Electrospun flexible self-standing silica/mesoporous alumina core–shell fibrous membranes as adsorbents toward Congo red

Yan Wang, Wande Ding, Xiuling Jiao* and Dairong Chen*
School of Chemistry & Chemical Engineering, National Engineering Research Center for colloidal materials, Shandong University, Jinan 250100, P.R. China. E-mail: jiaoxl@sdu.edu.cn; Fax: +86-531-88364281; Tel: +86-531-88364280

Received 29th April 2014 , Accepted 19th June 2014

First published on 19th June 2014

Flexible core–shell fibrous membranes for mesoporous alumina-based adsorbents have been fabricated via a one-step coaxial electrospinning, which was accomplished by electrospinning silica as the core phase and mesoporous alumina as the shell phase. The core–shell fibers could be directly electrospun in the form of membranes. After calcination, a mesoporous alumina shell formed to give a high adsorption capacity, and the core fibers became dense to provide good mechanical property of the membrane. The membranes exhibit good adsorption properties toward Congo red and also can maintain the membrane form during the cyclic test, which makes them easy to handle and retrieve. The good adsorption performance, high mechanical property, easy recovery and reuse characteristic of the fibrous membranes, as well as their easy to scale-up fabrication process all facilitate their practical application in environmental remediation.

1. Introduction

Water decontamination technology is attracting increasing attention as water contamination poses an ever increasing health risk.1,2 Organic dyes such as Congo red, discharged from the printing and dyeing industries, are usually released into water and can cause severe water contamination.2 The removal of these pollutants has been investigated by various methods such as adsorption,3–6 biodegradation,7 and chemical oxidation.8 Among these techniques, adsorption attracts the most attention due to its efficient removal of the dyes' organics.9 Notably, alumina is an excellent candidate for water purification and has been widely studied as an adsorbent.6,10–12 To date, various kinds of alumina materials, including nanorod-like mesoporous alumina,6 hierarchical spindle-like γ-alumina materials,10 mesoporous alumina fibers11 and core–corona porous structured alumina,12 have been prepared as adsorption materials. The adsorption capacities of these materials are relatively high, but the macrographs of these materials are still limited to scattered forms like powder. These materials in powder form are usually suspended dispersions in water and need filtration or centrifugation to reclaim them after the adsorption process, leading to difficulties in their practical application. Moreover, recovery difficulties may bring about the loss of adsorbents and cause secondary pollution to the environment.13,14

Recently, the fabrication of self-standing alumina films, as well as SiO2@γ-AlOOH (Boehmite) core–sheath fibrous membrane, and their potential as adsorbents have been reported.9,15,16 However, the methods of preparing alumina films are confined to mixing the γ-AlOOH nanofibers with a suitable solvent, and then drying in an oven15 or filtered,16 or the electrospinning combined hydrothermal procedure.9 These preparation methods are time-consuming and hard to scale-up; moreover, the hydrothermal process is energy intensive and needs demanding equipment. In addition, the flexibility and recycling performance of the alumina films have not been investigated. Furthermore, γ-AlOOH particles that deposit on electrospun silica fibers' surface via the hydrothermal method are considered to easily come off during application. Therefore, more effort is needed to fabricate flexible membrane adsorbents in a simple and convenient way.

Electrospinning techniques offer a simple, versatile and low-cost way for manufacturing one-dimensional fibers. Notably, fabricating materials to non-woven membranes that are composed of continuous fibers by the electrospinning method is of great value and has attracted significant attention these past few years.17–20 However, to the best of our knowledge, there are no reports about alumina membranes fabricated by electrospinning, which are being used as adsorbents. Generally, adsorbents with higher surface area and larger pore volumes are expected to have a better adsorption performance. Recently, we fabricated mesoporous alumina fibers with good adsorption performance toward Congo red.11 The mesoporous alumina fibers possess a large surface area, and high chemical and thermal stability, but they are fragile due to their porous structure; thus, the fibers cannot retain the membrane form in other applications. Having considered the integrity of the membrane form and the porous structure of the fibers, as well as the scalability issues, herein, we use flexible dense silica fiber as the core and mesoporous alumina fiber as the shell to form silica/mesoporous alumina core–shell fibrous membranes (denoted as S/M fibrous membrane) via a coaxial electrospinning process. The membranes show high strength, attributed to the dense silica core fibers, as well as a high specific surface area, due to the mesoporous alumina shell. Therefore, silica/mesoporous alumina fibrous membranes are flexible and exhibit a good adsorption capacity, as well as they are easy to handle and retrieve. These characteristics, as well as the simple, energy-efficient and easy to scale-up preparation process, make the membranes good candidates for water decontamination.

