Tiefeng Xu,
Fei Wu,
Yan Gu,
Yi Chen,
Jinting Cai,
Wangyang Lu*,
Hongguang Hu,
Zhexin Zhu and
Wenxing Chen*
National Engineering Lab for Textile Fiber Materials & Processing Technology (Zhejiang), Zhejiang Sci-Tech University, Hangzhou 310018, China. E-mail: luwy@zstu.edu.cn; wxchen@zstu.edu.cn
First published on 30th September 2015
A visible-light responsive photocatalyst, polyacrylonitrile-dispersed graphitic carbon nitride nanofibers (g-C3N4/PAN nanofibers), was synthesized by electrospinning. The g-C3N4 is dispersed uniformly in the nanofibers, which helps it overcome the defects of easy aggregation and difficult recycling of powder catalysts. The model substrate, rhodamine B (RhB), could be adsorbed rapidly into the PAN nanofibers and decomposed efficiently in situ simultaneously in the presence of the g-C3N4 over a wide pH range under visible light irradiation. As a fibrous catalyst, the g-C3N4/PAN nanofibers were quite simple to recycle, and the catalytic activity maintained a high level without obvious decline after being reused several times. In addition, based on the intermediates detected by ultra performance liquid chromatography-mass spectrometry and gas chromatography-mass spectrometry, N-de-ethylation chromophore cleavage and ring-opening mineralization are the main processes in RhB degradation. Finally, a possible mechanism was proposed, in which the hole along with the superoxide radical mainly contribute to the oxidative degradation of RhB.
With the continuous research in semiconductor photocatalytic materials, the excellent performance of graphitic carbon nitride (g-C3N4) in many aspects has been discovered, such as thermal stability, chemical stability,6,7 and optical and photoelectrochemical properties.8–12 As an inorganic semiconductor material, compared to TiO2, g-C3N4 has a more suitable band gap (the former is 3.2 eV, the latter is 2.7 eV), which makes it absorb visible light better.5 Because of the outstanding performance described above, g-C3N4 has received much attention.13,14 Great progress in g-C3N4 has been made in the research in photochemical water splitting,15,16 bio-imaging,17–19 catalytic oxidation,20,21 catalytic hydrogenation,22,23 catalytic NO decomposition,24 photocatalytic degradation1,25,26 and as a basic catalyst.27 The photocatalytic degradation of pollutants into water, carbon dioxide and some other environmentally harmless small molecule compounds is an effective energy-saving and environmentally friendly method. Some common organic pollutants have been degraded by using modified g-C3N4 with enhanced photocatalyst properties.28 As a metal-free photocatalyst, g-C3N4 has also been used to activate H2O2 to produce hydroxyl radical (˙OH) to degrade organic pollutants under visible light irradiation.29 However, there are still some drawbacks of g-C3N4 which limit its future application: (a) poor dispersion in water and easy aggregation, which lead to a high photogenerated electron–hole recombination rate, and (b) difficulty in recycling, as a powder material.5 Researchers have made many attempts to improve their performance. For example, synthesising heterojunction composite photocatalysts is a good process to enhance the separation efficiency of photoexcited electron–hole pairs,30–32 and doping with some other elements or magnetic oxides could improve its performance and make it easy to recycle.5,33–35 Electrospinning has been recognized as an efficient technique for the preparation of polymer nanofibers. PAN-based electrospun nanofibers are hydrophobic with a too-low density, which makes them float easily on liquid. This point would maximize the photodegradation ability and optimize the irradiation efficiency of light by avoiding the hindrance of light source penetration.36,37
In this work, we report a novel visible-light responsive photocatalyst, polyacrylonitrile-dispersed graphitic carbon nitride nanofibers (g-C3N4/PAN nanofibers), which was prepared by electrospinning. This simple and effective method could overcome the aggregation of g-C3N4 in water, and recycle g-C3N4 from the reaction system. In addition, rhodamine B (RhB) was selected as a probe compound to evaluate the catalytic activity of g-C3N4/PAN nanofibers. The PAN nanofibers can adsorb RhB quickly and degrade it into small molecules in the presence of g-C3N4 in a wide pH range under visible light irradiation. By observing the effect of the active species capture agents on the degradation efficiency, and detecting the degradation products by ultra-performance liquid chromatography/high-definition mass spectrometry (UPLC Synapt G2-S HDMS) and gas chromatography/mass spectrometry (GC-MS), the possible mechanism and degradation process of RhB were proposed in the presence of g-C3N4/PAN nanofibers under visible light irradiation.
