Recent advances in the direct fabrication of millimeter-sized hierarchical porous materials

Abstract

Millimeter-sized hierarchical porous materials are easy to handle and recover in practical applications, and have no potential safety hazards. Such materials are very useful, and the fabrication of millimeter-sized products without additional binders and complex post-treatments is challenging and promising. This review focuses on the recently emerged techniques for the direct preparation of millimeter-sized hierarchical materials, where sedimentation polymerization, emulsion-template method, phase inversion method, direct carbonization and deposition method by using preformed biopolymer beads and microfluidics method are discussed and summarized.

---

1. Introduction

Applications of porous materials require macroscopic physical shapes including extrudates of porous materials (using binders), monoliths or spherical shapes depending on the particular application, such as for wastewater treatment, oil spill cleanup, catalysis and gas adsorption. These fabrication processes require extensive pre- and post-processing and are cost intensive. In addition, the physicochemical properties of the materials may change during forming.¹ Millimeter-sized spheres have their own advantages such as no potential safety hazards, and being easy to handle and recover compared with the micro- and nanospheres.² Natural polysaccharides such as alginate, chitosan and carrageenan can be easily shaped into various structures including millimeter-sized spherical beads. The polysaccharide-based composites have been intensively studied and well reviewed elsewhere.³ However, relatively few reviews have been devoted to the use of other biopolymers simply as a sacrificial template for the preparation of inorganic beads or as a support to synthesize hierarchical porous beads.⁴

Direct fabrication of millimeter-sized structured porous materials has attracted increasing attention, where hierarchical pores (micro-, meso- and macropores) are ready to be created. Hierarchical porous materials consist of multiple levels of pores with different diameters, in which case, they usually contains macropores (>50 nm) or mesopores (2–50 nm) together with micropores. Such hierarchical structure would combine the advantages of large pores (reduce transport resistance; thus enhancing permeability and diffusion) and small pores (enhance the surface area and pore volume); thus improved properties are usually achieved compared to materials that only contain macropores or micropores.

So far, millimeter-sized hierarchical porous beads have been used in many fields such as catalysis, separation, optics, adsorption and electrochemistry.^2,5–8 Apart from natural polysaccharide gel beads,^4,9 new materials such as polyacrylamide (PAM), polysulfone, polyethersulfone have been used in the direct fabrication of millimeter-sized porous composite polymers and other inorganic oxides. Such polymers act as either sacrifice agent to create pores or as porous support for inorganic particles. Pores can be directly formed via sedimentation polymerization, emulsion, phase inversion, direct carbonization or other reactions without complex fabrication that involves many steps (for example, hard template needs to be removed via chemicals¹⁰). Moreover, direct fabrication of millimeter-sized porous spheres via polymers usually have large-scale yield compared to other methods. In this mini review, millimeter-sized spheres were prepared by several emerged techniques, including sedimentation polymerization, emulsion-template method, phase inversion method, direct carbonization and deposition method using preformed biopolymer beads, microfluidic method and their applications were reviewed.

2. Sedimentation polymerization and emulsion-template method

The method of “sedimentation polymerization” was introduced by Ruckenstein and co-workers.¹¹ Droplets of an aqueous monomer solution were introduced with a syringe into a heated oil medium for the sedimentation. The droplets are partially polymerized during the sedimentation process and allowed to polymerize completely at the bottom of the reactor. The resulting water-swellable polymer networks (WSPN) bead size can be adjusted by changing the nozzle diameter or the injection rate, and beads in the size range of 0.5–2.5 mm were obtained with relatively narrow size distributions (5–35%).^11,12 In the preparation of porous metal oxide beads, a metal nitrate was loaded into the WSPN either during the polymerization of the water-soluble monomer or via the swelling of a WSPN resin in an aqueous solution of the nitrate salt. The pyrolysis of the precursor-loaded network generated interpenetrating networks of carbon and metal oxide. Millimeter-sized porous metal oxide beads were obtained after removing the carbon network by combustion of the composite in air.¹²

Emulsion template involves forming a high internal phase emulsion (HIPE) and locking in the structure of the continuous phase, usually by reaction-induced phase separation, followed by the removal of the internal phase (i.e., the emulsion droplets), leading to a porous replica of the emulsion.¹³ Highly porous, emulsion-templated organic polymers or “polyHIPEs” have been prepared by the method. In the process, individual droplets of monomer solution (e.g. acrylamide) were partially polymerized during sedimentation through an immiscible sedimentation medium to form the PAM beads (Fig. 1a).^13,14 The internal structure is uniformly porous and consists of a “skeletal replica” of the original O/W HIPE. The pore structure is highly interconnected and there are open pores on the bead surface that are connected to the bead interior (Fig. 1b). The size of the macropores was found to be in the range 2–15 μm.

