Xiaobing Yang†
a,
Juan Chen†b,
Huixian Laic,
Jiapeng Hua,
Ming Fangd and
Xuetao Luo*c
aCollege of Ecology and Resource Engineering, Wuyi University, Wuyishan, Fujian 354300, China
bDepartment of Pharmacy, Zhongshan Hospital, Xiamen University, Xiamen, 361004, China
cFujian Key Laboratory of Advanced Materials, College of Materials, Xiamen University, Xiamen, 361005, China. E-mail: xuetao@xmu.edu.cn
dShaoxing Testing Institute of Quality and Technical Supervision, Shaoxing, Zhejiang 312366, China
First published on 7th August 2017
Metal doped ZnO supported on porous materials have unique and highly attractive properties and have drawn worldwide attention. Herein, we explored the synthesis of Co/ZIF-8@silicalite-1 catalysts containing different amounts of Co by using metal organic frameworks as precursors and silicalite-1 as support. Then Co/ZIF-8@silicalite-1 catalysts were calcined at 550 °C to prepare the Co/ZnO@silicalite-1 photocatalysts. All samples were characterized by XRD, SEM, FT-IR and UV-vis diffuse reflectance spectra. The results show that the doping Co can increase the particle size of Co/ZIF-8 and decrease the content of Zn in Co/ZIF-8@silicalite-1. When ZnO@silicalite-1 is doped with Co from 0% to 20%, its band gap absorption is extended from 380 to 480 nm and a new band gap absorption at about 720 nm appears. The photocatalystic properties were investigated by degradation of rhodamine B (RhB) aqueous solution under UV light. In the series of Co/ZnO@silicalite-1 photocatalysts, 5Co/ZnO@silicalite-1 shows the highest photocatalytic efficiency.
With the rapid development of science and technology, it is important to improve the photocatalytic activity of ZnO. The photocatalytic activity of ZnO is influenced by many factors. The most important factors are band gap and electron–hole pair recombination. Both of them are all affected by the particle size of the catalyst. For example, the photocatalytic reactions mainly occurred at the interface between the catalyst surfaces and organic pollutants. When ZnO catalyst is nanosized, it can provide more surface area for the photocatalytic reactions. Further more, the recombination rate of the electron–hole pair in large size catalyst particles is higher than in small size catalyst particles.9 So many researchers have made effects to prepare the ZnO nanoparticles.10–13 However, the small size catalyst particles are easily to aggregate together and difficult to recycle. This restricts its application in photocatalytic systems. As a result, many researchers have made attempts to immobilize ZnO nanoparticles on solid support materials to overcome these drawbacks by synergistic effects, such as bentonite,14 graphene,15 glass,16 stainless steel,17 and zeolite.18
In the tradition research, the band gap of ZnO is 3.2 eV. It indicates that ZnO can be excited to produce the electron–hole pair to degrade the organic pollution under UV irradiation.19 In order to improve the photocatalytic activity of ZnO, many strategies have been used. For example, doping some ions on ZnO is a useful way, such as Cr,20 Fe,21 Cu,22 Ag,23 and Co.24 When ZnO is doped with some ions, it has more defects in its surface layer, which can make the photogenerated electrons and holes separation effectively.
In the last few years, metal–organic frameworks (MOFs), with the remarkable characteristics of well-defined channels and cavities of regular size and shape, have been widely studied and used for gas storage,25,26 small molecule separations,27 chemical catalysis and drug delivery,28,29 due to its large internal surface areas, uniform channels, (sub)nanometer-sized cavities, high thermal and chemical stability.30–34 Some attempts also have been made to explore the semiconducting properties of MOFs. Lin et al. suggested that band gaps of MOFs are between 1.0 and 5.5 eV.35,36 Bordiga et al. demonstrated that MOFs behave as metal oxide quantum dots.36 Zeolitic imidazolate frameworks (ZIFs), as a subfamily of metal organic frameworks (MOFs), are consisted of tetrahedral divalent metal ions (Zn2+, Co2+) and an imidazole derivative.37,38 For example, Zn2+ can be connected by imidazolate linkers to form ZIF-8.39 And ZIF-67 is consisted of Co and imidazole.40 Zn and Co also can be linked together by imidazolate linkers to form Co/ZIF-8.41 However, up to now, there are still no reports that use MOFs as precursor to synthesize the Co co-doped ZnO with porous structure by self-doping. In this work, we used silicalite-1 as support to load the Co/ZIF-8 nanoparticles on its surface and synthesized the Co/ZnO@silicalite-1 photocatalyst by sintering method. Then we explored the influence of Co doping for the photocatalyst and the performance of Co/ZnO@silicalite-1 by degradation rhodamine B solution.
