Synthesis of Tunable Band Gap Semiconductor Nickel Sulphide Nanoparticles: Rapid and Round the Clock Degradation of Organic Dyes

Controlled shape and size with tuneable band gap (1.92–2.41 eV), nickel sulphide NPs was achieved in presence of thiourea or thioacetamide as sulphur sources with the variations of temperature and capping agents. Synthesized NPs were fully characterized by powder XRD, IR, UV-vis, DRS, FE-SEM, TEM, EDX, XPS, TGA and BET. Capping agent, temperature and sulphur sources have significant role in controlling the band gaps, morphology and surface area of NPs. The catalytic activities of NPs were tested for round the clock (light and dark) decomposition of crystal violet (CV), rhodamine B (RhB), methylene blue (MB), nile blue (NB) and eriochrome black T (EBT). Agitation speed, temperature, pH and ionic strength have significant role on its catalytic activities. The catalyst was found to generate reactive oxygen species (ROS) both in presence and absence of light which is responsible for the decomposition of dyes into small fractions, identified with ESI-mass spectra.

of the choices as it is an important class of the metal sulphide family [41][42] and has diverse phases and wide applicability in lithium ion batteries 43 , supercapacitors 44 and dye-sensitized solar cells 45 . The nickel sulphide system contains a number of phases, for example NiS (α and β ), NiS 2 , Ni 3 S 2 , Ni 3 S 4 , Ni 7 S 6 , Ni 9 S 8 and α -Ni 3+x S 2 46 . In recent years, many research groups have attempted to prepare different forms of nickel sulphide with different phases and morphology for specific applications 46 . The challenging task is to synthesize nickel sulphide by the low-temperature wet chemical route with controlled phases 47 . However, controlled-phase nickel sulphide synthesis is also difficult using hot-injection 48 , hydrothermal 49 , solvothermal 50 and microwave methods 51 because this leads to the formation of other phases within the synthesized product. Moreover, structural changes with respect to their concentration and formation of different pure and mixed morphologies by the wet chemical methods have been less investigated. Herein, we report a green synthesis of tuneable band gap nickel sulphide NPs in water using thiourea or thioacetamide as sulphur sources and subsequently applied as a catalyst for the degradation of organic dyes in presence and absence of light.

X-ray diffraction (XRD).
Powder X-ray Diffraction (PXRD) of the nickel sulphide nanoparticles prepared at different temperatures using different sulphur sources with various capping agents are shown in Fig. 1a. The sharp and intense diffraction peaks reveal that the samplesNi1-Ni8 prepared by a hydrothermal with different capping agents (PVP, SDS, CA and β -CD) are well crystalline and the peak values matches with the literature reported data (Fig. S1). The peak values of Ni1-Ni4 prepared using thiourea match with the α -phase of the NiS (hexagonal phase, JCPDS Card No. 75-0613) whereas peak values of Ni5-Ni8 prepared using thioacetamide match with the β -phase of the NiS (rhombohedral phase, JCPDS Card No. 12-0041) [52][53] . When the reactions are carried out at lower temperature (60-100 °C) with SDS as capping agent (Ni9-Ni12) broad peaks (2θ) at 16-24° and 30-40° were observed. It was difficult to assign the exact phase due to peak broadening. The PXRD of the reused NiS NPs was nearly the same as fresh one. On comparison of PXRD pattern of all the samples it was found that the temperature and sulphur sources play a significant role in the formation of different phases of nickel sulphide nanoparticle.
