Coaxial solution blowing of modified hollow polyacrylonitrile (PAN) nanofiber Fe complex (Fe-AO-CSB-HPAN) as a heterogeneous Fenton photocatalyst for organic dye degradation

Abstract

As a novel photocatalyst supporting material, one-dimensional hollow PAN nanofiber mat (HPAN) was successfully fabricated via coaxial solution blowing method (CSB), which we named CSB-HPAN. Then modified hollow PAN Fe complex (Fe-AO-CSB-HPAN) was prepared by the amidoximation and Fe coordination of AO-CSB-HPAN which used for the heterogeneous Fenton degradation of textile dyes. SEM, TEM and the digital photo revealed that CSB-HPAN has hollow structure, three-dimensional curly and loosely fibrous morphologies, which is beneficial for Fe³⁺ complexation, dye adsorption and degradation. The experiments of photocatalytic activity indicated that Fe-AO-CSB-HPAN showed excellent photocatalytic performance in the degradation of textile dyes in the presence of H₂O₂ under light irradiation. Meanwhile, 99% of RR195 molecules were decomposed by Fe-AO-CSB-HPAN in 35 min, which was faster than by conventional modified electrospinning PAN nanofiber Fe complex (Fe-AO-ES-PAN), due to the large surface area and loosely fibrous structure of Fe-AO-CSB-HPAN.

---

1. Introduction

Dye-containing wastewater from different industrial fields is a principal source of environmental contamination. And these dyes are toxic and nonbiodegradable to aquatic animals and plants.^1,2 During the last two decades, Fenton and photo-assisted Fenton technologies have been widely used in the degradation of water-soluble organic dyes owing to their low cost and high performance in treating processes. Furthermore, because of its application over a wide pH range and easy separation of the catalyst after the reaction, active heterogeneous Fenton catalysis is gradually replacing the homogeneous system.³

In recent years, metal coordination has been used as an effective method to immobilize Fe(III) ions on different materials such as porous materials (pillared clays and zeolite, etc.),^4–10 Nafion membranes^11,12 and fibrous materials^13–17 for producing heterogeneous Fenton catalysts. It should be noticed that among these different materials, the modified PAN nanofiber prepared by convenient electrospinning method has been successfully used as supported materials which was coordinated with Fe ions for catalyzing the dye degradation. Moreover, Fe(III)–PAN nanofiber complex represents a kind of promising heterogeneous catalysts with unique properties, such as lower cost, easy separation, large surface area as well as convenient utility.¹⁸ However, this excellent catalyst has not been widely used because of the drawbacks of electrospinning method.

Electrospinning is a common and straightforward method for preparing nanofibers, in which nanofiber-forming liquid is forced by electrostatic potential to eject out through a concentric capillary.¹⁹ However, most published works were carried out using experimental laboratory-scale setups. The nanofibers are fabricated in very low yields with 1.0–5.0 mL h⁻¹ solution flow rate which limits the industrial applications of electrospinning.²⁰ Meanwhile, the nanofiber membrane made by electrospinning often presents straight morphology and tightly compressed in bulk, which restricts its application in catalysis and so on. Recently, solution blowing process was reported as an innovative method to produce micro/nanofibers with its high nanofiber productivity.²¹ In this process, the solution jet was attenuated by the high velocity gas flow, and then polymer nanofibers rapidly solidify after solvents evaporation. Polystyrene (PS), poly(lactic acid) (PLA), poly(methyl methacrylate) (PMMA), polyvinylpyrrolidone (PVP),²² polyvinylidene fluoride (PVDF)²³ and cellulose micro-and nanofiber mats²⁴ had been successfully produced by solution blowing process. Furthermore, some works were carried out by our groups via the solution blowing method for preparing the metal oxide ceramic fibers, such as zirconia fiber²⁵ and alumina fiber.²⁶ It is worth noting that the fiber mats obtained by solution blowing method often present three-dimensional curly and loosely fibrous morphologies, which is beneficial for the catalysis application.²⁷ Whereas, until now, detailed information on solution blown PAN nanofiber, especially the hollow solution blown PAN nanofiber used as supported materials by coordinated with Fe ions for catalyzing the dye degradation has not been reported.

