Solar-Light-Driven Improved Photocatalytic Performance of Hierarchical ZnIn2S4 Architectures.

In the quest for developing novel narrow band gap semiconductor materials, the research in metal chalcogenides has gained a strong attraction. In the present investigation, a surfactant-free hydrothermal route has been followed to design hierarchical self-assembled flower-like ZnIn2S4 structures through control over precursor concentration and hydrothermal processing parameters. Uniform hexagonal marigold flower-like ZnIn2S4 architectures (∼4 μm) were formed with self-assembly of petals (thickness ∼8-12 nm) forming rose-like structures and finally forming marigold flowers in 24 h duration. The hierarchical ZnIn2S4 flower structure has been used as photocatalysts for the degradation of dye and chlorinated phenols. Photodegradation demonstrates that the high surface area from the porous flower architecture (∼72 m2/g) with an enhanced visible light absorption giving low band gap energy (2.15 eV) is responsible for higher photocatalytic performance. Complete degradation of the organic pollutants has been observed within 90 min in the presence of natural sunlight. To understand the participating reactive species contributing to degradation, scavenger studies were performed for deducing the plausible photocatalytic degradation pathways. This study might open new insights into the design of novel hierarchical structures.

Introduction

Synthetic textile dyes constitute a large class of contaminated wastewater.[1,2] Azo- and fluorine-based dyes constitute a significant portion.[3,4] The degradation products of these dyes include aromatic amines that are highly carcinogenic. One of the widely used “green” technologies is heterogeneous photocatalytic oxidation processes for the decomposition of organic dyes soluble in wastewater.[5,6] Recently, attention has been paid to designing of new nontitania and nonoxides as effective materials that can harvest and show enhanced activity under visible light irradiation than under ultraviolet light.[7−10] Through these materials, the natural sunlight could be utilized more efficiently for the photocatalytic process. In a pursuit to develop such materials, self-assembled three-dimensional (3D) structures through one-dimensional and two-dimensional nanoscale building blocks have been used to form 3D hierarchical architectures.[11,12] Such hierarchical structures often lead to large surface area and open pore structures providing more active sites for favorable transport of electrons and holes to the surface for reaction to occur. Subsequently, this may lead to superior properties in comparison to the nanoscale materials.[13,14]

The ternary chalcogenide semiconducting compounds have acquired interest due to their tunable optical and electrical properties.[10,15] Among them, ZnIn2S4 has demonstrated potential applications in diverse fields such as thermoelectricity,[16] charge storage,[17] electrochemical recording, and as photocatalysis.[18−20] To achieve the optimum potentiality of the material, synthesis of ZnIn2S4 by various chemical methods plays a major role. The control of shape and size of these inorganic structures is of utmost importance because of the strong correlation between the synthesis parameters and physical/chemical properties.[21,22] Various researchers have reported the solution/soft chemistry route to be the best technique for synthesizing variant chalcogenide compounds with desired properties.[23,24]

Literature reports several different morphologies of ZnIn2S4 prepared via effective methods.[25] Shen et al. employed nanostructured ZnIn2S4 for hydrogen evolution, and the photocatalytic activity of ZnIn2S4 was greatly affected by the crystal plane spacing along the c axis.[11] One-dimensional ZnIn2S4 nanotubes and nanoribbons were synthesized by a solvothermal technique using pyridine as a solvent, and hollow microspheres were synthesized by the hydrothermal method using cetyl-trimethylammonium bromide or poly(ethylene glycol) surfactants.[26] In another study, porous polycarbonate membrane was used as a hard template to fabricate nanotubes and nanowires of ZnIn2S4, with the wall of the tubes 5 nm thick. ZnIn2S4 nanotubes imparted the highest visible light photoactivity for the degradation of methyl orange (MO) in 4 h.[27] Chen et al. studied the effect of ZnIn2S4 polymorphs for the degradation of rhodamine B and methyl orange, where the negative (0001) S plane enhanced the adsorption of cationic dyes over hexagonal ZnIn2S4 compared with methyl orange and affects the photocatalytic activity synergistically.[28] Similar studies indicated rhodamine B is degraded by 97%, and only 5% methyl orange is degraded under the same conditions.[29] Most of the studies reveal the poor photocatalytic activity of ZnIn2S4 for the degradation of dyes.

Apart from the dye pollutants, chlorinated phenols are also contemporary in the system that results in high toxicity and serious environmental issues.[30,31] On the basis of the above context, development of a hierarchical material with well-defined structures and morphologies without surfactants and with enhanced surface activity to degrade these pollutants would be of great research interest. In the present study, we have reported the design and optimization of hierarchical ZnIn2S4 flower structures without any additional reagents. The concentration of precursor contents, hydrothermal temperature, and time have been varied to obtain a stable hierarchical structure. The optimized hierarchical ZnIn2S4 structure has been studied for the photocatalytic degradation of methyl orange. The degradation studies of chlorinated phenols were also carried out to understand the diversified effect of ZnIn2S4 structures.

