Flower-like N-doped MoS2 for photocatalytic degradation of RhB by visible light irradiation

In this paper, the photocatalytic performance and reusability of N-doped MoS2 nanoflowers with the specific surface area of 114.2 m2 g−1 was evaluated by discoloring of RhB under visible light irradiation. Results indicated that the 20 mg fabricated catalyst could completely degrade 50 ml of 30 mg l−1 RhB in 70 min with excellent recycling and structural stability. The optimized N-doped MoS2 nanoflowers showed a reaction rate constant (k) as high as 0.06928 min−1 which was 26.4 times that of bare MoS2 nanosheets (k = 0.00262). In addition, it was about seven times that of P25 (k = 0.01) (Hou et al 2015 Sci. Rep. 5 15228). The obtained outstanding photocatalytic performance of N-doped MoS2 nanoflowers provides potential applications in water pollution treatment, as well as other related fields.


Introduction
Persistent organic pollutants in underground water have already become serious problems that influence the survival the environment of human beings and other creatures. Effluents from the textile industries are important sources of water pollution, because dyes in wastewater undergo chemical as well as biological changes, consume dissolved oxygen, destroy aquatic life and endanger human health. It is necessary to disintegrate textile effluents into the receiving-water standard [2][3][4][5][6][7]. Hence, in recent years, numerous investigations have been devoted to developing products to address the challenges of water pollution, such as TiO 2 [8], ZnO [9], N-TiO 2 @g-C 3 N 4 [10], N-doped ZnO@g-C 3 N 4 [11], TiO 2 hollow fibers [1], etc.
An alternative is to search for novel highly efficient photocatalysts with simple phase structure and synthesis process. In this paper, we report the excellent photocatalytic activities of N-doped MoS 2 nanoflowers in degrading the organic dye of RhB, synthesized by a simple sol-gel method. The fabricated N-doped MoS 2 nanoflowers have a high surface area of 114.2 m 2 g −1 , and possess a high adsorption property of small concentrations of RhB (see supplementary data S1 (stacks.iop.org/NANO/27/225403/mmedia)), as well as excellent photocatalytic activity in degrading 30 mg l −1 RhB. The outstanding photocatalytic activities of as-prepared flower-like N-doped MoS 2 samples could also be extended to degrade other organic dyes and heavy metal pollutants, providing potential applications in future water pollution treatment.

Experiment
N-doped MoS 2 nanoflowers were synthesized by an optimized sol-gel method as previously reported [29]. In brief, 2 g thiourea was mixed with 0.5 g MoCl 5 by dropwise addition of alcohol. Then the brown gel-like precursor powders were formed after drying. Next, the precursor powders were transferred into a quartz boat and heated in a tube furnace for 2 h under 0.1 L min −1 argon flow at 550°C. To get its bulk form, we changed the annealing temperature to 1050°C.
To compare, pure MoS 2 nanosheets were prepared by the hydrothermal method, where 1 mmol ammonium molybdate tetrahydrate and 30 mmol thiourea were dissolved in 40 ml deionized water under magnetic stirring. Then, the solution was transferred to a 50 ml reaction still and maintained at 200°C for 20 h before being cooled down in air.
α-Fe 2 O 3 @N-doped MoS 2 heterostructures were synthesized by the hydrothermal method, where 90 mg N-doped MoS 2 was dissolved in 32 ml deionized water. Then, 0.202 g Fe(NO 3 ) 3 · 9H 2 O and 0.3 g CO(NH 2 ) 2 were dissolved in the above solution under magnetic stirring. After that, 0.006 g sodium dodecyl benzenesulphonate (SDBS) was added into the above solution and continuously stirred in a water bath of 60°C for 30 min. Finally, the solution was transferred to a 40 ml reactor and maintained at 90°C for 12 h before being cooled down in air.
The crystal structure of the samples was measured by x-ray diffractometry (XRD) in a Philips/X, Pert PRO diffractometer with Cu Ka radiation. A scanning electron microscope (SEM, Hitachi S-4800) and high resolution transmission electron microscope (HRTEM, TecnaiTM G2 F30, FEI, USA) were used to observe the morphology and structure of the samples. In addition, x-ray photoelectron spectroscopy (XPS, VG Scientific ESCALAB-210) was employed to study the chemical nature of N, Mo, and S with Al Ka x-ray. The Brunauer-Emmett-Teller (BET) surface area and pore width were measured using a Micrometrics ASAP 2020 V403. Meanwhile, Raman spectra were measured at room temperature using a Jobin-Yvon HR 800 spectrometer.
The photocatalytic activity of the samples was measured by degradation of RhB with a 175 W halogen lamp. 50 ml RhB (30 mg l −1 ) was placed in a glass. Meanwhile, 20 mg photocatalyst was added under constant stirring. Photocalytic activity of the samples was evaluated under visible light irradiation. At certain time intervals, 4 ml solution was taken out, where the photocatalyst was removed by a centrifugal machine. Then, the filtrates were analyzed by recording variations of the absorption band maximum (553 nm) in the UVvis spectra of RhB. In addition, the recyclability of the samples was also investigated.

