Regenerable g-C₃N₄–chitosan beads with enhanced photocatalytic activity and stability

Abstract

In this study, a series of regenerable graphitic carbon nitride–chitosan (g-C₃N₄–CS) beads were successfully synthesized via the blend crosslinking method. The prepared beads were characterized by scanning electron microscopy (SEM), thermogravimetric analysis (TGA), X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FT-IR), diffuse reflectance spectroscopy (DRS), photoluminescence (PL) spectroscopy, and X-ray photoelectron spectroscopy (XPS). The structural characterization results indicate that the g-C₃N₄ granules were uniformly distributed on the surface of the chitosan matrix, and the structures of g-C₃N₄ and CS are maintained. In addition, the prepared g-C₃N₄–CS beads exhibited efficient MB degradation and stability. The optimum photocatalytic activity of our synthesized g-C₃N₄–CS beads was higher than that of the bulk g-C₃N₄ by a factor of 1.78 for MB. The improved photocatalytic activity was predominantly attributed to the synergistic effect between in situ adsorption and photocatalytic degradation. In addition, the reacted g-C₃N₄–CS beads can be regenerated by merely adding sodium hydroxide and hydrogen peroxide. Additionally, the regenerated g-C₃N₄–CS beads exhibit excellent stability after four runs, while the mass loss is less than 10%. This work might provide guidance for the design and fabrication of easily regenerated g-C₃N₄-based photocatalysts for environmental purification.

---

1. Introduction

Photocatalytic technology can utilize the inexhaustible solar energy, which is broadly applicable in the field of energy and environmental protection and has been widely applied to energy production.^1–8 Semiconductors, especially TiO₂,^9–11 have been widely studied as photocatalysts, which have advantages including chemical stability, nontoxicity, no secondary pollution and low costs.^12–16 They also have disadvantages based on in-depth research, which are their excitability only by ultraviolet light, wide bandgap energy, low photocatalytic efficiency and the electrons and holes being prone to recombination.^17–23 New materials should be explored to compensate for these shortcomings.

Recently, graphitic carbon nitride (g-C₃N₄) has attracted attention as a metal-free catalyst.^24–28 Compared with traditional inorganic semiconductors, g-C₃N₄ possesses a narrow band gap of ∼2.7 eV,^29–32 good resistance to thermal and chemical damage (acid, base and organic solvent),^4,33 and visible absorption at approximately 450–460 nm.^34–36 Nevertheless, in the stability experiment, there is a problem of mass loss due to the powdered g-C₃N₄.^37–39 In addition, g-C₃N₄ has general adsorption properties on account of the structure. As a natural alkaline cationic polymer, chitosan (CS) has excellent biocompatibility, nontoxicity, biodegradability and other advantages.⁴⁰ CS is a product of chitin deacetylated, containing a significant amount of amino and hydroxy groups in the molecule, which has excellent flocculation and adsorption characteristics. The CS has a chain-like structure that is easy to crystallize; and has a stable crystalline structure, that is insoluble in water, alkali and most organic solvents.^41–43

In previously reported studies, Nishad et al.⁴⁴ prepared stable beads by using nano-TiO₂ and CS and exhibited full adsorption performance for niobium of the +3 and +5 oxidation states. Yang Limin et al.⁴⁵ prepared an easily regenerated TiO₂ impregnated CS adsorbent, and at the same time, absorbed the photosensitizing effect of TiO₂ to achieve the adsorption and degradation of pollutants. Inspired by previous work, if a suitable method can be used to combine g-C₃N₄ and CS to design an easily regenerated composite catalyst with good photocatalytic ability and adsorption capacity, while reducing mass loss to improve the stability and utilization in cycle experiments, it will effectively improve the g-C₃N₄-based catalyst. The catalytic efficiency meets the requirements for practical application. To the best of our knowledge, few studies have combined g-C₃N₄ and CS to date.