Hence, in the present work, coaxial solution blowing process was proposed to fabricate mental coordinated nanofibers in the form of core–shell structure for photoassisted heterogeneous Fenton system. Morphologies and structures of CSB-HPAN and Fe-AO-CSB-HPAN were characterized, and the results indicated that CSB-HPAN had hollow structure, three-dimensional curly and loosely fibrous morphologies, which was beneficial for Fe³⁺ complexation, dye adsorption and degradation. The catalytic property of Fe-AO-CSB-HPAN complex in the degradation of three typical textile dyes in the presence of H₂O₂ was investigated. Meanwhile, a comparative experiment of the Fe-AO-CSB-HPAN and the Fe-AO-ES-PAN to decompose RR195 was also performed.

2. Experimental

2.1 Chemicals

Polyacrylonitrile (PAN) powder (M_w ∼ 90 000) was purchased from Sigma-Aldrich. Polyvinyl pyrrolidone (PVP) was purchased from Boai NKY Pharmaceuticals Ltd. N,N-Dimethylformamide (DMF), hydroxylamine hydrochloride, sodium hydroxide, ferric chloride were all from Tianjin Guangfu regent company. All reagents were analytical grade and used as received. Three kinds of textile dyes including Reactive Red 195 (RR195), Reactive Blue 4 (RB4), and Acid Blue 7 (AB7) were commercially available and purified by reprecipitation method in this experiment.

2.2 Preparation of CSB-HPAN, ES-PAN, Fe-AO-CSB-HPAN and Fe-AO-ES-PAN

In order to prepare the nanofibers, 20 g PAN powder was dissolved into 100 mL DMF as the outer solution. 12.5 g PVP (K90) was added into 80 mL DMF under magnetic stirring to form a homogeneous precursor solution (inner solution) at room temperature. In coaxial solution blowing process, the solution streams are pressed out of the concentric capillary and then stretched extremely by the high speed gas flow to the collector accompanied by solvent evaporation to form nanofibers. The process is similar to electrospinning but with different nanofiber formation driving force. For coaxial solution blowing, high speed gas flow deforms the solution streams, evaporates the solvent, and solidifies them into nanofibers, while electric force performs in electrospinning process. In our experiment, the coaxial solution blowing setup and the concentric spinneret we used were shown in Fig. 1(a) and (b), respectively. The main spinning parameters were shown below: air-blowing pressure (0.05 MPa), solution feeding rate (inner and outer were both 16 mL h⁻¹), inner spinneret diameter (0.4 mm), outer spinneret diameter (0.6 mm), width of the width of the air slot (0.8 mm), humidity (25%), temperature (25 °C) and collection distance (1000 mm). On the other hand, in order to prepare electrospinning (ES) PAN nanofiber, the obtained PAN solution was added to a 10 mL glass syringe with a blunt needle. The PAN solution flow rate was controlled by a microinfusion pump to be 0.6 mL h⁻¹. The high-voltage supplier was used to connect the grounded collector and metal needles for forming electrostatic fields. The used voltage was 20 kV and a piece of flat aluminum foil on the grounded collection roller was placed about 15 cm below the tip of the needle to collect the nanofiber, humidity (25%), and temperature (25 °C). A jet of PAN solution came out from the needle tip at a critical voltage and was collected on the aluminum foil. The PAN nanofibers were obtained after DMF evaporation, which we named ES-PAN.

Fig. 1 Schematic of the coaxial solution blowing apparatus (a) and the concentric spinneret (b).

