1. Introduction
Oil and gas resources remain an important source of energy for social development, but there are still many problems in the development process [
1,
2,
3]. Block Mu146, a medium-high permeability structural reservoir, has reached a water cut of 98.20% and an oil recovery of 50.70% since its commissioning in 1976 [
4,
5]. Persistent challenges include elevated water–oil ratios, severe ineffective water cycling, suboptimal waterflood utilization, and significant injection–production imbalances. Additionally, while localized residual oil potential remains substantial, recovery enhancement is technically demanding [
6,
7]. Transitioning development strategies to enlarge sweep volume and enhance displacement efficiency is therefore critical for optimizing reservoir performance [
8,
9,
10,
11].
The key mechanism of nitrogen foam huff and puff in production wells is that nitrogen injection can increase reservoir pressure, surfactant injection and polymer injection can control nitrogen flow and reduce oil–water interfacial tension [
12,
13]. As the nitrogen injection pressure and the soaking time increase, the oil recovery of nitrogen puff and huff increases [
14,
15]. Bubbles initially form as spheres and expand until they contact the fracture walls, at which point they elongate along the length of the fracture. The rate of bubble growth is influenced by local mass transfer from the liquid phase to the gas phase, as well as gas volume expansion resulting from pressure reduction. The efficiency of the huff-and-puff process is contingent upon the solubility and miscibility of the injected fluid with oil. High gas solubility facilitates increased nucleation, growth, and expansion of bubbles during depressurization cycles [
16]. The nitrogen foam huff-and-puff technique presents a viable solution by effectively enhancing sweep coverage, mobilizing trapped oil within fractures or pores, interconnecting with larger fractures that are sealed at their upper ends, reducing residual oil saturation, and improving well productivity [
17,
18,
19]. From a microscopic pore perspective, foam-assisted nitrogen huff-and-puff not only enhances oil production in macropores but also yields a modest amount in micropores and mesopores. Oil recovery from macropores and mesopores significantly contributes to overall recovery factors. The intricate natural fracture structure and its permeability are critical factors influencing oil production in Bakken reservoirs [
20,
21,
22,
23].
The presence of fractures significantly influences water cut control and oil recovery in bottom water reservoirs. The application of high-strength starch gel to block these fractures can further enhance the effectiveness of CO
2/N
2/N
2 foam huff-and-puff operations. In the case of pure CO
2 huff-and-puff, the primary mechanisms contributing to increased oil production include extraction, viscosity reduction, and the escape of oil droplets. For pure N
2 huff-and-puff processes, the expansion of N
2 serves as the main driving force for displacing crude oil. When utilizing a gas mixture for huff-and-puff operations at a ratio of 7:3 (CO
2 to N
2), a synergistic effect between CO
2 slugs and N
2 slugs is observed, yielding optimal results [
24,
25,
26,
27,
28]. The drive mechanisms associated with gas cap drive, dissolved gas drive, and miscible drive during CO
2 huff-and-puff demonstrate greater efficiency compared to those observed during N
2 huff-and-puff processes. Variations in molecular diffusion rates lead to notable changes in oil recovery factors. While simultaneous injection of CO
2 and N
2 continues to enhance recovery rates, pure CO
2 has been shown to yield superior results. In field pilot tests, cost-effective CO
2 huff-and-puff processes have been successfully implemented in heavy oil reservoirs characterized by an API gravity as low as 4 °API, reservoir depths reaching up to 1985 m, and pay zones measuring just 12.2 m thick. Specifically, CO
2 utilization can be minimized to approximately 4.2 Mscf/Stb. Numerical simulation studies have produced highly accurate results that align well with both experimental data and pilot test findings [
29,
30]. The foam huff-and-puff mode, characterized by a constant gas injection rate or single slug injection, has proven inadequate in effectively regenerating foam, and its stability is challenging to enhance. Consequently, the effective implementation period on-site remains limited [
31,
32,
33,
34].
