1. Introduction
As oil and gas exploration has advanced to greater depths (> 4500 m), deep oil and gas resources have become an important strategic resource for large-scale reserves and production growth [
[1],
[2],
[3],
[4]]. Water flooding is a key method for efficient oil extraction in deep and ultradeep oil reservoirs. However, during reservoir development, high-permeability zones such as fractured bands and microcracks often form planar and vertical flow channels, leading to the direct flow of injected water to wells, which results in inefficient or ineffective circulation and significantly reduces the sweep volume and recovery rates [
5,
6]. Therefore, enhancing oil recovery (EOR) and increasing oil production have emerged as critical technical challenges in contemporary reservoir development.
Significant progress has been made in chemical profile-control agents used in reservoirs to EOR [
[7],
[8],
[9]]. The main agents used for profile control include water-soluble gels and freeze gels; the methods based on these agents primarily use physical blocking, trapping, and adsorption to block high-permeability reservoirs, and are suitable for fractured reservoirs [
[10],
[11],
[12]]. However, they present challenges in deep reservoirs with high temperatures (> 140 °C), high salinities (> 1.0 × 10
5 mg·L
−1), and complex media conditions, particularly in ultradeep layers (> 9000 m) with ultrahigh temperatures (> 180 °C) and ultrahigh salinities (> 2.0 × 10
5 mg·L
−1), as their thermal stability and salt resistance are poor. This issue makes EOR in such reservoirs particularly difficult [
12,
13].
In recent years, nanomaterials have gradually become a new research hotspot in the field of EOR [
[14],
[15],
[16]]. Ma et al. [
17] developed a high-temperature resistant and mechanically stable composite hydrogel (STP) by incorporating hard glucan (Slg) and 2,2,6,6-tetramethylpiperidine-
N-oxyl (TEMPO)-oxidized cellulose nanofibers (TOCN) into a polyacrylamide (PAM) hydrogel. STP demonstrated good water-blocking performance in a 140 °C saline environment. In addition, as an alternative to traditional silicate gels, the long-term water-blocking performance of nano-silica at 95 °C, demonstrates its significantly higher sealing efficiency compared with PAM/polyethyleneimine (PEI) gels [
18]. However, these materials still have limitations under high-temperature and high-salinity conditions. Nanographite, a two-dimensional (2D) carbon-based nanomaterial, has excellent properties, such as high temperature resistance, flexibility, self-lubrication, thermal stability, and ease of modification [
19,
20]. When nanographite particles are embedded in branched polymer chains or participate in cross-linking reactions in gel systems, their high-temperature and high-salinity resistance improves significantly; this is attributed to the carbon-based backbone structure and nanoscale size of the nanographite, which has been confirmed through extensive research [
21,
22]. Interestingly, we found that nanographite particle systems are thermodynamically unstable due to interparticle interactions, which readily promote aggregation, leading to a transition of the particle size from the nanoscale to the microscale [
23,
24]. This phenomenon may hold significant application potential in deep-reservoir high-permeability-zone water-channel control and EOR.
The application potential of nanographite in EOR through profile control in deep reservoirs is due to its unique carbon- and oxygen-based backbone structure, which exhibits excellent stability under the complex and harsh conditions of deep reservoirs [
25]. The flexible and self-lubricating layered structure of nanographite, in combination with its nanoscale size, provides excellent injectivity and mobility in low-permeability reservoir pore media [
26]. The large surface area and high surface energy of nanographite allow it to form aggregates in liquid-phase environments, resisting formation flow scouring in water flooding. However, several key scientific and technical challenges must be overcome to enable the practical application of nanographite in profile control: ① The aggregation process is uncontrollable. Owing to the inherent thermodynamic instability of nanographite, aggregation can occur before it reaches the targeted reservoir area, affecting its regulatory effect. Controlling the transition from dispersion to aggregation and utilizing the aggregation characteristics to block water channeling is a key challenge. ② The aggregation characteristics and reservoir regulatory mechanisms are unclear. When in contact with formation brine, nanographite is significantly influenced by salt ions. Understanding its aggregation characteristics, dynamic distribution, and occurrence in complex porous media is critical for its application in EOR.
This study addresses the challenges of EOR in deep oil reservoirs under high-temperature and high-salinity conditions by proposing a strategy based on the thermosalinity-responsive aggregation of functionalized nanographite. This strategy aims to improve the high-temperature dispersibility of nanographite, delay its aggregation process, and thus enhance its EOR after deep migration in reservoirs. First, a green and controllable electrochemical oxidation method is used to prepare nanographite units with multiple active grafting sites (EGO), and modified nanographite (MEGO) is synthesized via amidation reactions. The structural properties of the obtained MEGO are then analyzed. Next, MEGO is combined with a disodium naphthalene sulfonate (DNS) surfactant to form the MEGO system. The system’s dispersibility and responsiveness under high-temperature and high-salinity conditions are systematically investigated, and its aggregation dynamics at the molecular level are elucidated. Finally, the flow profile-control process in porous media is described, and the pore mobilization behavior before and after profile control is quantitatively analyzed to reveal the EOR mechanism in the reservoir. This study provides a new method for EOR and offers theoretical and technical support for the efficient development of deep oil reservoirs.