Fig. 1 Optical image of emulsion-templated PAM beads (a), and SEM image of an individual bead showing porous surface (b).^13,14

Uniform, hierarchically porous inorganic beads (SiO₂, Al₂O₃, TiO₂, and ZrO₂) have been produced using emulsion-templated polymer beads as templates.¹⁵ The inorganic beads were prepared by simply immersing the polymer scaffold beads in a range of inorganic precursor solutions, followed by sol–gel condensation in air and subsequent calcination of the polymer phase. The hierarchical structures are composed of mesopores (diameters 2–5 nm), micropores, and large emulsion-templated macropores of around 5–10 μm. All of the pores are highly interconnected. Polymer/silica high-internal phase emulsion (HIPE)-templated porous beads were prepared with an average bead diameter of 2.00 mm and high intrusion volume (5.68 cm³ g⁻¹).¹⁶ After calcinations, highly porous silica beads were obtained with an average bead diameter of 1.34 mm. The HIPE structure was retained in the silica beads and the material has high surface area (421.9 m² g⁻¹) and high pore volume (5.81 cm³ g⁻¹). By the same method, Li et al. prepared 3-dimensionally interconnected, highly porous silica materials with ordering on three different scales via a dual-templating approach.¹⁷ The direct emulsion droplets formed macropores (10–30 mm) and interconnecting windows (3–5 mm), while interfacially adsorbed microgels promoted the formation of nanoscale porosity throughout the macroporous walls (Fig. 2). Millimeter-sized gold beads, silica–gold nanoparticle (GNP) beads and alumina–GNP beads were also prepared by the similar method.^18,19

Fig. 2 SEM images (a and b) of the hierarchical macropores and inter-connected windows templating from HIPE droplets. High-magnification SEM images (c and d) revealing the nanoscale pores throughout the cell walls due to the removal of microgel particles after the calcinations.¹⁷

In order to reduce the quantity of organic waste in the process and allow simple separation and isolation of the products, sedimentation polymerization in compressed fluid solvents (e.g. supercritical CO₂) was introduced to prepare polyacrylamide beads.²⁰ Tailored macroscopic (organo)silica spheres with particle diameters of 0.3–2.5 mm have been prepared in a one-step synthesis by varying the surfactant concentration, the silane mixture, the solvent as well as the method to remove structure directing agents.²¹ The hierarchical structure is mainly controlled by the reactivity of the precursor solution and the rate of water diffusion into the sphere after initial (macroscopic) phase separation. The silica spheres provide specific surface areas up to 900 m² g⁻¹ and adjustable porosity in the meso- and macroporous region. The texture includes particulate structures with controllable size between 80 nm and 2.5 μm of the silica particles on the inside of the spheres and a macroporous foam-like shell (0.6–44 μm thickness) with macropores in the range between 0.2 and 4.0 μm. Novel hierarchically structured macroscopic silica spheres have been synthesized by combining silanes, surfactant and solvent with emulsion-based sol–gel processing in a biphasic oil-in-water system. The spheres were prepared in a one-step process by injection of the precursor solution into water, which leads to controlled hydrolysis and condensation of the silanes, resulting in the formation of millimeter-sized mesoporous silica spheres with tunable morphology and porosity (Fig. 3).¹ The spheres have a hierarchically layered morphology, high mechanical stability, specific surface areas up to 700 m² g⁻¹ with mesopores in the range of 9–27 nm.