Infrared spectroscopy (FT-IR). From FT-IR spectra (Fig. 1b) of nickel sulphide NPs (Ni9, Ni10, Ni12 and reused Ni12) peak at ~3154-3203, 1645-1627, 1105-1071 and 611-602 cm −1 was observed. A weak peak at 3154-3203 and 1645-1627 cm −1 can be assigned to the bending and stretching vibration of SDS and adsorbed  particle was noticed in case of Ni6 while Ni7 and Ni8 have network type morphology. When samples were prepared at 80 °C (Ni9 and Ni10), a non-uniform shape and size with aggregated form of particles was observed. For Ni12, a sponge like morphology was observed. The difference in the morphology of samples having different capping agents depend upon several factors like crystal-face attraction, electrostatic and dipolar fields associated with the aggregate, hydrophilic interactions, hydrogen bonds and van der Waals forces. A combination of these factors may have effects on the self-assembly to form final structure. TEM study confirmed that Ni12 has sponge like structure and comprised of very small NPs whereas Ni2 and Ni9 have mixed shape (Fig. 2b-e). The image of Ni12 (fresh and reused) is depicted in Fig. S3. The EDX result of all the samples confirmed that Ni and S are present in all samples (Table S2). The spatial regularity of elemental distribution of Ni12 was measured and composition was found to be Ni 0.82 S 1.00 (Table S2, entry 5). The EDX elemental mapping and line scanning further indicate the homogeneous distribution of Ni and S throughout the sample (Fig. 2a).

X-ray photoelectron spectroscopy (XPS) study.
To further confirm the elemental composition, XPS analysis was carried out. Figure 3a shows the typical survey spectrum of NiS. The binding energies were calibrated using C (1s) peak at 284.6 eV as reference. The strong peaks at 853.0 and 873.2 eV are assigned to Ni 2p 3/2 and Ni 2p 1/2 , respectively whereas peaks at 162.4 and 167.7 eV are assigned to the binding energy of S 2p 3/2 and 2p 1/2 . The presence of two strong satellite peaks at about 860.6 and 878.9 eV also corresponds to the binding energies for Ni 2p 3/2 and Ni 2p 1/2 respectively, which indicate the presence of the electron correlation in the system. Spectral study. The UV-vis absorption properties of all samples were measured. 2 mg of prepared samples were dispersed in 5-15 ml water under sonication. The entire sample showed similar type of spectra. Broad absorbance across visible region may be attributed partially to the scattering by NPs. The direct band gap of the sample was measured using Tauc-Mott equation [55][56] . From the plot (inset Fig. 3b), the band gap of Ni6, Ni9 and Ni12 was found to be 2.4, 2.05 and 1.92 eV. It was observed that the band gap is SDS concentration dependent. The photo-absorption properties of the samples were examined by diffuse reflectance spectrum (DRS). DRS measurement was carried out taking barium sulphate as reference (Fig. S4a). All the samples exhibit high absorption in UV-vis region (200-800 nm). The DRS spectrum of the reused Ni12 was nearly same and there was no Thermogravimetric analysis (TGA). The thermal stability of the nickel sulphide NPs and reused Ni12 was measured by TGA (Fig. S4b). All samples were heated up to 750 °C with heating rate 10 °C/min. Samples Ni2, Ni6, Ni9, Ni10, Ni12 and reused Ni12 showed multi-stage decaying pattern. The initial weight loss up to 200 °C is attributed to the loss of physisorbed moisture. The second step weight loss occurred between 260-390 °C which might be due to organics (surface bonded capping agent) whereas the decay above 450 °C might be due to sublimation of metal sulphides. As evident from the TGA curve, Ni6 shows maximum stability while Ni9 minimum.
BET measurement: surface area analysis. From the nitrogen adsorption-desorption isotherms surface area of NPs were measured (Fig. 4). Ni2, Ni6, Ni9, Ni10 and Ni12 show surface area values of 19.43, 13.04, 21.49, 29.13 and 30.52 m 2 /g respectively. It is clear from the data that the surface area increases with the increase in SDS concentration. Sample having same capping agent with different sulphur sources showed different surface area (Ni2 and Ni6) which is due to difference in reactivity of sulphur sources. The increase in surface area is believed to increase the catalytic activity of the nanoparticles. From the electron microscopy and surface area measurements, it was observed that upon decreasing particle size surface area and catalytic activity increases.  (Table S3). To assess the catalytic performance of all synthesized NPs, degradation reactions were carried out both in presence and absence of visible light (Sunlight, 200 W and 100 W tungsten lamp). The optimization of catalyst loading (Ni12: 1-5 mg) was performed using methylene blue (MB) under 200 W tungsten lamp (Fig. S5a-g). From the graph it was found that the degradation of MB was very effective in presence of 5 mg catalyst and 100% removal was achieved within a minute. In absence of catalyst, the concentration of MB (Fig. S5f) was almost constant throughout the experiments. All the reactions onwards were performed using 5 mg of catalyst.
