Supporting of mixed ZnS–NiS semiconductors onto clinoptilolite nano-particles to improve its activity in photodegradation of 2-nitrotoluene

Abstract

Photodegradation of a 2-nitrotoluene (2-NT) aqueous solution was studied using ZnS–NiS supported onto clinoptilolite nano-particles (ZnS–NiS–NCP) and a Hg lamp as the radiation source. A planetary ball-mill was used for preparation of nano-particles of clinoptilolite (NCP). The photocatalyst was prepared via sulfiding of the Zn–Ni exchanged NCP and characterized by Fourier transformation infra red spectroscopy (FT-IR), X-ray diffraction (XRD), UV-Vis diffuse reflectance spectroscopy (DRS) and transmission electron microscopy (TEM). UV-Vis spectrophotometric measurements were performed for the determination of degradation and mineralization of the pollutant. Degradation of the pollutant was also confirmed by chemical oxygen demand (COD) and high performance liquid chromatography (HPLC). The optimal operation parameters were found to be: 0.75 g L⁻¹ of the ZnS_9.7%–NiS_3.0%–NCP catalyst and 55 ppm of 2-NT at pH 11. Unsupported ZnS and NiS showed smaller degradation activities than the supported one. The degradation process obeyed first-order kinetics.

---

1. Introduction

Nitro derivatives of toluene such as 2-nitrotoluene (C₇H₇NO₂) cause contamination of groundwater and soil at numerous sites located in packing, explosive manufacturing plants, storage and military-related facilities. They are very toxic, mutagenic and carcinogenic.¹ For example, 2-nitrotoluene causes flushing of the face, dizziness, dyspnea (difficult breathing), cyanosis, nausea, vomiting, muscular weakness, increased pulse and respiratory rate, irritability and convulsions [2-nitrotoluene MSDS]. In general, industrial effluents containing toxic, volatile and refractory organic pollutants cause severe environmental problems and thus their removal or degradation is required.² In this regard, different methods have been developed for the removal of various volatile and non-volatile pollutants from water and wastewater. Common biological treatments are slow or ineffective, while the traditional physicochemical treatments such as adsorption on activated carbon and nano-filtration have inherent limitations in applicability, effectiveness and cost.³ They also transfer the pollutants from one phase to another, leaving a problem of disposal of the transferred materials.⁴

In recent two decades, advanced oxidation processes (AOPs) such as photo-Fenton, photo-oxidation, photo-reduction and photocatalytic technologies have been widely used for the removal of different organic pollutants from water.⁵ Although common homogeneous Fenton systems offer a cost-effective source of hydroxyl radicals, they have following drawbacks which can limit its industrial applications: (i) the tight working pH range, (ii) the need for recovering the precipitated catalyst after treatment and (iii) deactivation by some ion-complexing agents like phosphate anions. The resulting sludge may also contain organic substances as well as heavy metals and must be treated further, thus increasing the overall costs. Heterogeneous photocatalysis has proven to be the most suitable for the widespread environmental applications due to its chemical inertness, strong oxidizing power, cost effectiveness and long-term stability.^6,7 In this method, when a semiconductor catalyst is exposed to photons with at least as much energy as its band gap energy, excitation of an electron from the valence band to conduction band produces electron–hole pairs (e⁻/h⁺).^8,9 The produced holes can oxidize H₂O or OH⁻ (in aqueous solutions) to produce the most oxidative ˙OH radicals, while the conduction band electrons will trap at surface sites and finally absorb by dissolved oxygen to produce superoxide radical anions (O₂˙⁻).¹⁰ These radicals (especially hydroxyl one) are strong and nonselective oxidant species that react with the majority of organic pollutants to degrade them into smaller fragments and finally into water and carbon dioxide.^11,12

Among various semiconductors, ZnS¹³ and NiS¹⁴ with band gap energies about 3.7 and 0.5 eV, respectively, have potential applications as photocatalyst for the degradation of different organic pollutants.¹⁵ NiS is a potential transformation toughening agent for semiconductor materials, while nickel disulfide adopts the pyrite structure and exhibits semiconducting properties. Thus, the synthesis of the 3d transition metal sulfides has attracted great interests for several decades.¹⁶

Since the reactions mostly take place on the surface, supported semiconductor on a suitable adsorbent significantly increases the degradation efficiency, because suitable supports can concentrate molecules of pollutants near semiconductor particles and adsorb the generated intermediates.¹⁷ Among the different supports, zeolites are important candidates due to their special properties.¹⁸ Clinoptilolite, the most abundant zeolite in nature, is iso-structural with heulandites with Si/Al > 4 and monoclinic framework consisting of a ten-membered ring (7.5 × 3.1 Å) and two eight member rings (4.6 × 3.6 Å, and 4.7 × 2.8 Å).¹⁹ Due to its abundant in Iran, it was used in this work for decreasing the cost of the proposed method. Also, using of natural zeolite instead of the synthetic one is environmentally friendly and in accordance with the goal of green chemistry.

In this work, ZnS–NiS supported onto clinoptilolite nano-particles was prepared via a precipitation process followed by an ion exchange process. The obtained composite was used as a photocatalyst in the photodegradation of 2-nitrotoluene. Effects of different experimental parameters including: amount of the catalyst, initial concentration of 2-NT, initial solution pH and concentration of ion exchange solution were studied on the degradation extent of 2-NT.

2. Experimental

2.1. Materials

Natural clinoptilolite (CP) tuffs, belong to the Semnan region in the north-east of Iran, purchased from Afrand Tuska Company (Isfahan, Iran). 2-Nitrotoluene (C₇H₇NO₂), zinc nitrate, nickel nitrate, sodium sulfide and other used reagents with analytical grade purity obtained from Merck. pH of the solutions was appropriately adjusted by hydrochloric acid or sodium hydroxide solution (not buffered). Distilled water was used throughout the experiments.

