Research for Flow Behavior of Heavy Oil by CO₂ Foam Viscosity Reducer-Assisted Steam (CFVAS) Flooding: Microscopic Displacement Experiment Study

Abstract

Steam displacement is prone to cross-flow, small swept area, large oil–water ratio, large oil–water interfacial tension, and low oil displacement efficiency. Compared with steam flooding, foam flooding can effectively reduce the residual oil in the small throat of the main flow channel and the small hole in the near flow channel and increase the overall recovery factor. Therefore, researchers carried out CO₂ and chemical agent-assisted steam displacement. However, at present, there is a lack of research on the occurrence mechanism and model of residual oil. Steam flooding often encounters challenges such as cross-flow, limited sweep area, and high oil–water ratio. Foam flooding offers a promising alternative by effectively reducing residual oil in narrow throats and the near flow channel, thereby enhancing overall recovery rates compared to steam flooding alone. Therefore, chemical agent-assisted steam flooding was applied to enhance heavy oil recovery. However, the occurrence mechanism and model of residual oil after chemical agent-assisted steam is not clear. To fill this gap, the CO₂ foam viscosity reducer assisted steam (CFVAS) flooding technology has been adopted and carried out in several studies. First, the foam viscosity reducer was prepared and its foam properties (viscosity reduction effect, foam volume, and half-life) were tested. Subsequently, the CFVAS displacement experiments after steam flooding were carried out, and the flow behavior of the remaining oil in multiple regions (main flow channel, near flow channel, and far flow channel) was analyzed. Finally, the shape and number of remaining oil under different displacement stages were compared, and the occurrence mode of remaining oil under CFVAS displacement was determined. The results indicate the following: (1) During steam flooding, the amount of near flow channel residual oil decreased with injected pore volumes (PV), transforming into columnar structures in small perforations and film-like formations in far flow channels. (2) CFVAS flooding, including the foam stability mechanism, flow channel adjustment mechanism, and emulsification and dispersion mechanism, can improve overall recovery rates by 55.2% by driving the remaining oil in near flow channels. (3) During CFVAS flooding stage, crude oil mobility notably improved and flooding front expanded more evenly. Residual oil primarily existed as oil-in-water (O/W) emulsions with discontinuous columns. (4) In the CFVAS flooding stage, residual oil mainly formed O/W emulsions through emulsification and dispersion, with foam-filled large and medium pores, concentrating residual oil in thick and middle throats. This work can provide important references for injecting CO₂ gas into reservoirs to enhance heavy oil recovery and promote carbon sequestration.

1. Introduction

In the middle of the 20th century, research scholars disclosed that adding foam agents during the oil production process could effectively enhance oil recovery [1]. Through continuous and profound exploration, it was noticed that injecting CO₂, N_2, or other gases into reservoirs could considerably enhance the oil production of heterogeneous and low-permeability reservoirs [2]. At the 1997 International Symposium on Oilfield Chemistry, scholars deliberated on the performance of CO₂ foam flooding and affirmed its capability to effectively improve the efficiency of crude oil recovery [3]. Studies have been conducted on the mechanism of reducing viscosity by foam chemical agents in ultra-heavy oil [4], and it is found that too little injection volume of chemical agents or gas does not affect the flow of ultra-heavy oil. To further enhance crude oil recovery using CO₂ foam flooding, its viscosity reduction effect and recovery mechanism have gradually drawn attention. Through conducting experiments on the mixed-phase displacement of CO₂ with heavy oil [5,6], it was found that CO₂ injection could effectively reduce the viscosity of heavy oil and enhance its mobility.

Under reservoir conditions, the use of aqueous surfactants containing 0.70 gas fraction to stabilize CO₂ foam can provide a significant mobility reduction factor of 340 compared to pure CO₂ injection [7]. Compared with conventional gas foam chemical agents, researchers have gradually explored the advantages of combining CO₂ with chemical agents [8]. On the one hand, CO₂ can effectively dissolve in crude oil to decrease viscosity [9,10]. CO₂ foam can expand the application range of viscosity reducers, and then push chemical agents deeper into reservoirs. On the other hand, CO₂ injection efficiently mitigates greenhouse gases [11]. Therefore, CO₂ foam has better economic and social benefits compared to N₂ foam [12,13,14]. Similarly, surfactant solutions can be added to CO₂ foam [15,16,17] to improve foam performance. For example, non-ionic surfactants [18,19] enhance foam stability in high-temperature and highly mineralized heavy oil reservoirs [20,21,22]. In addition to surfactants, polymers can also act together with CO₂ foam. Li’s experiments, using micromodels simulating heterogeneous rocks, provided a fundamental understanding of phase behavior involving polymer-enhanced air foam [23]. During CO₂ foam flooding, the recovery factor of heavy oil without fractured cores was lower than that with complex fractured cores, the results indicated the minimal impact of fracture complexity on the ultimate recovery factor [24].