To address challenges such as low water flooding efficiency, reduced oil production rates, and declining recovery rates during periods characterized by high water cuts in the low-permeability reservoirs of the Mutou Oilfield, non-steady-state (NSS) CO
2 huff-and-puff oil recovery technology has been investigated. This NSS CO
2 huff-and-puff method demonstrates significant potential for improving development outcomes in low-permeability reservoirs by replenishing reservoir energy and substantially enhancing crude oil mobility. The fundamental mechanisms underlying NSS CO
2 huff-and-puff include dissolution, expansion, viscosity reduction, and an increase in swept volume—all effectively mobilizing residual oil from various pore throats within the reservoir. Field trials conducted on NSS CO
2 huffed-and-puffed operations within a low-permeability reservoir at Mutou Oilfield have resulted in a cumulative increase in oil production totaling 1134.5 tons. Furthermore, this NSS CO
2 injection approach provides valuable insights into nitrogen foam huff-and-puffs testing aimed at promoting effective regeneration and enhanced stability of foam [
35].
The alkyl polyglucoside derivatives utilized as foaming agents in foam flooding exhibit superior foaming characteristics. However, a significant issue arises from the fact that alkyl glycoside foaming agents tend to generate white precipitates when used in saltwater containing high concentrations of calcium and magnesium divalent ions. Although the incorporation of an alkyl glycoside sulfonate foaming agent enhances the ability of alkyl glycosides to resist these high levels of calcium and magnesium ions, its elevated production costs hinder its widespread promotion and application in medium-high permeability oil reservoirs [
36,
37]. Foam stabilizers composed of hydrophobic silica nanoparticles and bentonite have been shown to enhance foam stability by 2–5 times. Nevertheless, a prevailing challenge is the high cost associated with hydrophobic silica nanoparticle foam stabilizers, which limits their on-site promotion and application; additionally, bentonite nanoparticle foam stabilizers are not suitable for use in saltwater with elevated levels of calcium and magnesium divalent ions [
38,
39]. The engineered foam system demonstrates exceptional plugging characteristics and fluid diversion capabilities, maintaining both its plugging capacity and oil-displacement efficiency even after multiple cycles of dynamic adsorption [
40,
41].
In order to address the challenges associated with the nitrogen foam huff and puff oil production process, particularly its adverse effects on foam regeneration and stability enhancement, as well as the overall poor stability of the foam system, this study evaluates reservoir adaptability. Well Mu146-61 in Block Mu146 has been selected for field trials involving non-steady-state nitrogen foam huff-and-puff (NSSNFHP). A highly stable foam flooding system with excellent oil tolerance is developed specifically for this application. Core flooding experiments are performed to quantify system displacement efficiency, while systematic monitoring of field performance is undertaken to validate its effectiveness.
3. Results and Discussion
The Mu146-61 well in Block Mu146 was put into operation in January 2011, producing F4, 5, and 6 layers with a thickness of 11.8 m. Currently, it produces 12.13 tons of liquid and 0.54 tons of oil per day, with a water cut of 95.6%, an oil pressure of 0.9 MPa, a casing pressure of 0.18 MPa, a dynamic liquid level of 566 m, a cumulative oil production of 2947 tons, and a cumulative liquid production of 80,036 m3.
As the Mu146-61 well is not subject to secondary fracturing, it is characterized by low cumulative production, low production, a narrow south–north sweep range of water drive, a strong water-flooded strip, poor sweep in water drive layer and plane, rich residual potential, and great difficulty in tapping potential. Energy storage and expansion of sweep are urgently needed. NSSNFHF technology can meet geological needs, improve the problem of insufficient formation energy, effectively expand sweep volume, and improve single well productivity. According to the screening criteria for NSSNFHF technology, the reservoir conditions in Block Mu146 are more suitable for NSSNFHF technology, as presented in
Table 5.