Results and Discussion

Effect of Processing Parameters on Crystal Structure

To investigate the purity of the crystal phase as an effect of different processing parameters such as thiourea content, hydrothermal temperature, and time, X-ray diffraction (XRD) measurements were carried out, and the results are represented in Figure a–c. Figure a presents the composite XRD pattern of the hydrothermally synthesized powders obtained with various stoichiometric contents of thiourea at a temperature of 180 °C for 24 h. The powders synthesized with stoichiometric thiourea are designated as ZIS-TU, whereas those with double the stoichiometric amount are designated as ZIS-2TU. Similarly, those with 3 and 4 times the stoichiometric thiourea are named as ZIS-3TU and ZIS-4TU, respectively. The XRD pattern of ZIS-TU shows a series of characteristic peaks, with a few peaks having low intensity. These peaks can be attributed to the hexagonal phase of ZnIn2S4 (JCPDS No. 72-0773).[32] However, it can be presumed that phase formation is not complete and yet to occur. It is clear from the figure that ZIS-TU has the lowest crystallinity in comparison to the other samples. When the thiourea content was doubled, sharp and high crystalline hexagonal peaks were prominent, with main characteristic peaks indexed to (006), (105), (0010), (108), (1010), (112), (0012), (203), and (1112) corresponding to the space group P63mc of hexagonal crystal structure. No other crystal phase related to the impurity phases such as binary sulfides and oxides are present, indicating the high phase purity of ZIS-2TU. With the higher content of thiourea (i.e., ZIS-3TU and ZIS-4TU), some of the crystal planes of hexagonal crystal phase are also observed to grow. A significant increment in hexagonal crystal planes (004), (104), (105), and (112) is observed to develop from ZIS-2TU to ZIS-3TU to ZIS-4TU. There were no traces of impurity peaks apart from the extra developed peaks that also correspond to the hexagonal crystal phase of ZnIn2S4. It is known that ZnIn2S4 exhibits different polytypes depending on the difference in stacking fashion of sulfur atoms in the matrix.[33] The coordination environment of ions in the solution governs the formation of either the hexagonal or cubic phase.[34] In the precursor solution, thiourea can form three different complexes with In3+ and Zn2+, namely tetrahedral [In(TU)4]3+, [Zn(TU)4]2+, and octahedral [In(TU)6]3+. Under the hydrothermal conditions, the complexes [In(TU)6]3+ and [In(TU)4]3+ formed initially undergo Ostwald ripening and further react in situ with [Zn(TU)4]2+ to produce a thermodynamically stable hexagonal ZnIn2S4 containing both octahedral and tetrahedral coordination of In3+ ions.[35] Scherrer’s formula was used to calculate the crystallite sizes of the synthesized powders. The calculated crystallite sizes are ∼43, ∼66, ∼72, and ∼72 nm for ZIS-TU, ZIS-2TU, ZIS-3TU, and ZIS-4TU, respectively. There is not much difference between the crystallite sizes of ZIS-2TU and ZIS-3TU, indicating that apart from the growth of few planes due to high thiourea content, it does not significantly affect the crystallite size. In this study, pure and stable hexagonal crystal phase of ZnIn2S4 was obtained at twice the stoichiometric amount of thiourea (ZIS-2TU). Thus, experiments for further process optimization were carried out keeping this amount thiourea content.

Figure 1

Composite XRD patterns as an effect of (a) thiourea content (parameters: temperature 180 °C and time 24 h); (b) hydrothermal temperature (parameters: thiourea content 2TU and time 24 h); and (c) hydrothermal duration (parameters: temperature 180 °C and thiourea content 2TU).

Figure b presents the XRD pattern of hydrothermal experiments performed with double the stoichiometric content of thiourea at different hydrothermal temperatures (170, 180, and 190 °C) kept for 24 h duration. The obtained powders were designated as ZIS-170, ZIS-180, and ZIS-190 for different temperatures. It can be observed that at 170 °C, some of the hexagonal peaks are still to be developed and become highly crystalline. Upon comparing the XRD pattern obtained for 170 and 180 °C, the peaks for ZIS-180 are highly crystalline compared with ZIS-170. This has also been confirmed from the calculation of crystallite sizes, which is ∼44 nm for ZIS-170, which is similar to that for ZIS-TU obtained in our previous XRD study. There is a sharp increase in the crystallite size from ∼66 to ∼110 nm (ZIS-190) when the temperature is raised from 180 to 190 °C. The crystallite size almost doubles probably due to Ostwald ripening of the particles at a high reaction temperature.[36] Thus, the particles synthesized at 180 °C is found suitable to form a highly crystalline stable and lower crystallite size hexagonal ZnIn2S4.

To identify the influence of hydrothermal duration on the crystal phase formation, identical set of experiments were performed at different time durations for 6, 12, 18, 24, and 28 h under the constant temperature of 180 °C and twice the stoichiometric amount of thiourea content in the precursor solution. The powders obtained were designated as ZIS-6 h, ZIS-12 h, ZIS-18 h, ZIS-24 h, and ZIS-28 h, as shown in Figure c. There was no significant change in the crystal phase of ZnIn2S4 synthesized at 6 and 28 h. However, an increment in the crystallite size is observed from ZIS-6 h (∼61 nm) to ZIS-28 h (∼74 nm). It was difficult to conclude the optimum time due to similar hexagonal crystal phases, thus microscopic imaging of these synthesized powders was carried out for optimization on the basis of morphology formation.