Characterization
As shown in figure 1(a), the obtained products and the used sample (N-doped MoS 2 nanoflowers were used for photocatalytic activity testing) were measured by XRD. Results indicate that all the diffraction peaks can be indexed as hexagonal MoS 2 . For the used sample, the characteristic peaks were similar to primitive products, indicating our sample has a stable structure in the photocatalytic process, which is also confirmed by the further Raman study as illustrated in figure 1(b). As can be seen, the two distinct peaks located around 378 cm −1 and 402 cm −1 correspond to the MoS 2 characteristic signature, associated with in-plane E 1 2g (the inplane displacement of Mo and S atoms) and out-of-plane A 1g (out-of-plane symmetric displacements of S atoms along the c-axis) Raman mode, respectively [12,30]. SEM and TEM measurements were employed to study the morphology of the products. As shown in figure 1(c), the fabricated sample has a flower-like structure and each of the components shows nanosheet features. As illustrated in figures 1(d) and (e), the TEM images also show the nanoflower-structure of the product, which is consist with the SEM results. Meanwhile, the results also reveal the typical structure of the nanosheets, containing 3-5 layers from the curly edges. Energy-dispersive x-ray spectroscopy (EDS) mapping was carried out to verify the element distribution. It clearly shows the presence of elements Mo, S, and N in the product, and the N element was evenly distributed in the sample.
The XPS spectrum was employed to examine the surface electronic state and composition of the flower-like N-doped MoS 2 . The whole XPS spectrum further indicates the sample contains N, S, and Mo elements, as shown in figure 2(a), in agreement with the EDS mapping results. In addition, figure 2(b) shows a high-resolution spectrum in the binding energy range of 390-405 eV. Generally, the peak at 396.2 eV corresponds to Mo 3p 3/2 and there is a hump on the side of the Mo 3p 3/2 peak which originates from the N-Mo bond [29]. The crossover peak at 399.2 eV corresponds to N 1S peak from the Mo-N bond [31,32]. Besides, another peak at 402.1 eV is considered to be the N 1s peak attributed to the NO absorbed on the surface of the MoS 2 [33]. These results indicate the S sites were replaced by N on MoS 2 . In addition, BET nitrogen adsorption analysis was performed to further study the specific surface area of the sample. As shown in figure 2(c), when the relative pressure P/P 0 > 0.05, the amount of absorbed nitrogen increases rapidly with the increase of the relative pressure, indicating the process of adsorption of multi-layers [34]. Results indicate that the BET surface of the N-doping MoS 2 nanoflowers is 114.2 m 2 g −1 and the size of the pore width ranges from 1.7 nm to 30 nm, as shown in figure 2(d). These results indicate the fabricated sample has a large surface area and pore size distribution.
To investigate the optical properties of the fabricated N-doped MoS 2 nanoflowers, UV-vis spectra were considered and the results are presented in figure 3(a) (the result of the MoS2 nanosheets is also presented to compare). It can be seen from figure 3(a) that the samples exhibit an enhanced strong light absorption in the wavelength range of 200-800 nm. As shown in figure 3(b), the band gap of the samples is estimated from the plot of (ahv) n versus hv by extrapolating the straight line to the X axis intercept. Results indicate that N-doped MoS 2 nanoflowers (2.08 eV) have a narrow band gap in comparison with MoS 2 nanosheets (2.17 eV). The UV-vis diffuse reflectance spectra (DRS) results indicate that more photogenerated charges are generated when flower-like N-doped MoS 2 is excited under visible light irradiation, which enhances the photocatalytic performance [35,36].