3.1. Development and Performance Evaluation of Nitrogen Foam Flooding System
3.1.1. Development of Nitrogen Foam Flooding System
Under the condition of reservoir temperature of 40.0 °C and salinity of injected water of 7610.37 mg/L in the Mu146 block, the foaming property and stability of the foam flooding system are evaluated by the Warring stirring method. The self-developed foaming agents FP1688, FP2322, FP2320, FP2332, FP2330, FP2218, FP2268, FP2399, and FP2398 (The main compositions and type are presented in
Table 1) are selected for compounding optimization with foam stabilizers WP308, WP100, WP2209, WP131, and WP2366 (The main compositions and type are presented in
Table 2). The results are shown in
Figure 2 and
Figure 3 and
Table 6.
Based on the foaming agent foaming performance test results (
Figure 2), Betaine surfactants FP2218 and FP2268 do not comply with the technical requirements of the Q/SY17816-2021 standard [
42], which stipulates a foam volume greater than 800 mL. In contrast, the foam volume and foam stability time of fluorocarbon sulfonate surfactant FP1688 and polyoxyethylene ether sulfonate FP2399 meet the technical specifications outlined in the Q/SY17816-2021 standard [
42]; however, it is important to note that these data represent intermediate values. A similar situation applies to olefin sulfonates FP2322, FP2320, FP2332, and FP2330. FP2398 is selected as the foaming agent of the nitrogen foam huff-and-puff test foam oil displacement system, with a foaming volume of 1030 mL and a foam stabilization time of 328 s, which has excellent comprehensive performance. Foaming agent FP2398 is a new, high-foaming, efficient, environmentally friendly, sulfonate-type alcohol ether anionic-nonionic surfactant. Its molecular structure contains sulfonic acid groups and hydroxyl and ether bonds suitable for carbon chains. “Sulfonic acid groups” increase its ability to resist divalent ions, while “hydroxyl and ether bonds suitable for carbon chains” increase its foaming ability and temperature resistance.
In order to improve and strengthen the stability of foam, polymer and other foam stabilizers must be added to the foam flooding system to increase the foam stabilization time. A variety of oil and salt-resistant foam stabilizers have been developed for the reservoir conditions in Block Mu146, including biopolysaccharide polymers, polyacrylamide polymers, and salt-resistant polymers. The self-developed foaming agent FP2398 (0.4% fixed concentration) is selected for compounding optimization with different types and concentrations of foam stabilizers WP308, WP100, WP2209, WP131 and WP2366. As illustrated in
Figure 3, under identical apparent viscosity conditions, the type of foam stabilizer exerts minimal influence on the foaming ability of the foam flooding system; however, it significantly affects the stability of the foam within this system. Among the foam stabilizers selected in the experiment, the salt-resistant polymer foam stabilizer WP2366 has the best foam stabilization effect. When the concentration of 0.13% is preferred, it can extend the half-life of the liquid from 328 s to 1390 s and reduce the foaming volume from 1030 mL to 850 mL. In addition, the foaming volume of the foam oil displacement system with 10% oil content is 860 mL, and the half-life of liquid evolution is 1509 s higher than that without oil content, which is 850 mL and 1390 s, respectively, and the oil resistance is excellent (see
Table 3). Therefore, WP2366 is selected as the foam stabilizer of the nitrogen foam oil displacement system.
Based on the above experimental results of foaming agent and foam stabilizer optimization, a foam oil displacement system suitable for the NSSNFHP test in the Mu146 block is finally developed as 0.4% FP2398 + 0.13% WP2366.
3.1.2. Berea Core Flooding Experiment
The results of the core flooding experiments are illustrated in
Figure 4. The total oil recovery achieved through water flooding is 53.60%, as demonstrated by the pressure gradient and incremental oil recovery observed during foam injection. A robust foam is generated throughout the core, which effectively strips away residual oil trapped within the pores. When foam is formed, the hypertonic channel is blocked, and the injection pressure continues to rise. Furthermore, steady pressure is maintained during chase water flooding, indicating successful propagation of foam within the core’s pore structure, even under these conditions. Foam flooding results in an additional 33.20% incremental oil recovery compared to water flooding, bringing the total oil recovery to 86.80%. When the PV number is greater than 2.53, due to the continuous bursting of the foam, the sealing effect on the hyperosmotic channel is gradually weakened, and the injection pressure also decreases again. Consequently, it can be concluded that the foam formulation exhibited exceptional sweep efficiency, significantly enhancing overall oil recovery.