Herein, in this study, g-C₃N₄–CS beads were first synthesized via the blend crosslinking method. The effect of the ratio between the g-C₃N₄ and CS to the in situ adsorption and photocatalytic activity of the products was investigated. SEM, XRD, FT-IR, DRS, PL and XPS measurements were performed to determine the morphology and phase change. Photocatalytic decomposition of methylene blue (MB) was carried out to evaluate the photocatalytic properties of the g-C₃N₄–CS beads. The results show that there is an optimal ratio between g-C₃N₄ and CS to coordinate in situ adsorption and photocatalytic activity to achieve the best level.

2. Material and methods

2.1 Material and reagents

Melamine, chitosan (CS, 85% deacetylation), epichlorohydrin, sodium hydroxide, anhydrous acetic acid, methylene blue (MB), isopropanol (IPA), triethanolamine (TEOA), and p-benzoquinone (BQ) were purchased from Sinopharm Chemical Reagent Company (China). All chemicals were of analytical grade and used without any further purification unless otherwise mentioned.

2.2 Preparation of regenerable g-C₃N₄–CS-x beads

2.2.1 Preparation of g-C₃N₄. Graphitic carbon nitride was prepared via the thermal polycondensation of melamine. Melamine was put into an alumina crucible with a cover and calcined in air at 550 °C for 2 h with a ramping rate of 5°C min⁻¹ in a muffle furnace, followed by naturally cooling to room temperature. Finally, the harvested samples were crushed in an agate mortar and put into the fine for further use.

2.2.2 Preparation of the regenerable g-C₃N₄–CS-x beads. Regenerable g-C₃N₄–CS-x beads samples were prepared via the blend crosslinking method as follows: the CS powder was dissolved to a certain concentration of the acetic acid solution and stirred until the CS completely dissolved. The prepared g-C₃N₄ powder was added to the homogeneous colloidal solution. The cross-linking agent epichlorohydrin (2 mL) was subsequently added, reacting for 60 min at room temperature after sonicating for 60 min. The homogeneous colloidal solution was added as drops into the 1% sodium hydroxide solution under constant stirring. The drops were isolated and coagulated to form guttulate beads. Then, the beads were separated and washed several times with distilled water until the filtrate was measured to be approximately pH = 7; and were denoted as g-C₃N₄–CS-x, where the x value represent the CS and the g-C₃N₄ ratio (x = 1, 2, 3, and 4). The final g-C₃N₄–CS-x beads within a certain size range (1 ± 0.1 mm) were obtained by drying at 60 °C in a vacuum drying oven for 5 h. It is worth mentioning that the needle size determines the bead size.

The crosslinking process of the g-C₃N₄–CS-x samples are shown in Fig. 1. g-C₃N₄ with a heptazine ring as a structural unit was calcined at 550 °C by the melamine precursor. The CS molecules interact with each other via epichlorohydrin to form a network structure, which crosslinks with g-C₃N₄ to form g-C₃N₄–CS beads.

Fig. 1 The crosslinking process of the g-C₃N₄–CS-x samples.

2.2.3 Regeneration of the g-C₃N₄–CS-x beads. The regeneration of the g-C₃N₄–CS-x beads was carried out under a visible-light irradiation reaction which has a 300 W Xe lamp with a 420 nm cut-off filter as the light source. After the photocatalytic measurements, the g-C₃N₄–CS-x beads were filtered and placed in a 0.5 mol L⁻¹ sodium hydroxide solution. A total of 5 μL H₂O₂ was added, and then, the mixed system was magnetically stirred in the light for 3 h. Finally, the beads were separated and washed several times with distilled water until the filtrate was measured to be approximately neutral pH, and the g-C₃N₄–CS-x beads were regenerated after air-drying at 60 °C for 5 h. This process was repeated five times to measure the stabilities of the beads.