Morphological Analysis

A better understanding of the particle growth can be obtained from the microscopic imaging of the synthesized powders at different time intervals; thus, the powders were investigated by the field emission scanning electron microscopy (FESEM) technique. Figure a–f shows the images of the powders named ZIS-6 h, ZIS-12 h, ZIS-18 h, ZIS-24 h, and ZIS-28 h. A high agglomeration among particles with nonuniform shapes was observed for powders of ZIS-6 h (Figure a). Apart from these nonuniform particles, the presence of also plate-shaped particles indicates the development of some specific morphology. Upon increasing the hydrothermal duration to 12 h (Figure b), self-assembly of petal-like structures was observed to form rose-like structures along with other particles that are nonuniformly distributed in the system. However, further increasing the hydrothermal duration to 18 h results in hierarchically spherical flower-like structures having an average flower size ∼3–5 μm (Figure c). There are also some petals over the surface of the flowers that are softly agglomerated and self-assembled to form flower-like structures. A careful observation reveals that the powder synthesized at 24 h duration has uniform spherical marigold flower-like structures that are well formed from the organization of petal structures. The average size of the flower is observed to be in the range of ∼4–7 μm (Figure d), which is higher than the flower size for ZIS-18 h. An area of the spherical marigold flower ZIS-24 h has been zoomed in to determine the thickness of the petals forming these structures as shown in Figure e. The thickness of the petals ranges from ∼8 to 12 nm. Further increasing the hydrothermal duration to 28 h shows an increment in these flower sizes varying from ∼7 to 10 μm (Figure f), which could be attributed to the growth of the flowers via Ostwald ripening process.[37] In both parts, Figure c,d, we observe an opening mouth, which is indicative of the hollow flower structures based on the high-resolution images shown later. The enhancement in the aspect ratio of the flower can be due to the lower petal thickness and self-assembly. The hierarchical marigold flower structure was observed to have open cavity size of ∼500 nm. It can be predicted that beyond 24 h, flowers merely grow to form flowers of enlarged size. The specific Brunauer–Emmett–Teller (BET) surface areas of ZIS-6 h, ZIS-12 h, ZIS-18 h, ZIS-24 h, and ZIS-28 h are 23, 24, 47, 72, and 62 m2/g, respectively. It is quite evident that the surface area decreased after 24 h, which reveals that an increment in the flower size decreases the cavity distance between the petals, reducing the N2 adsorption during the surface area measurement. A more detailed morphological and structural analysis for ZIS-24 h has been carried out as a part of optimized powder characterization.

Figure 2

FESEM micrographs of powders synthesized at different hydrothermal duration (a) ZIS-6 h; (b) ZIS-12 h; (c) ZIS-18 h; (d, e) ZIS-24 h; and (f) ZIS-28 h.

Figure shows the high-resolution transmission electron microscopy (TEM) image and selected area electron diffraction (SAED) pattern of powder synthesized at 24 h. Figure a,b shows two distinct natures of the same flower morphology. Some of the spherical flowers are opaque; however, some flowers are hollow, which was also observed in FESEM images mentioned previously. The average flower size has been observed to be ∼4 μm from Figure a. Formation of these hierarchical structures in aqueous media without any surfactant under hydrothermal conditions is possible by geometrical building blocks forming an array of particles that get layer curvatures and develop into flower structures as reported by Kale et al.[25] On the other hand, the development of hollow flower structures could be due to the assembly of small flowers to form a bigger flower, leaving behind a cavity in between the assembly. The edges of the flower are observed to be transparent. The selected area electron diffraction (SAED) pattern taken from the flower edge shows a distinct array of spots demonstrating the single crystalline nature of the ZnIn2S4 petals (Figure c). The advantage of this unique type of hierarchical marigold flower structures is that they impart a higher surface area due to open access to the interspaces between the interweaved petals that boosts the light-harvesting ability of the hierarchical structures through the multireflection process on the incident visible light and provides more active sites for the photocatalytic reaction.

Figure 3

(a, b) TEM micrographs and (c) SAED pattern of ZIS-24 h.

Optical Properties

The optical properties of the ZnIn2S4 powders synthesized under different hydrothermal duration were investigated through UV–vis diffuse reflectance spectra. The reflectance data were converted to the Kubelka–Munk unit of absorption by means of Kubelka–Munk function,[35,38] and a plot of F(R)2 versus photon energy is plotted (Figure a). The intersection of the extrapolated linear portion to photon energy gives the estimation of band gap energy values (Eg). The Eg values varied with respect to hydrothermal duration and range from ∼1.77 to 2.49 eV. The Eg value for ZIS-24 h is estimated to be 2.15 eV, which is similar to the reported band gap values for ZnIn2S4 structures.[21,39,40] Furthermore, the conduction and valence band positions of the optimized ZnIn2S4 flowers were determined from the estimated band gap values using the following equation[41]where EVB and ECB are valence band and conduction band edge potentials, Ee is the energy of free electrons on the hydrogen scale (4.5 eV), Eg is the band gap energy, and X is the electronegativity of the semiconductor. The estimated valence band edge potential is 1.65 eV, and the corresponding conduction band edge potential is estimated to be −0.5 eV. On the basis of the calculated band positions, the possible mechanism for visible light photocatalysis over ZnIn2S4 flower structures is shown in Figure b. It can be well observed that the valence band of ZnIn2S4 is less positive than that of E° (•OH/OH–) (1.65 V vs normal hydrogen electrode (NHE)), indicating that the photogenerated holes are not able to oxidize OH– to give •OH.[28] The conduction band is more negative than E° (O2/O2–•) (−0.33 V vs NHE), indicating that the photogenerated electrons can reduce the surface chemisorbed O2 to yield the strong oxidizing species O2–•. From the above observation, it could be presumed that the photocatalytic reaction via ZnIn2S4 is favorable through photogenerated electron converting the adsorbed oxygen molecules on the surface of the flower to superoxide radical as an active species. To validate this hypothesis, photocatalytic reaction in the presence of different scavengers were performed, as discussed in the later section.

