Preparation and characterization of core-shell oil absorption materials stabilized by modified fumed silica

2.1 Materials

Methyl methacrylate (MMA; 98%; purchased from Sinopharm Chemical Reagent Co., Ltd., Shanghai, China) and lauryl methacrylate (LMA; 95%; purchased from Aladdin Industrial Corp., Shanghai, China) were used as monomers. Divinylbenzene (DVB; 55%; obtained from Aladdin Industrial) was used as a cross-linking agent. Benzoyl peroxide (BPO; 98%; J&K Scientific Ltd., Beijing, China) was used as initiator. Ethyl acetate (EA; 99.5%; purchased from Sinopharm Chemical Reagent) was used to make pores. Toluene (99%; purchased from Sinopharm Chemical Reagent) was used to test the oil absorption rate of OAMs. Gelatin (chemical pure) was purchased from Aladdin Industrial. Deionized water was used throughout the experiments.

Modified fumed silica particles were provided by Suzhou Huize Fine Chemical Co., Ltd. (Suzhou, China). The wettability of fumed silica particles was neutral and the median diameter was 4.63 μm (Figure 1A) as measured by Malvern Zetasizer NanoZS (Malvern, UK). The morphology of the modified fumed silica was investigated by an optical microscope (Figure 1B).

Figure 1:

Characteristics of modified fumed silica: laser diffraction particle size distribution curves (A) and optical micrograph (B).

2.2 Synthesis of core-shell OAM microspheres

Core-shell OAM microspheres were prepared by Pickering polymerization, and the synthesis process was as follows: 200 g deionized water was put into a 500 ml three-mouth flask, and a certain quality of modified fumed silica (2 g) was added in the flask and the mixture was stirred well. Then, a suitable amount of component (5.5 g MMA and 7 g LMA) was put into a beaker, wherein a cross-linking agent (0.06 g DVB), an initiator (0.2 g BPO), and a pore-forming agent (12.5 g EA) were added in sequence, and all of the compositions were stirred until the initiator (BPO) was completely dissolved. Then, the evenly stirred oil phase was dropped into the water phase at a stirring rate of 1000 rpm, and the stirring rate was slowed down to 300 rpm 10 min later after the oil phase was totally dropped into the water phase. The water bath temperature was set to 85°C and the reaction time was set to 150 min. Core-shell OAM microspheres were filtered after the reaction and washed three times with deionized water at 80°C. Finally, the microspheres were dried at 80°C for 24 h to vaporize water and EA.

2.3 Scanning electron microscope (SEM)

The morphology of core-shell OAM microspheres was observed under an S-4800 cold field emission scanning electron microscope (SEM). The microspheres were sprayed with gold before the test. SEM analyses were conducted at the State Key Laboratory of Heavy Oil Research at the China University of Petroleum (East China).

2.4 Fourier transform infrared (FTIR) spectrometry

FTIR spectra of core-shell OAM microspheres were recorded on a Bio-Rad FTS135 FTIR spectroscope at the State Key Laboratory of Heavy Oil Research at the China University of Petroleum. The test conditions were potassium bromide pellets and wavelength range of 4000 to 400 cm^-1.

2.5 Particle size distribution test

The particle size distribution of core-shell OAM microspheres was tested using Bettersize 2000 laser particle distribution instrument. Particles were dispersed in deionized water before the test and the concentration was 1 wt.%.

2.6 Oil absorption test

Because the cement slurry is alkaline and the pH value is approximately 13, core-shell OAM microspheres were alkali treated at 75°C and pH 13 for 300 min. Then, the oil absorption rates of the microspheres before and after alkali treatment were tested in toluene at 75°C. The measurement procedure was as follows: microspheres with a quantity of W₁ were placed into a nylon bag. The quantity of microspheres and nylon bag was weighed and the value was denoted as W₂. The nylon bag was put into toluene at 75°C. After 24 h, the nylon bag was taken out and suspended until no droplet is dripping. Then, the nylon bag was weighed again and the value was denoted as W₃. The oil absorption rate of microspheres is given by

where G is the oil absorption rate of OAMs (g/g).

2.7 Contact angle test

The contact angle can directly reflect the hydrophilicity or lipophilicity of core-shell OAM microspheres and indirectly explain their dispersion in cement slurry. The contact angles of microspheres synthesized using gelatin or fumed silica particles as dispersant were tested. Kostakis et al. [36] and Yang et al. [37] provided methods for testing the contact angle of powder. The method used in this paper was as follows: OAMs (0.5 g) were put into a compression molding device and compressed under a pressure of 5 MPa for 1 min into a test piece with a diameter of 15 mm and a thickness of 2 mm. The contact angle of the test piece was tested using PDE-1700 contact angle meter (KRÜSS Co., Ltd., Hamburg, Germany).