Fig. 1 TEM images of bulk g-C3N4 (a), the dispersion of g-C3N4 in PAN polymer solution (b), and 10% g-C3N4/PAN nanofibers (c). |
The X-ray diffraction (XRD) patterns for the as-prepared bulk g-C3N4, pure PAN nanofibers and g-C3N4/PAN nanofibers are shown in Fig. 2. The results indicated that the two peaks for g-C3N4 are in good agreement with the previous report.9,39 The strong XRD peak at 27.5° is indicative of interlayer stacking of the conjugated aromatic segment with distance of 0.324 nm, and the peak at 13.1° corresponds to in-plane ordering of tri-s-triazine units with an in-planar repeat period of 0.682 nm.22 For the pure PAN nanofibers, a strong diffraction peak centered at 17.1° and a weak diffusion diffraction peak centered at 27.9° were found, almost the same as in the previous report.40 The sharp diffraction peaks at 27.7° and 17.1° were observed in the g-C3N4/PAN nanofibers, however the intensities of the diffraction peaks were decreased due to the coupling with PAN. The surface area of the as-prepared samples was 12.07 m2 g−1 (Fig. S2†).
The FTIR spectra of the bulk g-C3N4, pure PAN nanofibers and g-C3N4/PAN nanofibers are shown in Fig. 3. For the pure PAN nanofibers, the peaks at 2928 and 2240 cm−1 are ascribed to stretch vibration of –CH2 and CN, and the characteristic peak at 1448 cm−1 corresponds to the bend vibration of –CH2. As can be seen in the FTIR spectra, the original g-C3N4 exhibits characteristic absorption peaks similar to those in the previous literature.15 The intense band at 810 cm−1 is ascribed to the breathing mode of triazine rings.41 The bands at 1323 and 1250 cm−1 are the stretching vibration of C–N–C. The peak at 1635 cm−1 can be ascribed to the C–N stretching vibration mode, while the four strong peaks at 1416, 1460, and 1570 cm−1 correspond to the CN heterocycle stretching of g-C3N4. Additionally, the 3000–3600 cm−1 broad band can be attributed to the stretching vibration of terminal –NH groups at the defect sites of the aromatic ring. For the g-C3N4/PAN nanofibers, the characteristic peaks of g-C3N4 could be seen clearly.
The optical absorption properties of the as-prepared samples were studied using UV-vis diffuse reflectance spectra. As can be seen from Fig. 4, the UV-vis diffuse reflectance spectrum of g-C3N4 shows typical semiconductor optical characteristics. The absorption edge of the g-C3N4 is around 450 nm, which signifies that the g-C3N4 can absorb visible light with a wavelength shorter than 450 nm. In comparison with the pure g-C3N4, the absorption edge of the g-C3N4/PAN nanofibers has no significant change.
Fig. 4 UV-vis diffuse reflectance absorption spectra of PAN nanofibers, bulk g-C3N4, and 10% g-C3N4/PAN nanofibers. |
Fig. 5 shows the TGA curves of the bulk g-C3N4, pure PAN nanofibers and g-C3N4/PAN nanofibers. It can be seen that there was no weight loss for the g-C3N4 until 500 °C and no residue of the sample left when heated to 750 °C, indicating that the as-prepared g-C3N4 shows high thermal stability. The pure PAN nanofibers have one significant weight loss at 300 °C, which is attributed to the release of small molecular gases such as NH3 and HCN. The PAN nanofibers started to degrade quickly above 300 °C, generating many gaseous compounds (e.g. HCN, CO and CO2) from the cyclization reaction. Due to the low quantity of g-C3N4 in the nanofibers, the thermal stability of the g-C3N4/PAN nanofibers presents a slight increase compared with the pure PAN nanofibers.
