1. Introduction
Carbon capture, utilization, and storage (CCUS) is one of the most effective strategies to mitigate climate change issues [
1],[
2]. In CCUS, CO
2 is captured from industrial sources and injected into saline aquifers, oil and gas reservoirs, coal beds, and so forth [
3],[
4]. For example, CO
2 flooding, a typical CCUS application, is a widely used technique for enhanced oil recovery (EOR) [
5],[
6]. The injected CO
2 can also be stored within reservoirs via solubility trapping and structural trapping mechanisms [
7]. However, early breakthrough and fingering of CO
2 may occur if there are high-permeability channels in reservoirs, such as fractures and high-permeability zones [
8],[
9]. It may take a shorter time for CO
2 to reach producers by flowing through those high-permeability channels. As a result, a large amount of residual oil in the lower-permeability zones cannot be swept by CO
2, and CO
2 storage performance will be severely compromised [
10],[
11]. One effective method for mitigating this issue is profile control, which is injecting plugging agents into the formations to plug those high-permeability channels [
12],[
13]. In this manner, more CO
2 can enter the lower-permeability zones to displace oil and can be trapped by pores and throats of the lower-permeability zones, which increases both oil recovery efficiency and CO
2 storage efficiency.
Among different plugging agents, gels are widely used because their particles can be readily delivered to the deeper parts of formations and yield
in situ plugging [
14]. Especially, CO
2-responsive gels gain great attention for profile control in CCUS applications since they will swell when being exposed to CO
2 and shrink when being removed from CO
2 [
15]. The particles of those CO
2-responsive gels can be injected into reservoirs, followed by CO
2 injection. This enables the particles to swell upon contact with CO
2. The particles with increased sizes yield an enhanced plugging on the pores and throats of the aforementioned high-permeability pathways [
16]. Many researchers devoted great efforts to developing CO
2-responsive gels for profile control in CCUS. For example, Pu et al. [
17] and Du et al. [
18] synthesized a hydrogel sample with an interpenetrating network using the monomers of
N,
N-dimethylaminoethylmethacrylate (DMAEMA) and acrylamide (AAm), in which DMAEMA is the most widely used monomer to prepare CO
2-responsive materials
[19],
[20],
[21]. Then, such prepared hydrogel was sheared into a suspension with an initial median size of 18 μm. After exposure to CO
2, the median size increased to 25.86 μm. Tian et al. [
22] prepared a thermal and CO
2 dual-responsive hydrogel by polymerizing the monomers of AAm and DMAEMA with [2-(methacryloyloxy) ethyl]dimethyl-(3-sulfopropyl) ammonium hydroxide (SBMA). The hydrogel could swell when CO
2 was purged or temperature surpassed 65 °C. Raj et al. [
23] synthesized a CO
2-responsive gel for mobility control using cotton pulp and nanocellulose. Multiple nanocelluloses could aggregate to form a gel when being exposed to CO
2, providing effective plugging on the pores and throats of CO
2 flowing channels. Zhang et al. [
24] prepared a CO
2-responsive gel using
N,
N-dimethyl octylamide-propyl tertiary amine (DOAPA) and sodium p-toluenesulfonate (SPTS). The experimental investigations indicated that the sample they prepared had a rapid CO
2 responsibility and a promising plugging performance. More CO
2-responsive gel recipes for CCUS were also reported in other studies
[25],
[26],
[27],
[28],
[29].
CO
2-responsive hydrogels contain functional groups that are sensitive to CO
2, such as amines, amidines, guanidines, and carboxylic acids [
30]. CO
2 can react with water to form a weakly acidic environment. Such acidic environment allows for protonation reactions between CO
2 and these functional groups [
31], adding protons (H
+) to the polymer chains [
32]. The protons will repel each other because they have the same positive electric charges. As a result, the distances between polymer chains will increase, enabling more water to enter the structure of the gel particles [
33]. Therefore, the gel particles will swell. Meanwhile, some small molecules with negative electric charges will also enter the gel particles to balance the charge, which further causes the swelling of the gel particles. The swollen gel particles yield a stronger effect on plugging the high-permeability gas migration channels. However, there are some drawbacks when using those CO
2-responsive gels in high-temperature CCUS applications. First, the protonation reactions between protons and CO
2-responsive functional groups are reversible at high temperatures [
34]. For example, deprotonation reactions will occur for some widely used CO
2-responsive gels when reservoir temperatures exceed 65 °C [
35],[
36], leading to a decrease in their particle size [
37]. As a result, the plugging efficiency of those gel particles will be significantly decreased. Second, the CO
2-responsive gel particles are easily to be thermally degraded at high temperatures, failing to form stable plugging on high-permeability channels. It is worth noting that some researchers developed irreversible CO
2-responsive hydrogels to improve the stability of plugging the CO
2 channeling pathways. Fang et al. [
38] developed an intelligent CO
2-responsive hydrogel for profile control in CCUS applications. This new material was irreversible and didn’t revert to solution phase after being N
2 bubbling. Moreover, its viscosity remained four times higher than that of the initial solution upon heating, showing a certain degree of irreversibility at high temperatures. Wu et al. [
39] developed a nonreversible hydrogel for profile control in CCUS applications by grafting poly(dimethylaminopropyl methacrylamide) onto the backbone of sodium carboxymethyl cellulose. The new hydrogel can be formed in the presence of CO
2, but cannot revert to copolymer solution after bubbling N
2 at 60 °C. However, we are still lacking such irreversible CO
2-responsive gels to enhance the profile control performance. Therefore, it is necessary to develop a new dispersed particle gel (DPG) product for profile control in high-temperature CCUS applications. The new DPG product is expected to exhibit a strong high-temperature resistance, and its particles can irreversibly swell at high temperatures.