2. Experimental section

2.1. Materials

All materials, which includes Al(NO3)3·9H2O (analytical grade), AlCl3·6H2O (analytical grade), aluminum isopropoxide (Al(O-i-Pr)3, industrial grade), nitric acid (HNO3, 65 wt%, analytical grade), aluminum powder (analytical grade), polyethylene oxide (PEO, MW = 500[thin space (1/6-em)]000), Pluronic P123 (poly(ethyleneglycol)-block-poly(propyleneglycol)-block-poly(ethyleneglycol)) (EO20PO70EO20) (MW = 5800), poly (vinyl alcohol) (PVA; MW from 85[thin space (1/6-em)]000 to 124[thin space (1/6-em)]000 g mol−1, Sigma), H3PO4 (analytical grade), and tetraethyl orthosilicate (TEOS; Lingfeng Chemical Co., Ltd., China), were purchased commercially without any further purification.

2.2. Preparation of alumina sol

In a typical synthesis, alumina sol with a molar composition of Al(NO3)3·9H2O[thin space (1/6-em)]: AlCl3·6H2O[thin space (1/6-em)]:[thin space (1/6-em)]Al(O-i-Pr)3[thin space (1/6-em)]:[thin space (1/6-em)]Al[thin space (1/6-em)]:[thin space (1/6-em)]H2O = 1[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]2[thin space (1/6-em)]:[thin space (1/6-em)]4[thin space (1/6-em)]:[thin space (1/6-em)]178 was prepared by hydrolysis and condensation under constant stirring at 80 °C. Moreover, an appropriate amount of HNO3 was added to the mixture to adjust the reaction rate and control the pH value of the final sol between 3.24 and 4.23. Then, 0.1 g of PEO and 6.0 g of P123 were added into 40 mL of as-prepared alumina sol, to improve the spinnability of the sol and to act as a pore structure directing the reagent, respectively. The mixture was stirred for 12 h to form a spinnable sol.

2.3. Preparation of silica sol

The silica sol was synthesized according to a previously reported procedure.21 In a typical synthesis, 22.3 mL of TEOS was added into 19.8 mL deionized water. Then, 58.2 μL of H3PO4 was slowly added into the mixed solution under stirring. Then, an equivalent weight of 10 wt% PVA solution was added into the silica sol as a spinnability additive. The mixture was constantly stirred until the spinnable SiO2 sol was obtained.

2.4. Fabrication of SiO2/Al2O3 core–shell fibrous membrane

The silica sol and alumina sol were loaded into 10 mL plastic syringes separately. The silica sol syringe was connected to a core needle with an inner diameter of 0.4 mm and outer diameter of 0.6 mm, and the alumina sol syringe was connected to the corresponding shell needle with an inter diameter of 1.0 mm and outer diameter of 1.2 mm. The flow rates of the core and shell sol were both 2.0 mL h−1. The distance between the spinneret and the aluminum collector was 17 cm, and the applied voltage was 18 kV. The spinning was conducted under ambient conditions. The electrospun xerogel core–shell fibrous membranes collected on the aluminum foil were dried at 90 °C for 12 h, and then calcined at 700 °C, 800 °C or 900 °C for 2 h at a heating rate of 10 °C min−1.

