Fast and effective catalytic degradation of an organic dye by eco-friendly capped ZnS and Mn-doped ZnS nanocrystals

Undoped and manganese doped ZnS nanocrystals encapsulated with thioglycolic acid (ZnS-TGA) were synthesized and characterized with different techniques, and finally tested in the photodegradation of a methyl orange in aqueous solution under UV and sunlight irradiations. FTIR and X-ray diffraction results confirmed the functionalization of these nanocrystal surface by thioglycolic acid and the formation of crystalline structures of ZnS and Mn-doped ZnS with cubic and hexagonal phases. Calculated average size of ZnS nanocrystals was in the range of 2–3 nm. It was observed a blue shift of the absorbance threshold and the estimated bandgap energies were higher than that of Bulk ZnS thus confirming the quantum confinement effect of charge carriers. Photoluminescence spectra of ZnS nanocrystals exhibited emission in the range of 410–490 nm and the appearance of an additional emission band around 580 nm (2.13 eV) connected to the 4T1 → 6A1 transition of the Mn2+ions. Photodegradation of methylene orange with undoped and Mn-doped ZnS-TGA nanocrystals was investigated. Dye adsorption prior to photocatalysis using nanocrystals was studied via kinetic and equilibrium experiments. The maximum dye adsorption capacity on doped ZnS-TGA was ~ 26.98 mg/g. The adsorption kinetic was found to follow the pseudo-second-order kinetic model. A statistical physics model was used to analyze the equilibrium data where the calculated adsorption energy was 17–18 kJ/mol. It was concluded that the dye adsorption was associated to the hydrogen interaction where the removal process was feasible and multi-molecular at 25 °C. The photocatalytic activity of undoped ZnS nanoparticles under UV irradiation showed better efficiency than doped nanocrystals thus indicating that manganese doping generated a dropping of the photocatalytic degradation of the dye. Dye degradation efficiency of 81.37% using ZnS-TGA nanocrystals was achieved after 6 min, which indicated that ZnMnS-TGA nanocrystals may be considered an alternative low cost and environmental friendly material for facing water pollution caused by organic compounds via photodegradation processes.


Introduction
Semiconductor nanoparticles (NPs) have found numerous applications in different technological fields going from photovoltaic devices to medical imaging, lasers, LEDs, and photocatalysis (Dao 2020;Rtimi et al. 2021;Meikle et al. 2020;Long et al. 2020;Su et al. 2020;Li et al. 2019b;Daskalakis et al. 2020;Rtimi et al. 2021). These applications are supported by their unique properties compared to their bulk counterpart, which arise from the confinement of the charge carriers as a result of their small sizes. This quantum confinement effect enlarges the energy bandgap, which makes the electronic and optical properties size dependent allowing the control of the absorption and the emission wavelengths of these materials. Additionally, semiconductor NPs have a large surface to volume ratio, which is responsible for a Responsible Editor: Sami Rtimi * Sabri Ouni Sabriouni1995@gmail.com * Naim Bel Haj Mohamed naimhajmed@gmail.com significant change in their physical and chemical properties. Indeed, the surface atoms are less coordinated than internal ones. This makes these NPs lesser stable against aggregation and more reactive with their environment in comparison to bulk materials. Additionally, the dangling bonds and surface defects introduce energy levels in the bandgap, which are usually responsible for non-radiative transitions and the emission quenching of semiconductor NPs. To prevent the agglomeration and the drawbacks of surface defects, NPs must be grown within a protecting medium such as glass and polymer or enrobed by encapsulating molecules that should be well fixed at their surface.
Recently, semiconductor NPs have emerged as promising photocatalysts to produce hydrogen by water dissociation or to degrade organic pollutants in wastewater being more effective than the conventional methods (Buthiyappan et al. 2016). However, these NPs must be dispersible in water to maximize their performance. Accordingly, the colloidal wet chemical synthesis has emerged as the most popular safe way to produce, at large scale and cost-effective, watersoluble semiconductor NPs with narrow size dispersion and promising optical and electronic properties (Jaldurgam et al. 2021).
Among the different semiconducting nanostructures, ZnS NPs are recognized as environmentally friendly and effective photocatalyst for the degradation of organic pollutants in wastewaters. This is due to their nontoxicity when they are used in small amounts and their unique electronic and optical properties. Moreover, ZnS NPs have a high absorption coefficient required for solar light harvesting, which is the first step for the photodegradation of organic pollutants at industrial level. They have also a high negative reduction-oxidation potential of excited electrons due to their higher conduction band position in an aqueous solution as compared to other extensively studied photocatalysts (Yang et al. 2014). Additionally, these NPs are able to rapidly generate great density of electron-hole pairs under photoexcitation because of their direct bandgap (Arao et al. 2009), and they show competitive photocatalytic activity due to trapped holes arising from surface defects (Saenger et al. 1998). The photocatalytic activity of these semiconductor NPs is due to the formation of OH * and O − 2 free radicals that results from the reaction of the photo-generated holes and electrons with water and free oxygen molecules, respectively. However, ZnS nanocrystals (NCs) that belong to large bandgap semiconductors (3.7 eV) can absorb only UV radiation, which reduces their photoactivity. One way to enhance the efficiency of these nanostructures and to improve their visible light harvesting efficiency is to extend the wavelength range photoactivity toward visible light region of solar spectrum by doping them by transition metal ions such as Mn 2+ , Ni 2+ , Cu 2+ (Kostrov et al. 2020;Kabachii et al. 2021;Wang et al. 2017;Chauhan et al. 2014;Othman et al. 2020). The doping impurities introduce trapping levels in the bandgap that reduces the carrier's recombination probability and allows them to diffuse faster to the NCs surface. With the aim of obtaining an effective, low-cost, and eco-friendly nanocatalysts that can be used at industrial scale, this paper reports the synthesis of ZnS and Mn-doped ZnS NPs via colloid method and their photocatalytic application. The physico-chemical properties of these nanostructures were investigated and their photocatalytic activity to decompose methylene orange (MO) dye was evaluated by conducting degradation experiments under UV and sunlight.