In this study, a novel DPG suspension containing modified particles of a double-network hydrogel is prepared for profile control in high-temperature CCUS applications. The double-network hydrogel consists of a PAAm network and a sodium alginate (SA) network. Then, a DPG suspension is prepared by shearing the synthesized double-network hydrogel in water. Finally, the new DPG suspension is developed by modifying the gel particles of the prepared DPG suspension by adding PMS and purging CO2. Herein, PMS is a water-soluble organic silicate that is widely used as a waterproof reagent to enhance water repellency to the building surfaces. Compared to the un-modified DPG suspension, the particle size of the new DPG suspension is significantly increased by over two times. Moreover, heating the new DPG suspension at 100 °C for 24 h will not decrease its particle size, suggesting that the swelling is irreversible at high temperatures. Besides, the coating layer generated on the gel particles by the modification has a good thermal insulation ability, thus improving the thermal stability of the gel particles. Core flooding experiments suggest that the new DPG suspension can offer a stronger plugging efficiency in blocking porous media with ultra-high permeabilities than the un-modified DPG suspension.
2. Experimental section
2.1. Materials
Acrylamide (99%), N,N′-methylenebisacrylamide (MBAA; 99%), potassium persulfate (KPS; 99%), ferric chloride (FeCl3; 97%), and SA are purchased from Sigma-Aldrich, Germany. Aqueous potassium methylsilanetriolate (PMS) solution with a solid content of 42 wt% is purchased from BOC Sciences, USA. All chemicals are used as received. Distilled water is made in-house. Quartz sands (diameter ≤ 0.3 mm) used for core flooding experiments are purchased from Target Product Ltd., Canada.
2.2. Preparation of double-network hydrogel and gel particle suspension
We add 0.3 g of SA into 15 mL of distilled water and slowly stir the solution using a magnetic stirrer for 6 h to allow the complete dissolution. Then, 1.5 g of AAm and 0.001 g of MBAA are sequentially dissolved into the solution. Following this, 0.0057 g of KPS is added to the solution. Subsequently, nitrogen gas is continuously purged into the solution for 10 min to remove oxygen. Next, the solution is transferred to an oven and heated at 65 °C for 12 h to form the bulk hydrogel. Finally, the bulk hydrogel is immersed in an aqueous FeCl3 solution with a concentration of 0.6 mol·L−1 for 24 h. As such, the bulk double-network hydrogel is obtained and coded as PAAm/SA-Gel. To prepare a DPG suspension of the double-network hydrogel, the bulk hydrogel is mixed with water at a gel-to-water mass ratio of 2%. Then, the mixture is sheared using a blender (NINJA, 1400 W) for 30 min. This DPG suspension containing the double-network hydrogel particles is coded as PAAm/SA-DPG.
2.3. Development of the novel DPG suspension
The novel DPG suspension for profile control in high-temperature CCUS applications is developed by modifying the PAAm/SA-DPG particles. We add 0.48 g of PMS solution to 200 mL of PAAm/SA-DPG suspension to obtain a mixture of PMS and PAAm/SA-DPG. The mixture is stirred for 10 min to become a homogenous solution with a PMS-to-SA mass ratio of 2:3. The solution is then transferred into a closed container, followed by continuous injection of CO2 to remove air in the solution. Finally, the closed container is heated in an oven at 50 °C for 24 h to allow for the complete modification of the PAAm/SA-DPG particles. The new suspension containing the modified hydrogel particles is coded as PAAm/SA-PMS2/SA3-mDPG.
2.4. Characterizations
2.4.1. Morphology
A scanning electron microscopy (SEM; Zeiss EVO M10) is used to characterize the microscopic structures of PAAm/SA-Gel, as well as the PAAm/SA-DPG and PAAm/SA-PMS2/SA3-mDPG particles. The element mappings of a PAAm/SA-PMS2/SA3-mDPG particle is obtained by SEM testing with energy dispersive X-ray (EDX) analysis testing. The PAAm/SA-Gel sample used for SEM scanning is prepared by freeze-drying its bulk hydrogel. The PAAm/SA-DPG and PAAm/SA-PMS2/SA3-mDPG particles used for SEM scanning and EDX testing are prepared by diluting the DPG suspensions in ethanol and then drying the diluted solution at room temperature.