2. Experimental procedure
2.1. Materials
High-purity graphite rods with a carbon content of 99.9% were obtained from Beijing Jinglong Special Carbon Technology Co., Ltd., China. Analytical grade anhydrous ethanol (99.7%), sodium hydroxide (96.0%), sodium chloride (99.5%), anhydrous calcium chloride (96.0%), and magnesium chloride hexahydrate (98.0%) were sourced from China National Pharmaceutical Group Chemical Reagents Co., Ltd. Analytical-grade polyether amine (PEA; 99%), with a molecular weight of 2070 g·mol−1, was supplied by Huntsman Chemical Trading (Shanghai) Co., Ltd., China. Analytical-grade 1-ethyl-(3-dimethylaminopropyl) carbodiimide hydrochloride (99%) and N-hydroxysuccinimide (NHS; 98%) were purchased from Shanghai Macklin Biochemical Technology Co., Ltd., China. Analytical-grade DNS (93%), with a molecular weight of 318.34 g·mol–1, was obtained from Guangdong Wengjiang Chemical Reagents Co., Ltd., China. Further details are provided in Section S1 in Appendix A.
2.2. Synthesis of the MEGO system
An aqueous solution of 0.3% NaOH was used as an electrolyte, with a graphite anode, a platinum cathode, and a direct current (DC) power supply connected and placed in a quartz electrolysis cell. A magnetic stirrer was added, and the system was set to 30 °C with stirring. Electrolysis was performed at 30 V for 6 h. The resulting solution was centrifuged, washed, and dried to obtain EGO. The optimization of the preparation process is detailed in the Sections S2–S6 in Appendix A. A certain amount of EGO was then dispersed in solution and treated with strongly acidic cation-exchange resin to protonate the material, thereby converting the carboxylate groups into carboxyl groups and removing the sodium ions. The pH was adjusted to 6 using a phosphate buffer. The treated EGO dispersion was then transferred to a three-necked flask, and 1-ethyl-3-(dimethylamino)propyl carbodiimide hydrochloride (EDC·HCl) was added under nitrogen protection at 25 °C. After stirring for 10 min, NHS was introduced, and stirring was continued for 1 h to complete the activation. Subsequently, PEA was added, heated, and refluxed, and the amidation reaction was carried out for 12 h, yielding MEGO.
Fig. 1 illustrates the synthesis procedure.
MEGO was used as the primary agent and was blended with eight types of surfactants at specific ratios. Deionized water was then added, and the resulting solution was stirred using a magnetic stirrer until the surfactant was fully dissolved. After various surfactants were screened, a DNS surfactant was selected to construct the MEGO system; the optimization process is described in detail in Section S7 in Appendix A. For comparison, EGO was also formulated with a DNS surfactant to construct the EGO system.
2.3. Characterization
The particle size of the samples was determined by dynamic light scattering (DLS) and laser diffraction measurements. The microstructure and elemental distribution were characterized by means of scanning electron microscopy (SEM) coupled with energy dispersive spectroscopy (EDS). Infrared spectra were recorded using a Fourier-transform infrared (FTIR) spectrometer. Raman spectra were obtained using a Raman spectrometer. Elemental analysis was performed using X-ray photoelectron spectroscopy (XPS). Thermogravimetric analysis (TGA) was conducted in the temperature range of 25–600 °C. The evolution of the surface morphology was examined using an atomic force microscope (AFM). The suitability of the samples for practical application was assessed using oil and gas flow analysis systems, nuclear magnetic resonance (NMR) spectroscopy, and mercury intrusion porosimetry. Further details are provided in Section S8 in Appendix A.
2.4. Evaluation of dispersibility in the MEGO system
Before the system reaches the target area for deep reservoir profile control, dispersibility is one of the key factors affecting deep migration. A series of experiments were conducted to evaluate the dispersibility of the MEGO system under high-temperature and high-salinity conditions. The MEGO and EGO systems were prepared in fresh water, and the particle size changes were investigated at different temperatures (140, 180, and 220 °C) under both static and dynamic conditions. Additionally, the particle size variations and particle stability were assessed under combined conditions (140 °C, 100 000 mg·L−1; 180 °C, 200 000 mg·L−1; 220 °C, 300 000 mg·L−1) to evaluate their performance in high-temperature and high-salinity environments. Fresh water with a mineralization of 400 mg·L−1 (Na+ concentration of 200 mg·L−1; Ca2+ and Mg2+ concentrations of 50 mg·L−1; Cl– and SO42– concentrations of 150 mg·L−1) was used.
