Selective Plugging and Flow Reallocation Characteristics of a Time-Controllable Silica-Reinforced Cationic Polymer Gel in High-Temperature, High-Salinity Heterogeneous Porous Media

Abstract

Preferential flow through high-permeability pathways remains a central challenge for polymer-based conformance control in reservoirs where high temperature, high salinity, and strong heterogeneity compromise gel stability and selectivity. Here, a delayed-gelation organic–inorganic gel was formulated from a microbranched cationic copolymer, P­(AM-co-DMDAAC), a urea–formaldehyde/phenol–formaldehyde prepolymer cross-linker, and silica nanoparticles. The precursor retained injectability and formed a grade H gel within 15 h at 90 °C in 240,000 ppm brine. Transmission electron microscopy and elemental mapping confirmed a continuous polymer network with uniformly distributed silica, indicating effective nanoscale reinforcement. Core flooding, computed tomography, scanning electron microscopy, and low-field nuclear magnetic resonance were used to correlate network formation with pore occupation and flow redistribution. After in situ gelation, steady-state water-flooding pressure drops increased from 0.052 to 0.28 MPa in a 100 mD core and from 0.015 to 2.25 MPa in a 1000 mD core, corresponding to plugging efficiencies of 81.4 and 99.3%, respectively. Imaging and relaxation responses revealed longitudinally extended gel zones, adherent layers on carbonate surfaces, and shorter pore-fluid relaxation times, demonstrating reduced mobility and enhanced confinement. In parallel cores with a permeability contrast of 10, the high-permeability/low-permeability flow ratio reversed from approximately 10:1 to 1:5. These results establish a structure–​morphology–transport relationship for silica-reinforced cationic polymer gels and show how nanoscale network reinforcement enables selective plugging and controllable flow reallocation in heterogeneous porous media.

1. Introduction

In strongly heterogeneous porous media, influenced by the combined effects of interlayer permeability differences, uneven pore-throat structures, and mobility contrasts, injected fluids are prone to preferential migration along high-permeability dominant channels. This produces nonuniform displacement phenomena, including viscous fingering, channeling, and local cycling, leaving low-permeability regions largely unswept. − As reservoir development enters the middle and late stages, this heterogeneity-induced flow imbalance intensifies, directly limiting displacement efficiency and residual-oil mobilization. , Particularly in high-temperature and high-salinity reservoirs, the complex fluid environment and strong heterogeneity further amplify the dominance of these preferential pathways, rendering conventional conformance measures unable to establish stable deep-formation flow control. − Therefore, effectively increasing the flow resistance in high-permeability channels, inhibiting fluid fingering behavior, and promoting fluid diversion into low-permeability regions in strongly heterogeneous porous media remains a key challenge in deep profile control and flow regulation. From a fluid-mechanics perspective, the central challenge is to selectively increase the hydraulic resistance of connected high-permeability pathways while preserving injectivity and enabling subsequent flow redistribution throughout the porous medium.

Polymer gels are a principal technology for deep profile control and water shutoff, as they form a three-dimensional network structure in situ within the formation, increase the seepage resistance of high-permeability channels, and improve the fractional flow behavior of subsequent fluids. , In particular, delayed-cross-linking polymer gels can penetrate deep into the reservoir while maintaining sufficient injectivity, and gradually form gels under appropriate conditions; thus, they have attracted considerable attention for conformance control in heterogeneous reservoirs. However, conventional polymer gel systems still face significant limitations under high-temperature, high-salinity conditions and in complex pore-throat geometries. High-salinity conditions promote chain coiling, which reduces solution stability and network integrity. Additionally, shear forces and spatial confinement within restricted pores weaken the mechanical strength and retention capacity of the gel network, limiting deep migration, plugging durability, and subsequent flow-control efficacy. Therefore, developing novel deep-plugging gel systems that are both time-controllable and stable under high-temperature and high-salinity conditions remains critically important. Beyond synthetic polymer gels, biopolymer-based systems have also demonstrated potential for conformance control in heterogeneous formations. For instance, gellan guma microbial polysaccharidehas been shown to form brine-triggered gels capable of self-selective placement in high-permeability water-flooded channels; treatment of a two-layered sand pack with a permeability ratio of 1:2.5 yielded a 9% incremental oil recovery, confirming its applicability for bullhead injection into multilayered reservoirs with sufficient cross-flow. ,

In recent years, targeting the inadequate stability and limited deep control capabilities of conventional polymer gels under extreme reservoir conditions, researchers have pursued advances in polymer matrix design, cross-linking system optimization, and the synergetic enhancement of nanoparticles. , For the polymer matrix, studies have introduced salt-tolerant functional monomers, rigid structural units, and branched or hydrophobically associating architectures to improve solution stability, shear resistance, and network-forming capacity under high-salinity conditions. For cross-linking systems, phenolic resins, urea-formaldehyde resins, and other organic cross-linkers, prepolymers, and delayed-cross-linking systems are widely used to regulate gelation rate, cross-link density, and ultimate gel strength, thereby balancing system injectivity with deep gelation capability. Concurrently, nanoparticles such as SiO₂, with their large specific surface area and strong interfacial interactions, have been employed to enhance network compactness, thermal stab , Overall, the composite of novel polymers, controllable cross-linkers, and nanoparticles has become a key research direction for improving gel-system adaptability to complex reservoir conditions. However, existing research largely focuses on the macroscopic gelation performance, temperature/salt resistance, and static plugging effects of these systems. Systematic investigation of the pore-occupancy characteristics, the establishment process of subsequent water flooding resistance, and the fractional flow reallocation laws of organic–inorganic composite gels in strongly heterogeneous porous media. , More fundamentally, the coupling among gel microstructure, pore-scale fluid confinement, hydraulic resistance buildup, and the resulting redistribution of preferential flow has not yet been quantitatively established.