2.3 Characterization

The thermogravimetric analysis (TGA) was conducted with a TGA Q500 (TA Instruments) under N₂ flow, from 50 to 780 °C, with a heating rate of 10 °C min⁻¹. The surface texture of the samples was investigated by emission scanning electron microscopy (SEM) (S-4800). The energy dispersive X-ray spectrum (EDX) was obtained from an attached Oxford Link ISIS energy-dispersive spectrometer fixed on a JEM-2010 electron microscope. The powder X-ray diffraction (XRD) of the g-C₃N₄–CS-x beads was measured by a Bruker D8 Advance diffractometer with Cu Kα1 monochromatized radiation (λ = 1.5406 Å). The FT-IR spectrum in the range of 4000–400 cm⁻¹ was recorded by using the KBr pellet technique. The diffuse reflectance spectroscopy (DRS) experiments were measured via a PerkinElmer spectrophotometer, which was equipped with an integrating sphere assembly with a nominal resolution of 1 nm using BaSO₄ as a reference. The photoluminescence (PL) spectrum was collected by a Hitachi F-7000 (Japan) spectrophotometer. The X-ray photoelectron spectroscopy (XPS) data were carried out on a Thermo ESCALAB250 instrument using a monochromatized Al Kα line source. Solid-state ¹³C Nuclear Magnetic Resonance (NMR) measurements were carried out on a Bruker Avance III spectrometer.

2.4 Photocatalytic test

The photocatalytic activity of the g-C₃N₄–CS-x beads was evaluated by the degree of degradation of the MB solution including a 100 mL MB solution (5 mg L⁻¹) and g-C₃N₄–CS-x (0.20 g). The reaction was performed in a photoreactor apparatus used to simulate sunlight (equipped with a 300 W xenon arc with a 420 nm cutoff filter) as a visible light source. The temperature of the reactant solution was kept at room temperature via a flow of cooling water. Before irradiation, the suspension was magnetically stirred in the dark for 60 min to achieve adsorption/desorption equilibrium. To quantify MB, a small sample (2.00 mL) of this suspension was taken out at given intervals and the absorbance of the removed filtrate was measured at 664 nm with an ultraviolet-visible spectrophotometer.

3. Results and discussion

3.1 Characterization of the g-C₃N₄–CS-x beads

Thermogravimetric analysis was measured to investigate the thermal stability of the g-C₃N₄–CS-x beads. Fig. 2 shows the TGA patterns of the g-C₃N₄, the CS and the g-C₃N₄–CS-3. The maximum rate of weight loss for g-C₃N₄ is observed at approximately 600–750 °C,⁴⁶ which is caused by the decomposition of carbon nitride.⁴⁷ The initial water loss for the CS is observed at near 260 °C, meaning the water adsorbed on the surface of the material was steamed. Due to the decomposition of CS, the hydroxyl and C–O bonds in the molecule are rapidly destroyed, and accompanied with a large loss of hydrogen, oxygen, and small amounts of carbon, the CS begins to decompose at 260 °C, the weight gradually decrease as the temperature increases, and there are no residual until 750 °C. The weight loss of the g-C₃N₄–CS-3 beads is observed before 260 °C, which can be attributed to the water loss, and the weight loss from 260 to 750 °C; is due to the CS decomposition; therefore, the g-C₃N₄ starts decomposing at 600 °C.

Fig. 2 TGA patterns of the g-C₃N₄, the CS and the g-C₃N₄–CS-3 samples.

Fig. 3 presents the XRD patterns of the g-C₃N₄, the CS and the g-C₃N₄–CS-3. It can be seen that the CS has two distinct diffraction peaks at 11° and 20°,⁴⁰ meaning the CS has a specific microcrystalline structure. For the g-C₃N₄, two characteristic peaks at 13.1° and 27.4° are observed,^24,25 corresponding to the (002) and (100) diffraction of the g-C₃N₄ structure. It is clear that the g-C₃N₄–CS-3 beads have two peaks at 20° and 27.4°, indicating that the g-C₃N₄ and CS were successfully assembled, among them, the diffraction peak at 2θ = 20° belongs to the microcrystalline structure of CS, and the diffraction peak at 2θ = 27.4° belongs to the inter-layer stacking of g-C₃N₄, corresponding to the (002) crystal plane.

Fig. 3 XRD patterns of the g-C₃N₄, the CS and the g-C₃N₄–CS-3 samples.