Fig. 3 Synthesis strategy for hierarchically structured macroscopic silica spheres.¹

Hahn et al. fabricated and systematically explored the synthesis, characterization, and experimental validation of millimeter-sized SiO₂ spheres.²² The larger and hierarchically structured SiO₂ spheres (in the range of 1–2 mm) offer advantages compared to the other materials with respect to the pore volume and pore size distribution. The SiO₂ spheres are dominated by large pores allowing fast access of CO₂ from the gas phase into the sorbent, followed by a mesoporous inner part enabling high CO₂ sorption capacities. Hydrophobic mesoporous silica spheres containing transition metals coordinatively bound to aminosilane groups are synthesized in a one-step procedure from functionalized silicon alkoxides.²³ Through tailoring the initial silicon precursor composition and injection rate it is possible to prepare spherical silica beads and control their macroscopic size as well as their pore structure. Amino silane functional groups introduced into the polysiloxane networks are not only important for the successful formation of beads and adjusting the textural properties and hydrophobicity of the resulting materials, but are also the sites for anchoring transition metals such as Co, Cu, Fe, Mn and V. The spherical shape of the polysiloxanes is illustrated in Fig. 4. The pure silicon precursors resulted in white spheres with uniform diameter of 0.5–1 mm. Incorporation of a transition metal into the precursor solution did not change the morphology or the size of the resulting spheres.

Fig. 4 Photographs of spherical materials in the absence (a) and in presence of metal ions Co (b), Mn (c) and V (d), and the schematic representation of the coordination of a metal species to the polysiloxane network (e).²³

Millimeter-sized spherical silica foam supports with hierarchical mesoporous–macroporous structure were prepared using agar addition, foaming, and drop-in-oil method, and they exhibited interconnected 3-dimensional network macropore structure (Fig. 5) leading to the enhancement of total porosity and BET surface area via the foaming process.²⁴ After being modified with polyethyleneimine (PEI) by a wet impregnation process for CO₂ capture,²⁵ the millimeter-sized spherical sorbents exhibited the highest CO₂ sorption capacity per unit mass of PEI, among the PEI-impregnated silica materials. Furthermore, the stability test results suggest that the spherical sorbent is robust enough and can provide a stable CO₂ capture capacity during multiple sorption–desorption operations.²⁵ Millimeter-sized spherical ion-sieve foams (SIFs) were prepared from spinel lithium manganese oxide (LMO) via a combined process of foaming, drop-in-oil, agar gelation and acid treatment to recover the lithium from natural seawater.^26,27 The SIFs exhibited hierarchical trimodal pore structure. Small and large bimodal mesopores were formed from the acid treatment-induced agar removal, and macropores from the bubble-template. The lithium adsorption in natural seawater of the SIFs was as high as 3.4 mg g⁻¹, and the adsorption efficiency was maintained at over 95% even after five adsorption–desorption cycles. The desorption efficiency was also maintained at ca. 86% after five adsorption–desorption cycles. By blending chitin with rectorite and organic rectorite in a NaOH/urea aqueous solution, millimeter-sized chitin/rectorite composite beads (BCRs) and chitin/organic rectorite beads (BCO) were successfully fabricated via an optimal dropping technology through a syringe needle into a sulfuric acid solution.²⁸ The pale gray composite beads with a diameter of about 2 mm were formed instantaneously. The chitin/layered silicate composite beads exhibited a high adsorption performance for the two dyes methyl orange (MO) and methylene blue (MB), and BCR and BCO sorbents could efficiently and selectively adsorb MO and MB, respectively, from wastewater.²⁸ A series of metal–organic framework (MOF) composite materials were prepared by depositing crystalline MOFs (CPO-27 and HKUST-1) into mm-sized macroporous polyacrylamide (PAM), SiO₂–Al₂O₃ or SiO₂–PAM beads using solvothermal methods or immersion in mother solutions.²⁹ The MOF@polymer beads are an attractive and viable alternative to bulk MOF in practical applications, and the flexibility of the polymer support enhances the mechanical stability of the crystalline MOF.

Fig. 5 A schematic diagram of the description for multi-modal pore structure in millimeter-sized SSF: (a) bubble pore by foaming process (large bimodal macropore), (b) interparticle-aggregated pore (small macropore), (c) surfactant P123 micelle pore (ordered mesopore) and (d) improved micropore by agar loading (micropore).²⁴