3.2. Field Application Effect of Nitrogen Foam Flooding System
3.2.1. Basic Information of Block Mu146
The Block Mu146 is located in the southern part of the Central Depression, within the Huazi Well Terrace of the Muto Structure. The reservoir is a faulted nose structure formed by three faults, with a lithology-structural oil reservoir. The target layers for development are the Fuyu oil layer and the Yangdachengzi oil layer. The Fuyu oil layer primarily consists of deltaic deposits, with the delta plain facies as the dominant feature. The 1st and 2nd sub-layers are delta front facies, and the strata are divided into 17 individual layers. The sandstones of the 4th, 5th, 6th, 7th, 9th, and 10th sub-layers of the Fuyu oil layer are laterally continuous throughout the region, with an average single-layer sandstone thickness of 4.0 to 8.7 m. The III and IV sand groups (8th to 12th sub-layers) produce oil only in the structural high areas of the eastern part of the block, while the structural low areas below the 7th sub-layer are water-bearing. The Yangdachengzi oil layer is developed only in the northern and western parts of the block at local well points. The reservoir depth ranges from 600 to 750 m. This area possesses a substantial oil-bearing zone characterized by significant geological reserves and a reserve abundance of 165.77 (104 t/km2). The average porosity is recorded at 25%, while the average permeability measures 201.8 × 10−3 µm2. The initial formation pressure stands at 7.5 MPa, with a saturation pressure of 6.2 MPa. At the surface, the crude oil exhibits a density of 0.882 g/cm3 and a viscosity of 77 mPa·s; conversely, the viscosity for formation crude oil is noted to be 6.8 mPa·s. Additionally, the wax content is quantified at 14.1%, and the pour point is determined to be at 16.2 °C.
3.2.2. Development Status and Existing Problems of Block Mu146
The Block Mu146 has undergone 45 years of waterflood development and rolling adjustments. Based on its development history, it can be divided into four development stages: the initial high production and low water cut stage, the stage of effective waterflooding with high and stable production, the stage of injection and production system adjustments, and the comprehensive adjustment stage. Initially, a diamond-shaped reverse nine-point area water injection well pattern was adopted with a well spacing of 600 m and a well density of 6.1 wells/km2. This was later changed to a linear water injection development method with a 200 m × 100 m spacing. Currently, Block Mu146 comprises a total of 69 oil wells and 46 water wells, yielding daily liquid production rates amounting to approximately 1717 tons alongside daily oil production reaching up to about 42 tons. Presently observed formation pressure has increased to an estimated value of around 8.5 MPa; however, it should be noted that the current oil recovery rate remains low at just about 0.3%. Furthermore, there has been an alarming rise in overall water cut levels, which have now reached 98.2%, coupled with a recovery degree standing at 50.9% and an unfavorable water-to-oil ratio calculated as 7.4. The efficiency regarding water injection practices appears suboptimal due to low utilization rates along with considerable conflicts between injection activities and production outputs being reported within this block’s operations—indicating that it has entered into a phase marked by rapid declines in production accompanied by high or extremely high water cuts.
The eastern part of the Block Mu146 is controlled by a reverse normal fault, with the main productive oil layers being the 4th, 7th, 9th, 10th, and 11th sub-layers. These layers have significant effective thickness, a continuous oil-bearing distribution, and multiple developed sets of strata. The reservoir has high abundance, high initial production, and significant oil and gas accumulation. According to core and saturation data, the remaining oil is widely distributed, with localized accumulations, and there is still considerable potential for further exploitation.
Currently, the Block Mu146 faces prominent development challenges, including severe water washing in the oil layers, significant intra-layer and inter-layer conflicts, imbalanced injection and production profiles, deteriorating sweep efficiency, high water consumption rates in recent years, and serious ineffective water circulation. The overall waterflooding effect is unsatisfactory, making it difficult to further extract remaining oil and stabilize or increase production. Enhancing the recovery rate is also challenging.