The texture of g-C₃N₄–CS-3 can be observed via SEM images presented in Fig. 4, and the yellow region in Fig. 4a is appropriately enlarged and shown in Fig. 4b. It is clear that the CS has a compact structure and smooth surface,⁴⁸ and many granules are observed (Fig. 4a) which is further determined to be the g-C₃N₄ (Fig. 4b). From Fig. 4b, it can be seen that the g-C₃N₄ displays a typical wrinkled morphology, with few mesopore caused by structural collapse.

Fig. 4 SEM images (a and b) of the g-C₃N₄–CS-3 samples, EDX elemental mapping (c) and EDX spectra of the rectangle region indicated by green border (upper) and red border (d) for g-C₃N₄–CS-3.

The elemental mapping indicated that the distribution of C, N, and O elements were relatively homogeneous in g-C₃N₄–CS-3 (Fig. 4c), breakpoints can be seen in C elemental mapping, and relatively concentrated points can be seen in N elemental mapping, which may be due to different C and N content of CS and g-C₃N₄. And the EDX measurement confirmed the co-existence of C, N, and O elements in g-C₃N₄–CS-3 (Fig. 4d). The granule is further determined to be the aggregation of g-C₃N₄ nanoparticles, based on the high N peak in the EDX spectrum at the spot.

Elemental analysis is an effective method to determine the ratio between g-C₃N₄ and CS. Table 1 shows that the content of N element in g-C₃N₄–CS-3 is increased, suggesting that g-C₃N₄ is successfully embedded in the CS matrix. Considering TGA, we can conclude that the ratio of g-C₃N₄ to CS is approximately 21.5 wt%, where the theoretical ratio of chitosan to g-C₃N₄ during the preparation was 25 wt%.

Table 1 Elemental analysis of the g-C₃N₄, the CS and the g-C₃N₄–CS-3 samples

Sample	C (wt%)	N (wt%)	H (wt%)	C/N (atomic ratio)
g-C3N4	34.43	61.43	1.78	0.65
CS	37.94	7.03	7.29	6.30
g-C3N4–CS-3	34.74	7.92	6.30	5.12

---

The FT-IR spectra of the g-C₃N₄, the CS and the g-C₃N₄–CS-3 samples are presented in Fig. 5. The characteristic peaks of CS and g-C₃N₄ were observed in the g-C₃N₄–CS-3 sample. For CS, typical strong absorption peaks of alcoholic hydroxyl groups C–O stretching vibration at 1024 and 1068 cm⁻¹ are observed.⁴⁰ The broadband from 3000 to 3600 cm⁻¹, corresponds to N–H breathing modes, which is likely to indicate the presence of surface N–H or the presence of NH₂. A series of peaks in the region from 1200 to 1600 cm⁻¹ were caused by aromatic CN heterocycle stretching vibration. Additionally, the typical bands at 810 cm⁻¹ originate from the breathing vibration of the heptazine units. Combined with XRD analysis, g-C₃N₄ has successfully combined with CS.

Fig. 5 FT-IR spectra of the g-C₃N₄, the CS and the g-C₃N₄–CS-3 samples.

DRS spectroscopy analyzed the optical absorption properties of the prepared samples. The light absorption edge of the g-C₃N₄ material is approximately 480 nm (Fig. 6); the absorption edge slightly shifts to 450 nm after doping CS, because CS has no light absorption. However, g-C₃N₄–CS-3 still has fair photocatalytic properties even though the absorption decreased.

Fig. 6 DRS spectra of the g-C₃N₄, the CS and the g-C₃N₄–CS-3 samples.

Photoluminescence emission spectra with an excitation wavelength at 320 nm were used to characterize the efficiency of the charge immigration and separation in the samples. The lower the PL density, the better the termination of the photogenerated charge carriers. As shown in Fig. 7, the CS has negligible PL density. The PL density of g-C₃N₄–CS-3 is appreciably lower than that of g-C₃N₄, it may be the incorporation of CS that leads to a decrease in the dispersion density of g-C₃N₄, resulting in a reduction in the density.

Fig. 7 PL spectra of the g-C₃N₄, the CS and the g-C₃N₄–CS-3 samples.