3. Phase inversion method

Phase inversion method is one of the most important means to prepare asymmetric membranes that was first reported by Loeb and Sourirajan.³⁰ Such method was introduced by our group for the direct preparation of ultrafine hollow fiber membranes from a polymer solution using a single orifice spinneret.³¹ Inspired by the successful attempts, the fabrication of porous millimeter-sized polymer beads in a one-step process has been developed. Li et al. successfully fabricated ZIF-8/polysulfone (ZIF-8/PS) composite beads via phase inversion method using a single orifice spinneret (Fig. 6). Surfactant F127 was added in the system to improve the porosity of the spheres. The polysulfone and ZIF-8 acted as binder and adsorbent, respectively, and the porous structure of the sphere made the gas easily diffuse. The millimeter-sized ZIF-8/PS composite spheres were easy to handle and recycle, and the gas sorption ability of the ZIF-8/PS 4 : 1 and ZIF-8/PS 2 : 1 composite spheres were comparable to the pure ZIF-8.⁸

Fig. 6 Schematic diagrams of the phase inversion method for polymer sphere preparation through a single syringe tip (a), and the formation process of the porous structure via solvent/water exchange (b).⁸

Floatable low-density millimetre-sized hollow carbon beads were prepared by carbonization of hollow polymer beads, which were directly obtained by a one-step phase inversion method through a traditional syringe using silica as the skeleton support.³² The resulting carbon beads were used to promote water evaporation under simulated sunlight. Hollow carbon beads with 1.5 mm diameter achieved an evaporation rate of 1.28 L m⁻² h⁻¹ that was around 237% of the rate attained without carbon beads when 714 g m⁻² of carbon beads were present. Hollow carbon beads of around 3.0 mm in diameter were prepared by a simple phase inversion method and subsequent carbonization. Due to the low density and hydrophobic property (Fig. 7a), the hollow carbon beads were floatable on water (Fig. 7b). Different organic solvents and oils were used to study the oil sorption properties of the carbon beads. Their toluene sorption capacity was as high as 55% of their own volume. For long-chain oil like motor oil, the hollow carbon beads still adsorbed 40% of their volume.² Moreover, the oil uptake capacity of these hollow carbon beads remained almost the same after five cycles of refreshment by heat treatment. Fig. 7c shows the carbon beads quickly adsorb paraffin oil droplet labeled with oil blue N in about 5 s. Similarly, millimeter-sized carbon/TiO₂ composite beads were prepared for oil and dye adsorption.³³

Fig. 7 Water contact angle measurement of a disc that was prepared by pressing carbon beads particles (a), photographic image showing that hollow carbon beads float on water surface (b), autonomous adsorbing process of a paraffin oil droplet labelled by oil blue N (c), the whole process took about 5 s.²

4. Direct carbonization and deposition method using preformed beads

Sun et al. prepared hollow carbon spheres on a large scale using phenolic resin (PR) as carbon precursor and polyvinyl alcohol (PVA) as template.³⁴ As depicted in Fig. 8, nonionic surfactant PVA is firstly formed micelles, and further self-assembled into spherical liquid crystal. PR is then dissolved and formed dispersive droplets under stirring. These PR droplets are adsorbed to the surface of spherical liquid crystal to build up the shells. By drying at 378 K, the spherical liquid crystal is collapsed and hollow phenolic resin spheres (HPRSs) are prepared, and hollow carbon spheres (HCSs) are obtained after the carbonization and activation. The as-prepared HCSs had a maximum external diameter of about 1.1 mm and average shell thickness of about 0.2 mm. In addition, the HCSs had a high specific surface area of 1036 m² g⁻¹ and total pore volume of 0.84 cm³ g⁻¹. They also prepared a series of carbon spheres with various porous texture parameters from polystyrene-based macroreticular resin spheres (∼Φ 0.8 mm) by carbonization and activation.³⁵ The as-prepared carbon spheres had a maximum specific surface area of 996 m² g⁻¹, total pore volume of 1.34 cm³ g⁻¹ and average pore size of 5.39 nm. Moreover, these carbon spheres exhibited high specific capacitance ranged from 153 to 182 F g⁻¹, excellent high rate performance and specific energy between 2.3 and 5.1 W h kg⁻¹. The polystyrene suphonate beads with particle size range 0.6–1.2 mm were used to prepare activated carbon spheres by carbonization of polymeric precursor at 800 °C followed by activation of resultant char with steam and carbon dioxide activation processes.³⁶ The resulting millimeter-sized carbon spheres had very high BET surface areas of pore volumes. Wang et al. prepared polystyrene-based activated carbon spheres by carbonization and steam activation method.^37–39 The activated carbon spheres has a size of about 0.5 mm with a surface area of 1672 m² g⁻¹ and a total pore volume of 163 cm³ g⁻¹. The adsorption capacity to sulfur-containing dibenzothiophene is up to 109 mg g⁻¹.