To address these issues, this study developed an organic–inorganic synergistic gel system utilizing a microbranched cationic copolymer, poly­(acryl­amide-co-diallyl­dimethyl­ammon­ium chloride) [P­(AM-co-DMDAAC)], as the primary matrix, a urea-formaldehyde/phenol-formaldehyde (UF/PF) resin prepolymer as the controllable cross-linking unit, and SiO₂ nanoparticles as the reinforcing component. Unlike previous studies that primarily focus on gel strength and static plugging performance, this work establishes a multiscale experimental framework linking material structure to pore-scale transport behavior and macroscopic flow redistribution. Under high-temperature and high-salinity conditions, the gelation behavior, pore-occupancy characteristics, plugging performance, and its regulatory effects on flow allocation in heterogeneous channels were systematically investigated through macroscopic gelation observation, microscopic gel morphology analysis, single-core flooding, rock surface characterization, low-field nuclear magnetic resonance (LF-NMR), computed tomography (CT) scanning, and parallel-core experiments. This study aims to reveal the mechanism by which preferential gel entry, in situ network formation, and selective resistance buildup collectively induce flow reallocation in strongly heterogeneous porous media.

2. Materials and Methods

2.1. Experimental Materials

2.1.1. Gel Materials

2.1.1.1. Polymer

Diallyl­dimethyl­ammon­ium chloride (DMDAAC), which possesses a cyclic structure, was selected as the functional monomer. Unlike conventional cationic monomers containing ester or amide groups, its five-membered ring structure exhibits high chemical inertness and resists hydrolysis under both alkaline and acidic high-temperature conditions. Via a controlled radical polymerization process, this cationic monomer was copolymerized with acrylamide to yield a high-molecular-weight, microbranched copolymer with elevated solution viscosity. The AM/DMDAAC feed molar ratio used for polymerization was 7:3. The repeating-unit structure of P­(AM-co-DMDAAC) is shown in Figure .

1.

Repeating unit structure of P­(AM-co-DMDAAC). The AM/DMDAAC feed molar ratio used for polymerization was 7:3.

2.1.1.2. Cross-Linker

A composite of urea-formaldehyde (UF) resin and phenol-formaldehyde (PF) resin prepolymers was employed as the cross-linking agent to construct a three-dimensional network featuring both flexible chain segments and rigid cross-linking nodes. The UF/PF prepolymer provides controllable cross-linking capability and facilitates the regulation of the gelation rate and ultimate network strength, thereby balancing precursor injectivity with subsequent intrapore gelation capacity. In the present formulation, UF and PF prepolymers were blended at a mass ratio of 2:1 (UF:PF). The cross-linking proceeds via condensation between the methylol (−CH₂OH) groups of the UF/PF prepolymers and the amide (−CONH₂) groups on the polyacrylamide backbone, forming −CH₂–NH–CO– bridges that interconnect polymer chains into a three-dimensional network. The UF component, with its higher methylol functionality, primarily governs cross-link density and gelation rate, whereas the PF component introduces rigid aromatic nodes that enhance thermal stability and mechanical strength of the resulting network.

2.1.1.3. Nanoparticles

Fumed SiO₂ nanoparticles with an average particle size of approximately 15 nm and a specific surface area of around 200 m²/g were selected as the inorganic reinforcing component. Their function is to enhance the compactness of localized cross-linking domains, improve the thermal and network stability of the gel system, and promote the formation of organic–inorganic composite structures.

2.1.1.4. Water

Simulated formation water was used, with a total salinity of 240,000 ppm and a calcium and magnesium ion content of 15,000 ppm.

2.1.2. Cores

The flooding experiments utilized carbonate cores with a diameter of 2.5 cm and a length of 10.0 cm, possessing gas permeabilities of 100 mD and 1000 mD, respectively. To ensure comparability across experiments, all cores were subjected to oil washing, salt washing, and saturation with simulated formation water prior to use.

2.2. Preparation of Polymer System

To ensure the uniform distribution of organic and inorganic components in a highly saline medium, the polymer system was prepared in the following sequence: First, 0.5 wt % SiO₂ nanoparticles were added to the simulated formation water and predispersed under mechanical stirring, supplemented by 10 min of ultrasonic treatment to weaken interparticle agglomeration and improve dispersion uniformity. Subsequently, 2.0 wt % microbranched cationic copolymer P­(AM-co-DMDAAC) was slowly added under low-shear conditions (400 rpm) and continuously stirred until the polymer was fully dissolved. Finally, 1.0 wt % UF/PF prepolymer was added, and stirring continued until the system formed a homogeneous and stable precursor solution.