2.5. Thermosalinity response validation of the MEGO system
The particle size variation under the combined influence of temperature and salinity has a decisive effect on reservoir profile control. A series of thermosalinity-response experiments were conducted under various conditions. The effects of different temperatures, salt concentrations, and salt ion types (sodium chloride, calcium chloride, and magnesium chloride) on the aggregation process were systematically investigated, providing a comprehensive analysis of the thermosalinity-responsive behavior and aggregation dynamics of these materials. The aggregation behavior was further evaluated under simulated formation water conditions of the East River sandstone reservoir in the Tarim Basin to assess the performance of the MEGO system in realistic reservoir environments. To quantitatively describe the particle size variation, the aggregation rate was introduced as a key parameter under different experimental conditions. This investigation revealed the aggregation dynamics induced by temperature and salt, as well as the thermosalinity-responsive mechanism of the MEGO system. Further experimental details are provided in Section S9 in Appendix A.
2.6. Molecular dynamics simulations of MEGO aggregation
In this study, molecular dynamics simulations of the aggregation processes of MEGO and EGO were performed using Materials Studio 8.0. The EGO, MEGO, and DNS molecular models were constructed based on the Lerf–Klinowski model to systematically investigate the aggregation behavior of the EGO and MEGO materials, as well as the impact of DNS on their aggregation. For the MEGO system, the effects of the dispersant DNS, reservoir temperature, salt ion type, and concentration on the aggregation characteristics and microdynamic behavior were explored. To simplify the calculations, the PEA molecular chain was reduced to include only the amide bond and one repeating unit. Because of computational resource limitations, the aggregation process of only two graphite molecules was simulated in this study. Further details are provided in Section S10 in Appendix A.
2.7. Potential EOR experiments with the MEGO system
This study utilized a core physical simulation displacement device to investigate the injection and profile-control effects of the MEGO system in cores with different permeabilities. The initial permeability of the core was measured by injecting simulated water, followed by injection of the MEGO system and aging. Then, simulated water was reinjected to measure the permeability after profile control. The resistance factor and residual resistance factor were calculated to evaluate the regulatory effect of the MEGO system on the core permeability.
To further analyze the oil-displacement effect of the MEGO system, NMR technology was used to study the mobilization of crude oil in cores with different pore sizes. A core saturated with simulated water (H2O) was placed in the core holder, followed by saturation with simulated oil. Initial T2 signals and images were collected. Then, simulated water (D2O) was injected at a constant flow rate for a water-flooding experiment, and T2 signals and images were recorded. Next, the MEGO system was injected, and the core was aged. Simulated water (D2O) was injected again, and water-flooding T2 signals and images were collected to analyze the EOR effect of the control system. Further details are provided in Section S11 in Appendix A.
3. Results and discussion
3.1. Characterization of MEGO
MEGO was successfully synthesized by covalently grafting PEA onto the surface of EGO via an amidation reaction between the carboxyl groups on the EGO surface and the amine groups at the terminus of the PEA (
Fig. 2(a)). The microstructure of the resulting MEGO was characterized using transmission electron microscopy (TEM). The EGO sample exhibited a smooth 2D sheet-like structure with clear edges, which is consistent with the literature [
27]. In contrast, the surface of the MEGO sample was rougher, primarily because of the hydrogen bonding and hydration layer effects in the aqueous solution, which caused the PEA chains to stretch. Upon drying during the sample preparation process, the PEA chains collapsed, giving rise to a rough surface (Figs. S1(a) and (b) in Appendix A) [
28]. As shown in
Fig. 2(b), the uniform distribution of carbon (C), nitrogen (N), and oxygen (O) in the MEGO sample confirms its successful preparation. The particle size distribution (
Fig. 2(c)) shows that the particle sizes of EGO and MEGO are 305 and 362 nm, respectively. The increase in the particle size of MEGO is attributed to the grafting of PEA chains onto the edges of the EGO via the amidation reaction [
29].