Synthesis of TGA capped ZnS nanoparticles
Undoped ZnS and Mn-doped ZnS NCs were prepared using the colloidal hydrothermal reaction method (Ouni et al. 2021), see Fig. 1. The experimental synthesis protocol is briefly described as follows. An aqueous solution was prepared by mixing dihydrated zinc acetate (2 mmol) with TGA (5 mmol) in 100 mL of water. Manganese sulfate was added at different percentages in weight (4, 8, and 12%) with respect to zinc for each synthesis. The pH of the resulting mixture was adjusted to 11 by drop addition of NaOH (1 M) with continuous agitation under N 2 for 30 min. The second step of this synthesis protocol was based on the injection of S 2− (0.08 mmol) at room temperature into the solution containing Zn 2+ -TGA complexes. The mixture was then heated at 100 °C for 3 h. Finally, the reaction mixture was cooled to room temperature to stop the NCs growth. After solution cooling and drying, undoped and Mn-doped ZnS NCs were purified by precipitation with ethanol, washed two times with ethanol, and stored in vacuum at room temperature.

NCs characterization
X-ray diffraction (XRD) characterization was performed using Philips X'Pert PRO MPD diffractometer equipped with a CuKα source of wavelength λ = 1,542 Å. Transmission electron microscopy (TEM) images were obtained using a FEI Tecnai G2 electron microscope operated at an accelerating voltage of 200 kV and equipped with Energy Dispersive X-ray (EDX) for element chemical analysis. Samples were analyzed by placing a drop of aqueous suspension of particles onto a carbon film on a copper grid. Fourier transform infrared (FTIR) spectra were recorded at room temperature with a Perkin Elmer version 5.3 within the wavenumber range of 400-4000 cm −1 and using KBr pellets. UV-visible absorption spectra were obtained at room temperature in the range of 200-800 nm using a SPECORD 210 Plus spectrophotometer with quartz cuvettes. Photoluminescence (PL) spectra were recorded at room temperature using conventional photoluminescence setup and a helium cadmium laser as excitation source (λ exc = 325 nm).