2.2. Preparation of nano-particles and catalysts

Natural clinoptilolite was mechanically pretreated by crushing in an agate mortar and sieving in analytical sieves for separating of the particles with mesh size of −352/+400 (44 to 37 μm). The obtained powder was used for preparation of nano-particles using a planetary ball-mill. In order to remove any water soluble and magnetic impurities, the obtained powder was heated at 70 °C in distilled water for 8 h on a magnetic stirrer (n = 3). In order to prepare a sample with fixed water content, after centrifuging, washing and drying, the pretreated powder was stored in a closed bottle containing saturated sodium chloride solution for 2 weeks.

For preparation of Ni(II)–Zn(II)-exchanged clinoptilolite nanoparticles (Ni–Zn–NCP), 10 g of NCP powder was added to 100 mL aqueous solution containing 0.1 M Ni²⁺ and Zn²⁺ cations (as nitrate salt) and the suspension was shaken for 48 h at room temperature. The procedure was repeated again to complete the ion exchange extent. The suspension was centrifuged and the solid material air dried. Finally, sulfiding of the ion-exchanged sample was carried out in a 0.1 M Na₂S solution. For this goal, 1 g Ni–Zn–NCP was added to 10 mL 0.1 M Na₂S solution and stirred for 2 h. The suspension was centrifuged and washed with copious amounts of water until the filtrate became free from the sulfide ions. Similar method was used for supporting NiS and ZnS onto micronized clinoptilolite particles. To investigate the effect of the extent of loaded zinc and nickel sulfides on degradation process, different catalysts were prepared by ion exchanging of the zeolite with different concentrations of Zn²⁺ and Ni²⁺ covering the rang from 0.05 to 0.3 M.

2.3. Characterization and instruments

All samples were analyzed by X-ray diffraction pattern (XRD) by using a Bruker diffractometer (D8 Advance) with a Ni-filtered copper radiation (Kα=1.5406 Å). Diffuse reflectance UV-Vis spectra (DRS) were recorded by using a Shimadzu UV-3101PC spectrometer equipped with an integrating sphere in the wavelength range of 190–900 nm. FT-IR spectra of the samples, on KBr pellets were recorded with a Nicolet single beam FT-IR (Impact 400D) spectrometer. An UV-Vis spectrophotometer (Carry 100 Scan) was used for recording the absorption spectra of samples. HPLC analysis of samples was performed by an Agilent Technologies 1200 Series instrument with absorbed electron detector, column XDB-C₁₈ (L = 15 cm, id = 4.6 mm and particle size = 5 mm) and UV detector. TEM images of samples were recorded using Transmission Electron Microscope S3500N with Absorbed Electron Detector S-6542 (Hitachi Science System Ltd). Atomic Absorption Spectrometer Perkin Elmer Analyst 400 (air–C₂H₂, λ = 232.0 nm for nickel and 213.9 nm for zinc) was used for measuring of zinc and nickel in solutions. The pH of point of zero charge of the catalyst, pH_pzc, was determined by the method reported in literature.²⁰

2.4. Photocatalytic degradation experiments

Before performing photocatalytic experiments, the surface adsorption extent of the pollutant on the surface of the ZnS–NiS–NCP photocatalyst was measured at dark condition. For this goal, an appropriate amount of the photocatalyst was added to 20 mL 55 ppm 2-NT aqueous solution to obtain a suspension containing the desired amount of the catalyst (for example: 0.75 g L⁻¹ of the catalyst as the optimized amount, see Section 3.2.3). For performing the photodegradation experiments, the same suspensions were added to a cylindrical Pyrex-glass cell (5 cm inside diameter and 10 cm height) as a reactor. The suspensions were irradiated with a medium pressure Hg lamp (55 W, Philips, located 30 cm above the reactor), under continuous magnetic stirring. The blank solutions had the same condition of analyte solution without the catalyst. During the irradiation, the suspension was sampled out at regular time intervals and the withdrawn suspension was centrifuged. Then absorbance of the supernatants was recorded at λ_max = 265 nm (using Carry 100 Scan UV-Vis spectrophotometer) and used to calculate the degradation extent by the following equation:^14,15where A_o and A_t are respectively the initial and final absorbance values of 2-NT, which respectively relate to the initial (C_o) and final (C) concentrations according to Beer–Lambert law. The following pseudo first-order kinetics (Langmuir–Hinshelwood) model was used for calculation of the photodegradation rate constants:^14,17,19

To confirm the mineralization of the mixture, remained COD of the photodegraded solutions was measured at regular time intervals using closed reflux titrimetric method.²¹ The degradation of the pollutant was also confirmed by HPLC²² (ZORBAX EDB C₁₈ column, L = 15 cm, id = 4.6 mm, particle size = 5 μm, injection volume = 20 μL, flow rate = 0.7 mL min⁻¹, 1 : 1 water/methanol as mobile phase and UV-Vis detector at λ = 254 nm).

3. Results and discussion

3.1. Characterization

3.1.1. Determination of loaded ZnS and NiS. The amount of loaded ZnS and NiS onto the NCP particles was determined by atomic absorption spectroscopy and the corresponding results are collected in Table 1. As expected, the amount of loaded zinc and nickel was increased with increasing in the concentration of both cations in the ion-exchange solutions. As shown, more zinc cations were loaded into the zeolite channels than Ni(II) because of different hydrated radii of cations. In Table 1, the abbreviations of the prepared catalysts are also shown. As shown later (see Section 3.2.3), the catalyst containing 7.1% wt of ZnS and 2.7% wt of NiS (C₈ = ZnS_9.7%–NiS_3.0%–NCP) showed the best degradation efficiency towards the proposed pollutant. Hence, this catalyst was used in characterization techniques.