Figure 4

(a) Plot of the F(R)2 vs photon energy for band gap determination and (b) calculated band positions of ZnIn2S4 and a possible mechanism for visible light photocatalysis.

Compositional Analysis

A significant investigation on the surface chemical states of the synthesized ZnIn2S4 flower using the X-ray photoelectron spectroscopy (XPS) measurement is essential, as the trace content of any impurity that cannot be detected in other analytical methods can have a synergistic influence over the photocatalytic properties of the material. Figure shows the wide survey of ZnIn2S4 and the corresponding elements Zn-2p, In-3d, and S-2p. The wide survey scan clearly shows the specific peaks for the presence of elements C, O, Zn, In, and S (Figure a). The binding energy for the C-1s peak at 284.6 eV was used as the reference for calibration, as it is attributed to the adventitious carbon species on the samples. The peaks in Zn, In, and S elements were deconvoluted to identify any trace impurities. The binding energies centered at 1021.2 and 1044.4 eV of Zn-2p (Figure b) correspond to the Zn-2p3/2 and Zn-2p1/2, respectively, which governs Zn to be in the Zn2+ form in ZnIn2S4.[42,43] The binding energies at 444.5 and 452.2 eV in XPS spectra of In-3d can be well attributed to the In-3d5/2 and In-3d3/2, respectively (Figure c), indicating In3+ to be the existing form. The XPS spectra of S-2p were deconvoluted into two separate peaks centered at 161.4 and 162.7 eV, as shown in Figure d indexed to S-2p3/2 and S-2p1/2, respectively, confirming the S2– form of sulfur.[44] The XPS signals obtained are consistent with the available literature values. The absence of any other elemental peaks and surface states other than C and O from atmospheric absorption confirms the chemical purity of the synthesized ZnIn2S4 flower structures.

Figure 5

(a) Wide survey; (b) Zn-2p core level; (c) In-3d core level; and (d) S-2p core level XPS spectra of ZnIn2S4 flower structures.

Photocatalytic Studies

We optimized the hierarchical ZnIn2S4 flower structure synthesized at 180 °C for 24 h with a double thiourea content (ZIS-24 h) to be the photocatalyst that will be used for the investigation of degradation of a model pollutant, methyl orange (MO). Apart from the optimization of the photocatalyst through morphology, the photocatalytic tests were also performed for individual ZnIn2S4 samples prepared under different reaction parameters prior to the systematic photocatalysis using the optimized photocatalyst. A series of degradation experiments were carried out under metal halide lamp with each of the synthesized catalyst with catalyst loading of 50 mg/50 mL and the corresponding degradation results has been tabulated in Table . The results depict that each of the synthesized ZnIn2S4 particles show variation in the specific surface area due to the hydrothermal processing conditions governing the morphology formation, which indeed have a significant effect on the photocatalytic degradation of MO. Prior to each of the photocatalytic experiment, adsorption–desorption equilibrium was achieved without light irradiation, and the absorbance obtained after adsorption was considered as the initial concentration for the photocatalytic experiments. The highest photoactivity of 96% was achieved with ZIS-24 h, which also has a high specific surface area among other synthesized photocatalysts, proving the superiority of the ZIS-24 h powders. The results also provide insight into the morphology-driven photoactivity following a trend: ZIS-6 h (nonuniform/plate particles, 54%) > ZIS-12 h (rose-like structure, 67%) > ZIS-24 h (hierarchical flower structure, 96%). The degradation with ZIS-18 h and ZIS-28 h is lower, as the surface area of ZIS-18 h is low due to the presence of some agglomerated plates on the surface of hierarchical flower structure and the phenomenon of Ostwald ripening in ZIS-28 h reduces the surface area because of prolonged hydrothermal duration. Thus, based on the photocatalytic tests, the hierarchical ZnIn2S4 flower synthesized at 180 °C for 24 h with double thiourea has been optimized.

Table 1

Specific Surface Area and Photocatalytic Degradation Results of ZnIn2S4 Sample Series Synthesized Under Different Reaction Conditions

sample name	surface area (m2/g)	% degradation
ZIS-3TU	46	75
ZIS-170	45	68
ZIS-190	54	79
ZIS-6 h	19	54
ZIS-12 h	24	67
ZIS-18 h	47	78
ZIS-24 h	72	96
ZIS-28 h	62	87

The surface charge of the photocatalysts plays an important role in the adsorption of dye pollutants. Although literature reports the surface charge of ZnIn2S4 to be negative,[29] confirmation of adsorption ability is much needed before detailed photocatalytic analysis. Therefore, prior to the systematic photocatalytic experiments, dye pollutant with contrary charges such as methylene blue (MB, a cationic dye) and methyl orange (MO, an anionic dye), which is the model pollutant in our case, was allowed to adsorb onto the ZnIn2S4 surface under dark conditions (50 mg catalyst/50 mL 20 ppm dye solution). We observed complete adsorption of MB on the ZnIn2S4 surface within 60 min, whereas only 22% MO adsorption takes place in 120 min, which remains constant afterward. The above results point toward the negative charge on the surface of flower-like ZnIn2S4 structures, which allows adsorption of the MB molecules on the surface of the catalyst. A similar result has been obtained by Chen et al.,[28] where they have reported that the exposed negative (0001) S plane of hexagonal ZnIn2S4 is responsible for the adsorption of cationic dye.