Then the as-spun shell/core nanofiber and ES-PAN nanofiber were dried for 12 h at 60 °C to remove residual solvents. Subsequently the former product was immersed into the deionized water with ultrasonic processing for 3 h to remove PVP until the weight of the nanofiber had no changed, as the frequency of ultrasonic process were 80 kHz (Kunshan Ultrasonic Equipment Co. Ltd). Then CSB-HPAN was obtained. The CSB-HPAN (1.0 g) was amidoximated using a mixed solution containing hydroxylamine hydrochloride solution (0.40 mol L⁻¹) and sodium hydroxide (pH = 9) in a 250 mL flask with a thermometer and agitator for 1–2 h at 68 °C. And then the resulting amidoximated hollow PAN nanofiber (AO-CSB-HPAN) was washed several times with deionized water until neutral and dried under vacuum at 50 °C. The obtained AO-CSB-HPAN was immersed into an aqueous solution of ferric chloride with the concentration varying from 0.08 to 0.12 mol L⁻¹ under continuous stirred at 50 °C for 2 h in order to produce the amidoximated hollow PAN nanofiber Fe complex (Fe-AO-CSB-HPAN), and the synthesis and schematic for Fe-AO-CSB-HPAN preparation were shown in Fig. 2(a) and (b). Similarly, the process of producing AO-ES-PAN and Fe-AO-ES-PAN were the same as the CSB one's.

Fig. 2 (a) The synthesis of AO-CSB-HPAN and Fe-AO-CSB-HPAN; (b) the schematic for Fe-AO-CSB-HPAN preparation.

Fig. 3 (a) SEM of CSB-HPAN; (b) cross-section and (c) TEM images of CSB-HPAN; (d) SEM of Fe-AO-CSB-HPAN; (e) SEM of ES-PAN; (f) SEM of Fe-AO-ES-PAN.

The residual concentration of Fe³⁺ in solution after coordination was determined using a WXF120 atomic adsorption spectrometry (AAS, Beijing Rayleigh Analytical Instrument Corp.) for calculating the Fe content (Q_Fe) of the complex.

2.3 Characterization

The morphologies of the CSB-HPAN and ES-PAN were characterized by a field emission scanning electron microscope (FE-SEM, Hitachi S-4800) and transmission electron microscope (TEM, Hitachi H-7650). The diameter distributions of nanofiber were determined by analyzing the SEM imagines with an image analyzing software (Image-Pro Plus). The compositions of CSB-HPAN and Fe-AO-CSB-HPAN were verified by a Fourier transform infrared spectrometer (FTIR, Bruker, TENSOR37). The crystalline phases were identified by X-ray diffraction (XRD) (D8 Discover with GADDS, BRUKER AXS co., USA) at 45 kV and 40 mA from 10° to 40° with a Ni-filtered Cu Kα radiation (λ = 0.1542 nm). The surface elemental composition of the Fe-AO-CSB-HPAN was analyzed using an X-ray photoelectron spectroscopy (XPS) (K-alpha X, Thermo Fisher co., USA). The photocatalytic activity towards the degradation of textile dyes was investigated by UV-vis absorption (Hitachi U-3900).

2.4 Dyes degradation procedure and analysis

Firstly, 0.2 g of Fe-AO-CSB-HPAN was immersed into 50 mL 0.05 mmol L⁻¹ RR195 solution and 15.5 μL 3.0 mmol L⁻¹ H₂O₂ in the beaker. The temperature in the beaker was kept at 25 °C. The solution in beaker was exposed to the irradiation of lamp in photoreaction system. The UV-vis spectra of the RR195 solution was monitored every 10 min. The degradation percentage of the dye was expressed as D% = (1 − A_t/A₀) × 100%, where A₀ is the initial absorbance value (max = 522 nm) and A_t is the absorbance value at t time. Meanwhile, the comparative study was carried out for the degradation of three kinds of textile dyes: azo dye, Reactive Red (RR195), anthraquinone dye, Reactive Blue (RB4), and triphenylmethane dye, Acid Blue 7 (AB7) from aqueous solution by Fe-AO-CSB-HPAN.