3.1 Pickering polymerization stabilized by modified fumed silica

The Pickering polymerization of core-shell OAM microspheres had two stages: dispersion stage and encapsulation stage. The dispersion stage corresponded to the high-speed stirring stage and lasted 10 min, whereas the encapsulation stage, which is also known as the polymerization stage, corresponded to the low-speed stirring stage and lasted 150 min.

The oil phase contained a monomer (MMA and LMA), a pore-forming agent (EA), a cross-linking agent (DVB), and an initiator (BPO). In the dispersion stage, the oil phase was cut into oil droplets under the given stirring rate. Fumed silica particles were absorbed on the oil/water interface and prevented the droplets from coalescence. This stage is the formation of Pickering emulsion. Interface adsorption energy is a quantity that reflects the reduction of interface energy when inorganic particles are adsorbed on the oil/water interface. For a spherical particle with radius R, its interface adsorption energy formula is [38], [39]

where E is the interface adsorption energy (×10^-8 J), R is the radius of a spherical particle (nm), γ_OW is the interfacial tension between oil and water (N/m), and θ is the contact angle of the particle. When the particle diffuses into the water, the sign in parentheses is negative; on the contrary, the sign is positive.

Binks and Lumsdon [40] studied the interface adsorption energy of a spherical particle (R=1×10^-8 nm) at a planar oil/water interface changed with the contact angle θ. The results showed that the interface adsorption energy E was much higher than the energy of thermal motion k_BT (where k_B is the Boltzmann constant and T is the thermodynamic temperature) when the contact angle θ was 90°. In this paper, fumed silica was modified by a silane coupling agent and the contact angle was a little <90°. Therefore, the adsorption of modified fumed silica was irreversible, and Pickering emulsion was prepared.

Figure 2A shows the Pickering emulsion under static condition after 10 min of high-speed stirring. Two layers were formed: the upper layer was the layer of oil droplets suspended in water and the lower layer was the layer of water phase containing fumed silica particles. Therefore, fumed silica particles were not totally adsorbed on the oil/water interface, and there were quite a lot of particles suspended in the aqueous phase. Frelichowska et al. [41] used to analyze the presence of excess silica coexisting with oil droplets in the emulsion by ultracentrifugation and infrared absorption spectroscopy. The excess fumed silica in the bulk phase could form a three-dimensional network surrounding the droplets [42], [43], [44], which prevented droplets from coalescence. The larger interface adsorption energy and the three-dimensional network of fumed silica were the two main factors that stabilized Pickering emulsion.

Figure 2:

Photographs of Pickering emulsion samples at different times: (A) 0 min, (B) 7 min, (C) 15 min, (D) 20 min, and (E) 25 min of the polymerization stage.

The samples were allowed to stand for 2 min at room temperature.

During polymerization, the monomer (MMA and LMA) was polymerized in the presence of BPO and then formed short-chain molecules. As time went by, short-chain molecules formed long-chain molecules and cross-linked by DVB and then formed a space network structure. Cheng et al. [45] found that nanocomposites could combine with the polymer because of attractive interactions, and the static thickness of the interfacial layer was found to increase with the molecular weight. Therefore, fumed silica particles would form a shell layer around the polymer. On the contrary, the viscosity of the oil phase was increasing during the stage, and solid microspheres were created at last.

As shown in Figure 2, it is clear that extra fumed silica particles suspended in the aqueous phase were reduced from the start to 25 min, and the aqueous phase became clear at 25 min. This means that fumed silica particles were constantly absorbed on the oil/water interface. This phenomenon was analyzed by the deformation of droplets and the viscosity of the oil phase.

Experiments [41] have shown that, under the same mechanical strength, the interfacial area per unit mass of silica was larger when the oil phase had a low viscosity than when the oil phase had a high viscosity. For the polymerization stage, the viscosity of the oil phase increased with time. At the beginning, Pickering emulsion had a large interfacial area per unit mass of silica, and then interfacial area should decrease with increasing viscosity of the oil phase. That is to say, fumed silica particles would desorb from the oil/water interface. However, the desorption described by Eq. (2) is not possible.