Dye adsorption experiments: kinetics and isotherms
Dye adsorption experiments were performed at 300 K in a batch system. In a typical experiment, 50 mL of dye solution with different initial concentrations and 50 mg of nano-catalyst (m) were used for these studies. The adsorption experiments were carried out at pH 7. Adsorption parameters like time contact and dye concentration were investigated. Kinetic studies were performed with MO dye concentrations from 10 to 50 mg/L with a sampling period of 120 min and magnetic stirring in the dark to measure the MO adsorption capacities. The concentration of dye solution was determined using SPECORD 210 Plus UV-Vis spectrometer at 465 nm. MO dye adsorption capacities Q e (mg/g) were calculated using the next equation where C 0 is the initial dye concentration (mg/L), C e is the equilibrium dye concentration in the solution (mg/L), V is the dye solution volume (mL), and W is the weight of the nano-adsorbent (mg). (1)

Photocatalytic degradation tests
The photocatalytic efficiency of the undoped and Mn-doped ZnS NCs was evaluated via the photodegradation of MO dye as a model molecule of an organic pollutant. Before starting the photocatalytic degradation experiments, the aqueous solutions containing both dye and NCs were submitted to a vigorous magnetic stirring in dark for 2 h in order to ensure the adsorption/desorption equilibrium (Chan et al. 2011).Then, they were submitted to irradiation by mercury lamp of 160 W to monitor the MO degradation. Solution samples were taken at regular time intervals and analyzed by UV-visible spectroscopy to determine the absorbance at 465 nm, which corresponded to the maximum absorption of MO molecule. Degradation experiments were also conducted under sunlight.