Each of the two sols was also electrospun for comparison purposes. Details of these processes have been presented elsewhere.11,21

2.5. Characterization

The morphology of the fibers was observed by a field-emission scanning electron microscopy (FE-SEM, JSM-6700F), transmission electron microscopy (TEM, JEM-1011), and high-resolution transmission electron microscopy (HR-TEM, JEM-2100, accelerating voltage: 200 kV). The phase of the fibrous membranes was characterized by X-ray diffraction (XRD, Rigaku D/Max 2200PC) with CuKα radiation (λ = 0.15418 nm) at room temperature with the applied tube voltage and electric current at 40 kV and 20 mA. Thermo-gravimetric and differential scanning calorimetry (TG-DSC) measurements were characterized on a Mettler Toledo SDTA851e thermo-gravimetric analyzer with a heating rate of 10 °C min−1 up to 1200 °C in an air atmosphere. The mechanical properties of the membranes were tested on a tensile tester (XG-1A, Shanghai New Fiber Instrument Co., Ltd., China) with a clamp distance of 5 mm and a drawing speed of 1 mm min−1. ICP-Elemental analysis was performed on an ICP-AES (IRIS INTREPID IIXSP, Thermo Electron) at a RF power of 1.15 kW and a plasma gas flow of 13.0 L min−1 (λAl = 308.2 nm).

The N2 adsorption–desorption isotherms were characterized on a Quadrasorb SI apparatus. The surface area was determined via the Brunauer–Emmett–Teller (BET) method, while the pore volume and pore-size distribution were determined via the nonlocal density functional theory (DFT) method. All samples were degassed at 120 °C under vacuum for 12 h before analysis. The adsorption properties toward Congo red were examined by adding 10 mg samples to 5–50 mL Congo red (50 mg L−1) solution at room temperature. The concentration of Congo red was analyzed by a UV-vis spectrometer (Perkin-Elmer, Lambda-35). The characteristic absorption of Congo red at 500 nm was chosen as the monitored parameter for the adsorption process.

3. Results and discussion

3.1. Morphology and structure of the samples

Fig. 1 presents the TEM images of the pure SiO2 and Al2O3 fibers after calcination at 700 °C. The silica fibers are dense and could keep the membrane form. The membrane is self-standing, flexible, and exhibits high mechanical strength (5.4 MPa),21 which is of great importance in practical applications. On the contrary, the mesoporous alumina fibers possess high BET surface area, and large pore volume with many mesopores (Fig. 1b and c), which makes the fibers fragile. Thus, the fibers exhibit a powder form in the macrograph (the inset in Fig. 1c), causing inconveniences in their practical application. However, these characteristics are in favor of their functional properties such as their adsorption properties.11 The formation of the mesoporous alumina shell involves several processes. Firstly, the acidic aluminum sources (Al(NO3)3·9H2O, AlCl3·6H2O) and alkaline aluminum sources (Al(O-i-Pr)3, Al powder) were hydrolyzed and polymerized in acidic solution to form the polymeric –Al–O– colloidal particles.11 The optimal molar ratio of Al(NO3)3·9H2O to AlCl3·6H2O was 1[thin space (1/6-em)]:[thin space (1/6-em)]1 from the experimental results, because the charge/size ratio of NO3 and Cl will affect the electro-spinnability of the sol. Considering the hydrolysis and polymerization rate, as well as the removal of carbon during the calcination procedure, aluminum isopropoxide (easy to hydrolyze and polymerize) and aluminum powder (hydrolyzes and polymerizes slowly and is carbon-free) were together chosen as the alkaline aluminum sources. Secondly, block copolymer Pluronic P123 was added into the sol as a pore structure directing reagent, due to its excellent self-assembly capability to construct mesostructures.22 During the electrospinning process, the solvent was evaporated rapidly and the mesoporous structure was formed via an evaporation-induced self-assembly.22–24 Thirdly, a small amount of PEO (a common spinning additive) was added to improve the spinnability of the sol. Theoretically, the shell thickness could be controlled by adjusting the electrospinning parameters, such as the feed rates of the core and shell solutions, or the ratio of the inter diameter to outer diameter of the coaxial needle. Since the adsorption property is mainly attributed to the mesoporous alumina shell, the adsorption capacity will be higher when the mesoporous alumina shell is thicker. However, the thicker shell will reduce the tensile stress of the membranes, since the core silica takes an important role in maintaining the tensile stress of the membranes. Herein, we adjust the electrospinning parameters as described in the typical synthesis, so that the membranes could simultaneously meet the requirements of adsorption capacity and membrane form with a certain tensile stress.
image file: c4ra03912b-f1.tif
Fig. 1 TEM images of silica fibers (a), mesoporous alumina fibers (b), and N2 adsorption–desorption isotherms of mesoporous alumina fibers with different amounts of HNO3 in the sol (c). The insets are optical graphs of the corresponding fibers (a and b) and the pore size distribution curves (c). All the fibers are calcined at 700 °C.