The oil droplets suspended in the aqueous phase are not always spherical. They are easy to deform in stirring. It is well known that a spherical oil droplet has the minimum oil/water interface compared to any other shape at the same volume. Therefore, the oil/water interface would increase after the deformation of droplets (Figure 3C and F). Binks and Whitby [46] found that the droplets of viscous silicone oil did not find it easy to go back to a spherical shape after their deformation in strong shearing. Therefore, the new oil/water interface was formed constantly during the polymerization stage, and it led extra fumed silica particles in aqueous phase adsorb on the interface to reduce the interfacial tension (Figure 3D and G). When the droplets relaxed back to spherical phase, their surfaces would reduce and fumed silica particles stacked together to form multilayers (Figure 3E and H). Unlike Pickering emulsion, extra fumed silica particles decreased or even dispersed in Pickering polymerization as time went by. This phenomenon could prevent viscous droplets from adhesion and coalescence, and it could stabilize the polymerization stage from the beginning to the end.

Figure 3:

Schematic illustration of the preparation of core-shell OAM microspheres with fumed silica shell based on Pickering polymerization.

During the encapsulation stage, fumed silica particles were constantly absorbed on the oil/water interface and combined with the inner polymer to form the shell layer. Because cement is an inorganic material, this layer would well improve the applicability of OAM in cement and increase the combination capacity between organic OAM and inorganic cement.

3.2 Particle size of core-shell OAMs

The concentration of silica particles in the bulk phase and the amount of oil phase are the two factors that influenced the droplet size [46], [47] and determined the size of core-shell OAM microspheres. The concentrations of fumed silica in the aqueous phase varied from 0.3 to 2 wt.%, and the ratio of oil to water was 12.5 wt.%. When the amount of silica was lower than 0.3 wt.%, Pickering emulsion was not stable and droplets were easy to adhere during polymerization.

Having more stable polymerization process, core-shell OAM microspheres had a narrow particle size distribution (Figure 4). As shown in Figure 5, the particle size of microspheres is reduced when the amount of fumed silica is 0.3 to 1 wt.%. It stops reducing when the amount is larger than 1 wt.% and reaches a plateau. For Pickering emulsion, Wiley [48] and Arditty et al. [49], [50], [51] have studied that the droplet size is reduced as the amount of particles increases. However, this trend would not last. Midmore [52] and Abend et al. [42] also observed a “plateau” in the silica concentration range. This is because the emulsification process needs energy to shear the oil droplets to smaller size, but ultrasonic dispersion could not provide enough energy. For Pickering polymerization, it is the same. With a large amount of fumed silica, there is a limit particle size for core-shell OAM microspheres under stirring condition.

Figure 4:

Particle size distribution curves: (A) cumulative distribution and (B) difference distribution.

The concentration of fumed silica used for preparing core-shell OAM microspheres was 1%.

Figure 5:

Relationship between silica content and mean diameter of core-shell OAM microspheres.

3.3 FTIR test

FTIR spectra were used to verify whether core-shell OAM microspheres contained silica. We prepared two kinds of OAMs: one used fumed silica as dispersant and the concentration was 2 wt.% (Figure 6B) and the other used gelatin as dispersant and the concentration was 1 wt.% (Figure 6A).

Figure 6:

FTIR spectra of polyacrylate (A) and core-shell OAM microspheres (B).

Analyzing the peaks of the two curves (Figure 6A and B), the characteristic absorption band at 2927 cm^-1 corresponds to the aliphatic C-H bond and unsymmetrical stretching vibrations of -CH₃ groups, and the characteristic absorption band at 2854 cm^-1 is due to the aliphatic C-H symmetrical stretching vibrations of -CH₂ groups of the 11 -CH₂ of LMA. It is clear that there is a characteristic absorption peak of the asymmetric bending vibrations of C-H at 1450 cm^-1, the characteristic absorption band of C=O appears at 1732 cm^-1, and that of C-O-C appears at 1149 and 1242 cm^-1. These characteristic absorption peaks confirmed that MMA and LMA are in a copolymerization reaction with OAMs. However, from the FTIR spectra of core-shell OAM microspheres, the characteristic absorption bands at 1101 and 471 cm^-1 are the characteristic absorption bands of SiO₂. It confirms that core-shell OAM microspheres are a composite polymer of SiO₂, MMA, and LMA.

3.4 Micromorphology

To explain the role played by fumed silica particles in the preparation process, two kinds of OAMs using fumed silica particles and gelatin as dispersant, respectively, were analyzed by SEM. Microspheres using fumed silica particles as dispersant do not adhere together and have high sphericity (Figure 7C). However, microspheres using gelatin as dispersant do not have the same properties, and the microspheres in different sizes cohere together (Figure 7A). Thus, it is clear that fumed silica particles play a more effective role than gelatin in polymerization. On the one hand, fumed silica particles can stabilize the polymerization process; on the other hand, they can make the particle size of microspheres uniform. As shown in Figure 7D, there is a fumed silica shell on the surface of microspheres using fumed silica particles as dispersant, whereas there is no shell on the surface of microspheres using gelatin as dispersant as shown in Figure 7B. This illustrates the combination of fumed silica shell and the inner core and the preparation of core-shell OAM microspheres.