Currently, the primary methods for developing heavy oil include steam drive and huff-and-puff techniques, but these methods will induce relatively low recovery efficiency. With the continuous exploitation of heavy oil resources, researchers have recognized the potential of adding CO₂ foam additives during steam flooding to further develop remaining oil reserves in late-stage oilfields. Experimental studies using CO₂-assisted steam huff-and-puff techniques have shown [25] that it is more effective than pure steam or pure CO₂ huff-and-puff, increasing the recovery factor by about 20%. Moreover, CO₂ foam flooding can reduce CO₂ consumption by 15.23% while increasing gas storage capacity by 3.53% [26]. In a comparative analysis of the development effects of N₂ and CO₂ to assist steam huff-and-puff, it is concluded that adding CO₂ during steam huff-and-puff significantly enhances heavy oil production rates [27,28]. Stone et al. employed a one-dimensional physical model to simulate Athabasca bitumen sand and discovered that injecting CO₂ into steam could strengthen the driving energy and improve crude oil recovery [29]. Based on these facts, asphalt mining was successfully achieved by mixing high-temperature steam with CO₂ and surfactant [30,31]. Of course, more efficient identification and exploitation of oil and gas resources and analysis of asphalt structure cannot be achieved without the application of advanced analytical chemistry technology and geochemistry [32]. Additionally, Mohammed et al. compiled production history data from Forest Oilfield and analyzed the effects of increasing the heavy oil recovery through CO₂-assisted steam huff-and-puff [33].

At present, it is generally recognized that CO₂ foam-assisted steam flooding can enhance the recovery of heavy oil, but its mechanism is still not clear enough. To fill this gap, we started our research by observing the flow behavior and existence mode of the remaining oil. This work focuses on observing the residual oil distribution and patterns in heavy oil reservoirs assisted by CO₂ foam during steam flooding, aiming to understand the mechanism of steam flooding combined with thermal enhanced oil recovery, revealing residual oil distribution characteristics, clarifying the state and patterns of heavy oil residual oil, and providing the theoretical basis and technical support for enhancing heavy oil recovery through combined thermal recovery.

2. Materials and Methods

2.1. Formation Water

Experimental water, CaCl₂, and MgCl₂ were prepared by simulated strata, with a total mineralization of 40,000 mg/L, a Ca²⁺ concentration of 1400 mg/L, and 500 mg/L in Mg²⁺ concentration.

2.2. Crude Oil

Experimental crude oil came from Xinjiang oilfield. The density was 0.966 g/cm³, saturated hydrocarbon 52.6%, aromatics 15.2%, and gum 32.2%.

2.2.1. Crude Oil Viscosity

The crude oil in the experiment is a crude oil after dehydration. Measurement of the viscosity of crude oil was carried out by Anton Pal MCR 302 (Anton Pache Group Co., Ltd., Austria Graz, Austria), Germany. The flow instrument parameters were set as follows: the gap was 1.00 mm, the shear rate was 7 S⁻¹, the normal force as 0.11 N, and the temperature gradually increased from 30 °C to 120 °C within 30 min. The 100 data points (Figure 1) show the relationship between viscosity and temperature.

2.2.2. Oil–Water Interfacial Tension

The mixture of stratigraphic water and crude oil was introduced into the TX-500C rotating IFT device (Shanghai Solon Information Technology Co., LTD., from Shanghai, China). During testing, temperatures were varied from 30 °C to 120 °C in 10 °C intervals, with a rotational speed set at 5000 rpm. The test data revealed a mathematical correlation between oil–water interfacial tension and temperature (Figure 2), expressed as σ = 16.447 T^−0.234, where σ represents the interfacial tension in mN/m, and T denotes the temperature in degrees Celsius in °C.