To analyze the surface functional groups and structural characteristics of the target product, FTIR, XPS, and Raman analyses were conducted. The FTIR spectrum of EGO shows an O–H stretching vibration peak at 3430 cm
–1, a C=O stretching vibration peak at 1720 cm
–1, a C=C skeletal vibration peak at 1629 cm
–1, and a C–O stretching vibration peak at 1050 cm
–1, confirming the presence of functional groups such as the carboxyl, hydroxyl, and epoxy groups, which provide active sites for subsequent chemical grafting. In the FTIR spectrum of PEA, the absorption peaks at 3431, 2884, 1634, and 1114 cm
–1 correspond to characteristic molecular structures, and the new peaks at 2922 cm
–1 (C–H stretching vibration) and 1631 cm
–1 (amide bond stretching vibration) in the MEGO spectrum, along with the C–O–C stretching vibration at 1081 cm
–1, further confirm that PEA was successfully grafted onto the EGO surface via amidation (
Fig. 2(d)) [
30,
31]. The newly observed N1s peak further confirms the successful grafting of PEA, with a binding energy of 401.0 eV corresponding to the amide bond (
Figs. 2(e) and
(f)). In the C1s spectrum of EGO, the peaks at 285.4, 287.3, 288.1, and 289.1 eV correspond to C=C, C–O, C=O, and O–C=O, respectively, indicating the presence of various oxygen-containing functional groups and unsaturated bonds on the EGO surface. These features not only enhance the stability and dispersion of MEGO in water but also provide conditions for the adsorption of dispersing agents. The C1s spectrum of MEGO shows a significant change in the functional group distribution: the fitting peaks at 285.7, 287.0, 288.1, and 289.4 eV correspond to C=C, C–O, C=O, and O–C=O, respectively (
Fig. 2(g)). The decrease in the C=C intensity and increase in the C–O intensity are likely due to the changes induced by the alkyl chains and ether bonds in PEA, while the decreases in the C=O and O–C=O intensities suggest that the carbonyl and carboxyl groups may have undergone reduction and amidation reactions, respectively [
32,
33]. Raman spectroscopy analysis revealed that the D peak at 1351 cm
–1 and the G peak at 1596 cm
–1 for both EGO and MEGO correspond to the disordered (sp
3 hybridization) and ordered (sp
2 hybridization) structures of the graphite lattice, respectively. After modification, the
ID/
IG ratio (the intensity ratio of the D peak to the G peak in Raman spectroscopy, indirectly reflecting changes in the ordered structural regions of the graphite surface) decreased from 0.97 to 0.89, indicating that thermal reduction during modification increased the degree of structural order and reduced the number of surface defects in the graphite lattice. Further calculations revealed that the crystallite size increased from 39.7 to 43.3 nm, providing more sites for surface adsorption of the dispersing agent (
Fig. 2(h)) [
34,
35].
The harsh conditions of deep oil reservoirs impose higher demands on the thermal stability and salt tolerance of nanographite materials. TGA results (
Fig. 2(i)) revealed that EGO exhibited a weight loss between 54.4 and 107.6 °C, corresponding to the evaporation of bound water, while a second weight loss between 182.6 and 223.0 °C was attributed to the decomposition of the oxygen-containing groups. The single-stage weight loss of PEA observed between 352.5 and 409.1 °C reflects its excellent thermal stability. For MEGO, the first stage of weight loss occurred between 188.9 and 225.0 °C, with an onset temperature 6.3 °C higher than that of EGO. The second stage of weight loss, observed between 307.5 and 391.3 °C, is attributed to the decomposition of the covalently grafted PEA, accounting for 33.33% of the total weight loss. This improvement is likely due to the synergistic effect of the physical protection and chemical grafting from PEA, which reduces the number of oxygen-containing groups and enhances the high-temperature stability of the graphite, significantly improving the thermal resistance of MEGO.
3.2. Dispersibility and stability of the MEGO system
The complex and harsh conditions of deep oil reservoirs impose stringent requirements on the deep migration capabilities and thermal and salt resistance of materials. The dispersibility of nanoparticles in liquid-phase environments directly influences their migration through pore channels. MEGO exhibits excellent dispersibility at room temperature; however, its performance deteriorates significantly under high-temperature conditions, as evidenced by the particle size increasing from 0.305 and 0.362 to 15.21 and 10.96 μm after 24 h of aging (
Fig. 3(a)); thus, it fails to meet the requirements for deep migration and profile control. Therefore, it is necessary to further enhance the dispersibility and stability of MEGO. DNS was selected as the optimal dispersant due to its ability to increase the zeta potential of the MEGO particles, promote electrostatic repulsion, and maintain dispersion stability under high-temperature conditions. Additionally, DNS is more compatible with the grafted PEA chains than other dispersants, ensuring stable dispersion even in complex reservoir environments. The optimal ratio of MEGO to DNS was determined: When the DNS concentration reached 1.0%, the particle size was significantly reduced, and the absolute value of the zeta potential increased with increasing DNS concentration until stabilization (
Fig. 3(b) and Fig. S2(a) in Appendix A), indicating a marked improvement in the high-temperature dispersibility. Conversely, as the MEGO concentration increased, the absolute value of the zeta potential gradually decreased (Fig. S2(b) in Appendix A)), and the particle size significantly increased (
Fig. 3(c)), suggesting that higher MEGO concentrations negatively impact dispersibility. Based on these results, the optimal composition of the MEGO profile-control system was determined to be 0.1% MEGO + 1.0% DNS.