Adopting the “SiO₂ nanoparticle–polymer–prepolymer” preparation sequence facilitates the predispersion of nanoparticles in high-salinity water and prevents premature viscosity spikes caused by the early addition of the cross-linking component, ensuring excellent fluidity and injectivity of the precursor during the preparation stage. The effective dispersion of SiO₂ nanoparticles within the polymer matrix is probably attributable to the formation of interpolyelectrolyte complexes between the positively charged quaternary ammonium groups of P­(AM-co-DMDAAC) and the negatively charged silanol groups on the SiO₂ surface, which provides electrostatic stabilization against particle aggregation in the high-salinity medium. Regarding the role of divalent cations, the permanently charged and chemically inert cyclic quaternary-ammonium groups of P­(AM-co-DMDAAC) remain stable in the presence of Ca²⁺ and Mg²⁺; rather than competing with cross-linking, these divalent ions compress the electrical double layer and promote a more compact network, while the silanol groups on the SiO₂ surface provide additional anchoring sites that further stabilize the organic–inorganic structure.

The resulting precursor maintained good fluidity at room temperature, satisfying the injection requirements. For microscopic characterization, selected gel samples were aged at 90 °C for 3 days to ensure complete network development, whereas the core-flooding experiments used an aging time of 15 h, as described below.

2.3. Macroscopic Gelation Observation and Microscopic Morphology Characterization

Macroscopic gelation behavior of the polymer system was evaluated using the inverted bottle method at 90 °C. The prepared precursor was sealed in glass sample vials and aged in a constant-temperature oven at 90 °C, with samples taken at different intervals to observe their flow states. Upon cooling to room temperature, the inverted bottle test was performed, and the gelation time and ultimate gel strength were classified according to the standard gel strength codes.

To characterize the microscopic network structure of the gelled system, samples aged at 90 °C for 3 days were dried and prepared for transmission electron microscopy (TEM) observation. TEM was utilized to analyze the micromorphology, including the gel network structure, dispersion uniformity, and the distribution characteristics of nanoparticles within the polymer network. In conjunction with energy-dispersive spectroscopy (EDS), the spatial distribution of elements such as Si, C, O, and N was analyzed to determine the dispersion state of the inorganic components within the organic network and the formation of the organic–inorganic composite structure.

2.4. Single-Core and Parallel-Core Flooding Experiments

Single-core flooding experiments were conducted at 90 °C to evaluate the gelation process, the development of subsequent water flooding resistance, and the stability of the system in porous media with different permeabilities. The procedure was as follows: First, initial water flooding was performed at a constant flow rate of 0.5 mL/min on cores saturated with simulated formation water until the pressure drop stabilized, and the initial steady-state pressure drop was recorded. Subsequently, 1.0 pore volume (PV) of the polymer precursor was injected, followed by aging at 90 °C for 15 h. After aging, subsequent water flooding was resumed, and the dynamic differential pressure was continuously recorded until a new steady-state pressure was reached.

To quantitatively evaluate the retardation effect of the gelled system on subsequent water flooding, the residual resistance factor F _rr was adopted as the characterization index, defined as

$$F_{\mathrm{rr}}=\frac{\mathrm{\Delta }{P}_2}{\mathrm{\Delta }{P}_1}$$ 1

where ΔP ₁ and ΔP ₂ represent the steady-state pressure drops during the initial and subsequent water flooding, respectively. To evaluate the fractional flow control capability of the polymer system under heterogeneous seepage conditions, a parallel dual-core model consisting of a 100 mD core and a 1000 mD core was constructed to simulate a competitive flow system with a permeability contrast of 10. The experimental process included initial water flooding, polymer precursor injection/aging, and subsequent water flooding. During the initial water flooding stage, the effluent flow rates of the high- and low-permeability channels were recorded to determine the initial fractional flow ratio. The precursor was then injected and aged at 90 °C for 15 h to allow in situ gelation within both channels. Finally, subsequent water flooding was resumed, and the dynamic changes in the effluent flow rates were continuously monitored to analyze the regulatory effect of the gelled system on the fractional flow behavior of the heterogeneous channels. A permeability contrast of 10 was chosen because it represents the moderate-to-strong heterogeneity typically encountered between the matrix and dominant high-permeability channels in the target carbonate reservoirs, while still allowing a measurable initial flow into the low-permeability channel so that the fractional-flow inversion could be reliably quantified.

The instantaneous fractional flow rate of each channel is defined as

$$f_i=\frac{Q_i}{Q_{\mathrm{high}}+Q_{\mathrm{low}}}\times 100\%$$ 2

where Q _i is the instantaneous flow rate of the ith channel, and Q _high and Q _low represent the flow rates of the high- and low-permeability channels, respectively.

A schematic overview of the displacement experimental apparatus is shown in Figure .

2.

Schematic diagram of the displacement experimental apparatus.