2.3. Foam Viscosity Reducer

2.3.1. Preparation of Foam Viscosity Reducer

The composition of the foam viscosity reducer is shown in

Table 1. Components of foam viscosity reducer.

Name	Chemical Formula	Mass Percentage
Sodium tripolyphosphate	Na5P3O10	3%
Sodium nonylphenol polyoxyethylene ether sulfate	C30H46O·OSO3Na+	12%
Sapindus saponins	C46H74O16	8%
Coconut oil diethanolamine	C4H11NO2	5%
Distilled water	H2O	72%

. The recommended value of mass fraction was obtained through more than 30 laboratory tests at 55 °C. Although a small difference in mass percentage will not have a significant impact on its foam performance, the recommended mass percentage error should not exceed 0.5%. The optimal concentration of foam viscosity-reducing agent needs to be further discussed through evaluation tests including viscosity-reducing effect and foam properties.

2.3.2. Foam Properties

1.

Viscosity-reducing effect

The foam viscosity reducer solution with a concentration of 0.2 wt%~1.4 wt% was prepared from crude oil. In the course of this test, the test temperature was 55 °C, and the viscosity was used by Anton Paar MCR302 viscometer. To evaluate the viscosity reduction effect of foam viscosity reducer solution at different concentrations, DRμ = (μ_a − μ_b)/μ_a was used to define the viscosity reduction rate, where DR_μ is the viscosity reduction rate %; μ_a is the viscosity before viscosity reduction, mPa·s; and μ_b is the viscosity after viscosity reduction, mPa·s. As shown in Figure 3, the optimal concentration of the foam viscosity reducer solution is greater than 0.8 wt%.

2.

Performance of foaming

The foaming volume and half-life of different concentrations of foaming viscosities were measured at 55 °C. As shown in Figure 4, the higher the concentration, the larger the foam volume and the longer the half-life of the foam. When the concentration reached 0.8 wt%, the foam volume and foam half-life did not increase significantly. Considering the economic benefits, the optimal dosage of the blowing agent is 0.8 wt%.

2.3.3. Optimization of CO₂ and Foam Viscosity Reducer

In the actual field operation, many factors affect the oil displacement effect of CO₂ foam. In this section, the concentration and gas–liquid ratio are tested to optimize the regulation parameters of CO₂ and foam viscosity reducer.

According to the viscosity-reducing effect and foam properties of foam reducer, the test results show that the optimal concentration is 0.8 wt%. The recommended gas–liquid ratio of CO₂ foam viscosity reducer solution was determined by the indoor evaluation experiment with the resistance factor as the evaluation index. The resistance factors of different gas–liquid ratio schemes (such as gas–liquid ratios of 1:1, 1.5:1, 2:1, 0.5:1, and 0.25:1) of CO₂ foam viscosity reducer show that the scheme with the highest resistance factor is the gas–liquid ratio 1.5:1 (Figure 5).

2.3.4. Mechanism of CO₂ Foam Displacement

Saponins feature a hydrophilic head (-O-, -OH, and -COO-) and a hydrophobic tail, exhibiting robust surface activity that allows for them to swiftly adsorb onto gas–liquid interfaces, generating numerous bubbles. The ether bond (-O-) in C₃₀H₄₆O⁺ and the hydrogen bond coordination with saponins form a resilient network structure, bolstering bubble stability at the interface (Figure 6a). This synergy not only enhances foam durability but also extends its transmission range, effectively blocking orifice and flow channel locations. By adjusting the liquid phase flow direction, particularly using surfactants and viscosity reducers, the potent foam coerces liquid to flow along pore walls, thereby cleansing and eliminating oil films (Figure 6b). Through emulsification and flow regulation, the aqueous solution interacts more effectively with oil films and large droplets on pore walls, dispersing residual oil in pores channels and pore throats, ultimately yielding a more fluid emulsion or microsphere state (Figure 6c–e).