Fig. 8 Schematic illustration of the formation mechanism of HPRSs and HCSs.³⁴

Sodium alginate is a natural biopolymer extracted from algae, and it has the capability to gel when the sodium ion is substituted by a multivalent cation (usually Ca²⁺) that builds crosslinking between the polymer chains.⁴⁰ In the past, many researchers used alginates as an in situ gelling template for making millimeter-sized ceramic beads.^4,40–45 Abramson et al. prepared millimeter-sized magnetic silica nanocomposite beads using alginate as a green biopolymer template, and the resulting silica beads exhibited a relatively good sorption capacity and a strong affinity for methylene blue in water.⁴ Meso–macroporous millimetre-sized TiO₂ beads were synthesized using sol–gel chemistry within preformed calcium alginate (CaAlg) beads, which were prepared by an extrusion external gelation technique whereby the NaAlg solution was manually dripped into aqueous Ca²⁺ baths.⁴³ The surface area of the TiO₂ beads reached maxima using template beads prepared in a 0.27 M Ca²⁺ bath. The loading of alendronate (an amino-bisphosphonate) on porous TiO₂ was highly dependent on surface area. By using sol–gel method using CaAlg as the hard template, hierarchically porous TiO₂/ZrO₂ millimeter-sized beads were synthesized.⁴⁵ The bead properties were varied by altering either Ti/Zr composition or calcinations temperature (500 or 700 °C). Uranyl adsorption was higher for the crystalline TiO₂/ZrO₂ beads (surface area: 52–59 m² g⁻¹) than the amorphous beads, reaching a maximum of 0.170 mmol g⁻¹ for the 22 wt% Zr sample (Fig. 9).

Fig. 9 Uranyl uptake adsorption isotherms (a), TEM images of crystalline (b) and amorphous (c) TiO₂/ZrO₂ beads and their digital image (d).⁴⁵

5. Microfluidic method

Microfluidics offers a convenient and finely controllable route to synthesize microparticles with advantages of using soft lithography to design desired geometries and of the ability to exploit fluid mechanics to tune particle morphology.^46,47 Recently, microfluidics has been used as a means to create polymeric microspheres with a narrow size distribution. Droplets of a dispersed phase that is either a low-melting-point lipid or a UV-curable polymer are pinched off into a continuous aqueous phase and are then either frozen or polymerized to create solid spheres.^48,49

The advantage of microfluidics is to precisely tune the particle size and morphology via designing devices and/or controlling operating conditions. Shao et al. designed capillaries-based co-flowing microfluidic devices and then fabricated for the controlled production of water-in-oil-in-water (W/O/W) double-emulsion droplets, where polystyrene dissolved in fluobenzene was employed as the oil phase.⁵⁰ Low-density porous millimeter-size poly(divinylbenzene) (PDVB) shells were produced by the use of a microfluidic device under ultraviolet (UV) irradiation at ambient temperature by Yang and coworkers.^51,52 They found that by controlling the monomer concentrations and illumination times, the pore structure, especially the mesoporous structure can be tuned. Fig. 10 gives the schematic illustration of growth mechanism of PDVB shells. During the polymerization, the assemblage of the particles was influenced by the inner diameter and external diameter of shells. More particles could gather at the inwall of the shells or nearby. Due to the surface vinyl groups, a mass of particles gave rise to the form of cluster aggregations. With the growing of the clusters, the inwall of the shells first formed the cross-linked structure. Then, the inwall network structure acted as the active point and gradually increased by capturing solvent oligomers or particles from the reaction medium. As a result, the inwall of the shells could form more compact structures than the outwall of the shells.

