The preparation of novel mineral-based mesoporous microsphere by microencapsulation technology and its application in the adsorption of dye contaminants

Azo dyes are prevalent in printing and dyeing wastewater and pose a potential threat to the ecological environment and human health due to their high mobility and persistence (Liu et al. 2017). Literatures indicate that long-term exposure to wastewater containing rhodamine B (RhB) can cause many health problems such as skin and eye injuries (Zhao et al. 2017). Although methylene blue (MB) is not a particularly dangerous compound compared with other toxic substances and dyes, MB can easily produce singlet oxygen and other oxidizing substances under white light, which can destroy DNA structure and pose a potential threat to life (Albadarin et al. 2017). Therefore, it is imperative to carry out the treatment of wastewater containing RhB or MB. The traditional treatment technologies of printing and dyeing wastewater are as follows: membrane, adsorption, reverse osmosis, catalytic ozonation, anion exchange, etc. (Zhao et al. 2017). Although photodegradation is also an efficient method for dyeing wastewater treatment, it is difficult to realize regeneration due to the loss of photocatalyst in the reaction process (Sharma et al. 2018; Naushad et al. 2019b). Adsorption technology is considered to be an effective way to control organic pollutants due to its low cost, convenient operation and efficient removal of pollutants.

As a natural adsorption material, montmorillonite is considered to replace the activated carbon due to its high ion exchange capacity, large specific surface area, structural plasticity, and low cost (Rathnayake et al. 2015; Zhu et al. 2016). However, the limited interlayer environment limits the adsorption capacity of organic pollutants onto montmorillonite. It is also difficult to separate montmorillonite from water solution after adsorption reaction due to small particle diameter and good hydrophilicity (Liu et al. 2011). It is noteworthy that the silicon ion (Si⁴⁺) or aluminum ion (Al³⁺) of montmorillonite can be substituted or partially replaced with inorganic ions or Stearyl trimethyl ammonium chloride (STAC) molecule due to the strong variability of silicon–oxygen–aluminum (Si–O–Al) bonds, which have an important effect on the active adsorption sites and electron cloud densities (Sharma et al. 2016). Through a series of modification on the surface charge and interlayer environment, there is a significant improvement in the adsorption performance of montmorillonite.

Microencapsulation technology has been widely used in many fields, such as food, medicine, and pesticides. Its core technology is how to implement the efficient encapsulation of core-shell materials (Sudipta et al. 2012). Nowadays, in situ polymerization, sol-gel reaction, interfacial polymerization, microemulsion polymerization, solvent evaporation/solvent extraction, and Pickering emulsion methods are the main methods of microsphere fabrication (Nesterova et al. 2011; Dong et al. 2015). Pickering emulsion method is widely applied in the microsphere fabrication because of its low toxicity, low cost, excellent droplet stability, and anti-emulsification (Sudipta et al. 2012; Yi et al. 2016; Yu et al. 2018). In recent years, silicon-based, organic silicon-based, and metal oxide-based porous microspheres have been used for the efficient adsorption of heavy metals or organic pollutants (Chakraborty et al. 2014). However, few studies on mineral-based microspheres have been reported. Microspheres have two advantages: first, the core material has no direct contact with the outside to avoid the influence of the external environment; second, the structural stability of shell material ensures that microspheres will not be damaged after repeated use. Literature indicates that chemical reactions involving active substances in microsphere are controlled in small regions, in which reactants and intermediates have higher concentrations, greatly improving the reaction efficiency (Liu et al. 2016; Dong et al. 2018; Kamble et al. 2018; Liu et al. 2018; Yu et al. 2018). Therefore, the interface structure regulation is the key factor affecting the adsorption performance of microspheres.

Our aim is to prepare the mesoporous microsphere through Pickering emulsion method to remove the dye contaminants like RhB and MB from solution. The microstructure characterization of microspheres was used to investigate the effect of preparation conditions on the adsorption of RhB and MB. The pH value, adsorbent dosage, reaction temperature, and time were also selected to investigate their effect on the adsorption behavior of two dyes onto microspheres.