3. Experimental Procedures

3.1. Experimental Equipment

The microscopic model temperature controller can control the temperature of 500 °C. The steam generator is used to produce steam; its temperature range is 100~300 °C. The experiment adopts the high-temperature and high-pressure displacement microscopic visualization experiment platform including the camera, observation window, and etched glass. The high-definition high-frequency camera system consisted of a computer and a digital microscope camera. The images were observed by Imageview software (Version 2.0) and analyzed by ImageJ software (Version 2.0), as shown in Figure 7a. The micro-etched glass model was placed in the high-temperature and high-pressure observation window. The appearance size of the glass model was 75 mm × 75 mm, the zone size with pore was 45 mm × 45 mm, the thickness of the model was 3.5 mm, the injection hole center distance was 92 mm, the diameter of the injection hole was 2.5 mm, and the glass model showed hydrophilicity (the contact Angle was 45°). (Figure 7b). The high-temperature and high-pressure observation window could withstand the pressure of 32 MPa and the temperature of 300 °C (Figure 7c); it was controlled by the temperature controller.

3.2. Experimental Procedures

As shown in Figure 8, the experimental steps are as follows:

• Connect the experimental devices and pipelines. In particular, it is necessary to prepare several pipelines at the entrance of the high-temperature and high-pressure visualization experimental device to avoid pollution and blockage of the pipeline when oil and other chemical agents are injected.
• Set the confining pressure to 2 MPa. Connect the pressure gauge to the confining pressure pipeline of the high-temperature and high-pressure visualization experimental device, inject water into the device with a hand pump, slowly increase the confining pressure to 2 MPa, and close the valve. The confining pressure is not lower than the input–outlet pressure system.
• An ISCO pump is used to saturate the micro-etched glass with water first (remove possible impurities inside the pore, and then use the temperature controller to dry the glass model at high temperature). After the model is dried, an ISCO pump is used to saturate the crude oil sample. The injection rate is set at 0.025 mL/min, and the pressure difference between the inlet and the outlet should not be too large.
• Simulate steam displacement. The microscopic model temperature controller is used to heat the visualization experimental device, and hot steam is injected into the microscopic glass saturated with oil samples. The injection speed of the ISCO plunger pump is set to 0.01 mL/min, and the flow and state changes of oil samples in the microscopic etched glass are observed.
• Simulate heavy oil combined with thermal recovery and displacement. The one-way valve, dryer, and gas flow meter are, respectively, connected to the gas tank containing CO₂. The CO₂ injection rate is set to 0.025 mL/min with the gas flow meter and 0.013 mL/min with the foam viscosity reducer solution (0.8 wt%) with the ISCO plunger pump. At the same time, the CO₂ and foam agent solution are injected into the foam generator. Foaming is carried out, and then the viscosity-reducing agent and foam are mixed and connected to the entrance of the high-temperature and high-pressure visualization experimental device, and the foam and viscosity-reducing agent are injected at the same time to observe the flow and state changes of oil samples in the micro-etched glass (for temperature fluctuations, the temperature controller is used to heat up the high temperature and high pressure visualization experimental device to 100 °C, and the ISCO pump is used to inject the crude oil at 0.025 mL/min).
• Video and record the heavy oil flow and state in the steam displacing to composite thermal recovery displacing stage, and analyze the flow form, remaining oil state, and occurrence mode of heavy oil in the model, using Imagej image analysis software (Version 2.0) of automatic area measurement function; calculate the etching glass model under different injection PV numbers within the area of remaining oil.

3.3. Experimental Parameters

According to the experimental scheme and experimental flow, two experimental schemes were used to connect the experimental equipment. Firstly, steam displacement was conducted to observe the state and mode of remaining oil according to the change in injected PV number. When the injection number was greater than 1.5, it was found that the state of the remaining oil did not change significantly, and the spread range of steam displacement no longer expanded. Therefore, a CO₂ foam viscosity reducer-assisted steam displacement experiment was conducted.

• During steam displacement, the temperature of the microscopic model temperature controller was set at 100~150 °C, the injection speed was 0.02 mL/min, and the injection PV number was from 0.5 to 1.5.
• For CO₂, the temperature range was the same as that of the scheme (i). The injection rate of 0.017 mL/min for 0.08 wt% foam viscosity reducer and 0.034 mL/min for CO₂ were used, both of which were injected simultaneously. The injection volume of this group of experiments was continued until 3.0 PV according to the cut-off injection PV number of the first group of experiments.