2.4.2. Fourier transform infrared (FTIR) spectroscopy
The FTIR spectra of the PAAm/SA-DPG and PAAm/SA-PMS2/SA3-mDPG particles are measured to determine the presence of specific bonds. The FTIR covers a wavenumber range of 4000 to 400 cm−1. The samples used for the spectral measurements are prepared by centrifuging PAAm/SA-DPG and PAAm/SA-PMS2/SA3-mDPG suspensions, followed by freeze-drying the centrifuged products.
2.4.3. Mechanical tests
The mechanical properties of PAAm/SA-Gel are measured at room temperature using an AGS-X universal tensile test machine (Shimadzu, Japan). The PAAm/SA-Gel sample is processed into stripes with a length of 20 mm, a width of 6.25 mm, and a thickness of 3.87 mm. These samples are stretched at a speed of 10 mm·min−1 during the tests. First, a tensile test is conducted on a sample until it is broken into two parts to obtain its nominal tensile stress, fracture stress, and toughness. Besides, several loading–unloading tensile tests are performed on other samples to investigate their energy dissipation characteristics. In these tests, strength, toughness, and dissipated energy of the hydrogel samples can be determined to gauge the mechanical properties of the hydrogel sample. Strength, corresponding to the peak stress in the stress–strain curve, is the maximum stress the sample can withstand before failure. Toughness can be determined by integrating the area under the stress–strain curve up to the fracture point. In loading–unloading tests, the dissipated energy can be obtained by calculating the area of the hysteresis loop being enclosed by the loading and unloading curves.
2.4.4. Rheology tests
We use an AR-G2 rheometer (TA Instruments) to measure the rheological properties of a cylindrical PAAm/SA-Gel sample with a thickness of 600 μm. The storage modulus (G′) and loss modulus (G′′) of this sample are measured over the angular frequency of 0.1–100 rad·s−1 at a fixed oscillation strain of 1%. In addition, the dynamic strain sweep is performed on this sample with strains varying from 1% to 10000% at a fixed frequency of 10 rad·s−1.
2.4.5. Thermogravimetric analysis (TGA)
TGA tests are performed on the freeze-dried PAAm/SA-DPG and PAAm/SA-PMS2/SA3-mDPG particles to measure their weight loss when they are heated from 25 to 500 °C with a heating rate of 10 °C·min−1.
2.4.6. Particle size distribution measurement
The particle size distributions of PAAm/SA-DPG and PAAm/SA-PMS2/SA3-mDPG are measured using a particle size analyzer (Mastersizer 3000, Malvern Panalytical, UK; 0.1–1000 μm).
2.4.7. Plugging tests
Two plugging tests are performed to investigate the performances of PAAm/SA-PMS2/SA3-mDPG and PAAm/SA-DPG in plugging high-permeability porous media. The detailed procedure of the plugging tests is given as follows. First, two sandpacks coded as #1 and #2 are prepared by compacting sand particles into two sandpack tubes (1 m in length and 0.05 m in diameter). Then, water is injected into the sandpacks to measure their physical properties, including pore volume (PV), porosity (), and permeability (K). The equations for calculating those properties can be found in Eqs. (S1)–(S3) in Appendix A. For sandpack #1, we inject 1 PV of PAAm/SA-DPG with a gel-to-water mass ratio of 5.26%, followed by 2.75 PV of chasing water. For sandpack #2, we inject 1 PV of a mixture containing 600 mL of PAAm/SA-DPG and 3.57 mL of PMS. In this mixture, the gel-to-water and PMS-to-SA mass ratios are 5.26% and 2:3, respectively. Next, CO2 is injected into sandpack #2 to reach a pressure of 60 psi (1 psi ≈ 6894.76 Pa). Once this pressure is reached, this pressure is kept for 48 h. Finally, 2.75 PV of chasing water is injected into sandpack #2. In these two tests, the pressures at the two locations are recorded along the sandpacks (i.e., 5 and 80 cm respectively away from the inlet). Herein, the pressure at 5 cm is used instead of the inlet pressure, since a reticular filter is mounted to the inlet of the sandpack tube, which generates additional pressure when fluid flows through the filter. The pressure at 80 cm is recorded to assess the performance of the DPG suspensions in plugging the deeper parts of the sandpack. The reduced permeabilities of the two sandpacks, plugging efficiencies, and residual resistance factors of the two DPG suspensions, can be determined based on the pressures recorded (Eq. (S4) in Appendix A).