Usually, the adsorption capacity increases with the surface area increasing. Herein, the impacts of the precursor's pH value on the surface area and pore volume were investigated (Fig. 1c). During the first step of the precursor preparation, HNO3 was applied to adjust the reaction rate, and the pH values of the final precursor varied with the different amounts of HNO3 added. The BET surface areas were 222.2 m2 g−1, 257.1 m2 g−1, and 344.5 m2 g−1 when the pH values of the precursor were 3.24, 3.53 and 4.23, respectively. The corresponding pore volumes were 0.387 cm3 g−1, 0.589 cm3 g−1 and 0.655 cm3 g−1, respectively. Considering that the high surface area corresponds to a large adsorption capacity, here the precursor with the pH value of 4.23 was chosen to fabricate the shell in the coaxial electrospinning process.

The core–shell fibers could be directly electrospun in the form of membranes. During the coaxial electrospinning process, the core silica solution and the shell mesoporous alumina solution were squeezed out simultaneously to form a stable Taylor cone. The same solvent in the core and shell streams could lead to a low interfacial tension between the two solutions, which is beneficial for successful coaxial electrospinning.25,26 In the calcination process, the H3PO4 added in the core solution may react with the shell alumina sol to form a small amount of aluminum phosphate at the core–shell interface, which could enhance the binding force between the core and shell and prevent the shell from coming off in application. Fig. 2a shows the optical image of the calcined membranes after bending several times. The flexibility of the membranes could be qualitatively identified by the bending test, as shown in the inset of Fig. 2a. The fibrous membrane keeps the integrity well and there are no cracks observed after bending the membrane several times, which favors the application of the membrane. Fig. 2b presents the mechanical properties of the core–shell membranes after calcinations at different temperatures. The tensile stress for the membranes calcined at 700 °C, 800 °C and 900 °C are 0.926, 0.672 and 0.538 MPa, respectively, which are high enough to keep the integrity of the membranes during adsorption process.

image file: c4ra03912b-f2.tif
Fig. 2 The silica/mesoporous alumina fibrous membranes after calcinations at 700 °C (a), and tensile stress–strain curves of the membranes after calcinations at different temperatures (b).

The microstructure of the silica/mesoporous alumina core–shell fibers after calcinations at 700 °C were investigated by TEM. As shown in Fig. 3, the distinctive phase in the fibers could be judged by the contrast that was created by the electron beam diffraction. The dark and bright regions represent the core and shell of the fibers, respectively. HRTEM images were obtained to get more information of the core–shell fibers (Fig. 3b). The mesopores are obvious and distribute randomly in the shell, while the core exhibits a compact structure. The core–shell fibers are not completely concentric, which may be due to the inevitable whipping motion of the charged jet and bending instability during the electrospinning process. Fig. 3c presents the selected area electron diffraction (SAED) pattern of the mesoporous alumina shell. The corresponding diffraction rings and weak spot on the SAED pattern prove the existence of γ-alumina with a low crystallinity. Moreover, the overall and localized structure of the fabricated membrane was further investigated by SEM (Fig. 4). It can be seen that the core–shell fibers are continuous with random distribution to form the nanopores between the fibers. The high-resolution SEM image also reveals the uniform size of the fibers with an average fiber diameter of ca. 220 nm.

image file: c4ra03912b-f3.tif
Fig. 3 TEM image (a), HR-TEM image (b) and SAED pattern of the part denoted by the circle in (b), (c) of the silica/mesoporous alumina core–shell fibers after calcinations at 700 °C. The inset in (a) is the corresponding low resolution TEM of the fibers.

image file: c4ra03912b-f4.tif
Fig. 4 SEM images of silica/mesoporous alumina fibers calcined at 700 °C at different magnifications.

TG–DSC curves of the as-spun silica/mesoporous alumina fiber membranes were investigated with those of the silica fiber membrane and mesoporous alumina fibers presented as references (Fig. 5). The mesoporous alumina fibers and silica fiber membrane show a total weight reduction of 67.28% up to 600 °C and 41.59% up to 700 °C, respectively. In the case of silica/mesoporous core–shell fiber membranes, the total mass loss is 61.47%, which is between that of the silica fiber membrane and mesoporous alumina fibers. The content of the mesoporous alumina in the silica/mesoporous alumina fibrous membrane is estimated to be 77.4% according to the following equation.