Table 1 Definition of the catalysts (averaged based on 3 replicates)

Abbreviation of catalysts	CM^2+ (ion-exchange solution (M))		meq cation/g of the catalysts		ZnS%	NiS%
	Zn^2+	Ni^2+	Zn^2+	Ni^2+
C1: ZnS4.1%–NCP	0.1	—	0.85	—	4.12 ± 0.06	—
C2: NiS2.9%–NCP	—	0.1	—	0.65	—	2.88 ± 0.05
C3: ZnS2.5%–NiS3.0%–NCP	0.05	0.1	0.51	0.67	2.53 ± 0.04	3.02 ± 0.05
C4: ZnS4.1%–NiS1.2%–NCP	0.1	0.05	0.84	0.27	4.07 ± 0.10	1.22 ± 0.03
C5: ZnS4.5%–NiS2.9%–NCP	0.1	0.1	0.93	0.64	4.46 ± 0.08	2.92 ± 0.06
C6: ZnS4.1%–NiS7.3%–NCP	0.1	0.2	0.85	1.63	4.08 ± 0.09	7.27 ± 0.11
C7: ZnS4.0%–NiS8.3%–NCP	0.1	0.3	0.81	1.83	4.03 ± 0.07	8.33 ± 0.13
C8: ZnS9.7%–NiS3.0%–NCP	0.2	0.1	1.98	0.67	9.72 ± 0.12	3.04 ± 0.08
C9: ZnS10.8%–NiS2.8%–NCP	0.3	0.1	2.21	0.61	10.83 ± 0.18	2.78 ± 0.06

---

3.1.2. X-ray diffraction studies. XRD patterns of the raw NCP, ZnS_4.1%–NCP (C₁), NiS_2.9%–NCP (C₂) and ZnS_9.7%–NiS_3.0%–NCP (C₈) samples are shown in Fig. 1A. Typical characteristic lines in the XRD pattern in Fig. 1A(a) are in good agreement with the XRD data of clinoptilolite according to JCPDS# 38-1389 and literature.^19,23 Based on the XRD results, the used natural tuffs include clinoptilolite phase as major component and slight amounts of quartz (3.8%) and cristobalite (7.1%). By loading NiS, ZnS and NiS–ZnS onto NCP, some changes took place in the XRD pattern. In pattern (b), the located peaks at 2θ = 30.9° and 33.9° can be indexed to rhombohedral phase of nickel sulfide (according to JCPDS 65-3686),²⁴ while in pattern (c), the located weak peaks at the 2θ = 29.1°, 48.6° and 56.5° can be indexed for cubic phase of zinc sulfide (according to JCPDS 77-2100).¹⁴ In all loaded samples especially ZnS_9.7%–NiS_3.0%–NCP (C₈), the intensity of XRD lines of NCP was decreased. In addition, more intense XRD lines for ZnS were observed in the pattern of the hybridized ZnS_9.7%–NiS_3.0%–NCP (C₈) sample with respect to ZnS_4.1%–NCP (C₁) because of higher ZnS loading in the hybridized sample. These observations can be considered as sufficient evidences for loading of NiS and ZnS by NCP.

Fig. 1 (A) XRD patterns of NCP (a), ZnS_4.1%–NCP (C₁) (b), NiS_2.9%–NCP (C₂) (c) and ZnS_9.7%–NiS_3.0%–NCP (C₈) (d). (B) FT-IR spectra of NCP (a), ZnS_4.1%–NCP (C₁) (b), NiS_2.9%–NCP (C₂) (c) and ZnS_9.7%–NiS_3.0%–NCP (C₈) (d).

The average particles size of samples was estimated by the following Scherer's equation using the XRD line broadening method:

D = kλ/β cos θ (3)

where λ is the wavelength of X-ray used (Cu Kα radiation λ = 1.5406 Å), β the width of the line at the half maximum intensity and k constant, 0.89.²⁵ Different peaks widths were used for the estimation of particles size but the most intense peak at 22.6° was selected and the corresponding sizes of 40 nm and 50–60 nm were respectively estimated for NCP and loaded samples.

3.1.3. FT-IR studies. FT-IR spectra were used to investigate the influence of zinc and nickel sulfides on the zeolite framework. Representative spectra of NCP, ZnS/NCP, NiS/NCP and ZnS–NiS/NCP in the wavenumber range of 450–2500 cm⁻¹ are shown in Fig. 1B. In general, four groups of bands including: the internal Si–O(Si) and Si–O(Al) (1200–400 cm⁻¹), the zeolitic water (1600–2500 cm⁻¹), pseudo-lattice vibrations of structural units (500–700 cm⁻¹) and the bands related with lattice vibrations (below 400 cm⁻¹) can be observed in the IR spectrum of clinoptilolite.^26–29

Comparison of the spectra of NCP and loaded NCP samples shows that all spectra are very similar, confirming that the NCP structure remained un-change during the ion exchange and precipitation processes which in turn show high stability of the NCP structure. Breck^30,31 showed that by entering transition metal cations no significant changes occurred in IR spectra of the loaded zeolite and only slight shifts occurred in the peak position for the located peaks below 1000 cm⁻¹. Our results have good agreement with Breck's results confirming relative loading of ZnS and NiS on NCP particles. Because of small amount of loaded NiS and ZnS in the NCP no significant peaks were observed in the IR spectra of loaded samples and this only caused to slight shifts in the peak position for loaded samples.

3.1.4. DRS studies. DRS spectra of the NCP, ZnS–NCP, NiS–NCP and ZnS–NiS–NCP samples were recorded and the corresponding results are shown in Fig. 2A. As shown, clinoptilolite shows absorptions maximum at the wavelengths of 216 and 245 nm due to (pπ to dπ) transitions of Si–O and Al–O.³² After NiS loading, slight shifts were observed at 208 and 252 nm and new weak peak was also observed at 600 nm (Fig. 2A(c)). When NiS was entered into the NCP framework the longest wavelength d → d weak transition was followed. The presence of ZnS into the zeolite only increased the absorption intensity which enhances the photocatalytic activity due to the enhanced formation of the photo-electrons and photo-holes.^33,34 In the ZnS–NiS/NCP spectrum, absorption intensity was increased along with broadening of the absorption bands towards 500 nm. Similar peak broadening was also observed for single loaded metal sulfide samples, all confirming formation of ZnS–NiS (and ZnS or NiS) onto the zeolite surface.

Fig. 2 (A) Diffuse reflectance spectra of NCP (a), ZnS_4.1%–NCP (C₁) (b), NiS_2.9%–NCP (C₂) (c) and ZnS_9.7%–NiS_3.0%–NCP (C₈) (d). (B) TEM images of NCP. (C) TEM images of ZnS_9.7%–NiS_3.0%–NCP (C₈).