In light of these challenges—including the existing development conditions as well as potential demands on this particular oilfield—it becomes imperative to reassess current developmental strategies and methodologies employed therein urgently. Research efforts aimed towards expanding swept volumes while simultaneously enhancing both oil displacement efficiencies and improving overall performance during waterflooding processes will play crucial roles in moving forward toward increasing recovery rates effectively.
3.2.3. NSSNFHP Scheme Design
The Mu146-61 well increases production by adopting the method of non-steady-state nitrogen foam energy-storage huff-and-puff and sweep-efficiency improvement. In the early stage, non-steady-state nitrogen foam is injected for energy storage to raise the formation energy and expand the swept volume. In the later stage, the injected foaming agent enhances the oil-washing ability. Finally, clean water is injected to displace the nitrogen and foaming agent in the wellbore into the interior of the formation, achieving a production increase through huff-and-puff.
Segment Plug Design
The segment plug design of non-steady-state nitrogen foam energy storage, wave adding and efficiency-enhancing huff-and-puff is divided into the following three parts:
- (1)
NSSNFHP slug: Improve reservoir energy and expand swept volume through non-steady-state N2 foam.
- (2)
Wash oil plug: Implementing foaming agent systems can significantly reduce interfacial tension associated with crude oils located within near-wellbore zones along deep matrix formations, thereby facilitating improved flow capabilities for said crude oils.
- (3)
Displace slug: Inject clean water to displace nitrogen in the wellbore and foam agent into the formation.
Injection Parameter Design
- (1)
The injection volume of non-steady-state nitrogen foam is 1340 m3 (underground volume, with 335 m3 of foaming liquid; 1005 m3 of underground nitrogen, which is equivalent to 73,900 m3 on the ground).
The design of the usage amount of the non-steady-state nitrogen foam system is calculated using the directional method:
In the formula: R—Foam sweep radius (m); H—Effective thickness of the oil layer (m); Φ—Porosity (%)
The effective thickness of the oil layer is measured at 11.8 m, with a porosity of 25%. The radius for foam sweeping has been determined to be 12 m. Calculations indicate that the total volume of the foam system amounts to 1340 m3 (calculated under subsurface conditions), comprising 335 m3 of foaming liquid and 1005 m3 of underground nitrogen. When converted to standard cubic volume at surface conditions, this equates to approximately 73,900 m3.
Calculation of nitrogen conversion coefficient:
Standard state: Temperature is 273.15 K (0 °C), and pressure is 0.1 MPa; Underground state of the Mu146-61 test area in November 2023: Temperature is 313.15 K (40.0 °C), and pressure is 8.43 MPa. The calculated conversion coefficient is 73.53 Nm3/m3, which means 73.53 Nm3 is converted into 1 m3 of underground volume.
- (2)
The volume of the foaming agent is 44 m3.
- (3)
The volume of the injected water is 22 m3.
Calculated according to the formula and combined with the geological conditions of this well, the injection volume design is shown in
Table 7 (During the construction process, the construction parameters shall be appropriately adjusted according to the pressure change).
- (4)
To ensure the safety of the wellhead and injection equipment, the anticipated construction pressure will not exceed 80% of the wellhead’s pressure-bearing capacity, which is 20 MPa.
