Ternary Metal Chalcogenide Heterostructure (AgInS2-TiO2) Nanocomposites for Visible Light Photocatalytic Applications.

Hybrid nanoarchitectures of AgInS2 and TiO2 photocatalysts were prepared by using a modified sol-gel method. The experimental results reveal that these nanocomposites display enhanced visible light absorption and effective charge carrier separation compared to their pristine parent samples (AgInS2 or TiO2). 0.5 wt % AgInS2 loading was found to be the optimum concentration for photocatalytic applications. More than 95% of doxycycline degradation was achieved within 180 min of solar light illumination. Similarly, the dopant concentrations at lower values (<2 wt %) exhibited 300 times higher H2 generation rate under visible light irradiation compared to AgInS2 and TiO2. The microbial strains (Escherichia coli and Staphylococcus aureus) exhibited a 99.999% reduction within half an hour of simulated solar light illumination. The computational investigation was employed to understand the structural, electronic, and the dielectric properties of AgInS2 and TiO2 composites. The improved photocatalytic results are explained as a result of the decreased rate of exciton recombination. The current investigation opens up new insights into the use of novel ternary heterostructure nanocomposites for improved visible light activity.

Introduction

Fossil fuels such as coal and natural gases have been the mainstay for the industrialized world for the past century.[1,2] However, their rapid depletion and ever-increasing demand raise some grave challenges.[3,4] Rising global temperature and population are the key elements of concern in relation to energy shortage and environmental pollution.[5,6] Hydrogen is considered as the fuel for the 21’st century.[7−9] However, processes such as steam reforming utilizing coal are still the major route to generate H2.[10−13] Similarly, the water plants across different cities around the globe utilize the conventional route of adsorption, sedimentation, and flocculation as secondary water treatment protocols.[14,15] This results in the disposal of untreated hostile effluents to the ecosystem, which further leads to channelization to a higher order of the food chain via biomagnification.[16,17] The lack of cost-effective and sustainable technologies has been a challenge for this region.[14,15] Nevertheless, sustainable technologies such as solar and wind energy have emerged to be efficient alternatives for modern civilization.[18,19] Among the various sustainable technologies, photocatalysis is identified as one of the long-term, stable, environmentally friendly alternative, to produce green energy and environmental remediation.[4,20−23] Semiconductor-based heterogeneous photocatalysis has the ability to utilize solar energy to convert and store into other renewable forms of energy, chemical fuels, degradation of harmful effluents, and even disinfection of microorganisms.[24−28] A photocatalytic process is defined by five basic steps: (1) Absorption of light on the semiconductor surface, (2) creation of photogenerated electron–hole pairs, (3) migration of these excitons or recombination, (4) adsorption of reactants and the desorption of the products, and (5) completion of redox reactions on the photocatalytic surface.[29−31] Thus, finding light-absorbing materials with semiconducting properties is not enough, but tailoring the design to effectively enhance the interfacial charge transfer and decreasing the rate of recombination is the key attribute. Various procedures such as band gap modulation, defect formation, dimensionality alteration, pore texture alteration, surface sensitization, heterojunction creation, etc. are attempted to reduce the recombination rate and enhance the visible light absorption.[24,29]

Titania is a traditional semiconductor material, which garnered significant attention in 1972 when Honda and Fujishima demonstrated its use in the electrochemical water splitting reaction to produce hydrogen.[32,33] Apart from the key properties, there exist major drawbacks such as low visible light absorption and wide band gap.[18,34,35] However, heterostructure formation of titania with other semiconductor materials has pushed them back into the spotlight. Ternary chalcogenides are another class of multicomponent elements, which have attracted attention for their small band gap and excellent photon absorption in the visible region of the electromagnetic spectrum.[25,36] These attractive multielement structures have been in the research realm for the past 3 decades, but their utilization as a possible heterostructure component for photocatalytic applications is still underexploited. Although copper-based ternary compounds have recently been explored for several photocatalytic applications such as degradation of emerging pollutants, real wastewater treatments, CO2 reduction, and energy generation,[37−40] silver-based ternary compounds as a potential heterostructure component are yet to be studied and there exists enough opportunity to explore.