Photocatalytic activity
Photocatalytic performances of the N-doped MoS 2 nanoflowers were evaluated by degrading RhB aqueous solution at room temperature under visible light irradiation. As shown in figure 4(a), the concentration of the RhB decreases as the test time increases for all the photocatalysts. As can be clearly seen, the degrading rate of RhB with the photocatalysts follows the order of: N-doped MoS 2 nanoflower > without light (flower-like N-doping MoS 2 heterostructure in a dark condition) > MoS 2 nanosheets > bulk N-doped MoS 2 . This result indicates that the prepared N-doped MoS 2 nanoflowers have better photocatalytic properties than others. Meanwhile, in the dark condition, the degradation efficiency of RhB for the N-doped MoS 2 nanoflowers is only 12%, indicating that light plays a key role in degradation of RhB. Plots of the absorbance versus wavelength for degradation of RhB for N-doped MoS 2 nanoflowers at various irradiation times are shown in figure 4(b). It can be seen that the intensity of the absorption peaks continuously decreases without any change in position during the degradation reactions. To further examine the role of the surface area in photocatalytic reactions, plots of ln(C/C 0 ) versus irradiation time are displayed in figure 4(c) (the initial concentration of the RhB suspension was measured and used as the initial concentration C 0 ; in addition, C is the actual concentration of RhB at the indicated reaction time). It can be seen that both the curves are linear, indicating photodegradation of the RhB goes through a pseudo-first-order kinetic reaction [35]. Besides, the potocatalytic activity of N-doped MoS 2 nanoflowers under visible light irradiation is higher than that of MoS 2 nanosheets because N-doped MoS 2 nanoflowers have a large surface area and large pore size distribution (see supplementary data S2 (stacks.iop.org/NANO/27/225403/mmedia)) [37]. N doping could also extend the spectral response to visible light and greatly improve the utilization of visible light [38,39]. The stability of photocatalysts is a crucial factor for their assessment and practical applications. Figure 4(d) shows the recycling reaction towards degradation of RhB with the catalyst of N-doped MoS 2 nanoflowers, where the sample was separated by a centrifuge after every 70 min of visible light irradiation. Results indicate the photocatalytic performance of the N-doped MoS 2 nanoflowers do not decrease obviously after four consecutive experiments, revealing its excellent recycling and structural stability.