3. Results and discussion

Fig. 1(a) shows the random orientation and smooth surface of CSB-HPAN, and its average diameter was 549.44 nm. The cross-section and hollow morphologies of the CSB-HPAN are shown in Fig. 1(b) and (c), which are SEM and TEM images, respectively. With the help of ultrasonic treatment, the core layer (PVP) was removed and CSB-HPAN was successfully manufactured and its cross-section was elliptical. The elliptical cross-section shape of CSB-HPAN was probably due to irregular interfacial tension between the core and shell solutions, the whipping and buckling phenomena during the coaxial solution blowing process, which was similar to the coaxial electrospinning process.²⁸ Moreover, the surface of CSB-HPAN was rarely affected by the modification and Fe coordination (Fig. 1(d)). However, the average diameter of the resulting Fe-AO-CSB-HPAN was 735.93 nm because of the swelling behavior of the CSB-HPAN during the chemical modification process.¹⁷ Meanwhile, the SEM of ES-PAN is shown in Fig. 1(e), with the average diameter of 349.44 nm. With the same reason, the diameter of Fe-AO-ES-PAN became thicker, which was 535.93 nm due to the swelling phenomenon occurred.

Fig. 4 exhibits the FT-IR spectrum of hollow CSB-HPAN. The typical characteristic band due to the stretching vibration of nitrile (2246 cm⁻¹) was clearly observed. The characteristic bands of the functional groups of PAN with additional peaks appeared at 3500–3000 cm⁻¹, 1654 cm⁻¹, 1110 cm⁻¹ and 939 cm⁻¹ which was due to the stretching vibration of O–H, C=N, C–N, N–O groups in amidoxime, respectively. The intensity of these peaks decreased after coordinating with Fe³⁺ ions in the spectrum of Fe-AO-CSB-HPAN. This demonstrated that the Fe³⁺ ions were chemically attached to the AO-CSB-HPAN.

Fig. 4 FT-IR spectrum of (1) CSB-HPAN; (2) AO-CSB-HPAN; (3) Fe-AO-CSB-HPAN.

Fig. 5 shows XRD patterns of the CSB-HPAN, AO-CSB-HPAN, Fe-AO-CSB-HPAN in comparison with that of the ES-PAN. Clearly, two diffraction peaks at the scattering angles 2θ of about 17° and 29° were evident for the ES-PAN. These peaks corresponded to the (100) and (101) diffraction planes.^29–31 Meanwhile, the diffraction pattern of the CSB-HPAN contained broader peaks at 2θ of 17° and 29°, suggesting a very low crystallinity of the fiber, which was beneficial to the amidoximation and Fe(III) ion coordination. In electrospinning process, the solution streams are always subjected to the effect of electric force, and the nanofiber prepared via this method has a higher molecular orientation degree compared with the coaxial solution blowing method. On the other hand, in the coaxial blowing process, the drafting force will decrease when the solution streams away from the spinneret, which make the molecular orientation and the crystallinity of the CSB fiber reduce, correspondingly. What's more, these peaks became lower intensities and broader after amidoximation and Fe(III) ion coordination. The decreased intensity indicates a possible reduction in the crystallinity, whereas the broadening represents a reduction crystal size³² as a consequence of the amidoximation and Fe(III) ion coordination.

Fig. 5 XRD spectra of ES-PAN, CSB-HPAN, AO-CSB-HPAN and Fe-AO-CSB-HPAN.

The XPS survey spectrum of Fe-AO-CSB-HPAN is presented in Fig. 6. It can be found that the main elements from the complex included C, O, Fe, N, Cl. Moreover, the changes in binding energy are summarized in Table 1.

Fig. 6 XPS survey spectrum of Fe-AO-CSB-HPAN.