Amid all, AgInS2(I-III-VI), a ternary semiconductor, crystallizes into two different polymorphs; a room-temperature stable tetragonal chalcopyrite structure and the other high-temperature orthorhombic wurtzite structure.[41] The phase transition occurs at 620 °C, and these nanocrystals are known to be synthesized using different liquid-phase reactions.[42−45] The orthorhombic AgInS2 has been reported for various photocatalytic applications.[46] The small band gap and excellent photon absorption in the visible region of the electromagnetic spectrum aid in several efficient processes. Nevertheless, these materials display reduced quantum efficiency and strong light etching that confines its significant application.[47] Yin and Cui et al. reported the synthesis of AgInS2 nanoparticle-sensitized TiO2 nanotube arrays and AgInS2 cluster films on the TiO2/FTO substrate, respectively. These composites were investigated for their photoelectrochemical properties.[48,49] Similarly, Liu et al. effectively studied the same composite and illustrated the degradation of 1,2 dichlorobenzene.[47] Similar heterostructural composites such as AgBiS2-TiO2 were also previously reported.[50] However, there still exists significant potential of these materials for their various functional applications, which remains unexplored. Studies of these composites for photocatalytic hydrogen production, doxycycline (DC) degradation, and antibacterial disinfection have not yet been reported.

Hence, in the current investigation, we outline the optimal solvothermal synthesis parameters of AgInS2 and developed a staggered heterostructure composite of TiO2 at different wt % (0.5, 1, 2, and 5). These as-prepared composites illustrate positive heterostructure formation in the Raman and the X-ray photoelectron spectroscopy (XPS) results. Moreover, the UV–vis diffuse reflectance spectroscopy (UV-DRS) spectra illustrate the enhanced visible light absorption by the composite samples. The decreased band gap of the material enabled in illustrating 300 times enhanced hydrogen output, 5-log bacterial inactivation within 30 min of illumination, and more than 95% degradation of doxycycline (DC) within 3 h of irradiation. The introduction of even a low dopant ratio of AgInS2 effectually harnessed the efficiency of the incoming light and decreased the recombination rate. A plausible photocatalytic mechanism and the interfacial charge transfer are discussed.

Results and Discussion

Computational Results

In this section, computational results concerning the structural properties, the electronic structure, the projected density of states (PDOS), and dielectric properties of TiO2 and AgInS2 are presented. This helped to understand the underlying photocatalytic mechanism. The initial structures of TiO2 and AgInS2 were obtained from the Materials Project library. The initial geometry of AgInS2 had the lattice constants a = 6.81 Å, b = 7.15 Å, c = 8.35 Å, and the angles were α = β = γ = 90°. After the structural optimization, there were no changes in the lattice angles, but the lattice constants slightly changed upon volume relaxation to a = 6.75 Å, b = 7.05 Å, and c = 8.25 Å. The volume was relaxed by 3.5% in the case of orthorhombic AgInS2. The X-ray diffractogram (XRD) pattern of the orthorhombic phase of AgInS2 obtained theoretically is also given in Figure S1. Both the Ag and In cations have tetrahedral coordination with S. The AgS4 and InS4 tetrahedra are connected to one another at the corners (see polygons diagrams in Figure ).

Figure 1

(Color Online) Optimized crystal structure of AgInS2. Color code: gray = Ag, light red = In, and red = S.

TiO2, on the other hand, had lattice constants a = b = 10.16 Å and c = 2.96 Å and the lattice angles α = β = γ = 90° at the end of the geometry relaxation where the volume was reduced by 3.77% from the initial structure. The initial structure of TiO2 had the lattice parameters a = b = 10.32 Å and c = 2.98 Å and angles α = β = γ = 90°. The Ti cations and O anions are connected to each other in an octahedral coordination. Alternate edge and corner sharing connect these octahedra to each other, as seen in Figure .

Figure 2

(Color Online) Crystal structure of TiO2 in the tetragonal phase. Color code: blue = Ti and red = O.