Sodium (Na)-bentonite was purchased from Wancheng bentonite Co. (Liaoning, China). STAC and liquid paraffin (PL) were offered by Sinopharm Chemical Reagent Co. Tetraethyl orthosilicate (TEOS), azodiisobutyronitrile (AIBN), methyl methacrylate (MMA), acetone, absolute ethanol, deionized water, and Span-80 were purchased by Tianjin Damao Chemistry Reagent Co. Ammonia (25%) and Tween-80 were obtained from Tianjin Beichenfangzheng Reagent Co. γ-Methacryloxypropyl trimethoxysilane (KH570, 98%) was provided by Dulai Biotechnology Co. (Nanjing, China). RhB was commercially obtained from Tianjin Institute of Reagents, and MB was provided by Tianjin Hengxing Chemical Reagent Manufacturing Co.

It is necessary for Na-bentonite to be purified for experimental use. The specific operations are as follows: 40 g Na-bentonite (particle size <75 μm) was dispersed into 800 ml deionized water, which was subsequently continuously agitated at 300 rpm and 60 °C for 2 h by a magnetic stir (HJ-6A, XRYQ, China). Through standing for 1 h, filtration, and drying, Na-montmorillonite (Mt) powder was obtained. Then, 2 g Mt powder (particle size <75 μm) and 1.8 g STAC were added to 200 ml deionized water and agitated at 300 rpm and 60 °C for 2 h to form the suspension. Through washing with deionized water three times and drying at 50 °C for 12 h, the modified Mt was successfully prepared for microsphere fabrication.

Silicon dioxide (SiO₂) nanospheres were prepared by the Stӧber method (Zhang et al. 2009). The prepared SiO₂ nanospheres were modified by KH570 to realize the encapsulation of montmorillonite core material. First, 2 g of the modified Mt and 8 ml MMA were mixed with 10 ml liquid paraffin as the oil phase (Figure 1). Then, 2 ml Tween-80 and 1 ml Span-80 were dissolved in 50 ml of water as the aqueous phase. The oil phase and water phase was mixed and emulsified at a high speed of 14,000 rpm for 5 min to form a stable emulsion. Subsequently, the modified SiO₂ nanospheres and 0.1 g AIBN were put into the emulsion, which was stirred at a speed of 1,000 rpm and reaction temperature 60 °C for 2 h. Finally, the produced microspheres were washed with acetone and deionized water three times. Through drying by vacuum freeze-drying for 24 h, the microspheres were successfully prepared. Then, the microspheres were calcinated in the muffle furnace at a calcination temperature from 200 °C to 560 °C for 5 h to form a mesoporous structure.

The microstructure and element compositions of samples were investigated by field emission scanning electron microscopy (FE–SEM, Gimin 300, Carl Zeiss AG, Germany) and energy dispersive spectrometer (EDS, Gimin 300, Carl Zeiss AG, Germany). A Fourier transform infrared spectrometer (FTIR, Nicolet iS50, TMO, USA) was used to analyze changes in the functional group information of material. Brunauer–Emmett–Teller (BET, v-sorb 2800, GAIC, China) was used to analyze the specific surface area, pore size, and pore volume of the prepared MBMM. Zeta potential (Zeta sizer Nano ZS, Malvern, UK) was adopted to investigate surface charge properties of MBMM samples before and after adsorption. Contact angle measurement (DSA100, KRUSS, Germany) was used to investigate the change of material contact angle.