The chemical composition and chemical states of the g-C₃N₄–CS-3 samples were examined by XPS measurements. The survey spectra reveal that the elements of C, N and O exist in the samples. The element O may come from the CS and the absorbed H₂O and CO₂ on the surface of the g-C₃N₄–CS-3 (Fig. 8a). Three peaks were observed in the C 1s spectrum³⁷ (Fig. 8b); the peak at 288.1 eV was assigned to the sp² N–C=N in the N-containing aromatic ring, the 286.2 eV peak is identified as the sp²-hybridized carbon in the C–N–C of g-C₃N₄ and the C–O bonds of CS,⁴⁹ and according to the literature, the 284.6 eV is ascribed to the sp² C–C bonds. The N 1s spectra (Fig. 8c) can be separated into four peaks at 398.4 eV, 399.1 eV, 399.6 eV, and 400.6 eV corresponding to the sp² C–N=C bonds, C–NH–C bonds, bridge N in the N–(C)₃ groups, and the sp²-bonded N in the aromatic heterocyclic, respectively. The O 1s spectra (Fig. 8d) show two distinct peaks at approximately 530.8 eV and 532.6 eV. The 530.8 eV peak is identified as tertiary amine oxidative C=O or HN=C=O.⁵⁰ In addition, the peak at 532.6 eV originates from the water molecules adsorbed on the surface.

Fig. 8 The XPS of the g-C₃N₄–CS-3 samples for the (a) survey spectrum, (b) C 1s, (c) N 1s, and (d) O 1s.

Solid-state NMR spectroscopy is one of the most powerful techniques for addressing issues of molecular structure, especially for insoluble or poorly materials. In the ¹³C NMR spectra of g-C₃N₄–CS-3 (Fig. 9), two common signals centered at 154.2 and 162.3 ppm are assigned to CN₃ and CN₂NH₂ groups of g-C₃N₄ material, respectively. This indicated that the existence of g-C₃N₄ of the sample. And disappearance of the signal at 20.5, 53.3, 58.3, 71.5, 73.3, 80.1, and 101.2 ppm are assigned to CH₃, OC(CH₂)₂, CHNH₂, CO, CHOH, CHO–, OCHO– groups of chitosan–epichlorohydrin.

Fig. 9 Solid-state ¹³C NMR of the g-C₃N₄–CS-3 samples.

3.2 Photocatalytic activity

The photocatalytic activity of the samples was evaluated via the photodegradation of MB under visible light irradiation. Fig. 10 shows the photocatalytic results for MB degradation (start lighting at 0 min. The inset is the adsorption–desorption equilibrium curves showing that the systems get an adsorption–desorption equilibrium after 60 min in dark). The higher the CS mixing is, the higher the absorption; the adsorption effect covers the highlight of the photocatalytic performance when CS mixes too much (Fig. 10a). To balance the effects of adsorption and photocatalytic activity, we regard g-C₃N₄–CS-3 as the mixing ratio, which can appropriately improve the in situ adsorption, while maintaining the maximum impact of the photocatalytic performance.

Fig. 10 (a) Photocatalytic degradation of MB over the g-C₃N₄ and g-C₃N₄–CS-x photocatalysts under visible light irradiation, where the inset is the adsorption–desorption equilibrium curves in the dark; and (b) the corresponding first-order-kinetic plots, where the inset shows the corresponding rate constants. (c) The stability test of g-C₃N₄–CS-3 under visible light irradiation. (d) TOC removal of the g-C₃N₄ and g-C₃N₄–CS-x photocatalysts under visible light irradiation.

The corresponding first-order-kinetic curves for the MB photodegradation over the g-C₃N₄ and g-C₃N₄–CS-x photocatalysts are shown in Fig. 10b. The experimental data correlate with the first-order reaction kinetic equation:

ln(C₀/C) = kt (1)

where k is the apparent first-order rate constant (min⁻¹), t is the illumination time (min), C₀ is the initial concentration of the solution, and C is the concentration at reaction time t under illumination. The inset is the corresponding rate constants showing that the first-order reaction kinetic constant of the g-C₃N₄–CS-3 beads is higher than those of the others by a factor of 1.78 (Fig. 10b).