Discussion of the photocatalytic mechanism of N-doped MoS 2 nanoflowers
In order to give further evidence to support the photocatalytic mechanism, the transient photocurrent responses of an N-doped MoS 2 nanoflowers electrode were recorded for several on-off cycles of irradiation. Figure 5(a) shows the photocurrent-time testing curves of the N-doped MoS 2 nanoflowers. Results indicate our photocatalyst has the highest photocurrent compared to graphene/C 3 N 4 composites [40], g-C 3 N 4 /Zn 2 GeO 4 heterojunctions [41], g-C 3 N 4 /NiS hybrid [42], and grapheme oxide/graphitic-C 3 N 4 nanosheet hybrid [43]. Generally, the value of the photocurrent indirectly reflects the ability to generate and transfer the photoexcited charge carrier under irradiation [44]. The higher the photocurrent is, the higher the e + −h + separation efficiency [40,45]. To further study the photocatalytic mechanism of the sample, radical trapping experiments were proposed. In radical trapping experiments, ammonium oxalate (AO, 5 ml), 1, 4-benzoquinone (BQ, 5 ml) and tertiary butyl alcohol (TBA, 5 ml) were used as scavengers of the photo-induced holes (h + ), superoxide radicals ( ) ⋅ -O , 2 and hydroxyl radicals (·OH), respectively [35,[46][47][48]. Displayed in figure 5(b) is the degradation efficiency of RhB from 100% to 5% in the presence of TBA compared to that with no radical scavengers under visible light irradiation. Meanwhile, the degradation efficiencies of RhB reach 15% and 40% in the presence of AO and BQ, respectively. Thus, it is reasonable to conclude that h + , ⋅ -O 2 , and ·OH as oxidation species were indeed photogenerated on catalyst surfaces and are responsible for the photocatalytic degradation. In general, the more positive   the valence band potential, the stronger the oxidation ability of photogenerated holes, which is favored for better photocatalytic activity [49]. So it can be concluded that direct oxidation by holes is crucial because the potential of photogenerated holes is so positive that it can effectively oxidize dyes directly. In addition, hydroxyl radicals are more important than other radicals for dye degradation due to transformation of the parts of ⋅ -O 2 into ·OH radicals. Based on above results, N-doped MoS 2 nanoflowers with excellent photocatalytic performance might be explained by the following factors. First of all, N-doped MoS 2 nanoflowers have larger BET areas (figures 2(c) and (d)), which can not only improve surface adsorption capacity of the reactants, but also expose more active sites, guaranteeing higher activity in degrading RhB [50,51]. In addition, as shown in figure 3, N-doped MoS 2 nanoflowers have a narrow band gap compared with MoS 2 nanosheets, which can extend the spectral response to visible light and greatly improve the utilization of visible light [38,39], guaranteeing higher activity in degrading RhB. Moreover, N doping extends the visible light absorption edge and electrons sre excited from the N impurity level to the conduction band, guaranteeing higher activity in degrading RhB [52]. Meanwhile, electrons in the CB of N-doped MoS 2 flowers are good reductants that could efficiently change the O 2 absorbed onto the catalyst surface into various reactive species ( ⋅ -O , 2 HO 2 , H 2 O 2 ), subsequently leading to the formation of ·OH and oxidation of RhB into CO 2 , H 2 O, etc. Based on the above results and discussion, we propose a possible mechanism (figure 6) to explain the degradation of RhB by N-doped MoS 2 nanoflowers under visible light irradiation. The radical production could be expressed by reactions as follows:  [54], etc. Here, α-Fe 2 O 3 @N-doped MoS 2 nanoflower heterostructures with strong magnetic properties were employed to magnetically separate our catalysts from the solution of RhB. As shown in figure 7, the degradation rate of RhB is almost the same for the catalysts of α-Fe 2 O 3 @N-doped MoS 2 heterostructures  and N-doped MoS 2 nanoflowers, but with magnetic separation in 10 s (shown in the upper right of figure 7). These results indicate that α-Fe 2 O 3 @N-doped MoS 2 heterostructure can not only serve as highly efficient photocatalysts, but also easily separate organic pollutants.

Conclusion
In summary, we investigated RhB removal with N-doped MoS 2 nanoflowers and α-Fe 2 O 3 @N-doped MoS 2 heterostructures. Results indicated that the as-prepared N-doped MoS 2 nanoflowers showed excellent photocatalytic activities and durability on the elimination of the organic pollutants under visible light irradiation. We also demonstrated that the α-Fe 2 O 3 @N-doped MoS 2 heterostructures can be easily separated from organic pollutants for recycling owing to their magnetic properties. This work helps us to deeply understand the uncommon photophysical processes necessary for the design of highly efficient photocatalysts for environmental applications in the future.