There is an underestimation in the band gap value with respect to the experimentally observed value even though the computed lattice parameters fall within a 1% error with respect to the experimental ones. This is not unexpected as the generalized gradient approximation (GGA) functional used is well-known to underestimate the band gap.[54] In the calculation of the band structure, the Perdew–Burke–Ernzerhof (PBE) functional under GGA was used owing to its computational speed. This functional is known to contain unphysical Coulomb repulsion, which leads to a poor description of the band gap.[51,52] There are works that report the PBE band gap of well-known compounds, which are shown to underestimate the experimentally observed band gap.[3] In fact, there are many insulators (nonzero band gap) such as FeO and GaSb that were estimated as metals (zero band gap) upon the usage of PBE functional.[53] The band structure of AgInS2 is presented in Figure a and shows a direct band gap of ∼0.5 eV. Both the valence band maxima (VBM) and the conduction band minima (CBM) lie on the Γ point in the Brillouin zone. The upper valence band (UVB) is dominated by S p-orbitals and Ag d-orbitals and the lower conduction band (LCB) by the In s-orbitals and S p-orbitals. Energy larger than the computed band gap value is expected for the electrons to transfer from VB to CB due to the p–p forbidden transitions. In the current work, the experimentally observed band gap of AgInS2 is 1.95 eV (Figure b), and the PBE band gap is 0.5 eV, which is within an error of 74.36%. The band structure and partial density of state (PDOS) plots were only used as a tool to illustrate the orbital occupancies near the band edges.

Figure 3

Calculated band structure of (a) AgInS2 and (b) TiO2 with the corresponding PDOS. The top of the valence band (horizontal line) is aligned at 0 eV.

Figure 11

(a) DRS spectra of AgInS2. (b) Band gap estimation of AgInS2.

TiO2, on the other hand, is an indirect band gap material with a band gap of ∼2.7 eV. The band structure and the PDOS of TiO2 are shown in Figure b. The VBM lie between Z and P points, and the CBM lie between the Γ and X points of the Brillouin zone. The O p-states dominate the UVB, and the Ti d-states dominate the LCB. The Ti p-states were also present at the UVB of TiO2.

Absorption spectra for AgInS2 and TiO2 are calculated in the independent particle approximation and plotted in Figure in the range of 0–7.5 eV input light. The first peak of the imaginary part of the dielectric function, which indicates the energy value at which the actual electron transition starts, of AgInS2 and TiO2, was found to be 2.04 and 3.43 eV, respectively.[55] AgInS2 is optically anisotropic, but along the x- and y-directions, the light absorption spectrum is isotropic for TiO2.

Figure 4

Imaginary part of the dielectric function (Im(ε)) of (a) TiO2- and (b) AgInS2-calculated infrared, visible, and ultraviolet regions.

The band gap and the optimized crystal structure of AgInS2 in the tetragonal phase were also calculated (Figure S2).The geometry was obtained from the Materials Project library with the lattice parameters a = b = 5.93 Å and c = 11.52 Å and the angles α = β = γ = 90°. After relaxation, the volume of the cell was reduced by 3.3% and the lattice constants were obtained as a = b = 5.85 Å and c = 11.42 Å without changing its angles. The band gap of AgInS2 in the tetragonal phase was found to be 0.36 eV. The crystal structure of AgInS2 attained via experimental synthesis exists in an orthorhombic phase, hence the computational conclusions drawn for the orthorhombic phase is discussed hereafter, except otherwise identified.

Synthesis and Crystal Structure

The synthesis process of AgInS2–TiO2 is illustrated in Figure . Typically, the TiO2 nanoparticles prepared by the sol–gel method undergo solvothermal process resulting in the in situ formation of the AgInS2 nanoparticles.[47] This results in a uniform distribution of the dopant particles and formation of close intimate heterojunctions as discussed in the latter sections. The challenges of synthesizing these semiconductor nanostructures are their stability, controlling the stoichiometric composition, and the crystal structure.[37,38,56] As similar to other bulk materials, the nanoparticles of the ternary compounds exhibit interesting electronic and optical properties compared to their bulk counterparts.[57] The formation of the low-temperature stable orthorhombic structure of AgInS2 could be ascribed to the solvent picked for the reaction and the balanced use of the two cationic precursors used in the synthesis process. The appropriate reaction temperature and time aid in tuning the reactivity of Ag and In precursors (as observed in the diffractograms of time and temperature studies of AgInS2). As Ag+ is a soft Lewis acid, therefore appropriate Lewis base such as ethanolamine (EA) is used as a suitable ligand required for the complexation process. Moreover, the presence of lone nitrogen chelating atom on an amine molecule makes EA, a potential structure forming mediator.[58,59]