The stability of the optimized g-C₃N₄–CS-3 beads was tested for five consecutive cycles under illumination by visible light (Fig. 10c). The sample activities decreases slightly after five repeat reaction cycles under the same experimental conditions, which may be because after several photocatalytic degradations, the active sites of the catalyst were reduced. However, after five times of repeated use, the degradation rate can still reach approximately 97%, and the mass change does not exceed 10%. The above data show that the g-C₃N₄–CS-3 beads have high stability and can meet the needs of practical application.

The removal efficiency of TOC by g-C₃N₄ and g-C₃N₄–CS-x photocatalysts can reached about 40% (Fig. 10d). The removal rate of g-C₃N₄–CS-3 is relatively high could be attributed to the synergetic effect between the adsorption of CS and photocatalytic activity of g-C₃N₄.

To research the possible mechanism of the photolysis performance, we identify the active species by adding radical scavengers, with BQ as the ·O₂⁻ scavenger,⁵¹ IPA as the ·OH scavenger,³⁷ and TEOA as the hole (h⁺) scavenger,⁵² during the photocatalytic degradation process. As shown in Fig. 11, the photocatalytic activity of the g-C₃N₄–CS-3 beads substantially decreased when TEOA and BQ were added, which indicates that h⁺ and ·O₂⁻ are the main active species in the process, respectively. The photocatalytic activity decreased slightly when IPA was dispersed into the aqueous solution. Thus, the influence order of the activated species in the process of MB photodegradation is h⁺ > ·O₂⁻ > ·OH.

Fig. 11 Photodegradation rates of MB over the g-C₃N₄–CS-3 beads in the presence of different scavengers.

From the above, the possible mechanism of the g-C₃N₄–CS beads is shown in Fig. 12. Photo-generated electrons migrate from the valence band (VB) to the conduction band (CB) under visible light irradiation, the holes can oxidize dyes to products such as CO₂ and H₂O, and the electrons can reduce O₂ to ·O₂⁻ and further react to produce ·OH. At the same time, it is unknown whether electrons can be transferred from the conduction band (CB) to CS, which need further experiments to prove.

Fig. 12 The possible mechanism of the g-C₃N₄–CS beads under visible light.

4. Conclusions

In summary, we have successfully fabricated the regenerable g-C₃N₄–CS beads via the blend crosslinking method. The mass ratio of g-C₃N₄ to CS had an important influence on the photocatalytic activity toward in situ adsorption and decomposition MB under visible light irradiation. At the optimal g-C₃N₄ content, the as-prepared g-C₃N₄–CS-3 sample exhibited the highest photocatalytic performance, exceeding that of pure g-C₃N₄ by a factor of 1.78. The enhanced photocatalytic performance was ascribed to the synergistic effect between in situ adsorption and photocatalytic degradation. The active species trapping experimental results demonstrated that the superoxide radicals (·O₂⁻) and holes (h⁺) were the main active species in the decomposition of MB. In addition, the reacted g-C₃N₄–CS beads can be regenerated by merely adding sodium hydroxide and hydrogen peroxide with a mass loss of less than 10% in recycles. This work might provide new insight for the smart design and fabrication of highly efficient regenerated g-C₃N₄-based photocatalysts for environmental purification.

Conflicts of interest

The authors declare that there is no conflict of interest.

Acknowledgements

The authors are grateful for the financial support from the National Science and Technology Major Project (NO. 2016ZX05040003) and the Fundamental Research Funds for the Central Universities (NO. 17CX06027, NO. 18CX06068A).