The variation in C/Co of MO (where C is the concentration at time t and Co is the initial concentration after adsorption equilibrium) as a function of time under different catalyst loading conditions is shown in Figure a. Before the initiation of catalytic experiments, the degradation of dye in the presence of visible light solely was carried out to understand the degradation in the presence of light and the observed photolytic degradation (PL) of MO was only 12%. Most of the studies have shown negligible adsorption of MO; however, in the present case, adsorption of methyl orange over ZnIn2S4 was observed to be 14, 22, and 25% with catalyst loading of 25, 50, and 100 mg/50 mL 20 ppm MO, respectively, which increases with increase in the catalyst loading. This increased adsorption could be attributed to the open cavities in the flower structures of ZnIn2S4, which facilitate the adsorption of dye. However, there is no significant increment in adsorption with the doubling of catalyst loading from 50 mg (22%) to 100 mg (25%). After the adsorption equilibrium, the visible light was switched on to carry out the photocatalytic experiments. We observe that the degradation increased from 92 to 96% when the catalyst loading is doubled from 25 to 50 mg/50 mL for 20 ppm MO. However, further doubling the loading from 50 to 100 mg decreases the catalytic activity to 88%. This decrease in catalytic activity could be attributed to the increased agglomeration among particles due to higher loading content, which reduced the active sites due to agglomeration. Another reason is that absorption of light by the particles reduces due to segregation of particles and thus reduces the activity.[45] The photocatalytic activity is comparable for all of the loadings. Therefore, first-order kinetic plot of −ln C/Co versus time was plotted to determine the kinetic rate constants with respect to catalyst loading, which would give insight into the choice of catalyst loading. Figure b presents the kinetic plot of photolysis and also loading effect on the photochemical reaction. The reaction follows the pseudo-first-order kinetics as shown in the figure. The kinetic rate constants with respect to different catalyst loading are 16.3, 26.4, and 17.2 (×10–3) min–1 for 25, 50, and 100 mg/50 mL of 20 ppm MO, respectively. The above data suggest that the rate of the reaction is almost halved with an increase in catalyst loading to 100 mg and also with a decrease in catalyst loading to half. There is a 1.6-fold increase in the rate constants of 25 and 50 mg catalyst loadings. On the basis of the above findings, the highest photocatalytic degradation with the highest rate constant is observed with a catalyst loading of 50 mg and has been used for further photocatalytic experiments.

Figure 6

(a) Degradation profile; (b) kinetic profile (parameters constant: MO concentration 20 ppm, and temperature 25 ± 2 °C); (c) degradation profile with respect to oxidizer H2O2; and (d) first-order rate constant vs oxidizer concentration in control system.

The presence of an oxidizer in the photochemical system acts as a booster for the production of reactive species and thus is an important additive for improving the photocatalytic activity of any material. Figure c shows the effect of the presence of the oxidizer H2O2 and its comparative photocatalytic degradation under the influence of visible light irradiation. The reactions carried out under different conditions are designated as follows: (a) in the presence of oxidizer H2O2 only, without a catalyst; (ii) in the presence of catalyst only, without H2O2, named as control; and (iii) in the presence of both H2O2 and catalyst. The concentration of hydrogen peroxide has been varied as 1 and 2.5 mM in the photochemical control reactions. The dye degradation in the presence of H2O2 under visible light irradiation is via photolysis of the oxidizer H2O2, which is through the hemolytic cleavage of H2O2 producing HO• radicals that contribute to the photochemical reaction as shown in eq .[46] In addition, H2O2 is also capable of accepting the photogenerated electron from ZnIn2S4 that inhibits the electron–hole pair recombination and also facilitates additional HO• radicals, as shown in eq The degradation observed in the presence of H2O2 having concentrations of 1 and 2.5 mM is 48 and 65%, respectively. This phenomenon indicates that the hydroxyl radical production for the degradation of MO in the presence of H2O2 is quite fast, and complete degradation of the dye can be achieved within less time. Comparing the photocatalytic degradation in the presence of H2O2 with control and control itself, complete degradation is achieved within 90 min in the presence of H2O2, where more than 90% degradation happens in the first hour. This suggests that increment in H2O2 concentration has a significant effect on the degradation and its presence greatly influences the photochemical system. Figure d presents the first-order rate constants of the photochemical reactions, which evidently indicate that the rate of the reaction increases significantly in the presence of 1 and 2.5 mM H2O2, which is 54.7 and 49.8 (×10–3) min–1, respectively. There is a 1.8 and 2.0 times increase in the rate constant in comparison to the control. The rate constant of control is higher than that of the reaction carried out with the oxidizer solely. It is well observed that complete photochemical degradation of MO also takes place in the presence of 1 mM H2O2. Because the photolytic degradation of MO in the presence of 1 mM H2O2 solely is also high, we presumed the degradation to happen much faster within less time. The above results suggest that there might be an ongoing competition between the oxidizer-produced hydroxyl radicals and other produced photoreactive species in the photochemical system for degradation of MO. To understand the reactive species active in degradation of MO, experiments were carried out in the presence of different reactive species scavengers without the presence of oxidizers.