Figure 5

Schematic illustration of the synthesis mechanism of AgInS2–TiO2 composites.

The formation of AgInS2 nanoparticles using a solvothermal technique could be explained as followsAt the beginning of the reaction, the Ag+ and In3+ ions form covalent complexes as given above (where x and y are positive integers), due to the use of EA as a solvent.[60,61] Additionally, these complexes react with thiourea (TH) resulting in the formation of complexes asThe formation of any free radicals of S2– from thiourea is proposed to be prevented to some extent, due to the use of EA as a solvent. This mechanism averts the creation of any additional binary impurity phases, for example, In2S3 or Ag2S. The stability of thiourea falls on enhancing the temperature and eventually results in the formation of more stable ternary chalcogenide phase to yield AgInS2. Meanwhile, the thiourea in the reaction plays a dual nature partly by aiding in the formation of a complex and also as a source of sulfur other than thiourea.[62,63]The structure and the crystallinity of the as-synthesized AgInS2–TiO2 composites were investigated using X-ray diffractogram (XRD). Figure displays the formation of the as-prepared AgInS2 and AgInS2–TiO2 composites. Orthorhombic phase of AgInS2 [a = 7.00 Å, b = 8.28 Å, c = 6.69 Å, and α = β = γ = 90°] with peaks defining at 24.90, 25.40, 26.80, 28.50, 36.70, 43.67, 44.78, 48.23, and 51.80° owing to (120), (200), (002), (121), (122), (040), (320), (203), and (042), respectively, is observed [JCPDS -00-025-1328].[46] The XRD peak intensity at 26.8° of the (002) plane shows a higher intensity. Although, the primary peak in the given case might reflect the formation of the tetragonal phase. However, the presence of prominent peaks at 24.9, 25.4, and 28.5° prove otherwise. The tetragonal phase of AgInS2 does not exhibit any peaks at <26° (2⊖ value). The slightly widened peaks could be attributed to the decreased nanocrystal size as observed in the transmission electron microscopy (TEM) images (Figure ) and also observed in the previous reports.[64−66] Moreover, the Raman results aid in confirming the orthorhombic phase as the peaks at 264 and 316 cm–1 correspond to the orthorhombic wurtzite structure, while the peaks of the tetragonal phase only appear at 240 and 345 cm–1, which are absent in this case (the inset of Figure ).[43,46,67,68]

Figure 6

X-ray diffractograms of AgInS2, TiO2, and AgInS2–TiO2 composites.

Figure 15

TEM pictures of (a) Titania; (b) AgInS2; and (c), (d) AgInS2–TiO2.

Figure 12

Raman analyses of AgInS2, TiO2, and AgInS2–TiO2 composites.

The diffractograms in Figure compare the as-synthesized AgInS2 with other peaks of Ag2S [JCPDS 00-014-0072], In2S3 [JCPDS 00-033-0624], AgIn5S8 [JCPDS 00-026-1477], and AgInS2 tetragonal [JCPDS 00-025-1330]. It displays the orthorhombic AgInS2 formation, deprived of any impurity.

Figure 7

XRD of as-prepared AgInS2 with other standard peaks of AgInS2 orthorhombic, AgInS2 tetragonal, Ag2S, and In2S3.