References

1. Z. Dong and Y. Wu, J. Photochem. Photobiol., A, 2017, 336, 156–163.
2. P. Djinović and A. Pintar, Appl. Catal., B, 2017, 206, 675–682.
3. S. Ye, L.-G. Qiu, Y. P. Yuan, Y. J. Zhu, J. Xia and J. F. Zhu, J. Mater. Chem. A, 2013, 1, 3008.
4. D. Masih, Y. Ma and S. Rohani, Appl. Catal., B, 2017, 206, 556–588.
5. H. Lin, Y. Liu, J. Deng, S. Xie, X. Zhao, J. Yang, K. Zhang, Z. Han and H. Dai, J. Photochem. Photobiol., A, 2017, 336, 105–114.
6. Y. Liu, W. Jin, Y. Zhao, G. Zhang and W. Zhang, Appl. Catal., B, 2017, 206, 642–652.
7. H. Sun, G. Zhou, Y. Wang, A. Suvorova and S. Wang, ACS Appl. Mater. Interfaces, 2014, 6, 16745–16754.
8. P. Liang, C. Zhang, H. Sun, S. Liu, M. Tadé and S. Wang, RSC Adv., 2016, 6, 95903–95909.
9. D. Dvoranová, M. Mazúr, I. Papailias, T. Giannakopoulou, C. Trapalis and V. Brezová, Catalysts, 2018, 8, 47.
10. Z. Zhu, M. Murugananthan, J. Gu and Y. Zhang, Catalysts, 2018, 8, 112.
11. S. Halevy, E. Korin and A. Bettelheim, RSC Adv., 2016, 6, 87314–87318.
12. W. Zhang, Z. Zhang, S. Kwon, F. Zhang, B. Stephen, K. K. Kim, R. Jung, S. Kwon, K.-B. Chung and W. Yang, Appl. Catal., B, 2017, 206, 271–281.
13. H. Li, T. Hu, R. Zhang, J. Liu and W. Hou, Appl. Catal., B, 2016, 188, 313–323.
14. L. Zhang, X. He, X. Xu, C. Liu, Y. Duan, L. Hou, Q. Zhou, C. Ma, X. Yang, R. Liu, F. Yang, L. Cui, C. Xu and Y. Li, Appl. Catal., B, 2017, 203, 1–8.
15. Y. He, M. E. Ford, M. Zhu, Q. Liu, Z. Wu and I. E. Wachs, Appl. Catal., B, 2016, 188, 123–133.
16. Y. Cui, Q. Ma, X. Deng, Q. Meng, X. Cheng, M. Xie, X. Li, Q. Cheng and H. Liu, Appl. Catal., B, 2017, 206, 136–145.
17. Z. Bian, J. Zhu, F. Cao, Y. Huo, Y. Lu and H. Li, Chem. Commun., 2010, 46, 8451–8453.
18. S. Sun, J. Zhang, P. Gao, Y. Wang, X. Li, T. Wu, Y. Wang, Y. Chen and P. Yang, Appl. Catal., B, 2017, 206, 168–174.
19. X. H. Yang, Z. Li, C. Sun, H. G. Yang and C. Li, Chem. Mater., 2011, 23, 3486–3494.
20. S. Khan, H. Cho, D. Kim, S. S. Han, K. H. Lee, S. H. Cho, T. Song and H. Choi, Appl. Catal., B, 2017, 206, 520–530.
21. X. Bai, L. Wang, R. Zong and Y. Zhu, J. Phys. Chem. C, 2013, 117, 9952–9961.
22. J. Jiang, S. Cao, C. Hu and C. Chen, Chin. J. Catal., 2017, 38, 1981–1989.
23. J. Fu, J. Yu, C. Jiang and B. Cheng, Adv. Energy Mater., 2018, 8, 1701503.
24. X. Chen, Y.-S. Jun, K. Takanabe, K. Maeda, K. Domen, X. Fu, M. Antonietti and X. Wang, Chem. Mater., 2009, 21, 4093–4095.
25. X. Wang, X. Chen, A. Thomas, X. Fu and M. Antonietti, Adv. Mater., 2009, 21, 1609–1612.
26. L. Zhou, H. Zhang, X. Guo, H. Sun, S. Liu, M. O. Tade and S. Wang, J. Colloid Interface Sci., 2017, 493, 275–280.
27. K. C. Christoforidis, M. Melchionna, T. Montini, D. Papoulis, E. Stathatos, S. Zafeiratos, E. Kordouli and P. Fornasiero, RSC Adv., 2016, 6, 86617–86626.