Fig. 3 is the XRD patterns of Co/ZIF-8@silicalite-1 catalysts. Fig. 3a is the XRD patterns of silicalite-1. From Fig. 3a, it can be seen that silicalite-1 shows characteristic peaks at 7.98°, 8.82°, 13.24°, 13.94°, 14.80°, 15.53°, 15.89°, 17.79°, 20.37°, 23.18° and 24.46°, which are ascribable to (101), (020), (300), (012), (301), (202), (040), (031), (501), (151) and (303) reflections of silicalite-1 (JCPDS: 48-0136).43 Fig. 3b is the XRD patterns of ZIF-8@silicalite-1. It can be seen that ZIF-8@silicalite-1 exhibits new peaks at 14.73°, 16.45°, 18.06°, which are ascribable to (101), (002), (112) reflections of ZIF-8 (JCPDS: 89-3739).44 Fig. 3c and e–g are the XRD patterns of 2Co/ZIF-8@silicalite-1, 5Co/ZIF-8@silicalite-1, 10Co/ZIF-8@silicalite-1 and 20Co/ZIF-8@silicalite-1. It can be seen that the diffraction peaks of Co/ZIF-8@silicalite-1 doping with different amounts of Co are almost the same with ZIF-8@silicalite-1. It indicates that the doping of Co does not change the diffraction peaks of ZIF-8@silicalite-1.
Fig. 3 XRD patterns of (a) silicalite-1, (b) ZIF-8@silicalite-1, (c) 2Co/ZIF-8@silicalite-1, (d) 5Co/ZIF-8@silicalite-1, (e) 10Co/ZIF-8@silicalite-1, (f) 20Co/ZIF-8@silicalite-1. |
Fig. 4 is the SEM images of ZIF-8@silicalite-1 and Co/ZIF-8@silicalite-1 samples. Fig. 4a is the SEM images of ZIF-8@silicalite-1. It can be seen that the surface of silicalite-1 is coated with a layer of ZIF-8 nanoparticles. The high magnification image (Fig. 4b) shows that ZIF-8 nanoparticles, with a narrow size distribution ranging from 100 to 200 nm, are clearly observed. Fig. 4c–f are the high magnification images of 2Co/ZIF-8@silicalite-1, 5Co/ZIF-8@silicalite-1, 10Co/ZIF-8@silicalite-1, 20Co/ZIF-8@silicalite-1. It can be seen that when the content of Co increase from 0 to 20, the average size of Co/ZIF-8 increases from 150 to 350 nm. It is in accordance with Zaręba's report.41
Fig. 4 SEM images of (a and b)ZIF-8@silicalite-1, (c) 2Co/ZIF-8@silicalite-1, (d) 5Co/ZIF-8@silicalite-1, (e) 10Co/ZIF-8@silicalite-1, (f) 20Co/ZIF-8@silicalite-1. |
To obtain the Co/ZnO@silicalite-1 photocatalyst, Co/ZIF-8@silicalite-1 catalysts were calcined at 550 °C for 2 h under air atmosphere to remove the organic ligand. Fig. 5 is the XRD patterns of Co/ZnO@silicalite-1 photocatalysts. Fig. 5a is the XRD patterns of silicalite-1, which is the same with Fig. 3a. From Fig. 5, it can be seen that after sintering, the characteristic peaks of ZIF-8 is disappeared. There are appeared new peaks at 31.77°, 34.42°, 36.25°, 47.54° and 56.59°, which are corresponding to (100), (002), (101), (103) and (110) reflections of ZnO (JCPDS: 36-1451). It indicates that ZIF-8 transforms into ZnO. Fig. 5c–f are the XRD patterns of ZnO@silicalite-1 coated with different amounts of Co. The doping of Co does not change the characteristic peaks of ZnO@silicalite-1.