2.5. Rock Surface Scanning Electron Microscopy (SEM) Characterization

A field emission scanning electron microscope (SEM) was used to scan the surfaces of pristine cores and the end-faces of cores postflooding. After drying and gold sputtering, the samples were observed for surface coverage, the formation of adhered layers, and changes in microscopic deposition morphology. This was combined with EDS point analysis to characterize the elemental composition of the surface.

2.6. LF-NMR Testing and CT Characterization

A low-field nuclear magnetic resonance (LF-NMR, 21.3 MHz) device employing the Carr–Purcell–​Meiboom–Gill (CPMG) pulse sequence was used to measure the transverse relaxation time (T ₂) distributions of the 100 mD and 1000 mD single cores, as well as the parallel core models, before and after treatment. For a porous media system dominated by surface relaxation, the T ₂ relaxation rate can be expressed as

$$\frac{1}{T_2}={\rho }_2\left(\frac{S}{V}\right)$$ 3

where ρ is the surface relaxivity constant of the rock, and S/V is the specific surface area of the pores containing the fluid. By measuring the changes in the T ₂ spectra before and after gelation, the influence of the polymer gel on the fluid occurrence state and confinement degree within the pore space was analyzed.

A computed tomography (CT) scanner was utilized to scan the cores before and after flooding to observe density variations in the longitudinal cross section and the distribution characteristics of the plugged regions within the core.

3. Results and Discussion

3.1. Gelation Behavior and Microscopic Network Structure

Figure illustrates the relationship between the gel strength code and aging time at 90 °C for the base gel system (without SiO₂ nanoparticles) and the composite gel system (with 0.5 wt % SiO₂ nanoparticles). The gel strength of both systems progressively increases with aging time and levels off within approximately 15 h, indicating that the cross-linking and gelation process primarily occurs within this time frame. Compared to the base system, the composite system incorporating nanoparticles maintains a similar gelation time scale while exhibiting a higher ultimate gel strength, ultimately reaching grade H within approximately 15 h. This suggests that the inorganic nanocomponents enhance the network-forming capability and postgelation structural stability of the system. The visual observations from the inverted bottle test are consistent with the variations in the gel strength code, indicating that the precursor maintains good fluidity after preparation and transforms into a self-supporting gel after aging at 90 °C for approximately 15 h. Notably, the composite system exhibits a transient plateau between approximately 5 and 7 h of aging (corresponding to gel strength code E), during which the gel retains slight fluidity. This plateau indicates that the nanoparticle-reinforced system requires a longer residence time to transition from code E to code F compared with the base system. The delayed progression can be attributed to the competitive adsorption of SiO₂ nanoparticles at incipient cross-linking sites, which temporarily retards further network densification until sufficient thermal energy accumulates to drive additional condensation reactions beyond the adsorption equilibrium.

3.

Variations of gel strength codes with aging time at 90 °C determined by the inverted bottle test: (a) base gel system (without SiO₂ nanoparticles); (b) composite gel system (with 0.5 wt % SiO₂ nanoparticles). All other components are identical between the two systems [2.0 wt % P­(AM-co-DMDAAC) and 1.0 wt % UF/PF prepolymer]; only the SiO₂ content differs.

As further shown in Figure , the aged gel samples form a continuous and interconnected network structure, indicating the formation of a relatively complete internal three-dimensional skeleton. The EDS elemental mapping results show that Si is relatively uniformly distributed together with C, O, and N, with no obvious local agglomeration. This indicates that SiO₂ nanoparticles are well dispersed within the polymer network and participate in the formation of the organic–inorganic composite structure.

4.

Microscopic morphology and EDS elemental mapping results of the gel samples.

3.2. Hydraulic Resistance Buildup and Longitudinal Plugging in Single Cores

As shown in Figure , the polymer precursor establishes substantial subsequent water-flooding resistance in cores with varying permeabilities after aging at 90 °C for 15 h. Upon resuming water flooding, the injection differential pressure for both core types rises rapidly and remains relatively stable during continuous injection. This indicates that the gel structure does not fail rapidly under the shear of subsequent flow but instead exerts a sustained retarding effect on fluid migration. It should be emphasized that the injected fluid is a gelling solution (gelant)i.e., a low-viscosity precursor comprising the polymer and cross-linkerrather than a preformed mature gel. Mature gels, owing to their high elastic modulus and inability to deform through microscale pore throats, cannot penetrate microporous media. Throughout this work, the term “precursor” or “gelant” denotes the injectable gelling solution prior to cross-linking, whereas “gel” refers exclusively to the cross-linked network formed in situ after aging.

5.

Evolution curves of differential pressure versus injected volume for single cores.

For the 100 mD core, the steady-state differential pressure during subsequent water flooding increases from 0.052 to 0.28 MPa, yielding a plugging efficiency of 81.4% and a corresponding residual resistance factor F _rr of 5.38. For the 1000 mD core, the steady-state differential pressure increased substantially from 0.015 to 2.25 MPa, achieving a plugging efficiency of 99.3% and a corresponding F _rr of 150. Compared to the low-permeability core, the high-permeability core exhibits a more pronounced pressure differential growth and a higher residual resistance post-treatment. This demonstrates that the gel system can still form a high-strength dense network in high-permeability porous media with larger pore-throat structures, thereby achieving effective deep plugging.

