Morphologies of nanostructured bismuth sulphide and Mn (II) doped bismuth sulphide nanoparticles: characterization and application

Bismuth nitrate, manganese acetate, polyoxyethylene 100 stearate (Myrj59), Na₂S (sodium sulphide), nitric acid, potassium ferricyanide (PFC), sodium thiosulfate (STS) and methyl violet were purchased from Sigma Aldrich. All the solutions were prepared using doubly distilled and demineralized water.

The experiments were carried out using ethyl violet dye as a model system. Each UV light assisted reduction experiment was carried out by taking 25 mg Bi₂S₃ nanoparticles into a 250 mL glass UV chamber containing water jacket with 32.5 × 10⁻⁴ mol·L⁻¹ of dye. Analogous experiments were performed with and without undoped Bi₂S₃ nanoparticles. The decolorization of the dye was measured by changes in its absorbance at λ_max using UV-Vis spectrophotometer.

The reaction mixture containing as-synthesized Bi₂S₃ nanoparticles (7 mg), 1 mL of 0.001 mol potassium ferricyanide (PFC) and 1 mL of 0.1 mol sodium thiosulfate (STS) were placed in a quartz cell with 1 cm³ path length. The reaction was monitored by UV-Vis spectrophotometer in the presence and absence of Bi₂S₃ NPs at 25±2 °C.

All experiments were repeated three times and the reproducibility was confirmed. The recyclability of the NPs was also studied. The NPs were recovered by centrifugation, washed with acetone, dried at 60 °C under vacuum and used for the next reaction.

Fig. 1 shows the XRD patterns of as-synthesized NPs. Fig. 1a is the XRD pattern of undoped Bi₂S₃NPs, whereas Fig. 1b and Fig. 1c show the spectra of 5 % and 10 % Mn ion doped Bi₂S₃NPs. Almost all peaks in the patterns could be indexed to a pure orthorhombic phase bismuth sulphide with the lattice parameters: a = 11.132 Å, b = 11.256 Å, and c = 3.976 Å, which is in a good agreement with the literature (JCPDS Card No. 17-320). The signals of Mn⁺² were not detected in the XRD spectrum of bismuth sulphide. There were no other diffraction peaks which indicates high purity and crystalline nature of the nanoparticles.

The peak due to Bragg’s reflection is observed at 2θ value corresponding to 25.3° and 28.9°. The particle size of undoped and Mn⁺² doped Bi₂S₃ nanoparticles calculated from the Debye-Scherrer equation was 20±2 nm: D=Kλ/(βcosθ)$$D = K\lambda /\left( {\beta \cos \theta } \right)$$(1)

where K is the Scherrer constant, λ is the wavelength of the X-ray, β and θ are the half width of the peak at half of the Bragg angle, respectively.

The size and morphology of the Bi₂S₃ nanoparticles were analyzed by TEM. The TEM images (Fig. 2) reveal that the undoped (Fig. 2a) and Mn⁺² doped Bi₂S₃ nanoparticles consist of nanorods-like structures. When doping increased, the thickness of the rods also increased, as shown in Fig. 2b and Fig. 2c. Fig. 2d and Fig. 2e are TEM images of typical Bi₂S₃NPs at higher magnification. The diameter of the nanorods is in the range of 20 nm to 30 nm, and the length is about 200 nm to 250 nm. The observed rod type morphology of the product is probably due to chain type structure of bismuth sulphide. It is known that Bi₂S₃ molecules form band-like structures as Bi units are connected via weaker van der Waals forces [11]. It seems that the formation of Bi₂S₃NPs may have originated from the preferential directional growth of Bi₂S₃ crystallites. The crystallinity of the product was also proven by SAED (Fig. 2f).

Morphology of the product was also analyzed by HR-SEM. The HR-SEM images (Fig. 3) show the formation of bar-shaped nanorods of Bi₂S₃. Fig. 3a shows HR-SEM image of pure bismuth sulphide nanoparticles which look like bundles of thin nanorods. Fig. 3b shows HR-SEM image of Mn doped Bi₂S₃NPs (5 %) with thick nanorods. Fig. 3c shows cubic shaped Mn doped Bi₂S₃NPs (10 %). When Bi₂S₃ was doped with Mn²⁺ metal ions, the shape of Bi₂S₃NPs has also changed. Rod type morphology of bismuth sulphide may be due to the inherent chain type structure.

The weight percentage of Mn²⁺ was determined by energy dispersive X-ray analysis (EDX) (Table 1, Fig. 4). EDX analysis indicates that the well-cleaned final product is mostly composed of Bi and S, with no other elements (Fig. 4a). Fig. 4b and Fig. 4c are EDX images of Mn²⁺ doped Bi₂S₃NPs with 5 % and 10 % doping of Mn ion, respectively.