3.4. Analytical Method

3.4.1. Flow Zone Division

Based on the flow channel observed in the micro-etched glass model, it is viewed as a diamond shape (Figure 9). Connect the diamond diagonals and tripartite with the red dotted line at the top left (as well as the red dotted line at the bottom right), connecting the other two vertices of the square according to the green dot of the tripartite, the innermost main channel, followed by the near channel, and finally the far channel.

3.4.2. Pore Structure Division

The pore channel width of the etched glass model is 10~50 μm, and the pore throat width is 90~450 μm. The pore channel and pore throat size are characterized by the pore/throat radius. According to the radius of the pore channel, it can be divided into three categories: large (≥35 μm), medium (15~35 μm), and small channel (<15 μm). According to the mean size of the pore throat, it can be divided into four categories: coarse (≥7 μm), middle (1~7 μm), fine (0.1~1 μm), and micro-fine throat (<0.1 μm) (Figure 10).

3.4.3. Remaining Oil State Division

According to the proportion of microscopic remaining oil in pore space, the remaining oil state is mainly divided into two categories: one is continuous remaining oil, and the other is dispersed remaining oil (

Table 2. Residual oil shape classification.

Continuous Residual Oil		Dispersed Residual Oil
Clusters form	Columniform	Film shape	Isolated island shape

).

4. Results and Analysis

4.1. Steam Flooding Phase

4.1.1. Main Channel Zone

(1)

Residual oil occurrence state

In the main channel, the water phase pushes forward to displace crude oil along the injection–production main channel, and the displacement front expands unevenly to both sides. The water phase occupies most of the volume of the channel, and most of the crude oil is displaced. In the main stream area, the remaining oil is mainly clustered in the continuous phase and columnar or film in the discontinuous phase (see the green circle in Figure 11).

(2)

Residual oil occurrence pattern

At the main channel zone, we select pore structures with pore coordination numbers (PCNs) of 3, 4, and 5, and the remaining oil is mainly short columnar. The pores are full of well-connected bubbles. As the number of PV injections increases, the viscosity of crude oil decreases, and bubbles in the fine noise gradually connect and gather in the pores. The bubbles in a PCN of 5 are almost unaffected. In PCNs of 3~4, part of the remaining oil is driven out. The amount of remaining oil in the film gradually decreases (Figure 12).

4.1.2. Near Channel Zone

(1)

Residual oil occurrence state

The location of the near channel is affected less. The effect of continuous steam injection on oil displacement is not obvious. In the area near the flow channel, the remaining oil occupies most of the pore area, and the remaining oil is columnar, membrane-like, and a few continuous phases cluster (see the green circle in Figure 13).

(2)

Residual Oil Occurrence Pattern

We select pore structures with PCNs of 3~4, in which there is a large area of residual oil, and the pore structure is mainly columnar, cluster, and a small part of the cluster. The bubbles are mainly distributed in large pores. With the increase of injected PV number, the small bubbles in the throat of large pores with larger radius gradually gather, the bubbles in adjacent throats gradually converge to exist in the pore, and the change of injected PV number has little effect on the remaining oil in the small radius of the bellows (Figure 14).

4.1.3. Far Channel Zone

(1)

Residual oil occurrence state

With the increase in injected PV, part of the crude oil in the flow channel is gradually displaced. The crude oil occupies most of the pore area in the far flow channel, the remaining oil is mainly in continuous sheet and cluster shapes, and a few are columnar and film shapes. The water phase slowly gathers in the form of dispersed droplets, and the steam can hardly spread to the far channel area (see the green circle in Figure 15).

(2)

Residual oil occurrence pattern

Crude oil occupies most of the pore area of the far flow channel. In the pore structure with PCNs of 3~5, the remaining oil is mainly continuous flake and cluster, and a small amount of columnar and film. Since the steam cannot reach the edge, the remaining oil in the pore with a coordination number of 5 is almost unchanged. In pore structures with PCNs of 3~4, with the increase in PV injection, the volume of bubbles gradually increases, and gradually dispersed bubbles are connected. As a result, the amount of the remaining oil in the longer column is reduced by steam flooding, the oil saturation in the middle throat decreases, and bubbles gradually gather in the fine throat (Figure 16).