Table 1 XPS analysis of AO-CSB-PAN, FeCl₃ and Fe-AO-CSB-PAN

XPS peaks	AO-CSB-HPAN	FeCl3	Fe-AO-CSB-HPAN
O1s	531.1	—	532.6
N1s	399.1	—	400.5
Fe2p	—	711.3	709.2

---

As can be seen from Table 1, the binding energy of the Fe2p in Fe-AO-CSB-HPAN is 709.2 eV, lower than that (711.3 eV) of the Fe2p in FeCl₃, revealing that Fe(III) ions has successfully coordinated with amidoxime groups of AO-CSB-HPAN to produce Fe-AO-CSB-HPAN in solution. Meanwhile, the binding energy of the O1s and N1s in Fe-AO-CSB-HPAN shifted to higher binding energy level, suggesting that oxygen and nitrogen atoms as the electron donors coordinated with Fe(III) to form Fe-AO-CSB-HPAN. Moreover, there may be a number of N–Fe and O–Fe coordinated bonds on the surface of Fe-AO-CSB-HPAN. This is in agreement with the results obtained by FT-IR method. Consequently, we believed that Fe(III) ions have been incorporated to AO-CSB-HPAN during our preparation. All the XPS results were similar to those published in literature.¹⁵

Fig. 7 shows the Fe contents (Q_Fe) of the coaxial solution blowing PAN nanofiber complex (Fe-AO-CSB-HPAN) and conventional electrospinning PAN nanofiber complex (Fe-AO-ES-PAN) with three different initial Fe³⁺ concentrations, respectively. It can be seen that the Fe³⁺ complexation of the AO-CSB-HPAN was higher than the AO-ES-PAN, which was generally ascribed to the excellent adsorption properties and higher specific surface area of the former. Moreover, comparing with the electrospun nanofiber membrane in which nanofibers are straight and tightly compressed in bulk (Fig. 7(b)), the solution blown nanofibers are curly in three-dimensional and result in loosely fibrous mat (Fig. 7(c)), which is beneficial for Fe³⁺ complexation, dye adsorption and degradation.³³

Fig. 7 (a) Different Fe contents of the electrospinning PAN nanofiber (ES-PAN) complex and coaxial solution blowing hollow PAN nanofiber (CSB-HPAN) complex with different initial Fe³⁺ concentrations; (b) digital photo of ES-PAN; (c) digital photo of CSB-HPAN.

To study the catalytic property of Fe-AO-CSB-HPAN as a heterogeneous Fenton photocatalyst, the degradation of RR195 was conducted under the various conditions. For comparison, a control experiment was also performed in the presence of the Fe-AO-ES-PAN, and the results were given in Fig. 8.

Fig. 8 Variations in D% values of RR195 in different systems (a) RR195 and Fe-AO-ES-PAN under dark, (b) RR195 and Fe-AO-CSB-HPAN under dark, (c) RR195, Fe-AO-ES-PAN, and H₂O₂ under visible irradiation, (d) RR195, Fe-AO-CSB-HPAN and H₂O₂ under visible irradiation. The experiments were carried out under conditions: Fe-AO-CSB-HPAN (Q_Fe = 5.12 mmol g⁻¹) (inset is its UV-vis spectra of RR195 degradation) and Fe-AO-ES-PAN (Q_Fe = 3.32 mmol g⁻¹) = 0.2 g, pH = 6, concentrations of RR195 and H₂O₂ were 0.05 mmol L⁻¹ and 3.0 mmol L⁻¹, respectively.