- (5)
The technical essence of the NSSNFHP test encompasses three core elements: 1. Implementation of a high-emulsification, high-foaming, and ultra-stable foam flooding system; 2. Daily alternating gas–liquid injection cycles with 24 h phase intervals; 3. Slug-stage stepped large-displacement gas injection followed by static diffusion, creating pulsed pressure fluctuations within the reservoir. This operational sequence promotes foam emulsification and high-efficiency regeneration, ultimately forming non-steady-state regenerative nitrogen foam. The resulting process effectively enhances swept volume while improving oil displacement efficiency through optimized foam propagation dynamics. Considering the economic benefits, the construction of this well avoids the peak electricity consumption period. From the 1st to the 13th day, the nitrogen foam huff-and-puff slug is injected. From the 1st to the 5th day, at night, 4860 m3 of nitrogen is injected at an injection rate of 600 m3/h for 8.1 h, and then the injection is stopped for more than 4 h to observe the wellhead pressure. During the day, after the pressure drops, 22 m3 of foaming liquid is injected at an injection rate of 5 m3/h for 4.4 h. From the 6th to the 13th day, at night, 6200 m3 of N2 is injected at an injection rate of 800 m3/h for 7.8 h, and then injection is stopped for more than 4 h to observe the wellhead pressure. During the day, after the pressure drops, 22 m3 of foaming liquid is injected at an injection rate of 5 m3/h for 4.4 h. On the 14th day, 44 m3 of the foaming agent for the oil-washing slug is injected at an injection rate of 5 m3/h for 8.8 h. On the 15th day, 22 m3 of clean water for the displacement slug is injected at an injection rate of 5 m3/h for 4.4 h. During this period, the pressure should be monitored at any time, and the dosage of the foaming agent should be dynamically adjusted according to the pressure change.
Soaking Time Design
The relationship between the pressure drop during soaking and the production increase of a single well has been evaluated in previous tests. When the pressure drop reaches 80% (the ratio of the final pressure after reduction to the pressure at the end of construction) [
14], it indicates optimal stimulation effects from these measures. Based on comprehensive analysis, an optimal soaking time is determined to be between 7 to 10 days. During this soaking process, changes in wellbore and reservoir pressures primarily result from gas diffusion; greater pressure drops facilitate better gas diffusion within the reservoir, leading to more effective contact with crude oil.
3.2.4. Evaluation of the Effect of NSSNFHP Test
Injection Situation
The construction, connection of on-site equipment and preparations for injection operations are completed in accordance with the injection process flowchart of the NSSNFHP test in Well Mu146-61 see (
Figure 5). Upon completion of injection operations for Well Mu146-61, the casing gas injection pressure was recorded at 13.07 MPa, while foaming liquid injection occurred at a pressure of 6.00 MPa. Consequently, wellhead tubing pressure increased significantly from 0.50 MPa to 5.70 MPa, demonstrating a notable effect on increasing pressure.
Chloride Ion Content
The changes in salinity in the test area before and after the NSSNFHP test in Well Mu146-61 are shown in
Table 8. After the NSSNFHP test, the average chloride ion content of five oil wells increased by 540.2 mg/L, and the average total salinity increased by 1194.2 mg/L. This indicates that the NSSNFHP test in this well has expanded the swept volume, mobilized the previously unswept oil-bearing areas, and played a “displacement” role in the oil well drainage [
35].
Test Results
The comprehensive development curve of Well Mu146-61 is shown in
Figure 6. The production of the Mu146-61 well group shows the characteristics of “three increases and one decrease”, with obvious oil production increase and a positive trend. Specifically, after the operation, the liquid production increased from 47.38 m
3 to 82.70 m
3, an increase of 74.55%. In February 2025, the water-cut ratio decreased from 97.64% to 96.46%, while daily oil production rose from 1.12 t to 2.92 t—an increase of 1.80 t (161%). Additionally, the flowing fluid level increased from −242 m to −220 m. For adjacent Well Mu128-64, following operations, daily liquid production escalated from 11.50 m
3 to 17.54 m
3; concurrently, the water cut reduced from 98.30% to 96.62%. Daily oil production for this well also saw an increase from 0.20 t to 0.60 t in February 2025—a rise of approximately 0.4 t (200%). The flowing fluid level experienced a decline from −83 m down to −266 m.
The validity period of the NSSNFHF field test is more than 12 months. Judging from the on-site sampling on 29 March 2024, the produced fluid was a khaki-colored emulsified oil liquid, which confirmed that the foaming agent has strong emulsifying ability. The NSSNFHP test has expanded the swept volume, mobilized the previously unswept oil-bearing areas, and played a “displacement” role in the adjacent wells of the oil well row (see
Figure 7).