4.2. CFVAS Flooding Process

4.2.1. Main Channel Zone

(1)

Residual oil occurrence state

In the main channel, O/W and foam block the channel, forcing the displacement fluid to expand to both sides, and the spread range is significantly expanded. With the increase in PV, a large part of crude oil is displaced out of the pore, and the remaining oil is very small. In the main channel area, oil–water emulsion and foam occupy most of the channel positions, and the remaining oil mainly exists in the O/W state (see the green circle in Figure 17).

(2)

Residual oil occurrence pattern

The pore structure with a PCN of 3 was selected, and the remaining oil was mainly characterized as O/W emulsion after emulsification and dispersion. The pores are filled with foam, and the remaining oil tends to gather in the larger radius of the roar. With the increase of PV number, the viscosity of crude oil will decrease, the driving force of foam in large pores will increase, the seepage resistance will decrease, and the foam in large pores can only be driven and dispersed in a small range in the pores. The O/W emulsion flows along the liquid film on the surface of the foam, which also demonstrates the efficacy of the CO₂ foam in improving the flow capacity of the remaining oil (Figure 18).

4.2.2. Near Channel Zone

(1)

Residual oil occurrence state

In the near flow channel, the displacement range expands greatly to both sides, and the remaining oil in the near passage is also used effectively. With the increase in PV number, most of the unused crude oil is displaced out of the channel. Emulsion and foam occupy most of the spread area, most of the remaining oil exists in an O/W state, and a small part is discontinuous columnar (see the green circle in Figure 19).

(2)

Residual oil occurrence pattern

For pore structures with PCNs of 3~4, the remaining oil was mainly characterized as O/W emulsion after emulsification and dispersion and a fully mixed state of oil, gas, and water, mostly distributed in the coarse and middle throats. The CO₂ foam gathered in large pores with a PCN of 4, and a few small bubbles intermingled in the throat. As PV increases, CO₂ foam collects between large pores. In the coarse and middle throats, O/W emulsions flow along the throat wall. Influenced by gas diffusion, the number of O/W emulsions decreases, and the emulsions displaced to adjacent throats gradually form the seepage of CO₂ and foam and the mixed state of oil, gas, and water through the transport of foam (Figure 20).

4.2.3. Far Channel Zone

(1)

Residual oil occurrence state

The remaining oil in the far flow channel is effectively used, and the oil displacement channel is formed in the edge area. Most of the remaining oil mainly exists in the gap between the water phase and the foam, and the remaining oil is in a discontinuous column shape and the oil–oil–water is fully mixed (see the green circle in Figure 21).

(2)

Residual oil occurrence pattern

For pore structures with PCNs of 3~4, oil primarily exists in O/W emulsion after emulsification and dispersion, along with a mixed state of oil, gas, and water, mainly in coarse and middle throats. Large foams gather in a PCN of 4, stripping and emulsifying oil into filament-like emulsion via foam action, reducing interfacial tension. With PV increase, it increases the effect of surfactant, the big hole in the foam increases, the residual oil amount decreases significantly, and the size of foam in the larger pores gels at a relatively stable state, leading to a lower amount of O/W emulsion (Figure 22).

4.3. Comparison

We carried out three tests of steam flooding to CFVAS flooding, and the oil recovery factor in each test was slightly different under different injection PV numbers (Figure 23). The oil recovery efficiency obtained in the three tests of experiments was better analyzed using averaging, and the results obtained can be seen in in Figure 24.

In the steam flooding process, the remaining oil occurrence state in the micro-etching model has the following characteristics (

Table 3. Comparison of remaining oil patterns of two displacement methods.

Displacement Stage	Flow Channel	Residual Oil Status	Residual Oil Pattern	Effect of Injection PV Number
Steam flooding	Main channel	Clusters form, columniform, film shape	Mostly in the fine throat	The amount of film remaining oil increases
	Near channel	Columniform, film shape	Mostly in the fine throat and small hole	Columnar remaining oil increases in the small hole
	Far channel	Clusters form, a few are columnar and film-shape	Mostly distributed in coarse, middle-throat	The clusters are transformed into film shapes and columns
CFVAS flooding	Main channel	O/W emulsion	Distributed in the thick and middle throat	The O/W emulsion in the pores gradually flows into the throat
	Near channel	O/W emulsions and columns		The number of O/W emulsions increases
	Far channel	Columnar and fully mixed state of oil, gas, and water