The aforementioned results indicate that the nanoparticle–polymer synergistic gel system can effectively reside and undergo intrapore gelation in porous media with varying permeabilities, establishing an extremely high plugging pressure and stable flow resistance. This characteristic of being able to construct a high-strength plugging barrier even in high-permeability media provides a reliable material foundation for subsequent controllable deep profile control and flow regulation under strongly heterogeneous conditions.

3.3. Adhesion Characteristics on Rock Surfaces

Figure presents the SEM images of the pristine rock surface and the rock surface after treatment. Compared to the pristine core, the treated rock surface exhibits distinct adherent layers and agglomerated deposits. The originally clear mineral grain boundaries of the rock matrix are covered by the polymer, indicating that the polymer gel was not entirely displaced by subsequent flooding but instead adhered to and remained on the rock surface and within adjacent pore regions.

6.

SEM images of the pristine rock surface and the rock surface after treatment.

From a micromorphological perspective, the surface of the pristine rock is relatively clean, with well-defined mineral grain boundaries. In contrast, the treated surface displays a dense coverage layer and irregular deposition morphology, indicating that the gel formed a residual adherent layer. This indicates that the gel–rock interface establishes sustained localized flow resistance near pore walls.

Overall, the adhesion and coverage observed on the treated rock surface provide direct morphological evidence for the retention and plugging effects of the gel system in porous media, offering a microscopic explanation for the maintenance of flow resistance during subsequent water flooding. Furthermore, EDS analysis of the treated rock surface confirmed the presence of Si, C, O, and N in the deposited layer, consistent with the elemental composition of the organic–inorganic composite gel. This elemental evidence corroborates that the adhered material is the polymer–SiO₂ composite rather than mineral debris, further supporting the interpretation that the gel forms a stable residual structure on rock surfaces.

3.4. Pore Fluid Confinement Characteristics in Single Cores

3.4.1. Gel Distribution Characteristics by CT Characterization

As shown in Figure , comparing the CT scan results of the cores before and after treatment reveals that both the 100 mD and 1000 mD cores exhibit significant pseudocolor variations in their longitudinal cross sections after the injection and aging of the polymer precursor. Under identical reconstruction and pseudocolor mapping conditions, certain regions that were originally yellow transitioned to orange-red after treatment, indicating an enhancement in the CT response. Based on the flooding process and the aging-gelation sequence, these color changes correspond to the distribution regions where the injected polymer precursor underwent in situ gelation within the pore space, suggesting that the system does not merely accumulate near the inlet face but forms a plugging zone with distinct longitudinal extension within the core.

7.

Longitudinal cross-sectional CT scan density reconstruction maps of single cores with different permeabilities before and after injection: (a) 100 mD low-permeability core; (b) 1000 mD high-permeability core.

Further comparison between cores of different permeabilities reveals that the orange-red regions in the high-permeability core are more extensive and continuous, whereas the variation regions in the low-permeability core are relatively small and localized. This indicates that in the high-permeability core, the polymer precursor more easily enters deeper regions along the well-connected dominant channels, leading to retention and gelation over a wider area. This occurs because high-permeability cores typically possess larger effective pore volumes and more developed interconnected pore-throat structures. Consequently, the precursor experiences lower migration resistance during injection and has more accessible space, making it more prone to remaining, adhering, and forming continuous gel regions within the pore walls and dominant flow channels.

Combined with the single-core pressure differential results, the continuous regions transitioning from yellow to orange-red in the CT images essentially reflect the occupancy and plugging effects of the polymer gel within the core. Especially in the high-permeability core, the more extensive and continuous gel distribution is consistent with the more significant pressure rise observed after treatment, indicating that the system can preferentially act on high-permeability dominant channels and establish strong, sustained flow resistance within the core. Overall, the CT scanning results directly reveal the spatial distribution differences of the gel system in cores of different permeabilities, providing structural evidence for explaining its permeability-selective plugging and subsequent flow regulation behavior.

3.4.2. Fluid Confinement Characteristics by LF-NMR Characterization

As shown in Figure , a comparison of the LF-NMR T₂ spectra for the 100 mD and 1000 mD single cores before and after treatment reveals distinct spectral changes for both types of cores following polymer injection and aging. Compared to the initial water-saturated state, the main peaks of the T₂ spectra for the treated cores shift globally toward the short relaxation time region. Concurrently, the signals in the short T₂ region are enhanced, while the signals in the long T₂ region are significantly weakened, indicating that the polymer system alters the occurrence state of the fluid within the pore space, transforming fluids originally in a high-mobility state into fluid components with a higher degree of confinement.

8.

Comparison of LF-NMR T₂ spectra for 100 mD and 1000 mD single-core flooding experiments.