As already predicted from the valence band and conduction band potential calculations in Section , the photocatalytic reaction via ZnIn2S4 is favorable through photogenerated electron converting the adsorbed oxygen molecules on the surface of the ZnIn2S4 to superoxide radical as an active species. Figure a shows the variation in C/Co of MO as a function of irradiation time before and after the scavengers were added to the photochemical system. The prime role of these active species was determined through the variation in C/Co of the dyes after the scavengers were introduced in the system. Only 57% of MO was degraded upon introducing benzoquinone (BZQN) to the photochemical system, and there was an obvious inhibition of the photochemical degradation. There was no significant suppression of the catalytic degradation with the addition of other scavengers such as tert-butyl alcohol (t-BuOH, OH• scavenger), glycerol (GLY, h+ scavenger), and dimethyl sulfoxide (DMSO, e– scavenger). It can be observed that with the addition of GLY, the concentration of MO decreased quickly in the first 60 min and then became slow. The results indicate that quite a large number of O2–• were generated when the catalyst was irradiated with visible light, and this played an important role in the degradation of the dye. The dye degradation was slightly suppressed when DMSO was added to the system in comparison to t-BuOH. Under the visible light irradiation of ZnIn2S4 in the system, electrons in the valence band transfer to the conduction band and form electron–hole pairs. The photoexcited electrons also have the possibility of the recombination either at the surface or directly that may result in reduced efficiency. However, the recombination was obviously restrained when glycerol was added into the system to scavenge the holes. So, during this time, more electrons get migrated to the surface of the photocatalyst and react with O2 to produce O2–•, enhancing the dye degradation initially. Figure b shows the first-order kinetic plot varying with time supporting the above observations. There is 3.4 times decrease in the rate constant when benzoquinone (7.7 × 10–3 min–1) was introduced into the system in comparison to the control (26.4 × 10–3 min–1). The photochemical reactions that follows during the degradation process are as followsThe catalyst recyclability and stability is an important factor in heterogeneous catalysis. Thus, recyclability of the synthesized ZnIn2S4 catalyst has been carried out to understand the variation in the reactivity of the samples with consecutive reuse, till 10 cycles (Figure c). After each of the photochemical experiments, the catalyst was separated from the dye solution via centrifugation, washed with water and ethanol, followed by drying at 70 °C for 12 h in hot air oven. The dried catalyst was then reused for the second experiment consecutively. However, it can be noticed that the dye degradation decreased slightly with each consecutive recycle, suggesting that first run has the highest degradation efficiency, which decreases with the increasing number of reuse cycles. This decrease is due to the loss of catalyst during the process of recycling and also due to the blockage of active sites, resulting in less activity compared to the pristine powder. A difference of ∼9% after the 10th run indicates that the catalyst is stable and recyclable. The surface area of the catalyst before and after reuse was measured, and a difference of 7 m2/g was observed. Thus, the above data confirm that the active surface site also decreases with decrease in the surface area; therefore, the photocatalytic activity drops in consecutive experimental runs.

Figure 7

(a) Degradation profile; (b) kinetic profile of photochemical reaction carried in the presence of various scavengers (catalyst concentration 50 mg/50 mL 20 ppm MO, temperature 25 ± 2 °C, scavenger concentration 1 mM); (c) catalyst reusability studies; and (d) degradation profile for MO in the presence of natural sunlight (solar).

ZnIn2S4 is a visible light active photocatalyst; thus, the experiments are also performed under natural solar light and compared with the above experiments to understand the difference in efficiency with reduced time interval. Figure d presents the comparative plot of C/Co for the photocatalysis performed under natural sunlight (solar) and visible light. Similar to the previous experiments, photolysis of the dye was carried out under sunlight to understand the degradation of dye under the influence of sunlight. The photolysis of MO dye under sunlight was ∼22%. MO degradation under natural sunlight was 90% in the first 60 min, out of which 70% degradation happens in 30 min. On the other hand, degradation under visible light was only 55% in 30 min and 75% in 60 min. The complete degradation of 20 ppm MO was achieved within 90 min in the presence of natural sunlight without any oxidizer using hierarchical ZnIn2S4 flower structure. The probable reason behind complete degradation within less time could also be attributed to the 3–5% of UV light being constituted in the radial distribution of the natural sunlight. Comparison of the photocatalytic results of different ZnIn2S4 morphologies/structures with the present work (Table ), it is promisingly efficient catalyst for visible light driven photocatalysis for the degradation of organic pollutants.

Table 2

Comparison of Photocatalytic Performances in Other ZnIn2S4 Structures and This Work