The variation in the aggregation particle size of a system under high-temperature conditions during seepage in deep oil reservoirs is a critical factor influencing its deep migration performance. In this study, the EGO and MEGO systems were prepared with fresh water to investigate the effects of static and dynamic conditions on aggregation behavior at different temperatures. Static experiments revealed that prolonged aging time and elevated temperatures accelerated the transition of particle size from the nanoscale to the microscale, reflecting reduced dispersibility at high temperatures. At the simulated reservoir temperatures, the particle size of the EGO system increased rapidly and stabilized between 46.77 and 49.21 μm, indicating poor high-temperature dispersibility and a limited ability to migrate through deep micropores and throats (
Fig. 3(d)). In contrast, the MEGO system exhibited significantly slower particle size growth at 140 °C, with a maximum particle size of < 10 µm at 200 °C, demonstrating superior dispersibility (
Fig. 3(e)). Considering the shearing effects in porous media, dynamic experiments further revealed that the EGO system undergoes rapid particle size growth under dynamic conditions, stabilizing at relatively large sizes ranging from 27.48 to 44.58 µm (
Fig. 3(f)). These findings indicate that the high-temperature dispersibility of the EGO system is constrained by its microstructure and that the physical effects of the dispersants provide limited improvement. Conversely, the MEGO system showed significantly reduced particle size growth under dynamic oscillation. Within the first five days, its particle size remained at the submicron level (< 1 µm); moreover, even at elevated temperatures of 200 or 220 °C, the maximum stable particle size remained as small as 4.61 µm, ensuring smooth transport through complex reservoir micropores and throats (
Fig. 3(g)). In summary, the synergistic effects of covalent grafting and noncovalent interactions effectively enhance the high-temperature dispersibility of the MEGO system, while the shearing effect under dynamic conditions effectively suppresses aggregation behavior [
36,
37].
Further investigations under high-temperature and high-salinity conditions revealed that the particle size of the MEGO system increased progressively with aging time and eventually stabilized, maintaining a range of 55.51–61.80 μm after 10 days (
Fig. 3(h)). Elevated temperature and salinity accelerated the aggregation process, shortening the transition time of the particle size from the nanoscale to the microscale. This phenomenon can likely be attributed to the increased molecular motion at higher temperatures and the salt-induced compression of the electrical double layer, which reduces interparticle electrostatic repulsion and promotes aggregation. However, no evidence of material decomposition or particle size reduction was observed, confirming the exceptional thermal and salt resistance of the system. As shown in
Fig. 3(i), after 30 days of aging, the system exhibited an aggregation factor greater than 140, with the formation of stable aggregates capable of effectively controlling water channeling. These findings highlight the significant potential of the MEGO system for profile control in deep oil reservoirs.
3.3. Thermosalinity-responsive properties of the MEGO system
In actual reservoirs, the leading edge of the injected MEGO system gradually encounters formation brine with varying salinities. Owing to temperature changes and osmotic pressure, ion diffusion occurs, resulting in complex aggregation behaviors induced by salt ions. This study systematically investigated the effects of salt ion type and concentration on the aggregation characteristics at different temperatures and elucidated the underlying mechanisms based on colloidal stability theory. As shown in
Figs. 4(a)–(e), the particle size of the MEGO system increased progressively with aging time at 140–220 °C. Salinity (i.e., NaCl concentration) significantly accelerated the transition from the nanoscale to the microscale, highlighting salt-induced aggregation behavior. This trend persisted consistently across different temperatures, indicating the universal influence of salinity on particle aggregation. This behavior is attributed to particle aggregation driven by collisions in the solution. As collisions persist, the aggregate size continues to grow. Once the gravitational potential energy of the aggregates becomes higher than their thermal kinetic energy, precipitation occurs, reducing the collision frequency and decelerating particle size growth.
Figs. 4(f)–(h) demonstrate that the aggregation process of the MEGO system is jointly regulated by the temperature and salt ion concentration. At different temperatures, the aggregation rate exhibits a distinct trend of rapid linear growth at low salinities and slower linear growth at high salinities. The intersection of these two stages corresponds to the critical aggregation concentration (CAC). At low salinity, the system undergoes reaction-controlled aggregation, characterized by a high repulsive energy barrier where only high-energy collisions can induce aggregation [
37]. At high salinity, the system transitions to diffusion-controlled aggregation, where irreversible aggregation occurs directly upon collision. As the temperature increases, the CAC value induced by NaCl decreases from 63 865 to 32 575 mg·L
–1, while the CAC values for CaCl
2 and MgCl
2 decrease from 2404 to 1180 mg·L
–1 and from 2763 to 1245 mg·L
–1, respectively. These results indicate that divalent ions (Ca
2+ and Mg
2+) have a stronger promoting effect on the aggregation of the MEGO system compared with monovalent ions (Na
+). This is primarily because divalent ions are able to compress the electric double layer surrounding the particles, reducing the electrostatic repulsion between them. Due to their higher charge density, divalent ions can more effectively neutralize the negative charges on the particle surfaces, thereby promoting interactions between particles and driving aggregation. In addition, divalent ions can induce a bridging effect between particles. When divalent ions form ionic bonds with functional groups on the particle surfaces, they effectively link multiple particles together, forming larger aggregates [
38].