3.1.5. TEM analysis. TEM images of NCP and ZnS_9.7%–NiS_3.0%–NCP (C₈) samples are respectively shown in Fig. 2B and C. As shown, particles size of raw NCP was remained in nano scale after loading of ZnS–NiS semiconductors. In general, TEM results are in accordance with particles size determined by XRD results, confirm the production of clinoptilolite nanoparticles during ball mill process.

3.2. Photodegradation studies

3.2.1. Effect of semiconductor and support. In preliminary experiments the effect of surface adsorption and direct photolysis on the removal of the proposed pollutant was studied. As shown in Fig. 3, these removal methods do not have significant role in the removal of 2-NT. Removal of 2-NT due to surface adsorption was below 13%. In order to omitting surface adsorption, pre photodegradation process was performed via stirring the suspensions for 160 minutes in the dark for achieving adsorption–desorption balance. Direct photolysis, direct destruction of pollutants by light, had also photocatalytic activity yield lower than 16%. Irradiation of the energetic photons to 2-NT cleaves C–N bond and produce NO₂˙ and CH₃-ph˙ radicals. The single electron of CH₃-ph˙ radical can start resonance with aromatic ring, and generate a relative stable structure and cause the CH₃-ph˙ to be less active than OH radicals.^35,36 NO₂˙ radical has also lesser activity than hydroxyl radicals. In addition, two CH₃-ph˙ radicals can combine and deactivate the CH₃-ph˙ radicals. These reasons cause to smaller efficiency of direct photolysis in the degradation of 2-NT. In all calculated degradation percentages in this work, the effect of direct photolysis in the removing of 2-NT was eliminated as follows. The absorbance of the photocatalytic degraded solution (in the presence of the catalyst) was recorded as A_t at time ‘t’. At the same time the absorbance of the irradiated 2-NT solution without photocatalyst was recorded as A_o. Finally, using these absorbencies (and corresponding concentrations) and eqn (1) the degradation percentages were calculated and the effect of direct photolysis was eliminated.

Fig. 3 Photodegradation, photolysis and surface adsorption efficiency of 2-NT at the irradiation time of 20 min, [2-NT]₀ = 50 ppm, catalyst = 0.1 g L⁻¹, pH = 6.3.

In the second step, the effect of loading of NiS and ZnS onto micronized clinoptilolite (MCP) was studied on the degradation of 2-NT and the corresponding results are summarized in Fig. 3. Typical plots of ln(C/C_o) versus time (inset of Fig. 3) were used for calculation of the rate constants based on the slopes of linear segment of these curves (linear curves were drowning till 160 min irradiation time). Based on the rate constants in Table 2 the following trend was observed for the degradation activity of the used catalysts: NiS–MCP > ZnS–NiS–MCP > ZnS–MCP. The same trend for supported nano-particles obtained as: NiS–NCP > ZnS–NiS–NCP > ZnS–NCP (see Fig. 5A).

Table 2 Rate constant values for the degradation of 2-NT as a function of different parameters

Investigated factor	Value		k × 1000 (min^−1) (n = 3)
Semiconductors	NiS–MCP		3.197 ± 0.097
	ZnS–NiS–MCP		2.763 ± 0.032
	ZnS–MCP		2.600 ± 0.014
Dose of semiconductors	ZnS%	NiS%
	9.7	3.0	5.633 ± 0.023
	4.1	7.3	4.693 ± 0.105
	4.5	2.9	4.003 ± 0.127
	4.1	1.2	3.870 ± 0.010
	4.0	8.3	3.673 ± 0.146
	2.5	3.0	3.460 ± 0.080
	10.8	2.8	3.243 ± 0.160
Catalyst mass (g L^−1)	0.75		6.483 ± 0.140
	1.00		5.723 ± 0.213
	0.50		4.975 ± 0.258
	0.25		3.977 ± 0.221
	1.25		3.617 ± 0.250
C2-NT (ppm)	55		12.120 ± 0.390
	60		9.990 ± 0.323
	65		9.243 ± 0.252
	50		7.503 ± 0.285
	45		4.137 ± 0.284

---

Fig. 4 Schematic diagram of energy levels of the semiconductors.

Fig. 5 (A) Effect of role of support on the photodegradation efficiency of 2-NT at the same conditions with Fig. 3, (B) effect of dose of semiconductors on the photodegradation efficiency of 2-NT, [2-NT]₀ = 50 ppm, catalyst = 0.1 g L⁻¹, pH = 6.3.

Comparison of the results showed that the activity of NiS–MCP is more than ZnS–MCP because of lower band gap energy of NiS (0.5 eV)³⁷ than ZnS (3.7 eV).³⁸ The conduction and valence bands potentials of these semiconductors are larger that the corresponding redox potentials of H⁺/H₂ and H₂O/O₂ and the photogenerated electrons and holes can be separated efficiently. In fact, valence band electrons of NiS can be excited more easily than ZnS which contributes the production of more electron–hole pairs. A typical schematic diagram of energy levels of the semiconductors is presented in Fig. 4.

It would be expected that hybridized ZnS–NiS–MCP system to be more effective than ZnS–MCP and NiS–MCP systems because combining of two semiconductors creates a new energy gap which lays between the band gap energies of single semiconductors.³⁹ But in our experiments, NiS–MCP was more active than ZnS–NiS–MCP. In our idea, lower activity of ZnS–NiS–MCP can be related to un-correct mass ratio of NiS/ZnS. Hence, the effect of mass ratio of NiS/ZnS was studied to improve activity of the hybridized photocatalysts which of results will investigate in Section 3.2.2.