Mesoporous alumina content% = (WS/MWS)/(WMWS) × 100%
where WS/M, WS, and WM represent the weight loss of the silica/mesoporous alumina fiber membrane, silica fiber membrane, and mesoporous alumina fibers. The weight losses of the three samples all mainly derive from the removal of water, and the decomposition of inorganic salts and organic additives. There is no further mass loss at temperatures higher than 700 °C. In the temperature range of 700–1000 °C, the exothermic peak in the DSC curve of mesoporous alumina fibers is attributed to the phase transformation from amorphous alumina to γ-alumina, and there is no obvious peak in the curve of the silica fiber membrane, which indicates that no phase transformation occurs for silica fibers from 700 to 1000 °C. The phase transformation peak for silica/mesoporous alumina fibers is weakened compared to alumina fibers, which may be caused by the curve superposition of silica fibers and mesoporous alumina fibers.

image file: c4ra03912b-f5.tif
Fig. 5 TG curves (a) and DSC curves (b) of as-spun core–shell membranes and corresponding neat fibers.

XRD was used to further illustrate the structure of the samples (Fig. 6). As for the as-spun xerogel fibers, no peaks were observed but broad humps were from the XRD patterns (Fig. 6a). The results reveal that all the three types of as-spun xerogel fibers are amorphous. After calcination at 700 °C, the peaks of γ-alumina appear in the XRD pattern of the mesoporous alumina fiber membrane while there are no obvious peaks except a broad hump observed in the pattern of the silica fiber membrane, which indicates that the amorphous alumina-based gel fibers transformed to γ-alumina fibers but the silica fibers still retain the amorphous structure after calcination. Compared to pure mesoporous alumina, the crystallization of γ-alumina for silica/mesoporous alumina core–shell fiber membrane is weak, which could be identified by the broad peak at 67°. The weak crystallization may be caused by the delayed effect of silica toward the crystallization of alumina. For the fibers calcined at 800 °C and 900 °C, the XRD patterns of the core–shell fibrous membranes show the reflections of γ-alumina, while the silica fibrous membranes still retain the amorphous structure. The crystalline sizes of Al2O3 for the samples calcined at 800 °C and 900 °C have been calculated by Scherrer's equation according to Scherrer line width analyses of the (440) reflections (Table 1), and they are 6.9 nm and 7.2 nm, respectively.

image file: c4ra03912b-f6.tif
Fig. 6 XRD patterns of the xerogel core–shell fibrous membranes compared with corresponding neat fibers (a), and those calcined at 700 °C (b), and at 800 °C (c), and 900 °C (d).
Table 1 The BET surface areas and porosity of the samples
Sample Crystalline size (nm) BET surface area (m2 g−1) Pore width (nm) Pore volume (cc g−1)
S/M-700 °C 134.010 5.300 0.191
S/M-800 °C 6.9 104.513 5.300 0.128
S/M-900 °C 7.2 79.491 5.300 0.114

N2 adsorption–desorption isotherms and the corresponding pore-size distributions of different samples calcined at 700 °C are presented in Fig. 7. The silica fibrous membrane shows type-III like adsorption isotherm, which represents the weak adsorption of N2 on nonporous materials. The corresponding surface area and pore volume of the fibers are only 7.23 m2 g−1 and 0.008 cc g−1, indicating the silica fibers are compact with few pores. In the case of silica/mesoporous alumina core–shell fibrous membranes, the isotherm changes to type-IV like adsorption with H1 loop, which is similar to that of pure mesoporous alumina fibers (Fig. 1c). The results correspond with the TEM images in Fig. 3, in which the mesopores are observed apparently in the shell. The surface area and pore volume increase to 134.01 m2 g−1 and 0.191 cc g−1, respectively. Furthermore, the N2 adsorption–desorption isotherms of the core–shell fibrous membranes calcined at 800 °C and 900 °C have also been investigated. As shown in Table 1, the BET surface area, as well as the pore volume of the S/M fibrous membrane, decreases with increase in the calcination temperature. The decrease is mainly attributed to the grain growth of the alumina shell at high temperatures, which is consistent with our previous study.11

image file: c4ra03912b-f7.tif
Fig. 7 N2 adsorption–desorption isotherms of different samples after calcination at 700 °C (a) and the corresponding pore-size distribution curves (b).