To quantify the influence of divalent ions on the aggregation of the MEGO system, the Schulze–Hardy rule exponent
n was calculated based on the CAC values (Section S11 in Appendix A). As shown in
Fig. 4(i), the Schulze–Hardy exponents for both divalent ions (5.23–5.71) are lower than the theoretical value of 6.0, suggesting that non-Derjaguin–Landau–Verwey–Overbeek (DLVO) forces are involved, in addition to double-layer compression [
39]. Analysis of the system composition indicates that the steric hindrance effect introduced by PEA and the dispersant DNS reduces the effective collision probability between particles, thereby delaying the aggregation process [
24,
40]. Furthermore, Ca
2+ has a stronger aggregation-inducing ability than Mg
2+, likely due to its larger relative atomic mass and specific adsorption effects [
41].
The aggregation behavior of the MEGO system was investigated under simulated conditions in the Donghe sandstone reservoir in the Tarim Basin using composite salt ion solutions. As shown in
Fig. 5(a), the particle size increased rapidly with aging time before stabilizing at 52.09–60.86 μm across different salinities. Compared with the aggregation rate under conditions with single salt ions, the aggregation rate under composite salt conditions increased significantly from 3.918 to 12.348 μm·d
–1 as the salinity increased, indicating a pronounced salt-ion-induced effect on the aggregation process (Fig. S3 in Appendix A). The stabilization of the particle size may be attributed to the limited number of particles, the prevention of further aggregation by diffuse double-layer repulsion, and the dynamic balance between the adsorption and desorption of the particles under high salinity (
Fig. 5(b)) [
42].
The aggregation morphology of the MEGO system was further characterized under conditions of deionized, low-salinity, and high-salinity water. In fresh water, the system remained well-dispersed and was dominated by 2D sheet-like structures with clear edges and a deposition height of 8.2 nm, reflecting strong electrostatic repulsion and hydration-layer steric hindrance that prevented particle aggregation (
Fig. 5(c)). In low-salinity water, partial aggregation occurred, and the overall height increased to 25.3 nm, suggesting that salt ions weakened interparticle repulsion via electrostatic shielding, leading to metastable aggregation (
Fig. 5(d)). In high-salinity water, aggregation became pronounced, increasing the deposition height to 43.2 nm. The aggregates formed irregular clusters, while 2D sheets became sparsely distributed. Electrostatic repulsion is fully screened under high salinity, reducing interparticle distances and promoting interlayer water expulsion and irreversible aggregation (
Fig. 5(e)). Based on these findings, fresh water is recommended for pre-flushing or as a pre-stage water slug to establish a salinity gradient. This approach can delay MEGO aggregation, enabling deep reservoir profile control.
3.4. Simulation of the aggregation kinetics of the MEGO system
Molecular dynamics simulations were employed to study the interactions between the components of the system. In the study of aggregation behavior, EGO molecules transitioned from “point contact” to “line contact” and eventually formed an “optimal face contact” aggregation structure driven by hydrophobic interactions (Fig. S4(a) in Appendix A). In contrast, due to PEA grafting on its edges, MEGO exhibited constrained contact pathways influenced by molecular entanglement (Fig. S4(b) in Appendix A). The centroid distance stabilized at 6.278 Å, with an interaction energy of −184.53 kcal·mol
–1, indicating a high binding strength that promotes unstable aggregation (Fig. S4(c) in Appendix A). The interlayer spacing increased to 7.949 Å, and the binding strength decreased to –138.03 kcal·mol
–1, delaying the aggregation process (Fig. S4(d) in Appendix A). This behavior can be attributed to the enhanced high-temperature dispersibility of the grafted PEA through steric hindrance and hydrophobic entanglement. At collision energies lower than the hydration layer resistance, steric hindrance dominated; when the collision energy exceeded the hydration resistance, hydrophobic alkyl chain entanglement locked the layer positions and expanded the interlayer spacing, further inhibiting aggregation [
43].
DNS further optimized the regulatory effect. Upon adsorption onto EGO surfaces, DNS caused the system to bypass the “point contact” stage through electrostatic repulsion and cluster blocking, directly transitioning from “line contact” to “face contact” (Fig. S5(a) in Appendix A). For MEGO, the synergistic effects of DNS adsorption and PEA entanglement created larger interlayer angles, slowing the expulsion of water molecules and DNS (Fig. S5(b) in Appendix A). The stabilized centroid distance for the EGO system with DNS was 8.752 Å, primarily due to the increased interlayer spacing induced by DNS via electrostatic repulsion and steric hindrance (Fig. S5(c) in Appendix A). This significantly increased the interlayer spacing to 11.586 Å, whereas the interaction energy decreased to −66.24 kcal·mol
–1, which was notably greater than the value of −111.76 kcal·mol
–1 for EGO (Fig. S5(d) in Appendix A) [
44]. The combined effect of DNS and PEA significantly reduced interlayer attraction, enhanced electrostatic repulsion, increased steric hindrance, and reinforced hydration effects, effectively delaying aggregation and improving system dispersibility. These findings are in partial agreement with the experimental results.