3. Results and discussion
3.1. Fabrication of hydrogel and modification of gel particles
A double-network hydrogel containing a crosslinked PAAm network and a crosslinked SA network is pre-synthesized to develop the new DPG suspension.
Fig. 1(a) schematically illustrates how to synthesize the hydrogel. The skeletal formulae of chemicals used are presented in
Fig. 1(b). As shown in
Fig. 1(c), the PAAm network is developed by the free radical polymerization of AAm with MBAA as the crosslinker. The SA network is formed by crosslinking SA chains through the interactions between Fe(III) ions and carboxyl groups. Hydrogen bonds that exist between the amide groups of AAm and the hydroxyl groups of SA link the two networks.
Figs. 1(d) and
(e) display the digital images of PAAm/SA-Gel before and after the crosslinking of the SA chains, respectively.
Fig. 2 shows the synthesis process of the new DPG suspension (PAAm/SA-PMS/SA-mDPG). As depicted in
Fig. 2(a), PAAm/SA-PMS/SA-mDPG is obtained by modifying the PAAm/SA-DPG particles using PMS and CO
2.
Figs. 2(b)–(d) illustrate the modification mechanisms. The dissolution of CO
2 provides an acidic environment in PAAm/SA-DPG, which will facilitate the hydrolyzation of PMS. The hydrolyzation will generate methylsilanetriols (
Fig. 2(b)). The silicon hydroxyl groups of methylsilanetriols can react with the hydroxyl groups of the gel particles through condensation reactions (
Fig. 2(c)). Then, a coating layer can be generated on the surface of gel particles through the self-condensation reactions between silicon hydroxyl groups (
Fig. 2(d)) [
40]. In this manner, we can obtain the PAAm/SA-PMS
2/SA
3-mDPG suspension.
Fig. 2(e) displays the digital images of PAAm/SA-DPG and PAAm/SA-PMS
2/SA
3-mDPG suspensions.
3.2. Characterizations
The FTIR spectra of the PAAm/SA-DPG and PAAm/SA-PMS
2/SA
3-mDPG particles are revealed in
Fig. 3(a). The adsorption peaks at 3324 and 3187 cm
−1 are attributed to N–H stretching and O–H stretching, respectively [
41]. The adsorption peaks in the range of 1600–1670 cm
−1 indicate the C=O stretching [
42]. The adsorption peaks ranging from 1400 to 1450 cm
−1 correspond to the stretching vibrations of –CH
3, –CH
2, and –CH groups, while those ranging from 1000 to 1150 cm
−1 are attributed to the stretching vibrations of C–O, C–C, and C–N groups. All these adsorption peaks can be found in the spectra of the PAAm/SA-DPG and PAAm/SA-PMS
2/SA
3-mDPG particles, suggesting that PAAm/SA-Gel is successfully developed. Additionally, we can observe a new peak at 1271.22 cm
−1 in the spectrum of the PAAm/SA-PMS
2/SA
3-mDPG particles. This peak is the result of the Si–CH
3 stretching, indicating the presence of PMS in the modified gel particles [
43].
Fig. 3(a) also shows that the intensity of the peaks ranging from 1000 to 1150 cm
−1 of the PAAm/SA-PMS
2/SA
3-mDPG particles is significantly improved compared to that of the PAAm/SA-DPG particles. This indicates the presence of Si–O–C in the PAAm/SA-PMS
2/SA
3-mDPG particles as the adsorption peak of the stretching of Si–O–C is also detected in this region [
44], which proves that PMS is successfully grafted to the PAAm/SA-DPG particles. Meanwhile, two adsorption peaks at 767.44 and 485.94 cm
−1 emerging in the spectrum of the PAAm/SA-PMS
2/SA
3-mDPG particles result from the Si–O–Si stretching [
45],[
46]. This suggests the occurrence of the self-condensation reactions between silicon hydroxyl groups. The FTIR characterizations indicate the successful modification of the PAAm/SA-DPG particles.
Fig. 3(b) presents the particle size distributions of PAAm/SA-DPG and PAAm/SA-PMS
2/SA
3-mDPG. It can be seen that PAAm/SA-DPG has two peaks at 0.23 and 92.05 μm, respectively, and PAAm/SA-PMS
2/SA
3-mDPG has two peaks at 0.33 and 197.99 μm, respectively. After the modification, the proportion of the particles smaller than 1 μm is greatly decreased, and a large number of particles with sizes ranging from 10 to 40 μm and from 200 to 700 μm appear. These findings suggest that the modification causes remarkable swelling of the PAAm/SA-DPG particles.
Fig. 3(c) depicts the microscopic pore structure of PAAm/SA-Gel, showing a dense structure with few small pores.
Figs. 3(d) and
(e) show the microstructures of a PAAm/SA-DPG particle and a PAAm/SA-PMS
2/SA
3-mDPG particle, respectively. It can be observed that the PAAm/SA-PMS
2/SA
3-mDPG particle has a rougher surface than the PAAm/SA-DPG particle. A more detailed comparison of the surface roughness of two gel particles can be found from Fig. S2 in Appendix A.