In the next step, the effects of raw zeolite, bulk and supported semiconductors and clinoptilolite nano particles (NCP) on the removal of 2-NT were studied and the corresponding results are summarized in Fig. 5A. Results show that the raw MCP and NCP have no significant photocatalytic effect. Bulk zinc and nickel sulfides have also a little photocatalytic effect with respect to the supported one (Fig. 5A). Bulk ZnS–NiS has slight enhancement in activity with respect to the bulk semiconductors alone which confirm the positive role of hybridation. Aggregation of semiconductors particles in the bulk samples decreases the effective surface area of particles available for absorbing photons, results a decrease in the generation of hydroxyl radicals. But, in the supported samples, zeolite prevents to aggregate ZnS, NiS or ZnS–NiS particles due to loading of them on definite ion exchange sites of zeolite, hence the degradation efficiency was increased. These observations are in agreement with literature.^40–42 Pure ZnS, NiS or NiS–ZnS have only a simple bond (Ni)Zn–S, but when it forms in the zeolite structure the new bonds of (Ni)Zn–O–Al will form. In this condition, ZnS–NiS is being a part of aluminosilicate framework which has a high thermal and chemical stability. Fitting of ZnS–NiS in the zeolite channels causes to migration of produced photoexcited electrons throughout the zeolite structure and hence prevent from recombination of electron–hole pairs which increases the degradation efficiency. In addition, due to high adsorption capacity of the zeolite, more 2-NT molecules can reach near the catalyst surface where hydroxyl radicals produced. This in turn enhances the chance of OH radicals to attack the 2-NT molecules.

As shown, supported ZnS–NiS onto the nano-particles of clinoptilolite (ZnS_4.5%–NiS_2.9%–NCP) has higher efficiency with respect to micronized one due to increase in the effective surface area of nano-particles.

3.2.2. Effect of the dose of semiconductors. As reported, the efficiency of the photocatalytic processes depends on the amount of semiconductor supported on zeolite surface.⁴³ As mentioned in Section 3.2.1, it would be expected that the activity of the mixed NiS–ZnS–NCP photocatalyst can change with changing the NiS/ZnS ratio. Hence, as shown in Table 1, different NiS–ZnS–NCP catalysts were prepared by ion exchanging of NCP in Ni(II)–Zn(II) solutions with different concentrations covering the range from 0.05 to 0.3 M with respect to each cation. According to atomic absorption results of the zinc and nickel determination, their ZnS and NiS contents were calculated and summarized in Table 1. Photodegradation efficiencies of the prepared catalysts were studied and the corresponding results are shown in Fig. 5B. Rate constants are also collected in Table 2.

As shown, changing the dose of semiconductors led to change in the degradation efficiencies of the hybridized catalysts, so the catalyst containing 9.7% wt of ZnS and 3.0% wt NiS showed the best efficiency in the degradation of 2-NT.

By increasing in the concentration of Ni(II) and Zn(II) in solutions from 0.05 to 0.3 M, entered cations into NCP was increased. Correspondingly, the degradation activities of the prepared catalysts in 0.05 to 0.2 M solutions were increased and thereafter decreased. We believe that the electron–hole pairs were produced in NiS particles (due to the ability to attract a wider range of electromagnetic waves because of its lower energy gap) and then the produced conduction band electrons migrate to ZnS energy levels. This prevents the electron–hole recombination and hence increasing in the NiS dose till to 7.3% wt (in 0.2 M Ni(II) and 0.1 M Zn(II)) caused to increase in the degradation efficiency to 57% (at 160 min). On the other hand, in this trend more NiS particles are available for photons and hence more e–h pairs have been generated. While, when the concentration of Ni(II) in solution was constant, the efficiencies of the obtained catalysts increased with increasing in the Zn(II) concentration to 0.2 M, so the catalyst containing 9.7% ZnS showed 63% degradation efficiency (at 160 min). In this trend, due to the presence of more ZnS particles they play important role to prevent the electron–hole recombination. Hence, ZnS_9.7%–NiS_3.0%–NCP catalyst showed better activity (63%) than NiS–NCP (49% at 160 min).

To have an estimation for the produced e–h and finally hydroxyl radicals depending to ZnS/NiS ratio in the catalysts, the activities of ZnS_9.7%–NiS_3.0%–NCP (C₈: as the most efficient hybridized catalyst) and ZnS_10.8%–NiS_2.8%–NCP (C₉: as the hybridized catalyst with the lowest efficiency) were compared in the presence of iso-propanol (i-PrOH). In general, because of very weak adsorption power of short aliphatic alcohols on surface of catalysts in aqueous media, direct oxidation of them by photogenerated holes is negligible.⁴⁴ So alcohols are usually used as a diagnostic tools of ˙OH radicals mediated mechanism. i-PrOH, a good scavenger like methanol, is more easily oxidized by ˙OH radicals.⁴⁴ Hence, the activities of the catalysts C1 and C2 were studied in the presence of 0.01 and 0.1 M i-PrOH and the photodegradation activities of these catalysts were respectively decreased from their initial values of 61% and 36% to (40% and 17%) and (20% and 6%) in the presence of 0.01 and 0.1 M i-PrOH after 160 min ([2-NT]₀ = 50 ppm). As shown by increasing in the concentration of i-PrOH from 0.01 to 0.5 M a sever decrease in the degradation extent of the pollutant was observed. This confirms significant role of hydroxyl radicals to degrade 2-NT. As shown, more sever decrease in the degradation extent was also observed for catalyst C2 in the presence of i-PrOH. This confirms lesser hydroxyl radicals generated by this catalyst. On the other hand, the production of e–h and finally hydroxyl radicals is ZnS/NiS ratio dependent in the hybridized catalysts.

The decrease in the efficiency of the process at higher ZnS and NiS loadings can be considered as the fact that when the concentration of the ZnS and NiS rises, the solid particles increasingly block the penetration of the photons due to aggregation of ZnS and NiS particles. Hence, less ZnS–NiS particles are available, due to decrease in the effective surface area, for receiving photons, so fewer OH radicals were produced. In addition, at higher concentrations of semiconductors, activated molecules collide with ground state molecules and hence deactivated molecules cannot generate electron–hole pairs and finally active hydroxyl radicals.⁴⁵

3.2.3. The effect of amount of catalyst. The variation in the photocatalyst dosage covering the range from 0.25 to 1.25 g L⁻¹ was studied and its effect on the degradation extent of 2-NT aqueous solution (50 mg L⁻¹, initial pH 6.3) was followed. As shown in Table 2, with increasing in the mass of the catalyst till 0.75 g L⁻¹, the degradation rate constants for the degradation of proposed pollutant were increased and thereafter decreased. By increasing in the catalyst dosage, more ZnS–NiS particles are available and hence more hydroxyl radicals can generate due to the activation of more semiconductor molecules.⁴⁰ The negative effect at higher catalyst dosage can be related to reducing the penetration of the photons by the aggregated solid particles.⁴⁶ High catalyst dosage can also increase the scattering of photons lead to reaching lesser photons to the semiconductors.⁴⁷ Final effects are producing lesser OH radicals.