The curve (a) and curve (b) present the adsorption of RR195 by Fe-AO-ES-PAN and Fe-AO-CSB-HPAN, respectively. It shows that the D% value is less than 25% after 35 min of the adsorption time under dark. Moreover, compared with Fe-AO-ES-PAN, Fe-AO-CSB-HPAN performed better effect of dye adsorption due to its hollow and three-dimensional curly fibrous structures. The curve (c) and curve (d) show the concentrations of RR195 decreased along with time with the existence of catalysts, which indicated that the two kinds of catalysts had significant catalytic performance. It's noteworthy that Fe-AO-CSB-HPAN possessed higher catalytic degradation rate for RR195 compared with Fe-AO-ES-PAN. The degradation rate of Fe-AO-CSB-HPAN was approximately 100% at 35 min while that of Fe-AO-ES-PAN was only 70%. The main reason was hollow structure, three-dimensional curly and loosely fibrous morphologies make the CSB-HPAN had higher specific surface area and better adsorption properties, which were favor of the complexation of Fe³⁺ leading to excellent catalytic performance. Moreover, the above mentioned properties of CSB-HPAN also increased the active sites of the Fe-AO-CSB-HPAN, which accelerate the decomposition of the dyes in solution. Additionally, the initial Fe³⁺ concentration was 0.12 mol L⁻¹, the Fe contents (Q_Fe) of Fe-AO-CSB-HPAN and Fe-AO-ES-PAN were 5.12 mmol g⁻¹ and 3.32 mmol g⁻¹, respectively.

The comparative study was carried out for the degradation of three kinds of textile dyes: azo dye, Reactive Red (RR195), anthraquinone dye, Reactive Blue (RB4), and triphenylmethane dye, Acid Blue 7 (AB7) from aqueous solution by Fe-AO-CSB-HPAN (Q_Fe = 5.12 mmol g⁻¹) and 3.0 mmol L⁻¹ H₂O₂ at pH 6 under light irradiation, and D% values of the three dyes were calculated and presented in Fig. 9. As shown in Fig. 9, D% values of the three dyes were all above 90% within 35 min. These results suggest that chromophores of three dyes have been almost completely destructed after the degradation. Consequently, Fe-AO-CSB-HPAN was proved to be a universal and efficient catalyst for degradation of the three textile dyes with different molecular structures.

Fig. 9 Degradation of three dyes with different molecular structures. The experiments were carried out under conditions: Fe-AO-CSB-HPAN (Q_Fe = 5.12 mmol g⁻¹) = 0.2 g, pH = 6, concentrations of RR195 and H₂O₂ were 0.05 mmol L⁻¹ and 3.0 mmol L⁻¹, respectively.

4. Conclusions

Three-dimensional hollow PAN nanofiber mat (CSB-HPAN), for the first time, was successfully fabricated using coaxial solution blowing method, which has the curly and loosely morphology. The heterogeneous catalyst of Fe-AO-CSB-HPAN with average diameter of 735.93 nm was obtained by amidoximation and Fe coordination of CSB-HPAN. 99% of RR195 molecules were decomposed by Fe-AO-CSB-HPAN in 35 min, which was faster than by conventional modified electrospinning PAN nanofiber Fe complex (Fe-AO-ES-PAN). Furthermore, Fe-AO-CSB-HPAN was proved to be a universal and efficient catalyst for degradation of the three textile dyes with different molecular structures. The enhanced photocatalytic activity of the heterogeneous catalyst can be ascribed to the improvement of large surface area, the enhancement of adsorption ability of surfaces and loosely fibrous structure of Fe-AO-CSB-HPAN. This facile method for producing hollow nanofiber mat may be extended to prepare other complex heterogeneous catalyst for degradation of organic dyes.

Acknowledgements

The author would like to thank National Natural Science Foundation of China (51102178 and 51173131) for their financial support.