28. A. H. Reshak, S. A. Khan and S. Auluck, RSC Adv., 2014, 4, 11967–11974.
29. A. Akhundi and A. Habibi-Yangjeh, RSC Adv., 2016, 6, 106572–106583.
30. Y. Zhao, X. Liang, Y. Wang, H. Shi, E. Liu, J. Fan and X. Hu, J. Colloid Interface Sci., 2018, 523, 7–17.
31. Z. Zhang, Y. Zhang, L. Lu, Y. Si, S. Zhang, Y. Chen, K. Dai, P. Duan, L. Duan and J. Liu, Appl. Surf. Sci., 2017, 391, 369–375.
32. S. Sahar, A. Zeb, Y. Liu, N. Ullah and A. Xu, Chin. J. Catal., 2017, 38, 2110–2119.
33. Y. Xu, H. Xu, L. Wang, J. Yan, H. Li, Y. Song, L. Huang and G. Cai, Dalton Trans., 2013, 42, 7604–7613.
34. J. Wang, Y. Xia, H. Zhao, G. Wang, L. Xiang, J. Xu and S. Komarneni, Appl. Catal., B, 2017, 206, 406–416.
35. E. Fayyad, A. Abdullah, M. Hassan, A. Mohamed, C. Wang, G. Jarjoura and Z. Farhat, Coatings, 2018, 8, 37.
36. S. A. Younis, A. Abd-Elaziz and A. I. Hashem, RSC Adv., 2016, 6, 89367–89379.
37. Y. Wang, H. Wang, F. Chen, F. Cao, X. Zhao, S. Meng and Y. Cui, Appl. Catal., B, 2017, 206, 417–425.
38. J. Fu, B. Zhu, C. Jiang, B. Cheng, W. You and J. Yu, Small, 2017, 13, 1603938.
39. L. Zhou, H. Zhang, H. Sun, S. Liu, M. O. Tade, S. Wang and W. Jin, Catal. Sci. Technol., 2016, 6, 7002–7023.
40. K. V. Harish Prashanth and R. N. Tharanathan, Carbohydr. Res., 2006, 341, 169–173.
41. L. Zhou, Z. Liu, J. Liu and Q. Huang, Desalination, 2010, 258, 41–47.
42. J. Berger, M. Reist, J. M. Mayer, O. Felt, N. A. Peppas and R. Gurny, Eur. J. Pharm. Biopharm., 2004, 57, 19–34.
43. P. Xu, X. Liang, N. Chen, J. Tang, W. Shao, Q. Gao and Z. Teng, J. Colloid Interface Sci., 2017, 507, 353–359.
44. P. A. Nishad, A. Bhaskarapillai and S. Velmurugan, Carbohydr. Polym., 2014, 108, 169–175.
45. L. Yang, L. Jiang, D. Hu, Q. Yan, Z. Wang, S. Li, C. Chen and Q. Xue, Carbohydr. Polym., 2016, 140, 433–441.
46. X. J. Wang, C. Liu, X. L. Li, F. T. Li, Y. P. Li, J. Zhao and R. H. Liu, Appl. Surf. Sci., 2017, 394, 340–350.
47. Y. Lu, C. Ji, Y. Li, R. Qu, C. Sun and Y. Zhang, Mater. Lett., 2018, 211, 78–81.
48. Z. Ji, Z. Shengtang, W. Guangpeng, W. Wenqiang, G. Sufang, Z. Ji and W. Yunpu, J. Bioact. Compat. Polym., 2016, 22, 429–441.
49. I. F. Amaral, P. L. Granja and M. A. Barbosa, J. Biomater. Sci., Polym. Ed., 2005, 16, 1575–1593.
50. R. S. Vieira, M. L. M. Oliveira, E. Guibal, E. Rodríguez-Castellón and M. M. Beppu, Colloids Surf., A, 2011, 374, 108–114.
51. B. Chai, J. Yan, C. Wang, Z. Ren and Y. Zhu, Appl. Surf. Sci., 2017, 391, 376–383.
52. B. Wang, J. Di, P. Zhang, J. Xia, S. Dai and H. Li, Appl. Catal., B, 2017, 206, 127–135.

---

Footnote

† These authors contributed equally to this work.

---