Figure 2 indicates that SiO₂ nanosphere were very regular in shape, with a sphere diameter of about 200 nm (Figure 2(a)). The lamellar structure of the modified Mt can be clearly seen in Figure 2(b), and this unique structure implied that it had a high specific surface area. When STAC was successfully implanted into the interlayer of Mt, there was a significant increase in its interlayer spacing. The introduction of an alkyl group also caused a remarkable conversion of the interlayer microenvironment from hydrophilicity to lipophilicity. This modification method made it possible to prepare mineral-core microsphere through the oil/water method. The microsphere and its surface texture are displayed in Figure 2(c) and 2(d), respectively. It can be seen that microspheres were generally spherical in shape and size between 150 µm and 200 µm, with many folds on the surface (Figure 2(d)), which was caused by volume shrinkage due to low temperature and vacuum. After removing the oil phase core material through calcination in muffle furnace, it is not difficult to see that a small number of macropores was formed on the microspheres surface (Figure 2(e) and 2(f)). The reason for the macroporous structure formation was that the steam from the thermal decomposition of the oil phase broke out from the shell. Pores on the microsphere surface enabled the pollutants to enter into the microsphere easily, which was beneficial to improving the mass transfer efficiency. It was also noteworthy that the folds of the microsphere surface disappeared. As seen from a broken microsphere (Figure 2(g)) and a close-up of its internal structure (Figure 2(h)), the removal of oil phase led to the appearance of hollow structure in the microspheres, which enhanced its adsorption capacity. Seen from the broken microsphere section, the wrinkles on the inner surface of microsphere shell are more obvious than that of modified Mt. The results indicated that the preparation of microspheres was successful.

FTIR analysis (Figure 3) indicated that the symmetrical stretching vibration absorption peaks of C–H and anti-symmetrical stretching vibration absorption peaks of C–H in the modified Mt appeared at 2,849 cm⁻¹ and 2,919 cm⁻¹, while the bending vibration absorption peak of C–H appeared at 1,467 cm⁻¹. The appearance of new diffraction peaks indicated that STAC entered into the interlayer of Mt successfully (Meng et al. 2016). The anti-symmetric stretching vibration peaks of Si–O–Si, Si–O bending vibration peaks, and Si–OH bending vibration peaks of the modified SiO₂ nanospheres were redshifted from 1,081 cm⁻¹, 452 cm⁻¹, and 953 cm⁻¹ to 1,061 cm⁻¹, 444 cm⁻¹, and 950 cm⁻¹, respectively. The anti-symmetric stretching vibration peaks of structural water –OH were wide and dispersed, and blue shifted from 3,215 cm⁻¹ to 3,248 cm⁻¹, indicating that the modification of SiO₂ was successful. All the above absorption peaks appeared in the microspheres before calcination, indicating that the modified Mt and SiO₂ nanospheres were successfully loaded onto the microspheres.

The calcination temperature was an important factor on the mesoporous structure of microspheres prepared by a soft template synthesis method. BET results indicated that the increase of calcination temperature was beneficial to improving the specific surface area, pore volume, and pore size of MBMM (Table 1). When the calcination temperature was between 200 °C and 500 °C, there was a marked increase in the specific surface area and pore size of MBMM from 8.39 m² g⁻¹ and 11.40 nm to 167.77 m² g⁻¹ and 28.25 nm, respectively. It was also noteworthy that the pore volume increased by nearly 30 times for MBMM before and after calcination. These phenomena were due to the decomposition of soft template molecular and other modifiers used in the material preparation. For example, the liquid paraffin began to decompose at 260 °C, while the decomposition temperature of polymethyl methacrylate (PMMA) in the mixed template was in the range of 270–400 °C. With the increase of the calcination temperature, the liquid paraffin was completely decomposed, while PMMA was partially decomposed until it was completely decomposed. The removal of these organic phases made it possible to form the mesoporous structure and increase the specific surface area of microsphere.

The adsorption–desorption isotherms (Figure 4) of nitrogen (N₂) were type IV and the hysteresis loop belonged to type H₃. The material in the higher relative pressure region in the isothermal adsorption curve does not show any adsorption limitation. The existence of the desorption–adsorption hysteresis loop was an obvious characteristic of mesoporous materials, while the adsorption–desorption curve has not been closed, indicating that the shape and size of mesoporous microspheres vary greatly (Wang et al. 2017). Combined with Figure 2(e) and 2(f), it can explain the existence of mesopores and macropores on the surface of microspheres, which will facilitate the diffusion of contaminants.