ZnIn2S4 structures	synthesis route and chemicals	surface area (m2/g)	pollutants concentration	light source	degradation in time	ref
ZnIn2S4 microspheres	hydrothermal and ZnCl2, InCl3·4H2O, C2H5NS	85.5	methyl orange 10 ppm, Congo red 20 ppm, rhodamine B 30 ppm	500 W W-halogen lamp	100% in 240 min, 100% in 300 min, 100% in 180 min	(47)
ZnIn2S4 nano/micropeony	solvothermal and CH3OH, Zn(NO3)2·6H2O, In(NO3)2·5H2O, CS2, CH2Cl2	78.2	methylene blue 50 ppm	300 W W-halogen lamp	100% in 90 min	(37)
ZnIn2S4 microspheres	solvothermal and ZnSO4·7H2O, InCl3·4H2O, CH3CSNH2, C2H5OH	86.2	benzyl alcohol	300 W Xe lamp	80% selectivity in 120 min	(18)
ZnIn2S4 nanoparticles	hydrothermal and Zn(NO3)2·6H2O, In(NO3)2·5H2O, C2H5NS		methyl orange 20 ppm	500 W W-halogen lamp	52% in 300 min	(48)
ZnIn2S4 flower-like microspheres	ZnCl2, InCl3·4H2O, C2H5NS				100% in 300 min
hierarchical ZnIn2S4 flower structure	hydrothermal and Zn(NO3)2·6H2O, In(NO3)2·5H2O, CH4N2S	72	methyl orange 20 ppm, 2,4-dichlorophenol (2,4-DCP) 50 ppm, 2,4,6-trichlorophenol (2,4,6-TCP) 50 ppm	natural sunlight	100% in 90 min, 100% in 120 min, 100% in 75 min	this work

Photocatalytic Studies of Chlorinated Phenols

Chlorinated phenols are typical organic pollutants possessing serious threat to the water systems, as they are colorless pollutants and nonbiodegradable. The photocatalytic activity of the optimized hierarchical ZnIn2S4 flower structure has been investigated for the degradation of 2,4-dichlorophenol (2,4-DCP) and 2,4,6-trichlorophenol (2,4,6-TCP) under irradiation of natural sunlight. Figure a illustrates the degradation profile of 2,4-DCP and 2,4,6-TCP, respectively. To understand the possibility of degradation by sunlight irradiation, the blank experiment was carried out where the organic contaminants were degraded without any catalyst. The photolytic results suggests that the photosensitized self-decomposition of these colorless pollutants under the present experimental conditions is nearly 12 and 15% for 2,4-DCP and 2,4,6-TCP, respectively. Moreover, there is an appreciable adsorption of 60 and 38% of 2,4-DCP and 2,4,6-TCP on the catalyst surface. Similar to the experiments carried out for dye degradation, herealso, the absorbance values for adsorption are considered as the initial concentration of the chlorinated phenols. Almost 80% of the degradation takes place within 15 min, and then gradual complete degradation of the pollutant has been observed, where 2,4-DCP and 2,4,6-TCP is completely degraded within 120 and 75 min, respectively. The photodegradation of these nonbiodegradable colorless compounds has been found to be effective and faster in the presence of natural sunlight. The above results also suggest that the degradation becomes faster with increase in substitution.

Figure 8

(a) Degradation profile of chlorinated phenols under natural sunlight; (b) schematic photocatalytic mechanism via hierarchical ZnIn2S4 flower structures; and (c) time-resolved total organic carbon (TOC) plot for MO, 2,4-DCP, and 2,4,6-TCP.

A comparison of the degradation efficiencies of both the chlorinated phenols shows that the degradation of 2,4,6-trichlorophenol (2,4,6-TCP) was much faster than that of 2,4-dichlorophenol (2,4-DCP). These divergences in the same reaction could be due to the efficient attack at the chlorinated positions by reactive oxygen species and also due to no expected chain reaction of the generated chlorine radicals (Cl•) during the photochemical reaction. There is also a possibility of autocatalysis, which is a simultaneous attack from the Cl• radical.[49−51] The above analysis indicates that the degradation of highly chlorinated phenol is more rapid and favorable than the less-chlorinated phenols because of the increase in the production of chloride ions. Consecutively, the as-synthesized ZnIn2S4 acts as a potential photocatalyst in the degradation of dye colorants and also phenolic systems for efficient remediation toward environmental contaminants.

It is quite usual to point out that the high surface area of hierarchical structures mainly contributes to the improvement in the photocatalytic activity, and the increase in photoabsorption due to the interspace between the petals also increases the efficiency of degradation. Another major advantage is the enhanced adsorption that enables the fast photocatalysis via superoxide radicals preventing the recombination of electron–hole due to proficient photochemical reactions, as shown in Figure b. All of these considerations suggest that the as-synthesized hierarchical ZnIn2S4 flower structures are potentially useful photocatalysts for many versatile organic pollutants.

The photochemical degradation reaction can undergo both decolorization and mineralization processes. To further confirm these occurrences, the total organic carbon (TOC) content in the solution taken at regular intervals was evaluated for both the dye and chlorinated phenols. The TOC/TOCo values are plotted against solar irradiation time and have been shown in Figure c. It can be observed that the total organic carbon decreased nearly 70, 55, and 63% for MO, 2,4-DCP, and 2,4,6-TCP after 90 min of solar irradiation, respectively. The TOC results suggest that the photochemical process in case of MO dye not only undergoes decolorization but also mineralizes the dye solution to a greater extent, whereas the colorless chlorinated phenols are also observed to be mineralized. Thus, it is clear from the figure that not only decolorization but also mineralization takes place during photocatalytic reaction using ZnIn2S4 flowers as photocatalyst under natural sunlight.

Conclusions

The processing parameters and the precursor concentration play a major role in the development of hexagonal hierarchical ZnIn2S4 marigold-like flower architectures. The enhancement in the photocatalytic degradation of MO under visible light can be credited to the increase in the dye adsorption due to open cavities in the flower structure facilitating active sites for adsorption and catalysis. The photocatalytic degradation of MO in the system was investigated to be via the superoxide radical that acts as the key active species in the oxidation process. The catalytic activity can be improved by adding hole scavengers to inhibit the recombination of holes and electrons, resulting in more electrons reducing oxygen to superoxide radicals. Complete degradation of both dye and chlorinated phenols was observed within 90 min under natural sunlight irradiation. The catalyst showed efficient recyclability and reusability.