Fig. 5 Thermogravimetric analysis for bulk g-C3N4, PAN nanofibers, and 10% g-C3N4/PAN nanofibers in air. |
Fig. 6 The adsorption and photocatalytic degradation of RhB under visible light irradiation (λ > 400 nm). (RhB: 2 × 10−5 M, pH 5.6, 10% g-C3N4/PAN: 1 g L−1). |
The pH value determines the surface charge properties of the photocatalyst, and influences the photocatalytic degradation and the adsorption behavior of pollutants. Fig. 7 displays the effect of pH on the adsorption and photocatalytic degradation of RhB over the g-C3N4/PAN nanofibers. Here, both the adsorption and photocatalytic degradation rate of RhB increased gradually with the decrease of pH. In Fig. 7b, the degradation rate of RhB was beyond 90% within 40 minutes at pH 3. Under neutral conditions, RhB can also be degraded completely within 3 h. However, in an alkaline environment, the degradation efficiency of RhB decreased significantly and the degradation rate of RhB was only 60% after 3 h. Therefore, the adsorption of RhB on the photocatalytic nanofibers played an important role in the photocatalytic degradation of RhB. In addition, the positive holes are more easily produced and are considered as the major oxidation species at low pH. In order to determine the optimal amount of g-C3N4, we studied the influence of the amount of g-C3N4 in the g-C3N4/PAN nanofibers on the photocatalytic activity under the same conditions. As can be seen in Fig. S5,† 10% g-C3N4/PAN nanofibers performed the highest photocatalytic activity. This is because the catalyst became aggregated more easily with the increasing amount of g-C3N4 (Fig. S1†).
Fig. 7 Effect of pH on adsorption of RhB in the dark (a) and photocatalytic degradation of RhB under visible light irradiation (λ > 400 nm) (b). (RhB: 2 × 10−5 M, 10% g-C3N4/PAN: 1 g L−1). |
To further investigate the reusability of the g-C3N4/PAN nanofibers, five repeated experiments of RhB degradation were carried out under the same conditions. Each cycle the fibers were separated from the solution, rinsed with ultrapure water and then dried at 70 °C. The results of the cycling experiments for the degradation of RhB driven by visible light irradiation are shown in Fig. 8, and there was no evident deactivation of the photocatalyst after five cycles. Therefore, it can be speculated that the g-C3N4/PAN nanofibers exhibited excellent self-reliant regeneration and recycling activity, indicating that the polyacrylonitrile-dispersed g-C3N4 nanofibers have broad application prospects for removing organic pollutants.
Fig. 8 The cyclic photocatalytic degradation of RhB under visible light irradiation (λ > 400 nm). (RhB: 2 × 10−5 M, pH 5.6, 10% g-C3N4/PAN: 1 g L−1). |
In view of the practical application of the g-C3N4/PAN nanofibers in the degradation of organic pollutants, direct sunlight was used to degrade RhB. The g-C3N4/PAN nanofiber membrane can float on the liquid surface, however, the majority of the g-C3N4 powder settled to the bottom of the solution in the case of no stirring (Fig. S7†). As shown in Fig. 9, no photocatalytic degradation of RhB under solar irradiation was observed in the absence of catalysts, and the removal rate of RhB based on the g-C3N4/PAN nanofibers was higher than that of the pure g-C3N4 after irradiation for 7 h in the sunlight. As for the g-C3N4 powder, the agglomeration and subsidence of the powder catalyst led to decreased photocatalytic performance. In addition, the dye solution can also absorb a part of sunlight, resulting in a reduced intensity of light on the bottom of the catalyst. Compared with the g-C3N4 powder, the g-C3N4/PAN nanofiber membrane floated on the surface of the solution to overcome the disadvantageous agglomeration and subsidence, resulting in significantly enhanced photocatalytic activity.