Structural and morphological characterization of NCs
FTIR spectra of the different samples are shown in Fig. 2. All these spectra showed an absorption band at 563 cm −1 that was attributed to the stretching vibrations of Zn-S (Amaranatha Reddy et al. 2014). The spectra of ZnS NCs doped with different concentrations of Mn 2+ showed the presence of the same absorption bands at wavenumber values close to those recorded with pure ZnS and no other apparent bands to other vibration modes of doping impurities or other elements were observed. These spectra also showed a wide band at ~ 3360 cm −1 , which was associated to the elongation vibration of (O-H) groups of TGA ligand (Liu et al. 2008). FTIR spectrum of TGA showed the presence of a characteristic band at 2560 cm −1 related to the S-H bond (Dehghan et al. 2018). This absorption band was absent in TGA-capped ZnS NCs spectra thus indicating the rupture of S-H bond and the fixation of the sulfur atom on the NCs surface (Xie et al. 2011). This finding can be explained by the strong interaction of the sulfur electron pair with zinc  (Ouni et al. 2021). In addition, these spectra contained absorption bands around 570-700, 850, 1429, 1508, and 1627 cm −1 assigned to (C-S), (C-H), (COO − ), (C = O), and (-OH) bonds, respectively . These results clearly demonstrated that NCs were well functionalized by TGA ligand, which allowed the control of their growth during synthesis and prevented NCs aggregation (Singh and Chauhan 2009). Figure 3 shows the XRD pattern of the undoped and Mn-doped ZnS samples. In the case of undoped ZnS NCs, XRD diffraction patterns contained peaks at 2θ = 28.95°, 33.76°, 47.58°, and 55.27°. These peaks were attributed to the diffraction plans (111), (200), (220), and (311) of the zinc blende phase (ZB) of ZnS according to JCPDS (80-0007). However, it was also observed the existence of a diffraction peak at 2θ = 31.08° that corresponded to the (101) plan of the wurtzite phase (WZ) of ZnS (JCPDS 80-0020). This structural anomaly made the distinction between the cubic and hexagonal structures very difficult. The most intense peak attributed to (111) suggested that NCs growth in this direction was preferred and confirmed that the structure of ZnS NCs was almost cubic. XRD patterns were adjusted with Gaussian profiles to calculate the full-width-at-half-maximum (FWHM) of identified peaks and these values were used to estimate the crystallite size according to Scherrer formula (Jothibas et al. 2018): where D is the average crystallite size (nm), is the wavelength of X-ray (1.5402 Å) Cu Kα radiation, K is the shape factor (0.9), β is the measured full width at half maximum of the diffraction peak, and θ is the Bragg diffraction angle. The average size of the undoped ZnS NCs was estimated to be ~ 7.15 nm, while the lattice parameter was 5.38 Å.
For Mn-doped NCs, the peaks at 29.52°, 31.82°, 47.31°, and 56.12° were assigned to the plans (002) JCPDS hexagonal ZnS standard sheet (80-0007). However, the presence of the peak associated with the crystallographic plane (200) was around 34.77° thus indicating the presence of the cubic phase of ZnS. Similar results were reported for ZnS NCs synthesized by precipitation in an aqueous medium (Chauhan et al. 2014). The estimated crystallite sizes are given in Table 1. NCs size decreased from 7.15 to 2.5 nm with the increment of Mn doping content up to 12%. This result indicated that Zn 2+ ions were replaced by Mn 2+ ions in the ZnS matrix. The reduction of ZnS NCs size with the doping concentration could be explained by the induced NCs growth and enhancement of surface area to volume ratio in the system (Mote et al. 2013). This result confirmed the incorporation of manganese into the host lattice of ZnS and the formation of Mndoped ZnS NCs. However, the Scherrer formula takes into account only the broadening effect of the spectra coming from the different diffracting domains and does not include the intrinsic micro-strain effect created within the lattice by stacking fault, point defect, and grain boundary (Poornaprakash et al. 2018;Pandya et al. 2016). Therefore, the change of NCs size may be correlated to different structural parameters including the lattice strain (ε), dislocation density (δ), and stacking fault (SF). These parameters can be estimated by using the next formula (Nath et al. 2020;Jothibas et al. 2016;Bel Haj Mohamed et al. 2021) and the calculated values are summarized in Table 1. The value of dislocation density can be associated to the amount of defects and vacancies in the crystals. Results showed that δ increased with the doping Mn content from zero to 12%. This was due to the inversely proportional relationship between the dislocation and the size of NCs. Thus, the smallest particles (ZnS: Mn 12%) have the highest possibility of dislocations since they have a tendency to minimize their higher surface energy (Gaceur et al. 2012). Undoped ZnS NCs with the highest size were the most ordered compared to Mn-doped ZnS. This finding made hkl obvious the effect of doping with Mn on the change of the high orientation plane from (111) to (101) as well as the improvement of crystallinity. Both micro-strain and stacking fault showed the same trends when they decreased with increasing the doping ratio except for the case of Mn 8%. The decreasing of micro-strain indicated that the Mn-doped ZnS samples contained lattice defects and vacancies. On the other hand, SF describes the disordering of crystallographic planes. Therefore, the decreasing tendency of SF suggested that the defects increased by substituting zinc by manganese. Figure 4 shows TEM images of tested samples and the corresponding size distribution histograms. NCs exhibited a spherical shape and were agglomerated probably due to the insufficient ligand amount to cover all ZnS NCs and inhibited the aggregation and formation of large clusters. This agglomeration may be also due to high particles concentrations used in TEM experiments. From the histograms of undoped and Mn (8%)-doped NCs, it was found that the average NCs diameters were 5.9 and 4.1 nm (± 0.5 nm), respectively. These results demonstrated that Mn doping decreased the ZnS particle size. The inter-planar distance was determined from digital micrograph for undoped and doped ZnS, and it was 0.36 nm, which corresponded to the plane (111) of the ZB phase and 0.2 nm for the plane (101) of the WZ phase, respectively. These results are consistent with the values found by XRD measurements. EDX spectra for ZnS and Mn-doped ZnS NCs are shown in Fig. 5. In particular, Fig. 5a indicates the presence of Zn and S that were the major elemental components with also the presence of Si related to the detector signal. It was observed the presence of the carbon, oxygen, and copper peaks that were related to TEM grid and ligand stabilizer. Results of Mn (8%)-doped NCs (Fig. 5b) showed the presence of Zn, S, and Mn, which confirmed the formation of doped NCs. This diagram also indicated the existence of carbon, oxygen, and silicon related to TGA ligand and TEM devices. Others peaks related to Al, Ca, K, Ni, and Fe were also detected probably due to residues coming from the main precursors during the synthesis. Hence, TEM results indicated the formation of undoped and doped ZnS NCs that were properly stabilized by TGA.