The temperature and time studies were accomplished, with the purpose of optimizing the reaction parameters for solvothermal synthesis. Figure a illustrates the temperature study of AgInS2 at altered temperatures (150–250 °C) for 12 h. The crystallinity of the sample increases with the rise in temperature from 150 to 180 °C (the peak observed at 26.80°). The intensity of the identical peak declines on increasing the temperatures after 200–250 °C. Therefore, 180 °C was applied as an ideal temperature for further assessing the time period for this solvothermal synthesis. Figure b exhibits the outcome of time study, which demonstrates the surge in the crystallinity of the peak at 26.10° on enhancing the time duration from 6 h to 12 h. The intensity of the peak at 26.80° decreases on assessment up to 72 h of the synthesis period. Therefore, 180 °C for 12 h is reflected as the optimum solvothermal parameters for AgInS2.

Figure 8

(a) Temperature and (b) time studies of AgInS2 solvothermal synthesis.

Figure also exhibits the diffractograms of TiO2 and its composites with AgInS2. Peaks corresponding to AgInS2 are not observed in the current diffractogram, which is endorsed to the high crystallinity of the TiO2 nanoparticles and the small doping amount of AgInS2. Sharp peaks of TiO2 [a = b = 3.78 Å and c = 9.50 Å] at 25.40, 38.50, 48.00, and 55.00° to (101), (112), (200), and (211) correspondingly are likewise displayed in the diffractogram.[69,70]

The composite structures as exhibited in Figure neither showed any change in crystallinity nor there is any indication of rutilation (impurity phase) of TiO2. Moreover, the lattice constant acquired for the bare samples presented a decent agreement to the computationally obtained values. The 0.5 wt % AgInS2–TiO2 composites demonstrated the top photocatalytic outcomes (as shown in the latter segments) amid their parent and other dopant % samples. Therefore, henceforward, it is used for additional characterizations except otherwise identified.

Optical Properties

To examine the optical absorption behavior of the as-synthesized materials, UV–vis diffuse reflectance spectroscopy (DRS) spectra were employed. Figure shows the diffused reflectance spectra of titania and AgInS2–TiO2 composites. TiO2 exhibits a maximum at 350 nm (UV region), which corroborates the computationally obtained first peak of the imaginary part of the dielectric spectra at 3.8 eV. A substantial improvement in the absorption in the visible region is detected in AgInS2–TiO2 materials.[70] The composite samples exhibited the signature sharp hump of TiO2 at 380 nm; however, the absorption edge is extended up to 800 nm.

Figure 9

DRS spectra of TiO2 and AgInS2–TiO2 composites.

The indirect band gap measurements of the composite samples and titania are assessed, as shown in Figure . Thus, this substantiates the results computed theoretically (Figures b and 4a). The composite samples show a decreased band gap value with an increase in the AgInS2 content. This effectively proves the successful doping of the sample and the formation of the heterostructure.[47]

Figure 10

Band gap estimation of titania and AgInS2–TiO2 composites.

Figure a illustrates the DRS spectra of AgInS2, it shows an enhanced visible light absorption, and the edge extending up to 800 nm. The corresponding band gap is also calculated and shown in Figure b. As observed through the absorption spectra, a high visible light absorption resulted in a small band gap of 1.95 eV, which is consistent with the theoretically computed results as discussed in an earlier section (Figures a and 4b). The first peak of the imaginary part of the dielectric spectra obtained at 2.04 eV restates the experimental findings of absorption maxima observed at 600 nm (Figure b).

The Raman analyses of titania, AgInS2, and AgInS2–TiO2 are illustrated in Figure . Characteristic vibrational peaks of the wurtzite structure of AgInS2 are observed at 264 and 316 cm–1 (as shown in the inset).[43,46] Apart from that, characteristic bands for titania are detected at 147, 197, 396, 516, and 638 cm–1 for A1g, 2B1g, and 3Eg correspondingly attributed to the anatase phase.[69,71]

Sharp and symmetric peaks of anatase appear in the case of the composite sample and a red shift is also detected for the composite maxima (as specified in the inset of Figure ), and this successfully proves the efficient doping of the TC atoms into the TiO2 host material. Moreover, the Raman spectra of the composite do not show the existence of any other impurity peaks, which corroborates to the inferences observed in XRD.[72,73]

The recombination rate of the solar light-generated electron–hole pair is evaluated using the photoluminescence (PL) measurement. The PL spectra of AgInS2–TiO2 and TiO2 are displayed in Figure (mass normalized). The emission peak intensity of AgInS2–TiO2 is reduced compared to TiO2. The lesser PL intensity specifies the poor rate of recombination of electron–hole pairs. This suggests that the introduction of AgInS2 results to retardation in the recombination rate of the excitons, which overwhelmingly improves the photocatalytic efficiency.