3.2.4. Effect of pH. Because of the important role of solution pH on the photodegradation rate of some organic compounds,⁴⁸ effect of the initial solution pH (3–12) was studied on the degradation extent of 2-NT and the corresponding results are summarized in Fig. 6A. Based on the inset of this figure, the pH of point of zero charge (pH_pzc) for ZnS_9.7%–NiS_3.0%–NCP is about 7.1. In this pH, the catalyst surface has zero charge, while it respectively positively and negatively charged at more acidic and basic pHs than pH_pzc.⁴⁹ Hence at pH 3, less 2-NT molecules can reach near the catalyst surface because of repulsive forces between the protonated 2-NT molecules and positively charged catalyst surface. In this pH, probably dissolving of ZnS and NiS from the catalyst surface can reduce the degradation efficiency.⁵⁰ To confirm this speculation, leached zinc and nickel cations was followed as a function of solution pH. The results confirmed negligible leached zinc and nickel cations at higher pHs while it was qualitatively recognized at pH 3 by atomic absorption spectroscopy. Hence, it would be expected that by increasing pH up to 7 these problems begin to eliminate and the degradation efficiency increases. The possible reason for increasing in the efficiency at basic pHs till 11 may be attributed to the formation of more OH radicals and the attack of produced hydroxyl radicals to the rich electron sites (nitro groups) in 2-NT molecules. At a high ⁻OH concentration (pH 12), the H₂O₂˙ and HO˙₂ radicals and H₂O₂ may form due to the reaction of ˙OH with −OH. But HO⁻₂ (as conjugate base of H₂O₂), reacts with non-dissociated H₂O₂ molecule (HO₂⁻ + H₂O₂ → H₂O + O₂ + ˙OH⁻) which leads to oxygen and water, instead of hydroxyl radicals under UV irradiation. On the other hand, according to the following reactions and the corresponding dissociation constants, reaction (5) is predominant reaction which finally causes to production superoxide radicals with lesser activity than hydroxyl radicals. Simultaneously, the instantaneous concentration of ˙OH is lower than expected value in these conditions, resulting to decrease in the degradation efficiency at strong basic conditions.^46,51

H₂O₂ ↔ HO₂⁻ + H⁺, pK_a = 11.6 (4)

HO₂ ↔ O₂⁻ + H⁺, pK_a = 4.8 (5)

OH⁻ ↔ O⁻ + H⁺, pK_a = 11.9 (6)

Fig. 6 (A) Effect of pH on the photodegradation efficiency of 2-NT at the irradiation time of 20 min, [2-NT]₀ = 50 ppm, ZnS_9.7%–NiS_3.0%–NCP = 0.75 g L⁻¹. (B) Effect of initial 2-NT concentration on its photodegradation efficiency at pH = 11 and the same conditions with (A). (C) Typical diagram of ln C/C_o versus time as a function of pollutant concentration. (D) Degradation results in terms of mmol of degraded 2-NT pollutant as a function of its initial concentration.

3.2.5. Effect of pollutants concentration. One important factor to guarantee the effective degradation of a pollutant is substrate concentration. The corresponding results for study of the effect of concentration of 2-NT in the range of 45 to 65 mg L⁻¹ on the photocatalytic degradation of 2-NT are presented in Fig. 6B. Fig. 6C shows the typical plot of ln C/C_o versus time as a function of the pollutant concentration. Based on the slopes of the curves rate constants were estimated and collected in Table 2. Due to very short lifetime of hydroxyl radicals (only a few nanosecond), at lower pollutant concentration (here 45 mg L⁻¹) less 2-NT molecules can reach near the catalyst surface, where hydroxyl radicals were generated and hence the produced radicals will deactivate before reaction with molecules of the pollutant, resulting a decrease in the degradation efficiency.⁵² Hence it would be expected that increasing the pollutant concentration overcomes to this problem and hence the efficiency of the process was increased. But, at high 2-NT concentrations (beyond 55 mg L⁻¹), more quantity of 2-NT molecules were adsorbed on the surface of the ZnS_9.7%–NiS_3.0%–NCP catalyst. In this case, lesser photons can absorb by the covered semiconductors and more photons can also be absorbed by high present pollutant molecules before they can reach to the catalyst surface. Hence the generation of hydroxyl radicals was reduced.⁵² In our previous work,³¹ we suggested that if the photodegradation results state in terms of the mmol of the degraded pollutants instead of degradation%, the better and more real results will be obtained. Hence, the mmol of 2-NT molecules were calculated and the results are shown in Fig. 6C. Based on the results, the best degradation efficiencies were obtained for 55 and 65 mg L⁻¹ of 2-NT when the degradation efficiencies were respectively calculated as percentage and mmol of degraded molecules. The best rate constant was observed for 55 mg L⁻¹ of 2-NT and thus this concentration was chosen for the next experiments.

To have an idea about applicability of the proposed method for real samples, a solution containing 55 mg L⁻¹ of 2-NT was prepared in tap water and it was subjected to photodegradation experiment at the optimized conditions. Satisfactory results with respect to distilled water (Fig. 7A) were obtained which confirm that the proposed method can be used as an effective method for the photodegradation of 2-NT in real samples containing common calcium, potassium, sodium, chloride, nitrate and sulfate ions.

Fig. 7 (A) Effect of real sample on the photodegradation efficiency of 2-NT at the irradiation time of 180 min, [2-NT]₀ = 55 ppm, ZnS_9.7%–NiS_3.0%–NCP = 0.75 g L⁻¹, pH 11. (B) Reusability of the ZnS_9.7%–NiS_3.0%–NCP catalyst in photodegradation of 2-NT at the same conditions with (A).