The effects of the dispersant DNS, reservoir temperature, and salt ions on the microscopic dynamic behavior of aggregation in the MEGO system were further investigated. As the DNS concentration increased, the MEGO system transitioned from a “face-to-face” configuration to a more stable “layered face-to-face” configuration (
Fig. 6(a)). DNS adsorption significantly reduced the mobility of MEGO, decreasing the collision frequency and effectively delaying aggregation (
Fig. 6(b)). When 25 DNS molecules fully covered the MEGO surface, enhanced electrostatic repulsion and hydration layers reduced the interaction energy to −0.95 kcal·mol
–1, completely preventing layer contact and aggregation (
Fig. 6(c)).
The reservoir temperature had a pronounced effect on the stability of the MEGO system. As shown in
Fig. 6(d), increasing the temperature significantly enhanced the MEGO mobility and collision frequency. At 140 and 160 °C, electrostatic repulsion and hydration-layer resistance dominated, with interaction energies stabilizing at −0.95 and −2.15 kcal·mol
–1, respectively, preventing aggregation. However, at temperatures exceeding 180 °C, the weakened hydration layer resistance accelerated layer contact, causing the absolute interaction energy to increase significantly to −39.34 kcal·mol
–1 and suggesting a tendency to aggregate [
44]. Nevertheless, the entanglement of the grafted PEA and the physical barrier provided by DNS prevented the layers from forming the optimal “face-to-face” contact configuration (
Fig. 6(e)).
As shown in
Fig. 6(f), increasing the NaCl concentration caused the absolute interaction energy to increase from −10.14 to −104.37 kcal·mol
–1, with high salinity accelerating layer contact and aggregation [
45]. Compared with monovalent ions (Na
+), divalent ions (Ca
2+ and Mg
2+) exhibited stronger charge shielding and double-layer compression capabilities, significantly accelerating layer contact. Specific adsorption by Ca
2+ further enhanced the aggregation-inducing effect (
Fig. 6(g)) [
46]. As shown in
Fig. 6(h), salt ions reduced the MEGO mobility. At low concentrations, the MEGO system maintained good dispersibility and increased mobility. At high concentrations, the salt ions induced layer aggregation, increasing the molecular weight and significantly reducing mobility. The centroid distances for Na
+, Ca
2+, and Mg
2+ were 17.251, 13.605, and 14.016 Å, respectively (
Fig. 6(i)), indicating that divalent ions reduced the interlayer spacing and enhanced the binding strength more significantly than monovalent ions, which is consistent with the experimental results [
47].
3.5. Potential EOR capacity of the MEGO system
The injection performance and profile-control effects of the MEGO system in cores with varying permeabilities were evaluated through physical simulation displacement experiments. The experimental results showed that, during the subsequent water-flooding stage, the resistance factor of the core with a permeability of 1.0 × 10
–2 μm
2 increased sharply after injecting 1 pore volume (PV) of the MEGO system, indicating that the accumulation of MEGO particles caused a reduction in permeability. In contrast, the core with a permeability of 0.5 μm
2 exhibited the lowest residual resistance factor. This suggests that the profile-control effect is primarily influenced by the relationship between the aggregate particle size and the pore throat size (
Fig. 7(a)).
Fig. 7(b) presents the NMR
T2 curves of the artificial cores, reflecting their real pore structures, which are mainly distributed between small and medium-sized pores. Based on the
T2 relaxation time, the pores were classified into three categories: small pores (
T2 < 1.98 ms), medium pores (
T2 between 1.98 and 28.99 ms), and large pores (
T2 > 28.99 ms). These pore classification criteria were developed based on the relationship between the actual pore size distribution and the
T2 relaxation time in the experiment.
Fig. 7(c) further reveals a strong linear correlation between the
T2 relaxation time and the pore diameter
r, indicating that the pore diameter is less than 0.32 μm, between 0.32 and 2.31 μm, and greater than 2.31 μm for small, medium, and large pores, respectively. Further analysis revealed that, as permeability increased, the blockage rate first increased and then decreased, with optimal performance observed in the core with a permeability of 1.0 × 10
–2 μm
2 (
Fig. 7(d)). With a blockage rate greater than 70% as the preferred criterion, this indicates that the MEGO system achieves the best profile-control effect in reservoirs with a permeability not exceeding 0.1 μm
2.
NMR and magnetic resonance imaging (MRI) were used to monitor signal changes during different displacement stages to evaluate the EOR effect of the MEGO system. Initially, the simulated oil was uniformly distributed within the core, with high NMR signal amplitudes primarily concentrated in the pores with diameters greater than 0.32 μm (
Fig. 7(e)). During water flooding, the NMR signal amplitude decreased significantly, particularly in large pores, indicating that water primarily flowed through large-pore pathways, forming water channels and leaving the oil in small pores untouched. After water channeling, the NMR signal further decreased, suggesting the recovery of residual oil, although the oil in the small pores remained trapped. MRI images (
Fig. 7(f)) revealed that, after primary water flooding, the residual oil signals in the front and middle regions of the core decreased significantly, while those in the rear region remained high, indicating the formation of water channels that caused oil retention in the rear region. After water channeling profile control, the residual oil signals in the rear region decreased significantly, demonstrating that the MEGO system effectively improved oil recovery in water-flooding optimization.