Literatures indicated that the solid surface is hydrophilic when the contact angle was less than 90°. The smaller the angle, the stronger the hydrophilicity and the harder it is to separate from water. When the contact angle was more than 90°, it indicated that the solid surface was lipophilic. The larger the angle, the stronger the lipophilic (Erbil 2009). The experimental results showed that the contact angle of Mt was about 25° (Figure 5(a)), while that of MBMM was about 77° (Figure 5(b)). Compared to Mt, the lipophilicity of MBMM in aqueous solution was strengthened after microspherization, indicating that it was easier for MBMM to be separated from the solution, and MBMM was a potential adsorbent for organic contaminants.

Figure 6(a) indicates that the adsorption capacities of RhB and MB onto MBMM increased from 19.63 mg g⁻¹ and 7.79 mg g⁻¹ to 49.37 mg g⁻¹ and 41.63 mg g⁻¹, respectively, when the calcination temperature ranged from 200 °C to 500 °C. Based on BET analysis, it was obvious that the significant increase in the adsorption capacity of contaminants onto MBMM was attributed to the change in the specific surface area, pore size, and pore volume. The gradual decomposition of soft template molecules and other organic modifiers resulted in the formation of mesoporous structures on the surface of microspheres, providing more adsorption sites for contaminants. However, the adsorption capacities of two target contaminates onto microspheres remained basically unchanged when the calcination temperature continued to rise. To ensure the comparability of test data, MBMM prepared at a calcination temperature of 500 °C was used in subsequent experiments.

The adsorbent dosage and the initial concentration of target contaminants are two important factors affecting the removal of contaminants in solution. When the initial concentration increased from 50 mg L⁻¹ to 300 mg L⁻¹, the adsorption capacities of RhB and MB increased from 16.85 mg g⁻¹ and 17.33 mg g⁻¹ to 45.19 mg g⁻¹ and 41.48 mg g⁻¹ (Figure 6(b)), respectively. It was also noteworthy that when the concentrations of target contaminants in solution were higher than 300 mg L⁻¹, their adsorption capacities onto MBMM fluctuated within 2 mg g⁻¹. As seen from Figure 6(c), the adsorption capacities of RhB and MB remained almost unchanged when the adsorbent dosage was in the range of 0.05–0.25 g. Based on the above two factors, the initial concentration 300 mg L⁻¹ and the adsorbent dosage 0.1 g were used for the subsequent experiments.

pH is an important factor to investigate the effectiveness of microspheres in removing contaminants from different alkaline or acidic solutions. Figure 6(d) indicates that there was a significant increase in the adsorption capacity of RhB when pH was between 3 and 7. However, high pH value (pH >7) caused an obvious reduction in RhB. The results were attributed to the competitive ability between H⁺ and RhB to the active adsorption sites on microsphere. For the solution with high pH value, the competition of H⁺ for the active sites was reduced (Durairaj et al. 2019), which was conducive to RhB adsorption. As seen from the adsorption curves of MB (Figure 6(d)), there was a slow increase in MB adsorption between pH 3 and 11, indicating that the change of H⁺ concentration in the test solution had no adverse effect on MB adsorption.

The adsorption temperature is an important factor affecting the adsorption behavior of RhB and MB onto MBMM. Figure 6(e) indicates that the adsorption capacity of RhB and MB onto microspheres increased from 49.17 mg g⁻¹ and 41.63 mg g⁻¹ to 56.50 mg g⁻¹ and 55.67 mg g⁻¹, respectively, when the adsorption temperature was in the range of 25–55 °C. The results indicated that the adsorption of two contaminants was an endothermic reaction.