3.2.6. Reusability of the photocatalyst. Due to importance of the cost effectiveness of the method, the possibility of reusing of the proposed photocatalyst was studied in four consecutive experiments by using fresh pollutant solution at the optimized conditions (55 mg L⁻¹ pollutant concentration, pH 11, 0.75 g L⁻¹ of the catalyst for 160 min). At the end of the degradation process, the reused catalyst was dried at 100 and 240 °C. As seen from Fig. 7B, a small and gradual decrease in the activity of catalysts at the first two cycles is due to adsorption of organic intermediates and by-products of the photodegradation process on the surface of the photocatalyst.⁵³ It would be expected that at higher temperatures due to removing of the adsorbed materials from the catalyst surface more degradation centers are available, so, the recovered catalyst at 240 °C is more active than the other during successive uses.

3.2.7. COD, UV-Vis and HPLC studies. HPLC was used for confirming the degradation of 2-NT and the corresponding chromatograms are shown in Fig. 8. As shown, a sharp peak at retention time of 14.5 min was observed for 2-NT solution before the degradation process which its area (2506) was significantly decreased during the irradiation process. During 85 min irradiation of the system, the peak area of the original 2-NT peak was decreased to 1257, while some new peaks were appeared in the range from 1.77 to 4.76 min (corresponding peak area are: 176 and 115). By continuation of the irradiation process to 160 min the peak area of the main peak at retention time of 14.5 min was significantly decreased to 359 while the peak area of the located peaks at 4.2 and 4.76 min were significantly decreased and the peaks area at 1.77 min and around it was increased (corresponding peak area is: 207). Based on the decrease in the peak area of main peak located at retention time of 14.5 min, the degradation percentages of 49.8 and 85.6% were respectively obtained during 85 and 160 min irradiation times. These agree with the UV-Vis results confirming the degradation of 2-NT molecules to smaller fragments during the proposed irradiation process. Unfortunately, our laboratory does not equipped enough to determine the degradation products, but for C–N containing compounds in a typical photocatalytic process, it has been reported that the attack of hydroxyl radicals cleavage C–N bond into NO₃⁻ and NO₂⁻ ions.^40,54,55

Fig. 8 HPLC chromatogram of 2-NT solution before and during 85 and 160 min photodegradation of the sample at the same conditions with Fig. 7A.

The initial and final COD values for the system were determined because the chemical oxygen demand is a measure the organic strength of wastewater.^14,47 The corresponding results are shown in Fig. 9A. The changing in COD of the 2-NT solution during 0, 60, 120 and 160 min irradiation times of the sample were respectively about 719, 479, 342 and 161 mg L⁻¹ corresponding to degradation percents of 0, 33.4, 52.0 and 78.0, respectively. The COD removal during the photodegradation process indicates the mineralization of 2-NT molecules.

Fig. 9 (A) Change of 2-NT concentrations was attained by COD method plotted versus irradiation time; (B) decrease in absorbance of 2-NT solution during 180 min photodegradation.

Fig. 9B shows the decrease in UV-Vis absorption spectra of 2-NT during the photodegradation experiments (55 ppm 2-NT, 0.75 g L⁻¹ of the catalyst, pH 11). No additional peaks appeared in the UV-Vis spectra during the irradiation process which confirm degradation of 2-NT molecules to smaller fragments. Thus the decrease in the samples' absorbance due to the decrease in the 2-NT concentration was recorded for measurement of the degradation extent in the all above investigated parameters.

In general, decrease in the remaining COD during the photodegradation process is in accordance with the UV-Vis spectrophotometric and HPLC results, all confirming the degradation of 2-NT molecules to smaller fragments.

4. Conclusion

The results confirm that hybridation of the ZnS–NiS increases its photocatalytic activity with respect to non hybridized ZnS and NiS semiconductors. The ratio of ZnS/NiS significantly affects the photocatalytic activity of the prepared catalysts, so the ZnS_9.7%–NiS_3.0%–NCP was the most active photocatalyst for the degradation of 2-NT. Also, supporting the hybridized system onto clinoptilolite nanoparticles significantly increased its photocatalytic activity with respect to unsupported one. This confirms the ability of the zeolite support to prevent the recombination of the produced electron–pair holes by migration of the conductance band electrons to zeolite structure. In addition, using zeolite nanoparticles improved the activity of the supported ZnS–NiS with respect to the micronized one because of the increased surface area of the nanoparticles.

Acknowledgements

The authors thank M. Alizadeh and M. H. Kazemzadeh for performing instrumental analysis of the samples, as experts in laboratory analysis in Shahreza Branch, Islamic Azad University. The authors also thank from Dr A. Sharifzadeh as the university president for supporting of this work.