Mechanistically, the T₂ relaxation time is closely related to the pore scale and the degree of fluid confinement. The shift of the main peak toward the short relaxation time region after treatment suggests that for fluids originally residing in larger pores or well-connected channels, their active space is compressed. This enhances the interaction between the fluid and the pore walls or the gel interface, resulting in shorter transverse relaxation times. This demonstrates that after the polymer precursor ages and gels inside the core, it not only occupies part of the pore space but also weakens the mobility of the fluid in the dominant flow channels, creating sustained flow resistance.

Comparing the different cores, the changes in the T₂ spectrum of the 1000 mD high-permeability core are more pronounced, indicating that the polymer precursor enters more thoroughly and distributes more widely in high-permeability media, exerting a more significant impact on the fluid occurrence state after aging. This result is consistent with the more extensive and continuous gel distribution observed in CT scans of high-permeability cores and aligns with the significant increase in resistance observed in the pressure differential results. In contrast, the 100 mD low-permeability core also shows a trend of the main peak shifting left and enhancement in the short T₂ region, suggesting the system can enter and effectively plug some pores in low-permeability media, though its scope of action is relatively restricted.

Synthesizing the LF-NMR, CT, and pressure response results, the role of the gel system in single cores is reflected not only in the macroscopic increase in injection pressure but also in the sustained regulation of the fluid occurrence state and flow capacity at the pore scale. That is, after the polymer precursor gels within the porous medium, it reduces fluid mobility through occupancy, adhesion, and localized plugging, thereby providing a pore-scale basis for the establishment of sustained flow resistance during subsequent water flooding.

3.5. Fractional Flow Adjustment Characteristics in Parallel Cores

As shown in Figure , the parallel dual-core flooding process can be divided into four stages: The light blue region represents the initial water flooding stage. The light yellow region represents the polymer precursor injection stage. The light red region is the aging and heating stage, and the light purple region denotes the subsequent water flooding stage after gelation. The dynamic changes in the flow rates of the high- and low-permeability channels across different stages reflect the processes of entry, aging-gelation, and subsequent fractional flow regulation of the polymer precursor in parallel heterogeneous channels.

9.

Flow rate evolution curves for the parallel dual-core flooding experiment.

During the initial water flooding stage, due to the significant permeability difference between the two channels, the fluid preferentially migrates along the high-permeability channel, which bears the majority of the system’s flow rate, while the fractional flow proportion of the low-permeability channel is extremely low, exhibiting typical dominant-channel flow behavior. Specifically, the flow rate of the high-permeability channel is approximately 0.27 mL/min, and the flow rate of the low-permeability channel is approximately 0.03 mL/min, yielding a flow ratio of about 10:1. Without treatment, injected fluids primarily bypass through the high-permeability channel, leaving the low-permeability channel largely unswept.

Upon entering the polymer precursor injection stage, the precursor preferentially enters the high-permeability channel, which has lower flow resistance, and undergoes varying degrees of migration and retention within both channels. Subsequently, during the aging and heating stage, the precursor further forms a gel structure under high-temperature conditions, gradually establishing the foundation for subsequent fractional flow adjustment. During the subsequent water flooding after gelation, the flow rate evolution trends of the two channels undergo significant changes: the flow rate of the high-permeability channel continuously decreases, while the flow rate of the low-permeability channel gradually increases. This indicates that subsequent fluids no longer continuously and preferentially pass along the original high-permeability dominant path, but begin to enter the low-permeability channel in greater quantities.

Ultimately, the flow rate of the high-permeability channel decreases from the initial ∼0.27 mL/min to ∼0.05 mL/min, whereas the flow rate of the low-permeability channel increases from ∼0.03 mL/min to ∼0.25 mL/min. The flow ratio between the two channels shifts from an initial ∼10:1 to approximately 1:5. The clear crossover of the two flow curves during subsequent water flooding demonstrates that the system has undergone a transition from “high-permeability-dominated fractional flow” to “low-permeability-enhanced fractional flow”. This result reveals that the nanoparticle–polymer synergistic gel system can preferentially increase the flow resistance in the high-permeability dominant channel and promote subsequent fluid diversion into the low-permeability channel, thereby achieving flow reallocation under strongly heterogeneous conditions. This transition represents a fractional-flow inversion, in which the originally dominant high-permeability pathway becomes hydraulically suppressed and the low-permeability pathway carries the majority of the subsequent flow.

3.6. Pore Fluid Responses in Parallel Cores

As shown in Figure , comparing the LF-NMR T₂ spectra of the 100 mD and 1000 mD cores in the parallel dual-core model before and after treatment reveals that, after the injection and aging-gelation of the polymer precursor, the spectral peaks of both types of cores still exhibit an overall trend of shifting toward the short relaxation time region. This indicates that under competitive seepage conditions, the gel system can still enter the pore space and alter the occurrence state of the pore fluid, transforming a portion of the fluid from a relatively movable state to a more confined state.

10.

Comparison of LF-NMR T₂ spectra for the 100 mD and 1000 mD cores in the parallel dual-core model.