Figure 7:

SEM images of polyacrylate microspheres (A and B) and core-shell OAM microspheres before alkali treatment (C and D) and after alkali treatment (E and F).

Core-shell OAM microspheres were alkali treated under 75°C.

3.5 Oil absorption rate

The oil absorption test was used to evaluate whether fumed silica shell affects the oil absorption performance of core-shell OAM microspheres. In this experiment, 1 wt.% concentration of fumed silica was used for core-shell OAMs microspheres, whereas 1 wt.% concentration of gelatin was used for the other kind of OAMs. The concentration of the pore-forming agent (EA) varied from 0 to 150 wt% relating to the total mass of monomer (MMA and LMA). Because polyacrylate microspheres using gelatin as dispersant coalesced together and the particle size was uneven, the lumps were dried and smashed, and the oil absorption rate was measured using particles of 150–250 μm.

As shown in Figure 8, the oil absorption rate of the two kinds of OAMs is close, and this explains that the kind of dispersing agent has little effect on the oil absorption rate. However, the oil absorption rate of core-shell OAM microspheres is smaller than polyacrylate microspheres overall. This is because fumed silica multilayer shells combined with the inner core and limited the swelling of polyacrylate while absorbing oil.

Figure 8:

Relationship between EA content and oil absorption rate of polyacrylate microspheres (A) and core-shell OAM microspheres (B).

3.6 Alkali resistance

To measure the alkali resistance of OAMs, the oil absorption rates of OAMs using gelatin and fumed silica as dispersing agent were tested after alkali (pH 13) treatment at 75°C for 300 min and compared to that before alkali treatment (Figures 9 and 10). The oil absorption rate of core-shell OAM microspheres after alkali treatment is higher than before treatment (Figure 9). This is because fumed silica shell on the surface of microspheres can react with alkali liquor and dissolve. Therefore, the limitation of fumed silica shell on the expansion of core-shell OAM microspheres was relieved partly while absorbing oil. The oil absorption rate of polyacrylate microspheres after alkali treatment is lower than before treatment (Figure 10). This illustrates the hydrolysis reaction of polyacrylate in an environment and that the structure of the polyacrylate was destroyed. Due to the existence of fumed silica shell, which prevented the inner polyacrylate core from contacting alkali liquor, the alkali resistance of core-shell OAM microspheres using fumed silica particles as dispersant was stronger than polyacrylate using gelatin as dispersant. This explained that core-shell OAM microspheres can be used directly in cement slurry downhole.

Figure 9:

Change of oil absorption rate of core-shell OAM microspheres before alkali treatment (A) and after alkali treatment (B).

Figure 10:

Change of oil absorption rate of polyacrylate microspheres before alkali treatment (A) and after alkali treatment (B).

Besides, the oil absorption rate of OAMs is mainly controlled by the amount of EA and is increased with the increasing amount of EA (Figures 8–10). It is because the space grid structure of polyacrylate is controlled by the amount of the pore-forming agent and is an important factor that affects the oil absorption rate of OAMs [53].

SEM was used to analyze the variation of the surface morphology of core-shell OAM microspheres before and after alkali treatment. The fumed silica shell was partly dissolved after alkali treatment (compare Figure 7D to F). This is because fumed silica was amorphous and could react with alkali liquor at high temperature [30]. However, fumed silica formed multilayer shells on the surface of polyacrylate microspheres, and the shells were too strong to be totally dissolved in alkali liquor. The multilayer shells provided a barrier between the inner polyacrylate core and the alkali liquor and prevented polyacrylate from hydrolyzing.

3.7 Contact angle

The cement slurry is hydrophilic, but the polyacrylate microspheres synthesized using conventional methods are lipophilic. Therefore, polyacrylate microspheres would be agglomerated if added directly in cement slurry and would not have the ability of sealing leaks due to the poor dispersing performance in cement slurry. The new core-shell OAM microspheres had broken this limit. The contact angle of core-shell OAM microspheres is 84.6° (Figure 11A) and the contact angle of polyacrylate microspheres is 120.3° (Figure 11B). The contact angle of polyacrylate microspheres is larger because polyacrylate was lipophilic. However, the contact angle of core-shell OAM microspheres is small, which is due to fumed silica shell combined with inner polyacrylate core, and the contact angle of core-shell OAM microspheres is near the contact angle of fumed silica particles. Comparing the contact angle of polyacrylate microspheres to that of core-shell OAM microspheres, core-shell OAM microspheres will have a better dispersion in cement slurry.

Figure 11:

Contact angle images of core-shell OAM microspheres (A) and polyacrylate microspheres (B).