Fig. 9 Removal of RhB under solar irradiation. (Average sunlight intensity: 46760 lux, RhB: 2 × 10−5 M, g-C3N4: 0.03 g L−1, 10% g-C3N4/PAN: 0.3 g L−1). |
To further identify the intermediates, the GC-MS technique was employed to detect the small molecule products. Five aliphatic acids were identified, including oxalic acid, malonic acid, maleic acid, succinic acid and malic acid (Table S3†). In general, it was inferred that N-de-ethylation chromophore cleavage and ring-opening mineralization were the main processes in RhB degradation. Based on the results of the UPLC Synapt G2-S HDMS and GC-MS, the possible degradation pathway of RhB is shown in Fig. 11. It’s very similar to the previous reports.43,44
Fig. 11 Photocatalytic degradation pathway of RhB over g-C3N4/PAN nanofibers (λ > 400 nm). (RhB: 2 × 10−5 M, pH 5.6, 10% g-C3N4/PAN: 1 g L−1). |
In order to evaluate the degradation mechanism of RhB over g-C3N4/PAN nanofibers, different active species capture agents such as isopropanol (IPA) benzoquinone (BQ)45,46 and potassium iodide (KI)25 were added to observe the effect on the degradation efficiency of dyes, as shown in Fig. 12. When adding IPA as a hydroxyl radical (˙OH) capture agent, the photodegradation efficiency only declined slightly, implying ˙OH was not the major active species in this photocatalytic degradation process. Nevertheless, when BQ as a superoxide radical (˙O2−) capture agent was added or N2 was continuously bubbled into the reaction solution, the photodegradation activity of RhB declined obviously, suggesting that ˙O2− plays a key role in this photodegradation system. With the addition of the hole (h+) scavenger KI, a significant inhibition effect of photocatalytic activity was observed, which confirms the important role of h+ in the photodegradation process.
Fig. 12 Effect of trapping agents on photocatalytic degradation of RhB under visible light irradiation (λ > 400 nm). (RhB: 2 × 10−5 M, pH 5.6, 10% g-C3N4/PAN: 1 g L−1). |
In addition, EPR was utilized to probe the possible reactive oxygen species generated in the catalytic reaction, and 5,5-dimethylpyrroline-oxide (DMPO) was used as the spin trapping reagent. In aqueous solution (Fig. S8a†), DMPO–˙OH signals were detected weakly in the systems, which means that ˙OH is not the major active species with the g-C3N4/PAN nanofibers. In methanol solution (Fig. S8b†), the g-C3N4/PAN nanofibers can produce the active species of ˙O2−, which reveals that ˙O2− plays an important role in the g-C3N4/PAN nanofiber photocatalytic system. The results were consistent with the above capture experiment results. Thus, it can be concluded that the photogenerated h+ and ˙O2− played a dominant part in the degradation process. It is quite different from the photocatalytic system of Yan and coworkers’.47 In their boron-doped g-C3N4 catalytic system under visible light irradiation, the degradation of methyl orange (MO) originated mainly from the photogenerated electrons, whereas the photodegradation of RhB is attributed mainly to the photogenerated holes. In our catalytic system, the central carbon of RhB could be attacked by h+ and ˙O2− to decolorize the dye and further degraded via an N-de-ethylation process. The N-de-ethylation intermediates DER, EER, DR, ER and R could be further carboxylated into aromatic acids with m/z values of 165 and 121. Finally, these aromatic compounds can be degraded into small molecular biodegradable acids by ring-opening.
Footnote |
† Electronic supplementary information (ESI) available: The synthesis of g-C3N4, g-C3N4/PAN nanofibers, spectral changes during photocatalytic degradation of RhB under visible light irradiation, the photograph of photocatalytic degradation of RhB by g-C3N4 powder and g-C3N4/PAN nanofibers under solar irradiation, the detailed condition for UPLC Synapt G2-S HDMS and GC-MS. See DOI: 10.1039/c5ra15973c |
This journal is © The Royal Society of Chemistry 2015 |