Further quantitative analysis revealed that, during the primary water-flooding stage, the contribution of small pores to oil production was only 6.65%, with most oil recovery coming from large pores. After water channeling profile control and subsequent water flooding, the contribution of small pores to oil production increased to 18.81%. The recovery rate during the subsequent water-flooding stage was 26.75%, resulting in an overall laboratory recovery rate of 56.01% (
Table 1). These results demonstrate that the MEGO system effectively improves microscale heterogeneity and mobilizes residual oil in small pores, significantly enhancing oil recovery.
3.6. EOR mechanism of the MEGO system
In the development of deep oil reservoirs, prolonged water injection often leads to the formation of high-permeability water channels, leaving large volumes of residual oil trapped in low-permeability regions. The MEGO system effectively regulates these dominant channels through particle aggregation and pore plugging (
Fig. 8(a)). During the initial injection, the MEGO particles in the nanoscale dispersed state adsorb onto the rock surface but exhibit limited blocking efficiency. As the system interacts with formation brine, salt ion diffusion induces an electrostatic shielding effect, promoting the aggregation of the MEGO particles into high-strength aggregates. Specifically, salt ions induce electrostatic shielding and compress the diffuse double layer of the MEGO particles, thereby driving aggregation. In the Stern layer, salt ions (Na
+, Ca
2+, and Mg
2+) adsorb onto the MEGO surface through charge neutralization, significantly weakening interparticle electrostatic repulsion [
48]. Additionally, salt ions compress the diffuse double layer, reducing the distance of the interaction between the particles and allowing them to enter the van der Waals force-dominated regime, thereby markedly promoting aggregation (
Fig. 8(b)). These aggregates accumulate and block narrow throats, with a diameter of approximately 15.1 μm in the throats, significantly increasing the flow resistance of the dominant channels. Concurrently, MEGO adsorption in small pores further reduces the cross-sectional flow area, forcing the injected water to bypass the blocked regions and expanding the swept volume (
Fig. 8(c)).
4. Conclusions
In this study, a functionalized nanographite (MEGO) system with significant thermosalinity responsiveness was successfully developed, and its coalescence behavior and reservoir profile-control mechanisms were systematically elucidated. By combining covalent grafting of PEA and noncovalent complexation with DNS, the MEGO system demonstrated excellent thermal stability and dispersibility under high-temperature and high-salinity conditions. Particle growth was significantly delayed, resulting in the formation of stable aggregates with sizes in the range of 55.51–61.80 µm, which are suitable for deep reservoir migration and water channeling blockage. Experiments and molecular dynamics simulations revealed that the dynamic coalescence behavior of the MEGO system was synergistically regulated by the temperature and salt ions (Na+, Ca2+, and Mg2+). Divalent ions effectively promoted coalescence through double-layer compression and bridging effects, exhibiting stronger coalescence-inducing ability than monovalent ions. Core flooding experiments further confirmed the reservoir profile-control performance of the MEGO system. In reservoirs with permeabilities ranging from 21.6 to 103 mD (1 mD = 0.986923×10−15 m2), the MEGO system formed throat aggregates (∼15.1 µm) that significantly increased the flow resistance, expanded the sweep volume, and improved the total oil recovery to 56.01%. This study provides a new technological pathway for reservoir profile control in deep oil reservoirs and reveals the synergistic thermosalinity dynamic regulatory mechanisms of functional material performance, providing both theoretical insights and practical guidance for deep reservoir development and nanomaterial design.
CRediT authorship contribution statement
Caili Dai: Writing – review & editing, Supervision, Project administration, Methodology, Funding acquisition, Formal analysis, Conceptualization. Wanlei Geng: Writing – original draft, Validation, Investigation, Data curation. Jiaming Li: Writing – original draft, Validation, Investigation. Guang Zhao: Writing – review & editing, Project administration, Funding acquisition, Formal analysis, Conceptualization. Bin Yuan: Writing – review & editing, Supervision, Project administration, Methodology, Funding acquisition, Formal analysis, Conceptualization. Yang Zhao: Writing – review & editing, Project administration, Funding acquisition, Formal analysis, Conceptualization. Tayfun Babadagli: Writing – review & editing, Supervision, Methodology.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
This study was supported by the General Program of the National Natural Science Foundation of China (52074335), the National Key Research and Development Program of China (2022YFE0129900 and 2019YFA0708700), the Fundamental Research Funds for the Central Universities (23CX07003A), and the Special Funding Program for the Operational Expenses of National Research Institutions (SKLDOG2024-ZYRC-01).
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.eng.2025.04.009.
PDF
(5892KB)