3.2. Adsorption property toward Congo red

Silica/mesoporous alumina fibrous membranes have potential applications in the field of water treatment as adsorbents due to the mesoporous alumina shell structure. Here, the adsorption performances of the membranes toward Congo red (a common dye) have been tested. The relative adsorption capacity was evaluated by C/C0 (C0 and C represent the concentrations of Congo red before and after treatment, respectively. C0 = 46 mg L−1). As shown in Fig. 8A, the absorption peaks clearly decreased when using the S/M fibrous membrane calcined at 700 °C as adsorbent, which indicates that the adsorption of Congo red on the membrane is efficient. Moreover, comparative studies were carried out for adsorption toward Congo red using the neat silica fibrous membrane as adsorbent. As shown in Fig. 8B, the neat silica fibrous membrane has almost no adsorption toward Congo red, even after an equilibrium time of 150 min, whereas Congo red was almost totally adsorbed on the silica/mesoporous alumina core–shell fibrous membrane with ∼97% removal of Congo red (C/C0 = 0.03) in 160 min. According to our previous study,11 the adsorption of Congo red on alumina mainly depends on the electrostatic attraction and the coordination effect of the aluminum atoms with amine groups and the sulfo groups of Congo red. The mesoporous alumina shell possesses a high surface area and large pore volume, which could provide more active sites toward Congo red. Thus, the mesoporous alumina shell plays a major role in this core–shell fibrous membrane adsorption. Furthermore, the adsorption capacity of the core–shell fibrous membrane toward Congo red is higher than that of other membrane adsorbents reported previously such as SiO2@γ-AlOOH (Boehmite) core–sheath fiber membrane9 and hierarchical films of layered double hydroxides.27
image file: c4ra03912b-f8.tif
Fig. 8 UV-vis absorption spectra of Congo red in the presence of the 700 °C S/M fibrous membrane after different time intervals (A) and the adsorption rates on samples calcined at 700 °C (B). Initial Congo red concentration: 46 mg L−1.

To further investigate the adsorbed capacity of the membrane toward Congo red, the volume of Congo red solution was further increased. As shown in Fig. 9, there was almost no adsorption for the silica fibrous membrane, while the adsorbed amount of silica/mesoporous alumina fibrous membrane continued to rise within 48 h. Considering the appropriate adsorption time in the practical application, the adsorbed amount was tested within the time limit of 48 h. The adsorbed amount of silica/mesoporous alumina fibrous membrane was 115 mg g−1 at 48 h, which is much higher than that of other membrane adsorbents reported previously, such as SiO2@γ-AlOOH (Boehmite) core–sheath fiber membrane toward Congo red (24.3 mg g−1),9 self-standing alumina nanofiber films toward methyl orange molecules (∼10 mg g−1),15 boehmite nonwovens toward stilbazo (∼58 mg g−1),16 SiO2–TiO2 composite porous nanofibrous membranes toward methylene blue (62.1 mg g−1),18 and hierarchical films of layered double hydroxides toward Congo red (20 mg g−1).27 Moreover, the adsorption capacities of the core–shell membranes calcined at 800 and 900 °C toward Congo red as a function of adsorption time were also tested. As shown in Fig. 9B, the adsorption of the two samples reached equilibrium within 17 h. The saturation adsorbed amounts of the S/M-800 °C and S/M-900 °C are 74 mg g−1 and 49 mg g−1, respectively. The adsorption capacities decreased with increase in the calcination temperature, which corresponds to the decrease in the BET surface area.

image file: c4ra03912b-f9.tif
Fig. 9 The adsorption amount toward Congo red as a function of adsorption time. 6 mg membrane immersed in 50 mL Congo red solution. Initial concentration of Congo red: 55 mg L−1. (A) for the membranes calcined at 700 °C, (B) for the core–shell membranes calcined at 800 and 900 °C.