References

1. D. Tomova, V. Iliev, S. Rakovsky, M. Anachkov, A. Eliyas and G. L. Puma, J. Photochem. Photobiol., A, 2012, 231, 1–8.
2. B. P. Nenavathu, A. V. R. Krishna Rao, A. Goyal, A. Kapoor and R. K. Dutt, Appl. Catal., A, 2013, 459, 106–113.
3. Z. Qiang, C. Liu, F. Tian and T. Zhang, Chemosphere, 2009, 76, 609–615.
4. U. Bali, E. C. Catalkaya and F. Sengul, J. Environ. Sci. Health, Part A: Toxic/Hazard. Subst. Environ. Eng., 2003, 10, 2259–2275.
5. Z. Lu, P. Huo, Y. Luo, X. Liu, D. Wu, X. Gao, C. Li and Y. Yan, J. Mol. Catal. A: Chem., 2013, 378, 91–98.
6. A. Nezamzadeh-Ejhieh and Z. Salimi, Appl. Catal., A, 2010, 390, 110–118.
7. Y. Zhou, H. Lin, Q. Gu, J. Long and X. Wang, RSC Adv., 2012, 2, 12624–12627.
8. A. Nezamzadeh-Ejhieh and Z. Salimi, Desalination, 2011, 280, 281–287.
9. G. L. Jincai Zhao, New J. Chem., 2000, 24, 411–417.
10. X. Li, G. Liu and J. Zhao, New J. Chem., 1999, 23, 1193–1196.
11. F. Chen, J. He, J. Zhao and J. C. Yu, New J. Chem., 2002, 26, 336–341.
12. A. Nezamzadeh-Ejhieh and M. Karimi-Shamsabadi, Appl. Catal., A, 2014, 477, 83–92.
13. O. P. Yadav, A. Eyasu and R. K. Bachheti, Int. J. ChemTech Res., 2013, 5, 1452–1461.
14. A. Nezamzadeh-Ejhieh and S. Moeinirad, Desalination, 2011, 273, 248–257.
15. A. Nezamzadeh-Ejhieh and H. Zabihi-Mobarakeh, J. Ind. Eng. Chem., 2014, 20, 1421–1431.
16. D. Q. Wang, D. R. Chen and X. L. Jiao, Chin. Chem. Lett., 2004, 15, 79–82.
17. T. Kou, C. Jin, C. Zhang, J. Sun and Z. Zhang, RSC Adv., 2012, 2, 12636–12643.
18. R. M. Mohamed and M. M. Mohamed, Appl. Catal., A, 2008, 340, 16–24.
19. A. Nezamzadeh-Ejhieh and K. Shirvani, J. Chem., 2013, 541736.
20. P. C. C. Faria, J. J. M. Órfão and M. F. R. Pereira, Water Res., 2004, 38, 2043–2052.
21. C. Pulgarín, N. Dela Cruz, J. Giménez, S. Esplugas, D. Grandjean and L. F. de Alencastro, Water Res., 2012, 46, 1947–1957.
22. G. Ledoigt, J. Wiszniowski, A. Ter Halle, C. Richard and A. Hitmi, Chemosphere, 2009, 74, 1224–1230.
23. A. Olad and B. Naseri, Prog. Org. Coat., 2010, 67, 233–238.
24. M. Banerjee, L. Chongad and A. Sharma, Res. J. Recent Sci., 2013, 2, 326–329.
25. A. Nezamzadeh-Ejhieh and M. Shahanshahi, J. Ind. Eng. Chem., 2013, 19, 2026–2033.
26. H. Faghihian, M. Talebi and M. Pirouzi, J. Iran. Chem. Soc., 2008, 5, 394–399.
27. M. Brazlauskas and S. Kitrys, Chin. J. Catal., 2008, 29, 25–30.
28. A. Nezamzadeh-Ejhieh and S. Khorsandi, J. Ind. Eng. Chem., 2014, 20, 937–946.
29. A. Nezamzadeh-Ejhieh and S. Hushmandrad, Appl. Catal., A, 2010, 388, 149–159.
30. D. W. Breck, Zeolite Molecular Sieves: Structure, Chemistry and Uses, John Wiley & Sons, New York, 1974.
31. A. Nezamzadeh-Ejhieh and M. Karimi-Shamsabadi, Chem. Eng. J., 2013, 228, 631–641.
32. Y. G. Basabe, I. R. Iznaga, L. C. Menorval, P. Liellyn, G. Maurin, D. W. Lewis, R. Binions, M. Autic and A. R. R. Salvador, Microporous Mesoporous Mater., 2010, 135, 187–196.
33. D. Wu, J. Xie, C. Li and L. Chi, Fuel, 2013, 103, 480–485.
34. R. John and S. Florence, Chalcogenide Lett., 2010, 4, 269–273.
35. A. Nezamzadeh-Ejhieh and E. Shahriari, Int. J. Photoenergy, 2011, 518153.
36. S. Pérez, L. Tong, Y. Wang, P. Eichhorn and D. Barceló, Chemosphere, 2011, 83, 340–348.
37. H. Guo, K. Lin, W. Weng, Y. Ke, R. Shen, D. Wang and J. Chen, J. Nanopart. Res., 2013, 15, 1475–1487.
38. R. C. Khandelwal, V. Sharma, N. Gandhi and A. Khant, Int. J. Chem. Sci., 2010, 8, 1965–1972.
39. A. Kar, S. Kundu and A. Patra, RSC Adv., 2012, 2, 10222–10230.
40. A. Nezamzadeh-Ejhieh and F. Khodabakhshi-Chermahini, J. Ind. Eng. Chem., 2014, 20, 695–704.
41. M. Subrahmanyam, P. A. K. Reddy, B. Srinivas and V. Durgakumari, Toxicol. Environ. Chem., 2012, 94, 512–524.
42. J. V. Tolia, M. Chakraborty and Z. V. P. Murthy, Pol. J. Chem. Technol., 2012, 14, 16–21.
43. A. Kar, S. Kundu and A. Patra, RSC Adv., 2012, 2, 10222–10230.
44. Y. Chen, S. Yang, K. wang and L. Lou, J. Photochem. Photobiol., A, 2005, 172, 47–54.
45. M. R. Hoffmann, S. T. Martin, W. Choi and D. W. Bahnemannt, Chem. Rev., 1995, 95, 69–96.
46. A. Nezamzadeh-Ejhieh and M. Khorsandi, J. Hazard. Mater., 2010, 176, 629–637.
47. C. Fotiadis, N. P. Xekoukoulotakis and D. Mantzavinos, Catal. Today, 2007, 124, 247–253.
48. A. Nezamzadeh-Ejhieh and M. Amiri, Powder Technol., 2013, 235, 279–288.
49. C. Shifu and L. Yunzhang, Chemosphere, 2007, 67, 1010–1017.
50. A. Nezamzadeh-Ejhieh and N. Moazzeni, J. Ind. Eng. Chem., 2013, 19, 1433–1442.
51. A. Aleboyeh, Y. Moussa and H. Aleboyeh, Dyes Pigm., 2005, 66, 129–134.
52. B. Krishnakumar, R. Velmurugan, B. Subash and M. Swaminathan, Indian J. Chem., Sect. A: Inorg., Bio-inorg., Phys., Theor. Anal. Chem., 2012, 51, 580–585.
53. A. Okte and O. Yilmaz, Appl. Catal., A, 2009, 354, 132–142.
54. H. Bagheri, J. Slobodonik and U. A. T. Brinkman, J. Sci., Islamic Repub. Iran, 2000, 11, 289–299.
55. S. Horikoshi, N. Watanabe, M. Mukae, H. Hidaka and N. Serpone, New J. Chem., 2001, 25, 999–1005.

---