The FT-IR spectra of Bi₂S₃NPs and pure polyoxyethylene 100 sterate (Myrj59) are given in Fig. 5. Both display the typical profile of C=O group of ester in the range of 1726 cm⁻¹. The peaks in the range of 1416 cm⁻¹ to 1420 cm⁻¹ are due to C–H bending. It is seen that the peak at 1726 cm⁻¹ shifts to 1731 cm⁻¹ in the Bi₂S₃NPs. The shifts observed in the spectra can be attributed to the interaction of Bi₂S₃NPs with the polyoxyethylene 100 sterate. The band at 2912 cm⁻¹ to 2930 cm⁻¹ is characteristic of C–H stretching. The broad band due to hydrogen bonded hydroxyl group (O–H) appears in the range of 3470 cm⁻¹ to 3490 cm⁻¹ and it is attributed to the complex vibrational stretching, associated with free, inter and intra molecular bound hydroxyl groups.

Many industries, such as textile, paint, ink, plastics, cosmetics, etc., use a variety of dyes. Dyes are lost in a waste water stream during dyeing operation. The removal of the color of waste water is a burning issue all over the world, as it creates water pollution [17].

Various chemical methods, such as oxidation, precipitation, coagulation, etc., physical processes, such as adsorption, filtration, biological techniques, such as microbial degradation, etc., are used for dye removal [18, 19]. Out of these, photocatalytic degradation process has been proven to be the most efficient way to decolor the dye effluents in an ecofriendly manner [20]. Photocatalytic degradation of pollutants by semiconductors is a new and effective technique for their removal from effluents of industries [21, 22]. Photocatalysis is defined as acceleration of a photochemical reaction in the presence of a catalyst.

In the present investigation, the photocatalytic decolorization of methyl violet dye has been carried out using undoped and Mn-doped Bi₂S₃NPs catalyst in a photoreactor. As a representative case, the catalytic activity of Bi₂S₃NPs was investigated in the decolorization of methyl violet dye. UV-Vis spectra in Fig. 6 show that before the addition of Bi₂S₃NPs, there was the maximum absorption band centered at 555 nm. UV light alone could not induce complete decolorization of the dye. But when irradiated in presence of Bi₂S₃NPs, UV light led to complete decolorization of the dye. The absorption band at 555 nm decreased and disappeared. When the size of bulk metals goes to the nanoscale, electron transfer is more efficient as compared to the bulk material [23, 24].

As shown in the UV-Vis spectra (Fig. 6) of methyl violet, without NPs photocatalytic decolorization did not take place completely even after 7 h. On the other hand, with NPs, the degradation was carried out in 180 min only. However, in the presence of 5 % and 10 % Mn doped Bi₂S₃NPs the degradation time was further reduced to 90 min (Fig. 6a) and 24 min (Fig. 6b), respectively.

The role of NPs in decolorization is as follows: when a photon of UV light strikes a surface of a semiconductor like Bi₂S₃NPs, a valence band electron moves to the conduction band, thus forming a positively charged hole in the valence band. The conduction band electrons and the valence band holes migrate then to the oxide surface and react with the chemisorbed O₂ or OH⁻/H₂O molecules to generate reactive oxygen species, such as O2−${\rm{O}}_2^ - $, HOO^· and ^·OH radicals which attack the dye molecules and cause their degradation [20, 24]. The mechanism for dye decolorization is similar to those discussed by Samira et al. [20] and Mohamed et al. [24].

Doping with a proper element is widely used as an effective method to tune surface states, energy levels, electrical, optical, structural and magnetic properties of semiconductor materials [25–27].

To control the behavior of materials, doping is used, which lies at the heart of many technologies. For this reason, researchers have begun to explore the use of dopants to influence the properties of semiconductor nanoparticles [28]. The energy from absorbed photons can be efficiently transferred to the dopants which is utilized in the excitation and increases the rate of reactions on the surface of NPs [29]. Although the role of Myrj59 is only to cap the Bi₂S₃ NPs and to rule out the possibilities of decolorization by Myrj59, an experiment was performed with using it. No significant decrease in the absorbance was observed with using Myrj59.

The electron transfer reaction between Fe(CN)63−${\rm{Fe}}\left( {{\rm{CN}}} \right)_6^{3 - }$ and S2O32−${{\rm{S}}_2}{\rm{O}}_3^{2 - }$ was catalyzed by a metal like Pt [30–32], Ni [33]. This reaction was performed for the first time in the presence of Bi₂S₃NPs. The characteristic absorption of Fe(CN)63−${\rm{Fe}}\left( {{\rm{CN}}} \right)_6^{3 - }$ was observed at 420 nm. When reaction took place between Fe(CN)63−${\rm{Fe}}\left( {{\rm{CN}}} \right)_6^{3 - }$ and S2O32−${{\rm{S}}_2}{\rm{O}}_3^{2 - }$, the intensity of the corresponding peak (420 nm) decreased with time. In both the reactions, absorption was measured at time intervals of 3 min. In presence of Mn doped Bi₂S₃NPs (10 %) (Fig. 7a), there was a continuous decrease in absorbance at 420 nm with respect to time and after 24 min, the peak completely disappeared. In case of 5 % Mn doped Bi₂S₃ (Fig. 7b) completion of the reaction took 39 min. However, in the absence of Bi₂S₃ the reaction was proceeding very slowly and did not take place even after 6 h (Fig. 7c). On the basis of these observations it can be concluded that in presence of Bi₂S₃NPs, redox reactions proceed faster owing to efficient electron transfer due to the larger surface area of Bi₂S₃NPs.