With the optimized condition, all NiS samples were tested with CV. It was observed that Ni1-Ni8 was able to degrade CV only upto 7-60% under 200 W lamp in 60 minutes (Fig. S7). When the same experiment was performed under visible light (100 W and 200 W lamp) in presence of Ni9, Ni10 and Ni12, almost quantitative degradation took place in 4 minutes. Under dark condition, Ni12 takes 15 minutes (Fig. S8). From time dependent plot (Fig. 5a,b) and Table S4 it was found that Ni12 has maximum catalytic activity while Ni8 has least. Very fast degradation was achieved with Ni12 when reaction was carried out under sunlight (Fig. 5g). Visual colour changes also support our observations (inset Figs 5 and S7).
In a similar way, same sets of reactions were performed with RhB (Fig. 5c,d) and MB (Fig. 5e). The catalysts are found to be equally effective. Details of degradation of RhB and MB with various catalysts under different condition are in Tables S5-S7 and in Figs S8-S10. When reactions were carried out with NB, MO, EBT and XO under visible light (200 W tungsten lamp), we observed fast degradation (4min) for NB (Fig. 6a). In case of MO and EBT, catalysts were not much effective. There was no improvement in degradation of MO (Fig. 6b) and EBT (Fig. 6c) with prolonging reaction time. XO (Fig. 6d) showed shift in absorbance peaks but no degradation, which might be due to the formation of complex 57 . Under sunlight, MB (Fig. 5f), NB (Fig. 5h) and RhB (Fig. S11) also show rapid degradations. Relative concentration variations of dyes under sunlight are represented in Table S8. We have repeated our experiments 3-4 times and the relative error was ~3-5%. The error bar is incorporated in time dependent plot.
On comparision of the catalytic activities of all NPs it was found that Ni12 is best amongst all (Tables S4-S6). The degradation of different sets of mixed dyes (Table S9) were carried out under 200 W lamp (Figs 7 and S12) and was found to be equally effective as in the case of individual dye. The degradation performance of all the catalysts showed the order of MB > CV ~ NB > RhB ≫ MO ≫ EBT (Fig. 6e, Tables S11-13). The effect of light, agitation speed, temperature, pH and ionic strength were also studied to check the efficacy of the catalyst under different conditions. The effect of light on the degradation rate was equated from relative concentration variations (Fig. 6f-h). It was observed that the degradation rate was faster in presence of light (Tables S14 and S15). Similarly, we also studied the effect of agitation speed and temerature for MB and CV under 200 W lamp. It was observed that degradation increases with increase in agitation speed (Fig. S13a) and temperature (Fig. S13b). Experiments were also carried out to check pH dependency (acidic, neutral and basic). The catalyst was highly active at neutral pH (Fig. S13c). Further experiments were done to check the effect of salts using 0.1 (M) NaCl solutions It was found that higher the salt concentartion the lower was the degradation (Fig.  S13d).
It is well known that photodegradation of organic dyes occur via radical formation. While designing the catalyst, it was thought worthwhile that the catalytic system should generate free radicals. To ascertain the active species in the catalytic process and to investigate the catalytic mechanism, electron paramagnetic resonance (EPR) study was carried out. The catalytic system was "EPR-silent" which might be due to the presence of non-Kramers system 58 and very short life time of Reactive Oxygen Species (ROS). Therefore, indirect measurement of ROS was done using terepthalic acid (TA). TA is a non-fluorescent compound which gets converted to fluorescent 2-hydroxy terephtalic acid (HTA) upon reaction with OH radical and increases in the fluorescent intensity with time on irradiation confirms the formation of OH radical in the presence of NiS. A linear characteristic was observed for fluorescent intensity versus reaction time and the inset shows the fluorescent spectra of TA at different time interval in presence of Ni12 (Fig. S14).