Optical characterization of NCs
UV-visible absorption spectra of the different NCs are reported in Fig. 6. They showed a blue shift of the absorption threshold when increasing Mn 2+ concentration. The absorption threshold was 307, 300, 298, and 290 nm for 0, 4, 8, and 12% Mn doping concentrations, respectively.
In accordance with the average sizes obtained from XRD analysis, this behavior was explained by the presence of small Mn-doped ZnS NCs, which induced the widening of gap energy (Meng et al. 1995).
The Tauc equation was utilized to estimate the gap energy of undoped and Mn-doped ZnS NCs (Kaur et al. 2015).
where α is the absorption coefficient, A is a constant,h is the photon energy, Eg is the optical gap of NCs, and n is a constant with n = 1/2 for direct semiconductors. The optical band gap energy of NCs was estimated from the intercept of the extrapolation of Tauc plots with the x-axis (Fig. 7) and the results of the optical gap energies are displayed in Table 2.
The calculated value of the optical gap energy of NCs increased from 3.75 to 3.92 eV with increasing manganese doping. This variation in E g was attributed to variations of NCs size by the incorporation of manganese ions in the zinc substitute position within the ZnS structure. At high doping concentrations, manganese ions caused a significant increment of E g . This finding was explained by the Burstein-Moss effect, which demonstrated that the excitation of an electron from the valence band to conduction band in a doped semiconductor required more energy due to the accumulation of electrons in the conduction band caused by the shallow donor states (i.e., the movement of the Fermi level to the conduction band by doping effect) (Brus 1986).
The average NPs sizes were also estimated by using the optical gap energy and the effective mass approximation according to the next equation (Jai Kumar and Mahesh 2017;Mansur and Mansur 2011): where E n g and E b g are the band gap energies of NPs and bulk material, respectively,m * e and m * h are the effective masses of the electron and hole, e is the elementary electric charge, and is the dielectric constant of ZnS. It was observed a reduction in the size of the doped ZnS NCs from 2.3 to 1.9 nm as the doping rate increased. The difference between the sizes found by absorption spectroscopy and DRX was attributed to the fact that calculation of the latter values was based on the effective mass approximation assuming that the NCs geometry was spherical, while the XRD measurements in polydisperse systems were weighted to the largest particles (Pejjai et al. 2017). Additionally, when the quantum dots are small enough, the carrier's wave functions overlap with the crystal boundaries thus resulting in energy dependent effective masses (Jai Kumar and Mahesh 2017;Mansur and Mansur 2011). This discrepancy may be also due to the fact that XRD results were obtained for powdered NCs, while absorption measurement involved NCs dispersed in solution. Figure 8 shows the room-temperature photoluminescence spectra of undoped and Mn-doped ZnS NCs with different Mn concentrations. PL spectra of pure ZnS nanocrystals showed two wide and asymmetrical bands. Accordingly, they were convoluted with three Gaussian curves, see Fig. 9. The observed emission spectra may be the result of several emission transition essentially caused by the size distributions of the ZnS NCs with a large contribution of surface defects. In fact, the stoichiometric ratio of synthetic precursor (Zn/S ~ 2.5) and the synthesis method were responsible for the formation of punctual defects (interstitials, substitutes) or extended (dislocations, aggregates, cavities) within the ZnS structure. Indeed, the use of excess zinc caused that defects present in the ZnS NCs correspond to zinc interstitials (Zn i ) or sulfur vacancy (Vs) (Peng et al. 2006). For the undoped ZnS NCs, the spectrum was dominated by a band around 426 nm responsible for the blue luminescence originated from sulfur vacancies Vs (Wang et al. 2011),while  (Tuan et al. 2018). The band at ~ 420 nm persisted and was related to radiative carriers recombination between the sulfur vacancy sites (Vs) and the valence band (Mansur and Mansur 2011   Mn 2+ ions was no longer negligible. Such coupling generated a reduction of the manganese emission band with a slight red shift caused by a non-radiative cross-relaxation between these ions (Fang et al. 2010). On the other hand, the decrement of the intensity of the overall emission may be because Mn 2+ ions settled at the superficial or interstitial positions of the ZnS host. Some of these defects acted as non-radiative recombination centers in the host as explained by Mohagheghpour et al. (Mohagheghpour et al. 2009). In order to provide a better understanding of the different emission bands of Mn-doped ZnS NCs, an energy level diagram was proposed where the possible radiative and non-radiative transitions of charge carriers are illustrated, see Fig. 10.

