the gas liquid ratio on recovery performance, three flooding tests (Runs 8, 11–12) were carried out. The gas liquid ratios were set to be 1:1, 2:1, 3:1, and 4:1, respectively. The total foam slugs in these tests were 0.4 PV. Figure 33 shows the oil recovery as a function of the gas liquid ratio. As GLR increased, the oil recovery first increased and then later decreased. The oil recovery was highest at a GLR of 3:1. Therefore, the optimized GLR was 3 for the nanoparticle stabilized foam flooding, to enhance the heavy oil recovery.

3.4.4. Effect of the Injection Pattern

To study the effect of the injection pattern on the recovery performance of nanoparticle-surfactant-stabilized foam flooding, three flooding tests (Runs 11, 13–14) were conducted by different injection patterns. The foam slug size and GLR were set to be 0.4 PV and 3:1, respectively. The nanoparticle SiO₂ dispersion slug and N₂ slug were injected through different patterns, as shown in Figure 34. For Run 11, the 0.3 PV N₂ slug and 0.1 PV nanoparticle dispersion slug were injected into the sandpack simultaneously. For Runs 13 and 14, the nanoparticle dispersion slug was injected firstly, and then the N₂ slug was injected. The nanoparticle dispersion slug size of each injection was 0.1 PV and 0.05 PV, respectively. Figure 35 shows the oil recovery for each test. It can be seen that the oil recovery of the test with the co-injection pattern (Run 11) was higher than that of the test with the nanoparticle dispersion alternating N₂ injection pattern (Runs 13 and 14). For the nanoparticle dispersion alternating N₂ injection pattern, the oil recovery of Run 14 is higher than that of Run 13. This indicates that the recovery performance of the nanoparticle stabilized foam flooding was better with a more frequent cyclic slug. Therefore, the injection pattern is recommended to select the N₂ and nanoparticle dispersion co-injection pattern.

4. Conclusions

(1)

The stabilities of the foam and O/W emulsion increase when silica nanoparticles were added. As a result, a large number of gas bubbles and oil droplets were stably dispersed in the porous media during the nanoparticle stabilized foam flooding.

(2)

The SiO₂ nanoparticles can increase the dilational viscoelasticity of the gas-water interface, which is an important phenomenon for improving the enhanced heavy oil recovery. The gas bubbles and oil droplets can plug pores through the mechanisms of capture-plugging and bridge-plugging, thereby increasing the sweep efficiency. The trapped residual oil could be pushed to the pores gradually by the elastic forces of gas bubbles, and subsequently, it could be pulled into oil threads by the flowing gas bubbles. As a result, a greater improvement in displacement efficiency is obtained.

(3)

The sandpack test results show that the tertiary oil recovery of nanoparticle stabilized foam flooding can reach about 27% using 0.5 wt % SiO₂ nanoparticles. The foam slug size of 0.3 PV and the gas liquid ratio (GLR) of 3 were found to be the optimum conditions in terms of the heavy oil recovery by nanoparticle stabilized foam flooding in this study. Continuous nanoparticle dispersion and N₂ could be more effective when compared with the cyclic injection pattern.