The effect of adsorption time on MBMM adsorption performance was investigated under the condition of pH 7, MBMM dosage 0.1 g, and an adsorption temperature of 55 °C. Figure 6(f) indicates that the curve slope was relatively large within 0–2 h, indicating that this stage was a rapid adsorption stage. These phenomena were due to the large number of active sites on the surface of microspheres and the higher diffusion rate (Chu et al. 2019). In the following 2–6 h, there was a linear increase in the adsorption capacities of RhB and MB onto the microspheres. The absorption equilibrium of RhB and MB onto MBMM was achieved when the adsorption time reached 7 h. Their equilibrium adsorption capacities were 57.79 mg g⁻¹ and 55.94 mg g⁻¹, respectively.

Langmuir and Freundlich adsorption isotherm models were used to investigate the adsorption behaviors of RhB and MB. The linear fitting formulas of Langmuir and Freundlich adsorption isotherm models are listed in the Supplementary Material.

As seen from Table 2, the Langmuir model can better describe the adsorption isotherm due to higher R². The fitting results showed that the maximum adsorption of RhB and MB onto MBMM were 59.14 mg g⁻¹ and 55.62 mg g⁻¹, respectively, which were in good agreement with the measured values of 57.79 mg g⁻¹ and 55.94 mg g⁻¹. It also showed that the adsorption process was a single-layer surface adsorption. Although the adsorbate concentration continued to increase, the adsorption capacity remained basically unchanged after reaching the saturation point (Shaban et al. 2017). Compared with other adsorbents (Table 3), it was found that the adsorption capacities of RhB and MB onto MBMM were better than that of some reported similar adsorbents, indicating that MBMM was a promising adsorption material for dye removal.

The pseudo-first-order-dynamics and pseudo-second-order-dynamics models were adopted to investigate the adsorption mechanism of RhB and MB onto MBMM. The fitting results indicated that the pseudo-first-order-dynamics model can well describe the adsorption process because the saturated adsorption capacities of RhB and MB were more close to the measured values in experiments (Table 4). This phenomenon indicated that physical adsorption was the main speed limiting step (Chang et al. 2018). The specific formulas and figures are listed in Section 2 of the Supplementary Material.

The Gibbs free energy ΔG, enthalpy change ΔH, and entropy change ΔS were calculated to investigate the thermodynamics behavior of RhB and MB onto MBMM in adsorption reaction. The specific formulas are listed in Section 4 of the Supplementary Material.

Literature indicated that the adsorption process was spontaneous when ΔG < 0. Moreover, ΔS > 0 indicated that the disorder of adsorbate increased in the adsorption process (Konggidinata et al. 2017). The thermodynamics calculation (Table 5) indicated that the values of ΔH in the adsorption of RhB and MB onto MBMM were 22.26 kJ mol⁻¹ and 36.34 kJ mol⁻¹, respectively, indicating that the adsorption process was endothermic and the temperature rise was beneficial to two dye adsorption. Usually, when the value of was in the range of 2.1–20.9 kJ mol⁻¹ the physical adsorption was predominant, while chemical adsorption was dominant when was within 80–200 kJ mol⁻¹ (Liu & Liu 2008; Zhang et al. 2019). Based on the above analysis, it was obvious that there was a coexistence of physical and chemical adsorption coexist in the adsorption reaction.