MO dye adsorption and its analysis
In recent years, II-VI semiconductors especially zinc sulfur (ZnS) have generated considerable interest in wastewater treatment due to their nontoxicity, physical and chemical stability and high catalytic efficiency (Chang et al. 2017;Bujňáková et al. 2017;Ajibade et al. 2020). ZnS nanoadsorbents have advantages in terms of their abundance, low-cost, good stability, non-toxicity, and good light UV absorption. Therefore, they are considered suitable for adsorption and photocatalysis applications. Herein, a simple colloidal chemical route for the synthesis of undoped and Mn-doped ZnS nano-adsorbents using thioglycolic acid (TGA) as a stabilizing agent is reported. This synthesis route offers a high purity, low-cost, simple implementation, and low energy consumption. The surface functionalization of nanocrystals by TGA molecules modifies their adsorption activities, morphology, particle size, optical properties, and mechanical stability (Wieszczycka et al. 2021). It is worth noting that the surface area and small particle size of ZnS contribute to enhance the adsorption efficiency. Furthermore, the combination of adsorption and photocatalyst is an interesting approach to degrade organic pollutants in wastewater treatment. Figure 11 reports the adsorption kinetic of MO dye on undoped and Mn-doped ZnS nano-adsorbent in dark conditions. Dye adsorption on the nano-adsorbent was fast for the first 5 min, then continued with a slower rate during 10-45 min, and reached the equilibrium condition after approximately 60 min. The experimental kinetic showed that doping with Mn ions increased the MO adsorption capacity. Mn-doped ZnS NCs showed a better adsorption than undoped ZnS NCs. This can be explained by the decrease in size under the effect of doping, which generated an increase in the surface-to-volume ratio that provided more sites on the surface. The pseudo-second-order kinetic rate equation where K 2 (mg/g.min) is the pseudo-second-order rate constant of adsorption, q e and Q(t) (mg/g) are the adsorption capacity at equilibrium and time t, respectively. Table 3 shows that K 2 increased as a function of the Mn doping amount. Overall, the adsorption kinetic of MO on undoped and Mn-doped ZnS nanocrystals could involve a rate limiting adsorption on the vicinity of Mn-doped ZnS surface. This process can be associated to an exchange interaction between MO-and Mn-doped ZnS adsorbent surface. This interaction increased with Mn doping in the ZnS.
On the other hand, the experimental isotherm showed that MO adsorption capacity of undoped ZnS increased from 6.21 to 25.91 mg/g when the dye concentration also increased from 10 to 50 mg/L, which corresponded to an adsorption efficiency of 64.56%. However, the Mn (12%)doped ZnS NCs showed an adsorption efficiency of 71.16%, which proved that the doping with manganese ions enhanced the adsorption. The adsorption capacity of undoped and Mndoped ZnS nanocrystals could be due to the ligand on the surface of inorganic TGA particles, which could contribute with positively charged sites on the surface of ZnS due to the zinc interstitials (Zn i ), Mn defects, and sulfur vacancies (Vs) that were generated during the synthesis (such as nonstoichiometric defects and dangling bonds). These sites generated attraction forces with the negatively charged molecule of MO (with sulfonate group R-SO 3 − ). The adsorption of the MO molecule on ZnS adsorbent surface, with different doping percentage of Mn, was analyzed with the Hill adsorption model obtained from statistical physics. This where n D is the number of MO molecules captured by the adsorption receptor site of adsorbent surface, D M is the density of receptor sites, and C 1/2 is the concentration at half-saturation of MO on the ZnS adsorbent surface with different doping of Mn. This last parameter can be associated to the adsorption energy using the next expression (Bouzid et al. 2016Mohamed et al. 2018;Atrous et al. 2019; Pang et al. 2020;Bel Haj Mohamed et al. 2021;Sghaier et al. 2021).
where C S is the MO dye solubility (i.e., 5.20 g/L). MO adsorption data were fitted with this model and their corresponding parameters are listed in Table 4, while the model fitting is illustrated in Fig. 12.
Note that the parameter n D can be used to describe the number of dye-adsorbed molecules per adsorption site and gives also information about the adsorption position. Table 4 shows that n D = 3.73 indicates that MO dye molecules could   (Li et al. 2019a(Li et al. , 2019bChaukura et al. 2017;Jafari et al. 2016).These results showed that ZnS NCs had better adsorption performance thus suggesting that these NPs could be promising in water purification and treatment. On the other hand, the calculated adsorption energy of MO on Mn-doped ZnS nanocrystal surface ranged from 17.05 to 18.05 kJ/mol thus corresponding to physical interactions associated to hydrogen bonding and van der Waals forces, see Table 4. Figure 13 shows the absorption spectra of MO solutions using undoped and Mn-doped ZnS nanocatalysts for various It was observed the color disappearance after 6 min due to the breaking of the azo groups (N = N) responsible for the orange coloration, which indicated that the aromatic compounds were attacked by hydroxyl radicals forming radical intermediates (Chan et al. 2011). These experiments demonstrated the high photocatalytic efficiency of pure ZnS NCs, which generated a rapid degradation of MO molecules and the solution decolorization.