Fig. 5 XRD patterns of (a) silicalite-1, (b) ZnO@silicalite-1, (c) 2Co/ZnO@silicalite-1, (d) 5Co/ZnO@silicalite-1, (e) 10Co/ZnO@silicalite-1, (f) 20Co/ZnO@silicalite-1. |
Fig. 6 is the magnification of XRD patterns of Co/ZnO@silicalite-1 photocatalysts from 30° to 40°. It can be found that there are existing distinct characteristic peaks at 31.77°, 34.42°, 36.25° for ZnO@silicalite-1 (Fig. 6b). When ZnO@silicalite-1 is doped with 2% Co, the intensity of characteristic peaks of ZnO is weakened (Fig. 6c). With the increase of Co from 2% to 20%, the intensity of characteristic peaks of ZnO decreases more. It indicates that the doping of Co can restrain the formation of ZIF-8.
Fig. 6 XRD patterns of (a) silicalite-1, (b) ZnO@silicalite-1, (c) 2Co/ZnO@silicalite-1, (d) 5Co/ZnO@silicalite-1, (e) 10Co/ZnO@silicalite-1, (f) 20Co/ZnO@silicalite-1. |
Fig. 7 is the SEM images of Co/ZnO@silicalite-1 photocatalysts. After sintering, the structure of Co/ZIF-8 collapses. Fig. 7a and b are the SEM images of ZnO@silicalite-1. It can be seen that the collapsed nanoparticles are about 50 nm and adheres on the surface of silicalite-1. With the increase of Co content, the collapsed nanoparticles increase too (from 50 to 200 nm). From Fig. 5 and 6, we can know that the nanoparticles are ZnO.
Fig. 7 SEM images of (a and b) ZnO@silicalie-1, (c) 2Co/ZnO@silicalite-1, (d) 5Co/ZnO@silicalite-1, (e) 10 Co/ZnO@silicalite-1, (f) 20Co/ZnO@silicalite-1. |
The FT-IR spectra of silicalite-1, ZnO@silicalite-1 and Co/ZnO@silicalite-1 with different Co content are shown in Fig. 8. From Fig. 8, it can be found that all samples show vibration bands at around 446, 546, 781 and 1103 cm−1. The band at 446 cm−1 is attributed to the bending vibration of SiO4.38 The band at 546 cm−1 is corresponding to the five membered ring of pentasil zeolite structure.23 The bands at 1103 cm−1 and 781 cm−1 are attributed to the internal asymmetric stretching vibration and external symmetric stretching of Si–O–Si bonds, respectively. Fig. 8b is the FT-IR spectra of ZnO@silicalite-1. It can be seen that there is existing new band at 459 cm−1, which is belonged to the characteristic band of ZnO. When ZnO@silialite-1 is doped with Co, it shows band at 671 cm−1 (Fig. 8c), which is ascribed to Co–O bond.
Fig. 8 FT-IR spectra of (a) silicalite-1, (b) ZnO@silicalite-1, (c) 2Co/ZnO@silicalite-1, (d) 5Co/ZnO@silicalite-1, (e) 10Co/ZnO@silicalite-1, (f) 20Co/ZnO@silicalite-1. |
Fig. 9 shows the UV-vis diffuse reflectance spectra for silicalite-1, ZnO@silicalite-1, Co/ZnO@silicalite-1 containing different amounts of Co. Silicalite-1 just shows a weak band gap absorption at about 300 nm. When silicalite-1 is coated with ZnO nanoparticles, it shows a highly band gap absorption at about 388 nm. When ZnO@silicalite-1 is doped with 2% Co, its band gap absorption is extended into the visible light region (about 450 nm). With the increase of the doping Co content, the band gap absorption of Co/ZnO@silicalite-1 is extended from 450 to 480 nm. Except that, there is appearing new band gap absorption at about 720 nm. It is probably due to the Zn1−xCoxO solid solution formation.45,46 And the intensity of the band gap absorption also enhances with the increase of Co content.