): The front is not uniformly advanced, and it is easy to form a channeling way. The degree of oil flow differences, largely affected areas concentrated in the mainstream, basic not wave, and near port and port range is small. The oil–water interfacial tension is large, and the oil displacement efficiency is low. The overall occurrence state of the remaining oil in the microscopic model is shown in Figure 24 (PV = 0.5~1.5). CFVAS flooding (PV = 2.0~3.0) and CO₂ foam viscous-reducing agent-assisted steam flooding can obviously improve the flow of crude oil, block the channeling channel, expand the spread area, and solve the problem of low oil displacement efficiency of steam flooding. The foam viscosity reducer can quickly adsorb to the gas–liquid interface and form many bubbles, make the bubbles more stable, and adjust the flow direction of the liquid phase. Foam agents can effectively block the channel, obviously increase the spread range, and expand the spread area. The CO₂ foam viscosity reducer can effectively reduce the interfacial tension of oil and water, the contact area between the solvent and the oil film and large oil droplets on the pore wall is larger, and it emulsifies them into smaller emulsions, which significantly improves the oil displacement efficiency and greatly increases the recovery rate.

In the etched-glass model (Figure 25), the white area is the calculated area of the remaining oil. The area of remaining oil decreases with the increase of injection PV number, and the area of remaining oil is still large at the end of steam flooding. During CFVAS flooding, the area of remaining oil decreases, and the displacement range and efficiency increase.

5. Conclusions and Discussion

5.1. Conclusions

Regarding the unclear understanding of CO₂ foam-assisted steam flooding, the flow process of heavy oil under steam flooding and CFVAS flooding in porous channels was simulated. Through the study on the occurrence state and mode of residual oil from steam drive to composite thermal recovery in a heavy oil reservoir, the distribution characteristics of residual oil were revealed, and the occurrence state and mode of residual oil were clarified, which provides the theoretical basis and technical support for the improvement of heavy oil recovery by composite thermal recovery. According to the state and mode of the remaining oil in each stage of the etched glass model, the following points were mainly observed:

• From the two aspects of viscosity reduction effect and foam performance, the optimal concentration of foam viscosity reduction agent is 0.8 wt%, and the ratio of CO₂ and foam viscosity reduction agent to gas and liquid is 1.5:1.
• The sweep range of steam flooding is small, and the oil displacement efficiency is low. The CFVAS flooding has a larger sweep area, can block the flow channel and reduce the interfacial tension, significantly improve the mobility of crude oil, and uniformly expand the displacement front. The recovery factor of CFVAS flooding is 40~55% higher than that of steam flooding.
• During CFVAS flooding, the remaining oil mainly exists in the form of an O/W emulsion or fully mixed state of oil, gas, and water, and a small part is in the form of discontinuous columns. The remaining oil form is related to the regional location, and the increase in injected PV number has a greater impact on CFVAS flooding but has little effect on the far and near flow channels
• After CFVAS flooding, crude oil in the fine pore throat is mostly emulsified and dispersed, while the residual oil in the coarse and middle pore throat is columnar with foam. In the main channel, the pores with PCNs of 3~4 were filled with foam and some O/W emulsions. In the near flow channel, O/W emulsions were mixed with a few small foams and gathered in pores with a PCN of 4. The remaining oil was characterized as O/W emulsions and columns. In the far flow channel, residual oil that is fully mixed with foam, oil, gas, and water is gathered in the pore with PCNs of 4~5.

5.2. Discussions

For environmental protection, the novel foam viscosity-reducing agent prepared, extracted from tea trees, has the characteristics of a long half-life, large foam volume, and a more obvious blocking effect. The saponins contained in it are mainly extracted from oil tea cake and waste.

• Saponin is mainly extracted from tea cakes and tea waste. Tea cakes contain irritating elements that deteriorate the quality of the land. People mainly deal with it by incineration. Through our extraction and utilization, we can effectively protect the quality of the land and reduce carbon emissions. Although the effect of saponin is quite good, the extraction rate of saponin for tea seeds is only 15%.
• In terms of CO₂, it can reduce CO₂ emissions and promote carbon sequestration. On-site, we also recommend the use of captured CO₂ injection to reduce the CO₂ originally present in the atmosphere. The disadvantage is that the economic cost of captured CO₂ is not low.