Photocatalytic activity
To quantify the photocatalytic activity of ZnS NCs, the degradation efficiency, degradation constant, and half-life were calculated using the following equations (Dhandapani et al. 2016): where A 0 , A t , C 0 , and C t are the initial absorbance, the absorbance after irradiation, the initial concentration of the solution, and the concentration of the dye after irradiation for time t, respectively.
The values of degradation efficiency of undoped ZnS NCs are given in Table 5. These results revealed a high degradation efficiency of MO after just 6 min of irradiation (~ 81.4%). After 120 min of irradiation, the degradation efficiency was 98.5% and the solution became uncolored. This high efficiency may be due to the small size of the TGA molecules used as stabilizer in addition to the largest specific surface of NCs, thus generating several available adsorption sites.
The effect of manganese doping on the photocatalytic activity of MO is illustrated in Fig. 13. The solution decolorization was obtained at higher contact time than that observed with the undoped ZnS NCs. Table 6 shows that doping generated an impending effect for the MO degradation and the corresponding efficiency was lower than that observed using undoped ZnS. The highest MO degradation efficiency was obtained for 8% manganese concentration of doped ZnS NCs. In terms of degradation time, this doping concentration of ZnS NCs required 120 min for obtaining an efficiency of 87.1%.
The pseudo-first-order kinetic equation was used to calculate the rate constant (k) for MO degradation using pure and Mn-doped ZnS NCs where the corresponding plots of ln (A 0 /A t ) versus irradiation time are given in Fig. 14. The rate constants ranged from 0.281 to 0.004 min −1 , see Table 6. Indeed, the trapping of charge carriers with pure ZnS was provided by sulfur vacancies, while multiple traps   (Zong et al. 2014). These defects produced discrete energy levels within the forbidden band (E g ) of ZnS NPs and, therefore, the possibility of recombination of electron and hole inside the NPs increased (Štengl et al. 2009), which resulted in a reduction of the photocatalytic activity of these NCs with low MO degradation. In view of that, it has been suggested that as the concentration of Mn 2+ ions increased, a competitive process of charge carrier trapping between V S and Mn 2+ occurred (Sergeeva et al. 2018). The rate of MO degradation depended on nanocatalyst properties such as specific surface, nanocrystal size, composition, crystallinity, and gap energy (Chaturvedi et al. (Ashkarran 2014). As a result, the presence of high photoluminescence generated a low photocatalytic activity. Consequently, undoped ZnS NCs are a promising nanocatalyst for the degradation of MO dye and its application can be extended to degrade other organic molecules.

Toward an industrial efficient solar-activated nanocatalyst
As stated, the degradation of organic pollutants is a necessity due to environmental pollution problems. Consequently, the industries request the most efficient degradation process but with a low cost. One way to achieve this goal is the use of solar-activated efficient photocatalyst. In this context, additional photodegradation experiments were carried out under solar light irradiation to evaluate the usability of undoped ZnS NCs for wastewater purification. Before the sunlight irradiation, the suspensions were magnetically stirred for 120 min to allow adsorption-desorption equilibrium between MO molecules and NCs. Then, the mixture was submitted to sunlight irradiation between 11 a.m. and 2 p.m. to evaluate the dye degradation. Figure 15 presents the UV-visible absorption spectra of MO degradation using ZnS NCs and the degradation rate as a function of time recorded after sunlight exposition. Results of Fig. 15 and Table 7 showed the total decolorization of the solution after 120 min of sunlight exposition. This experiment confirmed the degradation efficiency of these NCs. Overall, the percentage of dye degradation increased significantly with the exposition time where 97.4% was the maximum decolorization efficiency with a degradation constant K of 0.045 min −1 . In fact, there was no a significant difference with the efficiency obtained under UV irradiation (98.5%). It is convenient to note that sunlight is often exploited for the semiconductor irradiation with the aim of performing the photodegradation of pollutants thus contributing to reduce the cost of water treatment methods and the environmental pollution compared to UV lamps. These results proved that undoped ZnS NCs exhibited a strong photocatalytic activity for MO dye. This behavior may be explained by the large band gap energy of ZnS, which caused a higher redox potential of e-h pairs and, consequently, a higher photocatalytic performance (Herrmann et al. 1984