The silica/mesoporous alumina core–shell fibrous membrane could be manipulated and separated more conveniently due to its flexible self-standing membrane form compared to the conventional powdery form, which can been seen from Fig. 10 and 11. The structure of the silica/mesoporous alumina fibrous membrane after adsorbing Congo red was further investigated by SEM (Fig. 10a). The core–shell fibers are still continuous, which is good for maintaining the integrity of the membrane. Furthermore, the membrane could be regenerated by a simple thermal treatment in air at 450 °C for 2 h. Moreover, the reusable property of the core–shell membrane has been investigated. The experiments were carried out by adding 10 mg membrane into 10 mL Congo red solution with an initial concentration of 55 mg L−1. For the silica/mesoporous alumina fibrous membrane calcined at 700 °C, as shown in Fig. 11, the initial adsorption capacity was 36.56 mg g−1, which decreased slightly with increasing cycle times, and the adsorption capacity remained as high as 30.14 mg g−1, even after five regenerations. According to our previous study, the slight decrease of the adsorption capacity for mesoporous alumina during the cyclic test may be caused by the small amount of residual SO42− derived from Congo red, which is difficult to remove completely and thus may occupy some of the active sites of the alumina adsorbent.11 The inset in Fig. 11 shows optical photograph of the membrane that had adsorbed Congo red for the fifth cycle, which were dried at 70 °C for 24 h. The membrane form was kept well after being reused several times.

image file: c4ra03912b-f10.tif
Fig. 10 SEM images of the core–shell fibrous membrane after adsorption of Congo red (a), and optical images of a Congo red before (b) and after (c) adsorption by silica/mesoporous alumina core–shell fibrous membrane.

image file: c4ra03912b-f11.tif
Fig. 11 Relationship between the adsorption capacity and cycle times of the S/M core–shell fibrous membranes calcined at 700 °C. The inset is the optical image of the membranes after the fifth cycle of adsorption toward Congo red.

To determine whether the alumina shell was stable in aqueous conditions, 10 mg membrane (obtained at 700 °C) was immersed in 10 mL aqueous solution for several days, and then the Al element in the solution was analyzed on an ICP-AES apparatus. Theoretically, if all the alumina in the membrane dissolved in the solution, the Al concentration in the solution should be 774 ppm. Actually, the Al content was only 0.0433 ppm from the experiment. The result shows that the alumina shell is basically stable in aqueous conditions, and there is little dissolution of the alumina shell into the water. The high mechanical strength and good reusable performance in adsorption toward Congo red, as well as the high stability in aqueous solution of the membrane, favors its application in the treatment of water.

4. Conclusions

In summary, silica/mesoporous alumina core–shell fibrous membranes have been successfully fabricated through a facile and applicable one-step coaxial electrospinning technique. The calcined silica/mesoporous alumina core–shell fibrous membranes exhibit good adsorption performance toward Congo red with an adsorption of 115 mg g−1 within 48 h. Furthermore, the form of the silica/mesoporous alumina fibrous membranes was kept well throughout the reuse process, and the adsorption capacity decreased only slightly after several cyclic experiments. Moreover, the membrane form as an adsorbent is easy to handle and can be easily reclaimed in practical application. These outstanding features of the core–shell membranes favor their applications in the treatment of water.


This work is supported by the National Key Technologies R&D Program (no.2013BAC01B02) and the National High-tech R&D Program (863 Program, 2012AA03A210).