Experimental Section

Synthesis of ZnIn2S4 Architectures

In a typical synthesis process, stoichiometric ratios of zinc nitrate pentahydrate (Zn(NO3)3·5H2O) and indium nitrate pentahydrate (In(NO3)3·5H2O) were dissolved in 30 mL distilled water. A stoichiometric amount of thiourea (CH4N2S) was added to the above solution and stirred magnetically to obtain complete dissolution of the compounds. The resulting solution was then transferred to a Teflon-lined beaker having a volume capacity of 60 mL. The capped beaker was put inside a stainless steel autoclave and sealed tightly for hydrothermal treatment at a predetermined temperature in a muffle furnace. After certain time intervals, the autoclave was cooled to room temperature and the precipitate together with the solvent was centrifuged at 5000 rpm to separate the precipitate from the solvent. The precipitate was washed thoroughly with water and absolute ethanol. The centrifuged precipitate was dried at 70 °C for 12 h. The concentration of thiourea played a major role in the formation of crystal phase. Thus, hydrothermal experiments were conducted by varying the ratio of thiourea in the precursor solution. Furthermore, a series of experiments were carried out with varying hydrothermal temperature and time to optimize the crystal structure and morphology. The hydrothermally obtained powders were characterized for their crystallinity, crystal structure, and morphology by different techniques given below.

Characterizations

Powder X-ray diffraction (XRD) patterns were recorded on Rigaku Miniflex 600 X-ray diffractometer having Cu Kα target with a step size of 0.01 and at the scan rate of 4°/min. Scanning electron microscopic (SEM) imaging was performed using Ultra-55, FESEM Carl Zeiss. TEM samples were prepared by drop casting the dilute ethanolic solution of dispersed photocatalyst on the carbon-coated Cu grid followed by drying in a vacuum desiccator overnight. The specific BET surface areas of the synthesized photocatalysts were measured in a Smart Sorb 92/93 surface area analyzer by nitrogen adsorption/desorption. X-ray photoelectron spectroscopy (XPS) measurements were performed on an ESCA+, Omicron Nanotechnology, Oxford Instruments, Germany, equipped with monochromator Aluminum Source (Al Kα radiation hυ = 1486.7 eV). UV–vis diffuse reflectance spectra were measured using PerkinElmer UV/vis/NIR Lambda 750 spectrophotometer equipped with integrating sphere. Photoluminescence spectra were recorded on LS-55 with 450 W continuous xenon arc lamp as the excitation source and excitation wavelength of 350 nm.

Photocatalytic Studies

Methyl orange (MO) was taken as a model pollutant to evaluate the photocatalytic property of the optimized photocatalyst. The concentration of the standard stock solution of MO was 20 mg/L. For each of the photocatalytic experiments, catalyst concentration was maintained at 1 g/L. Metal halide lamp (Philips India Ltd.) with λmax around 510 nm was used to perform all of the photodegradation experiments. The lamp was cooled down using the surrounding water jacket connected to the chiller for maintaining the reaction at room temperature. The intensity and the photon flux of the metal halide lamp were determined to be 10–7 einstein/s and 220 μW/cm2, respectively. The reaction suspension containing catalyst and dye solution was continuously stirred in dark conditions to reach an adsorption–desorption equilibrium. During the catalytic reaction, 2 mL of suspension was taken out at certain time intervals followed by centrifuging at 3000 rpm to separate the catalyst from the dye solution. The absorbance of the clear dye solution was measured by a UV–vis spectrophotometer. The percentage of degradation was calculated as follows: % degradation = ((Co – C)/Co) × 100, where Co and C are the initial (after adsorption) and final absorbance (after photocatalysis) at time t, respectively. Dye solution was also degraded in the presence of light to understand the photolysis process. All of the sunlight experiments were carried out between 12 and 2 p.m., as the fluctuations in solar intensity are minimal in this duration. The variation in solar intensity was recorded using the Eppley PSP 32483F3 radiometer, Newport. The average solar intensity observed was 0.768 kW/m2. For investigating the active species involved in photodegradation of 20 ppm MO using the optimized catalyst, scavengers were added to the dye solution. The scavengers taken were t-butyl alcohol for hydroxyl radicals (OH•), benzoquinone for superoxide radicals (O2–•), glycerol for holes (h+), and dimethyl sulfoxide for electrons (e–). The scavenger of 1 mM concentration was added to the reaction solution in the dark reaction (before switching on the light).

Apart from the dye degradation, the catalyst was also used to monitor the degradation of chlorinated phenols, namely, dichlorophenol (2,4-DCP) and trichlorophenol (2,4,6-TCP). However, the standard stock concentration was 50 mg/L. The samples were taken at definite intervals after degradation, and the concentration was determined based on the absorbance at λmax of 280 nm with UV detector using high-performance liquid chromatography (Waters Inc.). C18 column (Poroshell 120) and 50/50 acetonitrile/water with 0.5% formic acid was used as the eluent at a flow rate of 0.4 mL/min.

The total organic carbon (TOC) of both the dye solution and chlorinated phenols at different time intervals of the photochemical reaction was analyzed with the Shimadzu, TOC VCSNanalyzer.