Fig. 10 Schematic illustration of growth mechanism of poly(divinylbenzene) shells.⁵¹

6. Conclusions

This short review has highlighted recent emerged techniques for the fabrication of millimeter-sized hierarchical porous materials. Sedimentation polymerization used aqueous monomer solution as the staring materials to prepare millimeter-sized polymer products, which can be further used to prepared inorganic porous materials. Emulsion-template method is a kind of modified sedimentation polymerization, where the emulsion will create the replica pores. Millimeter-sized PAM beads were widely studied and they could be used as templates to prepare inorganic porous beads and other functional composites.¹³ Phase inversion method is newly appeared, and it could be easily adapted to prepare polymer composites and carbon composites for various applications. Direct carbonization uses carbon-rich polymers to prepare porous carbon spheres by carbonization, and preformed millimeter-sized biopolymer beads were directly used as the sacrificial templates in the deposition method to prepare porous functional materials. Microfluidics could precisely tune the particle size and morphology via designing devices and/or controlling operating conditions. In conclusion, the fabrication of millimeter-sized porous materials could be achieved by above mentioned techniques, and the resulting porous materials with hierarchical structures could be directly used in practical applications, such as adsorbents to remove dyes, oils and nuclear waste.

Acknowledgements

The authors are grateful for the financial support of Jiangsu Specially-Appointed Professor Program and Natural Science Key Project of the Jiangsu Higher Education Institutions (15KJA220001).