Combined with the single-core flooding results, it can be further observed that the changes in the T₂ spectrum postgelation are not only manifested as the leftward shift of the main peak but also as an increase in peak height after the shift, signifying a rise in the hydrogen signal intensity in the short T₂ region. This phenomenon is quite distinct in single-core experiments; it is also present in parallel dual-core experiments, although the magnitude of enhancement is relatively smaller. This can be interpreted as follows: after the polymer precursor enters the pores, undergoes retention, and gels, the hydrogen-containing component signals associated with the gel within the pores increase, thereby raising the height of the shifted peak. However, under parallel competitive flow conditions, because the precursor is distributed between the two channels, the volume of polymer entering and ultimately retained within individual cores is lower than under single-core conditions; consequently, the degree of signal enhancement postshift is less pronounced than in single-core tests.

Further comparing the response differences of the low-permeability core between single-core and parallel-core conditions reveals that, under single-core flooding, the leftward shift of the main peak for the 100 mD core is relatively obvious. In contrast, under parallel dual-core conditions, although the low-permeability core still exhibits some degree of leftward shift, the extent of the shift is limited, and the change in peak height is not obvious. This indicates that during the competitive flow process, the polymer precursor preferentially enters the high-permeability channel with lower flow resistance and better connectivity, resulting in a relatively small amount of precursor entering the low-permeability core. Consequently, its retention and occupancy effects postgelation are also limited, leading to a weaker T₂ spectrum change compared to the single-core low-permeability condition. In other words, the weaker spectral peak response in the low-permeability core under parallel conditions does not reflect system failure, but is the result of differential precursor distribution driven by strongly heterogeneous fractional flow.

Synthesizing the flow rate evolution results in Figure and the LF-NMR responses in Figure , it can be concluded that the preferential allocation, retention, and in situ gelation of the polymer precursor in the parallel dual-core system primarily occur within the high-permeability channel, thereby preferentially weakening its flow capacity. Simultaneously, although the low-permeability channel is also affected to some extent, its pore-scale response is relatively limited due to the small entry volume. It is precisely this selective entry and selective plugging induced by competitive seepage that ultimately leads to a decrease in high-permeability channel flow, an increase in low-permeability channel flow, and the realization of flow reallocation within the parallel system.

3.7. Correlation Analysis Between Plugging Effect and Fractional Flow Adjustment

The interaction process of the polymer gel in strongly heterogeneous porous media essentially manifests as a sequential response: “preferential entry → postinjection gelation → selective plugging → fractional flow adjustment”. This sequence can be interpreted as a dynamic reconstruction of the internal hydraulic resistance field, which alters the stability of preferential flow paths and drives system-scale flow redistribution. Combining the aforementioned macroscopic and microscopic experimental results, it is evident that during the injection stage, the precursor more easily enters the internal porous media along the high-permeability channels characterized by lower flow resistance and better connectivity, and gradually forms a gel structure with a certain strength and spatial continuity under high-temperature aging conditions. This process not only alters the occupancy state of the pore space but also enhances the interaction between the fluid and the pore walls/gel interfaces, thereby reducing fluid mobility within the dominant channels.

The significant increase in differential pressure post-treatment indicates that the system can establish sustained flow resistance inside the core. The distribution regions with high CT response emerging post-treatment demonstrate that the polymer gel does not merely accumulate at the inlet face but forms a plugging zone with certain continuity within the longitudinal depth of the core. The LF-NMR results, showing the main T₂ peak shifting toward the short relaxation time region accompanied by enhanced short T₂ signals, suggest that the gel alters the fluid occurrence state and increases the degree of fluid confinement upon entering the pores. These findings corroborate each other, proving that the plugging effect of the system is not merely localized blockage, but rather the progressive establishment of stable seepage resistance through occupancy, adhesion, and flow restriction at the pore scale.

From the perspective of the parallel-core results, this plugging effect further translates into the adjustment of the fractional flow pattern. Because the high-permeability channel possesses a stronger capacity for the preferential entry of the precursor, the retention, adhesion, and subsequent gelation of the polymer system within it are more extensive, thereby preferentially weakening the flow capacity of the high-permeability channel. Accordingly, during subsequent water flooding, fluids that originally migrated primarily along the high-permeability dominant path gradually divert to the low-permeability channel. This is manifested macroscopically as a decrease in high-permeability flow, an increase in low-permeability flow, and a shift in the fractional flow ratio from an initial ∼10:1 to ∼1:5. The LF-NMR responses in parallel cores further reveal that this macroscopic fractional flow variation is consistent with the changes in the degree of fluid confinement at the pore scale; namely, the more extensive polymer distribution and gelation effects in the high-permeability channel serve as the foundation for the realization of subsequent flow reallocation.