Figs. 3(f) and
(g) display the SEM image of a PAAm/SA-PMS
2/SA
3-mDPG particle and the element mappings of this gel particle, respectively. The uniformly distributed Fe and Si elements imply the formation of the SA network and the successful development of the new DPG suspension.
3.3. Properties of PAAm/SA-Gel
To effectively plug the high-permeability channels, the gel materials used for preparing DPG suspensions are expected to be mechanically strong. The mechanical and rheological properties of PAAm/SA-Gel are measured and shown in
Fig. 4.
Fig. 4(a) shows the stress–strain curve of PAAm/SA-Gel and the digital image captured when this sample is stretched at a strain of ∼400%. It can be found that the stress of this sample gradually increases when the strain increases. Then, the stress suddenly drops when the strain increases to 525%. At this point, the sample is broken into two parts with a maximum stress of 0.259 MPa. The toughness of this sample is calculated to be 1.17 MJ·m
−2 (
Fig. 4(b)). The large strength and toughness values suggest that the PAAm/SA-Gel has a high mechanical strength. To further understand the energy dissipation characteristics of PAAm/SA-Gel, several loading–unloading tensile tests are performed.
Figs. 4(c) and
(d) show the stress–strain curves and dissipated energies of a PAAm/SA-Gel sample when it is stretched by five successive loading–unloading cycles with a constant strain of 200%. It is demonstrated that the sample exhibits a large hysteresis loop with a high dissipated energy of 168.04 kJ·m
−3 in the first cycle. However, the areas of the hysteresis loops and the dissipated energies significantly decrease in the subsequent four cycles. These results indicate that the polymer chains and networks of PAAm/SA-Gel sample are broken and cannot recover to their original status in a short time.
Figs. 4(e) and
(f) present the stress–strain curves and dissipated energies of the PAAm/SA-Gel sample when it is stretched for five cycles with maximum strains of 100%, 200%, 300%, 400%, and 500%, respectively. It can be found that the hysteresis loop becomes larger, and the dissipated energy gradually increases as the maximum strain increases. The results suggest that a larger strain imposed on the gel sample yields a more severe damage to its network structures.
Fig. 4(g) presents the measured storage modulus (
G′) and loss modulus (
G′′) of PAAm/SA-Gel in an angular frequency range of 0.1–100 rad·s
−1. The results illustrate that the
G′ and
G′′ of the sample increase to some extent as the frequency increases. At higher frequencies, the gel sample will be hardened because it becomes more difficult for the polymer networks to rearrange themselves quickly. Meanwhile,
G′ is consistently higher than
G′′ throughout the frequency sweep range, implying that the gel maintains its elasticity. Moreover,
Fig. 4(h) displays the measured moduli (
G′ and
G′′) of the sample over the strain range of 1%–2000% and those in a time sweep (300 s) at a strain of 1%. These tests can determine its linear viscoelastic region and the recovery ability. We can find that the sample maintains its elastic behavior when the strain is lower than 689.6% because
G′ is consistently higher than
G′′.
G′′ becomes higher than
G′ when the strain is larger than 689.6%, revealing that the networks of this sample are ruptured due to the severe dislocation of crosslinking points. The time sweep results show that
G′ becomes larger than
G′′ again, indicating that the sample recovers its elasticity after being damaged at high-strain conditions. These findings reveal that the gel has a good stability.
3.4. Properties of PAAm/SA-PMS/SA-mDPG
Another new DPG suspension is prepared by changing the PMS-to-SA mass ratio to 2.0:7.5 for investigating the effect of PMS concentration on the modification of the PAAm/SA-DPG particles. This new DPG suspension is coded as PAAm/SA-PMS
2/SA
7.5-mDPG.
Fig. 5(a) shows the particle size distributions of PAAm/SA-DPG, PAAm/SA-PMS
2/SA
7.5-mDPG, and PAAm/SA-PMS
2/SA
3-mDPG. Due to non-uniform shearing, PAAm/SA-DPG has a wide size distribution ranging from 0.1 to 153 μm with two peaks at 0.33 and 81 μm. When being modified by PMS with a PMS-to-SA mass ratio of 2.0:7.5, gel particles smaller than 10 μm almost disappear. Meanwhile, gel particles larger than 153 μm appear because of the swelling of gel particles. However, the peak value slightly decreases from 81 to 71 μm. In contrast, PAAm/SA-PMS
2/SA
3-mDPG contains gel particles with much larger sizes. Compared to PAAm/SA-DPG, the proportion of gel particles smaller than 81 μm is significantly decreased. Meanwhile, the peak value of gel particle with larger sizes remarkably increases from 81 to 224 μm, with a large swelling factor of 2.77. Therefore, from the perspective of plugging high-permeability pathways for profile control, using PMS with a PMS-to-SA mass ratio of 2:3 is the optimal scheme for modifying the PAAm/SA-DPG particles.