The adsorbent retained a complete morphology after adsorption, and the BET parameters (Table 6) indicated that the specific surface area, pore size, and pore volume decreased to some extent, which was attributed to the pore blockage caused by dyes. Literature indicated that RhB existed in different forms under different pH levels. When pH < 3, RhBH₂²⁺ (RhB molecule with two positive charges) and RhBH⁺ (RhB molecule with a positive charge) were the main forms of RhB, but when pH > 4, RhBH^± (when RhBH⁺ and RhBH⁻ exist at the same time) was the main form (Wu et al. 2018). Meanwhile, the dissociation of RhB carboxyl (–COOH) led to the increase of negative charge in the whole system. MB is an alkaline dye that can ionize Cl⁻, which causes an increase in the total negative charge in the system. After RhB and MB adsorption onto microspheres, zeta potential of MBMM decreased from −7.59 mV to −12.30 mV and −22.07 mV, respectively (Figure S3 in Supplementary Information). The results were due to the change of the total charge in the reaction system. Moreover, the conductivity was an important factor on the dye adsorption, and it was found that the conductivity of the microspheres was about 10 times lower than that before adsorption, from 0.45 mS cm⁻¹ to 0.05 mS cm⁻¹ and 0.04 mS cm⁻¹, respectively. It was caused by the adsorption of anions or the material exchange between dye molecules and montmorillonite in the microspheres. Zeta potential and conductivity analysis indicated that electrostatic attraction was one of the driving forces in the adsorption process.

The EDS analysis (Figure 7) showed that there was a slight increase in the mass percentages of oxygen and chlorine of MBMM before and after RhB adsorption from 45.86% and 0.02% to 47.01% and 0.03% respectively, indicating that RhB molecule had been successfully trapped by the microspheres. After MB adsorption, the mass fraction of sulfur in the microspheres increased from 0.00% to 0.03%, implying that MB molecules successfully entered into the microsphere shell.

FTIR analysis (Figure 8) indicated that when RhB was adsorbed by MBMM, the symmetrical stretching vibration peak of the C–C bond appeared at 1,591 cm⁻¹, while the peaks at 1,414 cm⁻¹ and 1,340 cm⁻¹ were due to the bending vibration of C–H. After MB adsorption onto MBMM, the symmetric stretching vibration peak of the C–C bond appeared at 1,601 cm⁻¹, the bending vibration peak of C–H appeared at 1,389 cm⁻¹ and 1,332 cm⁻¹, and the bending vibration peak appeared at 884 cm⁻¹ on the aromatic ring skeleton plane. The appearance of a new absorption peak indicated that there was a chemical reaction between dye molecules and adsorbents. Moreover, Si–O–Si stretching vibration peak became wider due to the hydrogen bonding of oxygen, nitrogen atoms with lone electrons on dye molecules with hydroxyl groups (Yang et al. 2013). Although dye adsorption was the interaction of physical adsorption and chemical adsorption, physical adsorption was dominant. The analysis of zeta potential and FTIR indicated that electrostatic attraction and hydrogen bonding were two driving forces in dye adsorption.

The adsorbent after dye adsorption was washed by ethanol and deionized water three times and then dried to carry out the adsorption experiments in order to investigate its repeatability. The results showed that the adsorption capacity of MBMM for RhB and MB decreased from 57.01 mg g⁻¹ and 54.89 mg g⁻¹ to 52.11 mg g⁻¹ and 50.09 mg g⁻¹ after five cycles, respectively, while the adsorption capacity of MBMM was still 91.41% and 91.26% (Figure S5 in Supplementary Material). In the whole adsorption experiment, MBMM showed good repeatability and reproducibility, indicating it was a high cost-effective adsorbent.

The Pickering emulsion method was adopted to prepare the mesoporous microsphere in the application of dyes adsorption. Microstructure characterization indicated that there was a lot of mesoporous material on the surface of the prepared microspheres, in which there was a hollow structure. The calcination temperature played an important role in the increase in specific surface, pore size, and pore volume. Batch experiments indicated that the optimum adsorption capacities of RhB and MB onto the microspheres were 57.79 mg g⁻¹ and 55.94 mg g⁻¹, respectively. The adsorption kinetics and thermodynamics analysis indicated that the electrostatic attraction was in the dominant position in the adsorption of RhB and MB onto MBMM, accompanied by a chemical reaction like hydrogen bonding. The regeneration of adsorbent showed that it was an economic and effective new adsorbent.

This work was financially supported by Liaoning Provincial Natural Science Foundation of China (grant number 20180510024).

The Supplementary Material for this paper is available online at https://dx.doi.org/10.2166/wst.2020.188.