Therefore, the action mechanism of the system in strongly heterogeneous porous media can be summarized as follows: The polymer precursor preferentially enters the high-permeability dominant channel and forms a gel inside the pores after aging; by occupying the pore space, adhering to pore walls, and increasing localized flow resistance, it achieves selective plugging of the high-permeability channel. Building upon this, subsequent fluids are forced to enter low-permeability regions in greater quantities, ultimately realizing flow reallocation and controllable deep conformance control. These results indicate that the plugging effect and fractional flow regulation are not independent, but exhibit a clear causal relationship: the former is the prerequisite for the realization of the latter, while the latter is the macroscopic manifestation of the former under strongly heterogeneous conditions. It is worth noting that the use of internal pressure taps distributed along the core length could further resolve the permeability reduction profile and quantify the longitudinal heterogeneity of gel placement. This approach is recommended for future experimental work to provide spatially resolved information on the evolution of flow resistance during and after gelation. From an engineering standpoint, the base materials [P­(AM-co-DMDAAC), UF/PF prepolymer, and fumed SiO₂] are all commercially available and used at low dosages (2.0, 1.0, and 0.5 wt %, respectively), which keeps the formulation cost moderate; on a field scale, the only additional operation relative to a conventional polymer gel is the predispersion of SiO₂, which can be realized by in-line high-shear mixing or preblended nanoparticle dispersions instead of the laboratory ultrasonic step. Moreover, the stable differential pressure sustained under continuous postgelation water flooding indicates that the system functions as a durable in-depth conformance-control plug; where later removal is required, the gel can in principle be degraded by oxidative or acidic breakers, which is left for future work.

4. Conclusions

This study developed an organic–inorganic synergistic gel system composed of a microbranched cationic copolymer P­(AM-co-DMDAAC), a UF/PF prepolymer, and nanoparticles, and systematically investigated its gelation behavior, plugging characteristics, pore fluid confinement effects, and flow reallocation role in strongly heterogeneous porous media. The main conclusions are as follows:

• (1)Under conditions of 90 °C and a total salinity of 240,000 ppm, the polymer system achieves grade H gel strength within approximately 15 h, demonstrating controllable gelation kinetics and stable network formation under harsh physicochemical conditions. Macroscopic gelation observation and microstructural characterization results indicate that the system can form a relatively uniform and dense composite network. This suggests that the synergetic design concept of “polymer skeleton–prepolymer cross-linking–nanoparticle reinforcement” can effectively enhance the adaptability of the gel system to complex reservoir conditions.
• (2)Single-core flooding results demonstrate that after the polymer precursor enters the core and ages, it can form a stable gel structure within the pores, significantly increasing the subsequent water flooding resistance. For the 100 mD core, the steady-state differential pressure increased from 0.052 to 0.28 MPa, yielding a plugging efficiency of 81.4%; for the 1000 mD core, the steady-state differential pressure surged from 0.015 to 2.25 MPa, achieving a plugging efficiency of 99.3%. These results indicate that the system possesses stronger selective plugging capability for high-permeability dominant channels.
• (3)CT scanning results reveal the emergence of continuously distributed high-response regions inside the treated cores, indicating that the polymer system does not merely accumulate near the inlet face but forms a plugging zone with certain continuity within the longitudinal depth of the core. Compared to the low-permeability core, these regions in the high-permeability core exhibit a larger extent and stronger continuity, demonstrating that the precursor more easily enters high-permeability channels to undergo retention, adhesion, and subsequent gelation. This is consistent with the more significant resistance establishment process in high-permeability cores. Furthermore, the rock surface SEM results indicate the presence of distinct adhered layers and deposition structures on the treated rock surface, providing direct morphological evidence for the retention and plugging role of the system in porous media.
• (4)LF-NMR results demonstrate that under both single-core and parallel-core conditions, the main peak of the T₂ spectrum for the treated cores shifts toward the short relaxation time region, indicating a decrease in pore fluid mobility and an increase in the degree of confinement. After single-core flooding, the leftward shift of the peak is accompanied by an increase in peak height, suggesting that the entry of the polymer into the pores not only alters the fluid occurrence state but also enhances the hydrogen signal contribution associated with the gel. This phenomenon also exists during parallel flooding but with a smaller magnitude of variation, indicating that under competitive flow conditions, the fractional flow effect leads to a reduced amount of polymer retention in individual cores. Particularly for the low-permeability core, both the leftward peak shift and the peak height variation under parallel conditions are weaker than those under single-core conditions, signifying that more precursor preferentially enters the high-permeability channel, resulting in relatively limited occupancy effects in the low-permeability channel.
• (5)Parallel-core experiments demonstrate that the system can significantly alter the flow allocation relationship under strongly heterogeneous conditions. For the dual-channel model with a permeability contrast of 10, the flow ratio between the high- and low-permeability channels shifted from an initial ∼10:1 to ∼1:5. This indicates that the system can preferentially increase the flow resistance in the high-permeability dominant channel and promote subsequent fluid diversion into the low-permeability channel. Synthesizing the macroscopic flooding, CT, SEM, and LF-NMR results, it can be concluded that the nanoparticle–polymer synergistic gel system integrates controllable gelation, selective plugging, and flow reallocation capabilities. It provides experimental evidence and material design references for controllable deep conformance control in high-temperature, high-salinity, strongly heterogeneous reservoirs.

These findings establish a mechanistic link between gel microstructure, pore-scale fluid confinement, hydraulic resistance buildup, and preferential-flow redistribution in heterogeneous porous media.

The data that support the findings of this study are available from the corresponding author upon reasonable request.

The authors declare no competing financial interest.