Mechanism of photocatalytic activity
The mechanism of photocatalytic degradation of methylene orange by undoped and doped ZnS nanocatalysts could be explained as follows. First, UV light excitation of ZnS NCs generated the electron-hole pairs (e − /h + ) in ZnS. The (OH * ) radical was formed from a simple oxidation of water or hydroxyl ion (OH − ) by photo-generated holes. Similarly, superoxide radicals (O 2 *− ) were formed in water from a reduction reaction of the dissolved dioxygen in water by photo-generated electrons. These highly reactive hydroxyl radicals (OH * ) and superoxide radicals (O 2 *− ) reacted with MO dye molecule adsorbed on ZnS nanoparticles ensuring the degradation. The next equations describe this mechanism: Figure 16 summarizes the different stages of this photocatalytic process. The first step corresponded to the OH * + methylene orange → degradation products + CO 2 + H 2 O transfer of the MO molecule from the aqueous phase to the surface of NCs. In the second step, the organic molecules were adsorbed on the surface of the catalyst followed by the movement of photo-generated e − and h + charges to the catalyst surface. The positive hole in the NCs reacted with a water molecule to produce hydrogen gas and free hydroxyl radical (OH*), which was the most important and powerful oxidizing radical for the mineralization of adsorbed or free MO on the catalyst surface. Photo-induced electrons were easily trapped by electronic acceptors, such as adsorbed O 2 , to produce a superoxide anion radical (O 2 *− ). This reaction limited the recombination phenomena and thus improved degradation efficiency. Finally, these different radicals reacted with the MO to form intermediate compounds before the mineralization to CO 2 and H 2 O.

Conclusion
Nanocrystals of undoped and Mn-doped ZnS were successfully synthesized by colloidal precipitation reaction method and applied in the photodegradation of MO dye. FTIR results confirmed the functionalization of nanocrystal surface by TGA. XRD analysis indicated the formation of nanocrystals with two crystalline phases, hexagonal and cubic, identical to the solid material and with a relatively large size distribution. The average size of ZnS nanocrystals calculated by using Debye-Scherrer formula varied from 2.5 to 7.1 nm. The optical absorption spectra showed the appearance of an absorption edge shifted toward short wavelengths compared to that of the solid-state semiconductor ZnS. PL spectroscopy indicated that the emission intensity increased with the doping rate up to 8%, and then, it decreased. Modeling results showed that the Hill model explained the adsorption of MO on the nanocrystal surface at 25 °C and the dye removal followed a pseudo-second-order reaction kinetics. Adsorption capacity of doped ZnS-TGA was 26.98 mg/g and this property contributed to the direct interaction of photogenerated electrons with adsorbed MO molecules and hence facilitated a quick photodegradation. Dye adsorption process was multi-molecular via physisorption involving hydrogen bonding and van der Waals interactions. The undoped ZnS-TGA nanocrystals exhibited high photocatalytic activity to degrade MO dye under UV and sunlight. Consequently, the pure ZnS-TGA nanocatalysts are suggested as a promising photocatalyst for wastewater purification. On the other hand, Mn-doped ZnS-TGA nanocrystals showed superior optical and physicochemical characteristics making them good candidates for fluorescence-based applications.
Data availability Data and materials are contained within the article.

Declarations
Ethics approval and consent to participate Not applicable for this study.
Consent to publish All authors consent to this publication.

Competing interests
The authors declare no competing interests.