  1. D. W. Kolpin, E. T. Furlong, M. T. Meyer, E. M. Thurman, S. D. Zaugg, L. B. Barber and H. T. Buxton, Environ. Sci. Technol., 2002, 36, 1202–1211 CrossRef CAS .
  2. C. Chen, P. Gunawan and R. Xu, J. Mater. Chem., 2011, 21, 1218–1225 RSC .
  3. J. B. Fei, Y. Cui, X. H. Yan, W. Qi, Y. Yang, K. W. Wang, Q. He and J. B. Li, Adv. Mater., 2008, 20, 452–456 CrossRef .
  4. X. D. Chichao Yu, L. M. Guo, J. T. Li, F. Qin, L. X. Zhang, J. L. Shi and D. S. Yan, J. Phys. Chem. C, 2008, 112, 13378–13382 Search PubMed .
  5. W. Cai, J. Yu, B. Cheng, B.-L. Su and M. Jaroniec, J. Phys. Chem. C, 2009, 113, 14739–14746 CAS .
  6. W. Cai, Y. Hu, J. Chen, G. Zhang and T. Xia, CrystEngComm, 2012, 14, 972–977 RSC .
  7. S. D. Kalme, G. K. Parshetti, S. U. Jadhav and S. P. Govindwar, Bioresour. Technol., 2007, 98, 1405–1410 CrossRef CAS PubMed .
  8. S. Chiron, A. Fernandez-Alba, A. Rodriguez and E. Garcia-Calvo, Water Res., 2000, 34, 366–377 CrossRef CAS .
  9. Y. E. Miao, R. Wang, D. Chen, Z. Liu and T. Liu, ACS Appl. Mater. Interfaces, 2012, 4, 5353–5359 CAS .
  10. W. Q. Cai, J. G. Yu and M. Jaroniec, J. Mater. Chem., 2010, 20, 4587–4594 RSC .
  11. Y. Wang, W. Li, X. Jiao and D. Chen, J. Mater. Chem. A, 2013, 1, 10720–10726 CAS .
  12. C. Wang, S. Huang, L. Wang, Z. Deng, J. Jin, J. Liu, L. Chen, X. Zheng, Y. Li and B.-L. Su, RSC Adv., 2013, 3, 1699–1702 RSC .
  13. A. S. Hashmi and G. J. Hutchings, Angew. Chem., Int. Ed., 2006, 45, 7896–7936 CrossRef PubMed .
  14. Z. Guan, L. Liu, L. He and S. Yang, J. Hazard. Mater., 2011, 196, 270–277 CrossRef CAS PubMed .
  15. Z.-G. Zhao, N. Nagai, T. Kodaira, Y. Hukuta, K. Bando, H. Takashima and F. Mizukami, J. Mater. Chem., 2011, 21, 14984–14989 RSC .
  16. N. Nagai, K. Ihara, A. Itoi, T. Kodaira, H. Takashima, Y. Hakuta, K. K. Bando, N. Itoh and F. Mizukami, J. Mater. Chem., 2012, 22, 21225–21231 RSC .
  17. F. Zhang, C. Yuan, J. Zhu, J. Wang, X. Zhang and X. W. D. Lou, Adv. Funct. Mater., 2013, 23, 3909–3915 CrossRef CAS .
  18. Q. Wen, J. Di, Y. Zhao, Y. Wang, L. Jiang and J. Yu, Chem. Sci., 2013, 4, 4378–4382 RSC .
  19. Y. Zhu, X. Han, Y. Xu, Y. Liu, S. Zheng, K. Xu, L. Hu and C. Wang, ACS Nano, 2013, 7, 6378–6386 CrossRef CAS PubMed .
  20. M. Shang, W. Wang, S. Sun, E. Gao, Z. Zhang, L. Zhang and R. O'Hayre, Nanoscale, 2013, 5, 5036–5042 RSC .
  21. X. Mao, Y. Si, Y. Chen, L. Yang, F. Zhao, B. Ding and J. Yu, RSC Adv., 2012, 2, 12216–12223 RSC .
  22. J. H. Pan, X. S. Zhao and W. I. Lee, Chem. Eng. J., 2011, 170, 363–380 CrossRef CAS PubMed .
  23. D. Grosso, F. Cagnol, G. J. d. A. A. Soler-Illia, E. L. Crepaldi, H. Amenitsch, A. Brunet-Bruneau, A. Bourgeois and C. Sanchez, Adv. Funct. Mater., 2004, 14, 309–322 CrossRef CAS .
  24. S. Zhan, D. Chen, X. Jiao and C. Tao, J. Phys. Chem. B, 2006, 110, 11199–11204 CrossRef CAS PubMed .
  25. J. H. Yu, S. V. Fridrikh and G. C. Rutledge, Adv. Mater., 2004, 16, 1562–1566 CrossRef CAS .
  26. D. Li, A. Babel, S. A. Jenekhe and Y. Xia, Adv. Mater., 2004, 16, 2062–2066 CrossRef CAS .
  27. Y. Zhao, S. He, M. Wei, D. G. Evans and X. Duan, Chem. Commun., 2010, 46, 3031–3033 RSC .

This journal is © The Royal Society of Chemistry 2014