To clarify the role of CO
2 in gel particle modification, the particle size distributions of several PAAm/SA-PMS-mDPG suspensions modified by different gases (including CO
2, N
2, and air) are compared. These suspensions are coded as PAAm/SA-PMS
2/SA
3-mDPG (CO
2), PAAm/SA-PMS
2/SA
3-mDPG (N
2), and PAAm/SA-PMS
2/SA
3-mDPG (air), respectively.
Fig. 5(b) shows such comparison. We can see that CO
2, N
2, and air can contribute to the swelling of the PAAm/SA-DPG particles to different extents. Compared to PAAm/SA-DPG, the peak values of larger particles (> 10 μm) in PAAm/SA-PMS
2/SA
3-mDPG (CO
2), PAAm/SA-PMS
2/SA
3-mDPG (N
2), and PAAm/SA-PMS
2/SA
3-mDPG (air) increase from 81 to 118, 153, and 224 μm, respectively. This indicates that CO
2 exhibits the most promising effect, while N
2 has the least effect on swelling gel particles. The distinct swelling effects arise from the different abilities of these gases to contribute hydrogen ions (H
+). Specifically, the dissolution of CO
2 in PAAm/SA-DPG can provide the largest amount of H
+, which can facilitate the hydrolysis of PMS and strengthen the reactions shown in
Figs. 2(b)–(d). In contrast, the dissolution of air can only provide a much smaller amount of H
+, and the dissolution of N
2 cannot provide H
+ at all. Thus, it can be concluded that the modification of the PAAm/SA-DPG particles is strongly responsive to CO
2, which is a particularly favorable mechanism in CCUS applications.
As aforementioned, we hope that the swelling of the particles used for profile control upon exposure to CO
2 is irreversible at high temperatures. PAAm/SA-PMS
2/SA
3-mDPG is heated at 100 °C for 24 h to examine the effect of high temperatures on the size of the PAAm/SA-PMS
2/SA
3-mDPG particles. The aged sample is coded as PAAm/SA-PMS
2/SA
3-mDPG (100 °C).
Fig. 5(c) compares the particle size distributions of PAAm/SA-PMS
2/SA
3-mDPG and PAAm/SA-PMS
2/SA
3-mDPG (100 °C). This figure proves that the gel particle swelling will not be reversed after being heated at 100 °C for 24 h. It is interesting to observe that the peak value of gel particle size distribution further increases from 224.95 to 425.89 μm. This result indicates that the new DPG suspension developed in this study can be used for profile control in high-temperature CCUS applications. Furthermore, the effect of water salinity on the gel particle swelling is also examined, as shown in Fig. S3 in Appendix A. In the experiments, the gel particles are modified in distilled water and saline water with a salinity of 2.5 × 10
5 parts per million (ppm), respectively. In the saline water, the concentration of divalent cations is 0.4 × 10
4 ppm. From the testing results, it can be concluded that a high salinity environment will inhibit gel particle swelling. Despite this, the modification enables the gel particles to swell about twice their original size in the saline water.
Fig. 5(d) presents the TGA results of the PAAm/SA-DPG and PAAm/SA-PMS
2/SA
3-mDPG particles. We can observe that the decomposition onset temperatures of the PAAm/SA-DPG and PAAm/SA-PMS
2/SA
3-mDPG particles are 200 and 240 °C, respectively. The PAAm/SA-DPG particles lose their weight by 12.97% when being heated to 200 °C and further lose their weight by 35.34% when temperature reaches 500 °C. In contrast, the weight losses of the PAAm/SA-PMS
2/SA
3-mDPG particles at 240 and 500 °C are 13.68% and 35.95%, respectively. The elevated decomposition onset temperature and reduced mass loss of the PAAm/SA-PMS
2/SA
3-mDPG particles demonstrate that they exhibit a stronger thermal ability than the PAAm/SA-DPG particles. In addition,
Figs. 5(e) and
(f) show the digital images of the two samples after being heated to 500 °C. It can be seen that the PAAm/SA-DPG particles crumble into pieces when being slightly touched by a tweezer. However, the PAAm/SA-PMS
2/SA
3-mDPG particles are quite hard and can be easily clamped by a tweezer. This comparison further indicates that the PAAm/SA-PMS
2/SA
3-mDPG particles have a much stronger high-temperature resistance than the PAAm/SA-DPG particles. This is attributed to the silsesquioxanes (SS, CH
3SiO
1.5) coating generated in the modification process, which has a good thermal insulation ability [
47]. It can protect the internal PAAm/SA-DPG particles from being decomposed at high temperatures, as schematically illustrated in
Fig. 5(g). This property is very useful because it helps gel particles effectively plug high-permeability channels in high-temperature reservoirs. In addition, the SS coating is widely used for fabricating waterproof materials due to its hydrophobic property [
48]. In our work, we drop a water droplet on the surface of the PAAm/SA-PMS
2/SA
3-mDPG particles to determine their wettability (Fig. S4 in Appendix A). The measured contact angle is around 138°, indicating that the particles are hydrophobic. This observation further proves the successful generation of the SS coating.