References

1. S. Scholz, S. R. Bare, S. D. Kelly and J. A. Lercher, Microporous Mesoporous Mater., 2011, 146, 18–27.
2. Y. Zeng, K. Wang, J. F. Yao and H. T. Wang, Carbon, 2014, 69, 25–31.
3. V. Singh, P. Srivastava, A. Singh, D. Singh and T. Malviya, Polym. Rev., 2016, 56, 113–136.
4. S. Abramson, C. Meiller, P. Beaunier, V. Dupuis, L. Perrigaud, A. Bee and V. Cabuil, J. Mater. Chem., 2010, 20, 4916–4924.
5. R. C. Jin, G. Liu, C. P. Liu and L. Sun, Mater. Res. Bull., 2016, 80, 309–315.
6. F. Ran, K. W. Shen, Y. T. Tan, B. W. Peng, S. H. Chen, W. J. Zhang, X. Q. Niu, L. B. Kong and L. Kang, J. Membr. Sci., 2016, 514, 366–375.
7. K. B. Xu, S. J. Li, J. M. Yang, H. Y. Xu and J. Q. Hu, J. Alloys Compd., 2016, 678, 120–125.
8. L. X. Li, J. F. Yao, P. Xiao, J. Shang, Y. Feng, P. A. Webley and H. T. Wang, Colloid Polym. Sci., 2013, 291, 2711–2717.
9. I. Akartuna, E. Tervoort, J. C. H. Wong, A. R. Studart and L. J. Gauckler, Polymer, 2009, 50, 3645–3651.
10. Y. Li, Z. Y. Fu and B. L. Su, Adv. Funct. Mater., 2012, 22, 4634–4667.
11. E. Ruckenstein and L. Hong, Polymer, 1995, 36, 2857–2860.
12. E. Ruckenstein and L. Hong, Chem. Mater., 1996, 8, 546–553.
13. H. F. Zhang and A. I. Cooper, Soft Matter, 2005, 1, 107–113.
14. H. Zhang and A. I. Cooper, Chem. Mater., 2002, 14, 4017–4020.
15. H. Zhang, G. C. Hardy, Y. Z. Khimyak, M. J. Rosseinsky and A. I. Cooper, Chem. Mater., 2004, 16, 4245–4256.
16. H. F. Zhang, G. C. Hardy, M. J. Rosseinsky and A. I. Cooper, Adv. Mater., 2003, 15, 78–81.
17. Z. Li, X. Wei, T. Ming, J. Wang and T. Ngai, Chem. Commun., 2010, 46, 8767–8769.
18. H. F. Zhang, I. Hussain, M. Brust and A. I. Cooper, Adv. Mater., 2004, 16, 27–30.
19. H. F. Zhang, I. Hussain, M. Brust and A. I. Cooper, Chem. Commun., 2006, 2539–2541,  10.1039/b604392e.
20. H. Zhang and A. I. Cooper, Macromolecules, 2003, 36, 5061–5064.
21. S. Scholz and J. A. Lercher, Chem. Mater., 2011, 23, 2091–2099.
22. M. W. Hahn, M. Steib, A. Jentys and J. A. Lercher, J. Mater. Chem. A, 2014, 2, 13624–13634.
23. T. Foerster, S. Scholz, Y. Zhu and J. A. Lercher, Microporous Mesoporous Mater., 2011, 142, 464–472.
24. Y. Han, J. Choi, M. Tong and H. Kim, Mater. Charact., 2014, 90, 31–39.
25. Y. Han, G. Hwang, H. Kim, B. Z. Haznedaroglu and B. Lee, Chem. Eng. J., 2015, 259, 653–662.
26. Y. Han, H. Kim and J. Park, Chem. Eng. J., 2012, 210, 482–489.
27. Y. Han, S. Kim, H. Kim and J. Park, J. Am. Ceram. Soc., 2011, 94, 2742–2745.
28. G. X. Li, D. C. Lu, X. D. Shi and M. Y. Cai, J. Appl. Polym. Sci., 2014, 131, 40905.
29. L. D. O'Neill, H. F. Zhang and D. Bradshaw, J. Mater. Chem., 2010, 20, 5720–5726.
30. S. Loeb and S. Sourirajan, Adv. Chem., 1963, 38, 117–132.
31. J. F. Yao, K. Wang, M. R. Ren, J. Z. Liu and H. T. Wang, J. Membr. Sci., 2012, 421, 8–14.
32. Y. Zeng, K. Wang, J. F. Yao and H. T. Wang, Chem. Eng. Sci., 2014, 116, 704–709.
33. X. Gu, K. Zhou, Y. Li and J. Yao, RSC Adv., 2016, 6, 16314–16318.
34. G. Sun, K. Li, J. Wang, H. He, J. Wang, Y. Li, Y. Liu and J. Cu, Microporous Mesoporous Mater., 2011, 139, 207–210.
35. G. Sun, J. Wang, K. Li, Y. Li and L. Xie, Electrochim. Acta, 2012, 59, 424–428.
36. A. Singh and D. Lal, J. Appl. Polym. Sci., 2010, 115, 2409–2415.
37. Q. Wang, X. Liang, W. Qiao, C. Liu, X. Liu, L. Zhan and L. Ling, Fuel Process. Technol., 2009, 90, 381–387.
38. Q. Wang, X. Liang, W. Qiao, C. Liu, X. Liu, R. Zhang and L. Ling, Appl. Surf. Sci., 2009, 255, 3499–3506.
39. Q. Wang, X. Y. Liang, R. Zhang, C. J. Liu, X. J. Liu, W. M. Qiao, L. Zhan and L. C. Ling, New Res. Carbon Mater., 2009, 24, 55–60.
40. E. Prouzet, Z. Khani, M. Bertrand, M. Tokumoto, V. Guyot-Ferreol and J. F. Tranchant, Microporous Mesoporous Mater., 2006, 96, 369–375.
41. M. Buaki-Sogo, M. Serra, A. Primo, M. Alvaro and H. Garcia, ChemCatChem, 2013, 5, 513–518.
42. T. Nozaki, T. Arima, K. Idemitsu and Y. Inagaki, J. Nucl. Mater., 2011, 412, 184–189.
43. M. C. Kimling and R. A. Caruso, J. Mater. Chem., 2012, 22, 4073–4082.
44. C. J. E. Santos, T.-S. Wei, B. Cho and W. M. Kriven, J. Am. Ceram. Soc., 2013, 96, 3379–3388.
45. M. C. Kimling, N. Scales, T. L. Hanley and R. A. Caruso, Environ. Sci. Technol., 2012, 46, 7913–7920.
46. D. Dendukuri, K. Tsoi, T. A. Hatton and P. S. Doyle, Langmuir, 2005, 21, 2113–2116.
47. S. Okushima, T. Nisisako, T. Torii and T. Higuchi, Langmuir, 2004, 20, 9905–9908.
48. D. Sin, J. Fitzpatrick, P. Luckman, E. J. Wolvetang and J. J. Cooper-White, Soft Matter, 2012, 8, 10712–10718.
49. S. Sugiura, M. Nakajima and M. Seki, J. Am. Oil Chem. Soc., 2002, 79, 515–519.
50. T. Shao, X. L. Feng, Y. Jin and Y. Cheng, Chem. Eng. Sci., 2013, 104, 55–63.
51. Y. Yang, J. Y. Zhu, X. J. Wang, J. Wang, Z. F. Li, X. Luo and L. Zhang, J. Mater. Sci., 2015, 50, 3638–3645.
52. Y. Yang, X. Luo, Z. F. Li, J. Wang, Z. J. Wei, X. J. Wang and L. Zhang, J. Appl. Polym. Sci., 2015, 132, 41625.

---