ESI-Mass study was carried out to analyze the intermediates formed during the degradation of methylene blue. The molecular ion peak of MB was obtained at m/z = 284 (Fig. S16). The peaks at m/z = 305-318 was due to consecutive addition of hydroxyl in the MB molecule (Fig. S17). The degradation of MB into small molecule was suggested by the presence of peaks at m/z = 235, 211, 132, 105 and 71. Detection of hydroxylated intermediate in ESI-MS spectrum also confirmed that the degradation of MB in presence of Ni12 proceeded through hydroxyl radical mechanism which was also confirmed from TA treatment. Finally, reusability of catalyst was performed under 200 W lamp for the sample Ni12. PXRD (Fig. 1a), IR (Fig. 1b), TEM and SEM (Fig. S3), DRS (Fig. S4a) and TGA (Fig. S4b) of reused catalyst were compared with the fresh one. It authenticated that there is no structural and morphological change after fifth cycles of its use. While scaling up reaction, concentration of MB was doubled (1000 ml) and catalyst amount was reduced 8 times under 200 W lamp. It took 20 minutes for degradation (Fig. S15).
It was observed that ROS is the active species for the degradation of dyes into small molecules as indicated from UV and mass spectroscopy. Low band gap of NiS NPs and stray light (that cannot be ignores during dark reaction) are helpful for the easy generation of electron-hole pair 59 . The conduction band electron may chemically form hydrated electron (e − hyd ) on the surface of NiS NPs 60 . The dissolved oxygen along with e − hyd forms ROS even in absence of light but faster under the visible light (Fig. S13). It was observed from the experimental results that the catalyst having high surface area showed higher catalytic activities. This might be due to the fast generation of electron-hole pair and the scavenging of the holes in presence of S 2− /S n 2− redox couples to maintain its stability during the degradation process (Fig. 8).
In conclusion, we have developed a new, simple, economical and green protocol for the synthesis of tuneable band gap (1.92-2.41 eV) nickel sulphide NPs in the presence of thiourea or thioacetamide as sulphur sources with the variations of temperature and capping agents. EDX and XPS analysis confirmed that the NPs were composed of Ni 2+ and S 2− . UV-vis and DRS reflectance spectra suggested that the samples were capable of absorbing visible light. We explored the rapid catalytic decomposition of organic dyes [series of individual dyes (positive, negative, neutral dyes with various functionality) and their mixture] such as crystal violet (CV), rhodamine B (RhB), methylene blue (MB), nile blue (NB), methyl orange (MO) and eriochrome black T (EBT) under dark and visible light. It was also observed that this catalyst is effective at neutral pH with high agitation speed even at room temperature with the fast generation of ROS. The generation of ROS even in dark is responsible for the fast degradation of dyes where scavenging of holes in the presence of S 2− /S n 2− redox couples maintain their stability throughout the experiments. Decomposition of MB into small fragments was identified using mass analysis. Catalyst is reusable, Preparation of NiS nanoparticles. For hydrothermal synthesis, these solutions (set 1 and set 2) were transferred into 25 ml Teflon lined sealed stainless steel autoclaves and temperature was maintained 200 °C for 8 h. It was then allowed to cool naturally to room temperature and the resulting black solid precipitate was then collected by centrifugation followed by washing with deionized water and finally with ethanol. The sample was dried in desiccator for 2 h and then it was collected. In order to synthesize the NiS NPs at different temperature, stock solutions were mixed and stirred at the required temperature (Table S1) for 2 h. Samples were collected similarly as in case of hydrothermal synthesis. Details of sample preparation are in supporting information (flow chart diagram).
Degradation of organic dyes. Crystal violet (CV), rhodamine B (RhB), methylene blue (MB), nile blue (NB), methyl orange (MO), xylenol orange (XO) and eriochrome black T (EBT) were purchased either from Sigma-Aldrich or Alfa Aesar and were used as model dyes to assess the catalytic performance of the prepared nickel sulphide in presence and absence of visible light (Sunlight,100 W and 200 W tungsten lamp). 5 mg NiS was dispersed in 14 ml aqueous solution of ~10 −5 (M) dyes. The suspensions were magnetically stirred under dark or in presence of light. At given time interval, 2 ml aliquots were taken and were centrifuged to remove the catalyst. UV-vis spectra were recorded with 1:1 dilution of experimental solution taken at certain interval. Blank experiments were also performed under identical conditions.