Furthermore, some modified DPG suspensions with PMS-to-SA mass ratios of 2:1.875, 2:1.5, and 2:1.25 are prepared. It is interesting to find that some precipitate-like substances appear in these suspensions (
Fig. 6(a)). In contrast, there are no precipitate-like substances in PAAm/SA-PMS
2/SA
7.5-mDPG and PAAm/SA-PMS
2/SA
3-mDPG (
Fig. 6(b)). One possible ingredient of the precipitate-like substances is iron(III) hydroxide (Fe(OH)
3), since the hydrolysis of PMS may produce hydroxyl ions (OH
−) which can react with the Fe(III) ions. To determine the compositions of the precipitate-like substances in PAAm/SA-PMS
2/SA
1.875-mDPG, we first centrifuge the suspension and freeze-dry the centrifuged sample. Then, the FTIR spectrum of the precipitate-like substances is obtained and compared with that of the PAAm/SA-PMS
2/SA
3-mDPG particles, as shown in Fig. S5 in Appendix A. The trends of the two spectra are almost identical, indicating that they contain similar functional groups. One way to determine whether the precipitate-like substances contain Fe(OH)
3 is to remove Fe
3+ in the modification process. Thus, a new hydrogel is prepared following the same procedure used for preparing PAAm/SA-Gel, without immersing the bulk hydrogel in FeCl
3 solution. As a result, the PAAm network in this gel is crosslinked by MBAA, but the SA network is un-crosslinked by Fe
3+, leading to the so-called semi-interpenetrating network [
49]. This gel is coded as PAAm/unSA-Gel. The structure of PAAm/unSA-Gel is schematically given in Fig. S6 in Appendix A. A PAAm/unSA-DPG suspension is prepared by shearing PAAm/unSA-Gel in water. Then, the PAAm/unSA-DPG particles are modified by adding PMS with different PMS-to-SA mass ratios and purging CO
2. The modified PAAm/unSA-DPG suspensions with PMS-to-SA mass ratios of 2:7.5, 2:3, 2:1.875, 2:1.5, and 2:1.25 are coded as PAAm/unSA-PMS
2/SA
7.5-mDPG, PAAm/unSA-PMS
2/SA
3-mDPG, PAAm/unSA-PMS
2/SA
1.875-mDPG, PAAm/unSA-PMS
2/SA
1.5-mDPG, and PAAm/unSA-PMS
2/SA
1.25-mDPG, respectively. Fig. S7 in Appendix A provides the images of these DPG suspensions. As seen in Fig. S7, precipitate-like substances still appear when the PMS-to-SA mass ratio is higher than 2:1.875. These observations prove that the precipitate-like substances are the re-crosslinked gel particles rather than Fe(OH)
3 precipitates, as PAAm/unSA-DPG does not contain iron(III) ions. Therefore, we can conclude that there are two modes of gel particle modification when varying the PMS-to-SA mass ratio. When the PMS-to-SA mass ratio is lower than 2:1.875, multiple SS coating layers can be formed on the gel-particle surfaces, as illustrated in
Fig. 6(c). In this case, the gel particles swell without forming the precipitate-like substances (
Fig. 6(b)). When the PMS-to-SA mass ratio exceeds 2:1.875, the precipitate-like substances will appear, as evidenced by
Fig. 6(a). This phenomenon is caused by the excessive PMS which acts as a re-crosslinker to congregate the suspended gel particles (
Fig. 6(d)).
3.5. Plugging performance
Two core flooding experiments are conducted to test the abilities of PAAm/SA-DPG and PAAm/SA-PMS
2/SA
3-mDPG in plugging high-permeability sandpacks. The apparatus used in the experiments is displayed in
Fig. 7(a).
Fig. 7(b) shows the pressures recorded at 5 and 80 cm respectively of sandpack #1 during three stages: initial water flooding, PAAm/SA-DPG injection, and chasing water flooding. At the end of the initial water flooding, the initial permeability, porosity, and pore volume of sandpack #1 are measured to be 20.52 μm
2, 31.47%, and 618 mL, respectively. The subsequent PAAm/SA-DPG injection leads to increases in the pressures at the two locations, suggesting that the PAAm/SA-DPG particles form plugging within the sandpack. The pressures at 5 and 80 cm gradually decrease with the injection of chasing water and finally stabilize at 5.6 and 0.6 psi, respectively.