References

1. B. Lodha and S. Chaudhari, J. Hazard. Mater., 2007, 148, 459–466.
2. P. Nidheesh, RSC Adv., 2015, 5, 40552–40577.
3. S. Parra, V. Nadtotechenko, P. Albers and J. Kiwi, J. Phys. Chem. B, 2004, 108, 4439–4448.
4. M. Hartmann, S. Kullmann and H. Keller, J. Mater. Chem., 2010, 20, 9002–9017.
5. J. Pérez-Ramírez, M. S. Kumar and A. Brückner, J. Catal., 2004, 223, 13–27.
6. B. Fan, H. Li, W. Fan, C. Jin and R. Li, Appl. Catal., A, 2008, 340, 67–75.
7. G. Sartori and R. Maggi, Chem. Rev., 2006, 106, 1077–1104.
8. A. C.-K. Yip, F. L.-Y. Lam and X. Hu, Chem. Eng. Sci., 2007, 62, 5150–5153.
9. M. Cheng, W. Song, W. Ma, C. Chen, J. Zhao, J. Lin and H. Zhu, Appl. Catal., B, 2008, 77, 355–363.
10. H. Jia, J. Zhao, X. Fan, K. Dilimulati and C. Wang, Appl. Catal., B, 2012, 123, 43–51.
11. J. Kiwi, N. Denisov, Y. Gak, N. Ovanesyan, P. Buffat, E. Suvorova, F. Gostev, A. Titov, O. Sarkisov and P. Albers, Langmuir, 2002, 18, 9054–9066.
12. S. Parra, I. Guasaquillo, O. Enea, E. Mielczarski, J. Mielczarki, P. Albers, L. Kiwi-Minsker and J. Kiwi, J. Phys. Chem. B, 2003, 107, 7026–7035.
13. V. Ishtchenko, K. Huddersman and R. Vitkovskaya, Appl. Catal., A, 2003, 242, 123–137.
14. X. Liu, R. Tang, Q. He, X. Liao and B. Shi, J. Hazard. Mater., 2010, 174, 687–693.
15. Y. Dong, Z. Han, C. Liu and F. Du, Sci. Total Environ., 2010, 408, 2245–2253.
16. Y. Dong, Z. Han, S. Dong, J. Wu and Z. Ding, Catal. Today, 2011, 175, 299–309.
17. Z. Han, Y. Dong and S. Dong, J. Hazard. Mater., 2011, 189, 241–248.
18. X. Zhao, Y. Dong, B. Cheng and W. Kang, Int. J. Photoenergy, 2013, 2013, 1–9.
19. A. Greiner and J. H. Wendorff, Angew. Chem., Int. Ed., 2007, 46, 5670–5703.
20. F. L. Zhou, R. H. Gong and I. Porat, Polym. Int., 2009, 58, 331–342.
21. E. S. Medeiros, G. M. Glenn, A. P. Klamczynski, W. J. Orts and L. H. Mattoso, J. Appl. Polym. Sci., 2009, 113, 2322–2330.
22. L. Zhang, P. Kopperstad, M. West, N. Hedin and H. Fong, J. Appl. Polym. Sci., 2009, 114, 3479–3486.
23. X. Zhuang, L. Shi, K. Jia, B. Cheng and W. Kang, J. Membr. Sci., 2013, 429, 66–70.
24. X. Zhuang, X. Yang, L. Shi, B. Cheng, K. Guan and W. Kang, Carbohydr. Polym., 2012, 90, 982–987.
25. B. Cheng, X. Tao, L. Shi, G. Yan and X. Zhuang, Ceram. Int., 2014, 40, 15013–15018.
26. L. Li, W. Kang, Y. Zhao, Y. Li, J. Shi and B. Cheng, Ceram. Int., 2015, 41, 409–415.
27. L. Shi, X. Zhuang, X. Tao, B. Cheng and W. Kang, Fibers Polym., 2013, 14, 1485–1490.
28. G. H. Lee, J.-C. Song and K.-B. Yoon, Macromol. Res., 2010, 18, 571–576.
29. P. Bajaj, D. Paliwal and A. Gupta, J. Appl. Polym. Sci., 1998, 67, 1647–1659.
30. J. Sutasinpromprae, S. Jitjaicham, M. Nithitanakul, C. Meechaisue and P. Supaphol, Polym. Int., 2006, 55, 825–833.
31. L. Quan, H. Zhang and L. Xu, J. Therm. Anal. Calorim., 2015, 119, 1081–1089.
32. R. M. El-Shishtawy and N. S. Ahmed, Color. Technol., 2005, 121, 139–146.
33. X. Zhuang, L. Shi, B. Zhang, B. Cheng and W. Kang, Macromol. Res., 2013, 21, 346–348.

---