To evaluate the photocatalytic activity of all samples, degradation rhodamine (RhB) was chosen as a model organic dye. The results of RhB removal by photocatalysts are shown in Fig. 10a. In order to eliminate the influence of the adsorption of photocatalyst, all experiments were operated in the dark for 30 min. After 30 min of continuously stirring in the dark, 20.4% of RhB is absorbed by silicate-1. It is attributed the high specific surface area of silicalite-1.47 When silicalite-1 is coated with ZnO nanoparticles, it can absorb 14.8% of RhB. The decrease of its adsorptive property is attributed to the decrease of silicalite-1. With the increase of Co content, the adsorptive property of photocatalyst decreases. It may be caused by the increase of Co in the photocatalyst and the decrease of silicalite-1 content. Then, all experiments are exposed under the UV light. After 90 min of UV light irradiation, only 5.4% of RhB is degraded by itself under the UV light. And silicalite-1 just degrades 6.3% of RhB. Silicalite-1 almost has not photocatalytic properties. When silicalite-1 is coated with ZnO nanoparticles, it can degrade 98.5% of RhB. When ZnO@silicalite-1 is doped with 2% Co, it can degrade 90.7% of RhB. The photocatalytic efficiency of 2Co/ZnO@silicalite-1 is lower than the photocatalytic efficiency of ZnO@silicalite-1, which is attributed to the significantly decreasing ZnO (Fig. 6). However, the photocatalytic efficiency of 5Co/ZnO@silicalite-1 is 94.0%, and then the photocatalytic efficiency of Co/ZnO@silicalite-1 decreases with the increase of Co content. It indicates the doping Co can enhance the photocatalytic efficiency of ZnO@silicalite-1 and 5Co/ZnO@silicalite-1 has the highest photocatalytic efficiency. The photocatalysis degradation kinetic reaction can be described by ln(C0/C) = kt, where k is a pseudo-first-rate kinetic constant and t is the irradiation time. And the results are shown in Fig. 10b. The calculated k value for 2Co/ZnO@silicalite-1, 10Co/ZnO@silicalite-1 and 20Co/ZnO@silicalite-1 is 0.023, 0.02 and 0.016 min−1, respectively. For 5Co/ZnO@silicalite-1, the calculated k value is 0.027 min−1, which is higher than 2Co/ZnO@silicalite-1, 10Co/ZnO@silicalite-1 and 20Co/ZnO@silicalite-1. The photocatalysis activity of the composite is improved greatly. For ZnO@silicalite-1, its k value is 0.05 min−1, which it higher than Co doped ZnO@silicalite-1. It is attributed to the decrease of ZnO content.
In order to investigate the stability of photocatalytic performance of Co/ZnO@silicalite-1, 5Co/ZnO@silicalite-1 was chosen to degrade RhB in four repeated cycles. The results are shown in Fig. 11. It can be found that after four photocatalytic degradation cycles, the photocatalytic efficiency of 5Co/ZnO@silicalite-1 reduces only by 3.6%. It may be the reason that 5Co/ZnO nanoparticles adhere tightly on the surface of silicalite-1.
Fig. 11 Four photocatalytic degradation cycles of RhB using 5Co/ZnO@silicalite-1 photocatalyst under UV light. |
Fig. 12 shows the schematic illustration of Co/ZnO@silicalite-1 for RhB degradation under UV light. Co/ZnO nanoparticles adhere tightly on the surface of silicalite-1. Silicalite-1 has high surface area. It can absorb RhB molecules on its surface or pores. When Co/ZnO is irradiated by UV light, it can generate electrons and holes. O2 can capture electrons from the conduction band to form O2−. And OH− can lose its electrons from the valence band to form ˙OH. The generated O2− and ˙OH can convert RhB molecules into CO2, H2O and so on.
Footnote |
† Xiaobing Yang and Juan Chen contributed equally to this work. They are co-first authors. |
This journal is © The Royal Society of Chemistry 2017 |