Fig. 7(c) displays the pressures recorded at the same locations of sandpack #2 during three stages: initial water flooding, mixture (PAAm/SA-DPG and PMS) injection, and chasing water flooding. The measured initial permeability, porosity, and pore volume of sandpack #2 are 30.78 μm
2, 30.56%, and 600 mL, respectively. The pressures increase during the mixture injection due to the plugging effect yielded by the PAAm/SA-DPG particles. CO
2 is subsequently injected and retained in sandpack #2 for 48 h, followed by chasing water flooding. The pressures at the two locations experience a remarkable increase at the earlier stage of the chasing water flooding. Further injection of chasing water decreases the pressure profiles. Finally, the pressures at 5 and 80 cm reach 22.5 and 9.8 psi, respectively. We can also see from
Figs. 7(b) and
(c) that the pressure difference between the location at 80 cm and the outlet location at 100 cm in sandpack #2 reaches 9.8 psi; this is much higher than that achieved in sandpack #1 (0.6 psi). Correspondingly, the permeabilities over the 80–100 cm section of the two sandpaks are 16.41 and 24.62 D (1 D ≈ 0.9869 × 10
−12 m
2), respectively. After being plugged by the gel particles in PAAm/SA-DPG and the modified gel particles, the permeabilities of over the same section of the two sandpaks are reduced to 8.21 and 0.50 D, respectively. This reveals that the deeper part (80–100 cm) of sandpack #2 is much more effectively plugged. To effectively plug the deeper part of a reservoir is essential for achieving in-depth profile control [
50].
Using the above experimental data, the reduced permeabilities of the two sandpacks and the plugging efficiencies of the two DPG recipes can be calculated (
Fig. 7(d)). The permeability of sandpack #1 is reduced from 20.52 to 3.69 μm
2 due to the injection of the PAAm/SA-DPG particles. The plugging efficiency of PAAm/SA-DPG is calculated to be 82.8%. In contrast, the use of the PAAm/SA-PMS
2/SA
3-mDPG particles in sandpack #2 reduces the permeability from 30.78 to 1.45 μm
2, providing a much higher plugging efficiency of 95.3%. Moreover, the residual resistance factors of the PAAm/SA-DPG and PAAm/SA-PMS
2/SA
3-mDPG suspensions are determined to be 5.56 and 21.23, respectively. These results reveal the promising performance of the PAAm/SA-PMS
2/SA
3-mDPG suspension in plugging the ultra-high permeability porous media.
Fig. 7(e) schematically illustrates how this new recipe works in the pore spaces. The PAAm/SA-DPG particles are distributed in the pores, forming mild plugging after the mixture injection. The particles are then modified to become PAAm/SA-PMS
2/SA
3-mDPG particles due to the exposure to CO
2 in the presence of PMS. The modified particles have much larger sizes than the unmodified particles, thereby providing a much stronger plugging performance.
4. Conclusions
In this work, a novel DPG suspension for profile control in high-temperature CCUS applications is developed. This new DPG suspension is developed by modifying the particles of a strong double-network hydrogel using CO2 and PMS with a PMS-to-SA mass ratio of 2:3. The modification swells the double-network hydrogel particles by over two times, and such swelling is irreversible at high temperatures. Meanwhile, the modification greatly improves the thermal stability of the double-network gel particles by forming SS coating layers on their surface. These two properties enable the new DPG particles to be more favorable for high-temperature CCUS applications than the conventional CO2-responsive gel particles. Interestingly, the suspended gel particles can be re-crosslinked if the PMS-to-SA mass ratio is further increased, since PMS can act as a re-crosslinker at higher concentrations. In addition, modifying the double-network gel particles within porous media helps yield a high plugging efficiency of 95.3%, whereas the plugging efficiency of the unmodified gel particles is only 82.8%. The unique characteristics of high swelling ratio, irreversible swelling at high temperatures, improved high-temperature stability, and high plugging efficiency of the new DPG suspension allow it to be a promising candidate for profile control in high-temperature CCUS applications.
CRediT authorship contribution statement
Lin Du: Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Conceptualization. Yao-Yu Xiao: Writing – review & editing, Validation, Methodology, Formal analysis, Conceptualization. Zhi-Chao Jiang: Writing – review & editing, Validation, Methodology, Investigation, Formal analysis, Conceptualization. Hongbo Zeng: Writing – review & editing, Supervision, Methodology, Formal analysis, Conceptualization. Huazhou Li: Writing – review & editing, Supervision, Project administration, Methodology, Formal analysis, Conceptualization.
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
Lin Du acknowledges the financial support provided by China Scholarship Council (CSC) via a Ph.D. Scholarship (202008510128). This work was financially supported by Core Technology Project of China National Petroleum Corporation (CNPC) “Research on Thermal Miscible Flooding Technology” (2023ZG18)
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.eng.2025.04.002.
PDF
(4445KB)