Figure 13

Photoluminescence spectra of AgInS2–TiO2 and TiO2.

Chemical Composition Analysis

The chemical environment and the bonding characteristics are analyzed using X-ray photoelectron spectroscopy (XPS). The survey spectra exhibit the presence of Ti 2p and O 1s in spectra of TiO2. Peaks of Ti 2p, O 1s, Ag 2d, In 3d, and S 2p were observed for AgInS2–TiO2 composites. Very low intensity peaks of C 1s and N 1s are also detected among all of the materials, which might be acquired through the experimental or analysis procedure. Figure A illustrates the Ti 2p spectra of titania (a) and the composite of AgInS2–TiO2 (b).

Figure 14

High-resolution deconvoluted spectra of (A) Ti 2p (a) titania and (b) AgInS2–TiO2; (B) O 1s (a) TiO2 and (b) AgInS2–TiO2; (C) Ag 3d (a) AgInS2 and (b) AgInS2–TiO2; (D) In 3d (a) AgInS2 and (b) AgInS2–TiO2; and (E) S 2p (a) AgInS2 and (b) AgInS2–TiO2.

Peaks of Ti 2p3/2 are observed at 457.8 eV, assigned to Ti+4. The Ti 2p peaks remained unaltered in both cases. Moreover, the O 1s spectra illustrated in Figure B show no change in the values of 528.59 and 530.40 eV consistent to crystal lattice oxygen of (O–Ti4+) and adsorbed O2 molecules correspondingly in TiO2 (a) and AgInS2–TiO2 (b) samples.[33,74,75] The Ag 3d high-resolution spectra for both AgInS2 (a) and AgInS2–TiO2 structure (b) are displayed in Figure C.

The signature points of Ag 3d3/2 and Ag 3d5/2 are detected at 372.73 and 366.72 eV correspondingly for the AgInS2 sample.[76,77] The peak shifts slightly to 372.1 and 366.1 eV for the AgInS2–TiO2 sample.[44] Nevertheless, the doublet separation of the Ag 3d oxidation state remained unaltered (6 eV).[78] A similar shift in peaks is observed for In 3d peaks at 451.52 and 443.97 eV corresponding to In 3d3/2 and In 3d5/2, respectively, to 451.24 and 443.63 eV for AgInS2–TiO2 (Figure D). The doublet separation of 7.55 eV was observed for the In 3d peaks, consistent with the previous literature.[79]Figure E illustrates the high-resolution spectra of S 2p. Peaks at 161.41 and 160.26 eV correspond to the S 2p1/2 and S 2p3/2 are observed for AgInS2 samples (doublet splitting of 1.15 eV).[80] Apparently, significant peaks of sulfur were not detectable in that same region; however, broad peak around 161.4 eV is noticed. The peak location of all of the elements is summarized in Table S2.

Thus, the effective interaction amid the heterostructure elements results in the rise in electron concentration, which leads to the interfacial charge transfer causing the shift of the peak detected. The effective electron screening results in the increase in electron concentration leading to a decrease in binding energy.[81−84]

Morphological and Microstructure Analysis

The transmission electron microscopy (TEM) is employed to evaluate the morphology and microstructure of the samples prepared. The TEM of titania and its selected area diffraction pattern is demonstrated in Figure a. The nanoparticles exhibited ovular features, and the typical size was observed to be around 25–40 nm. The interplanar spacing (d) values obtained for diverse areas of the image confirm the formation of the anatase phase of the nanoparticles.[85]Figure b illustrates the TEM image of pristine orthorhombic AgInS2. The nanoparticles likewise displayed ovular nature, and the clear lattice fringes were identified of (002) and (121) planes with respective interplanar spacing values of 0.314 and 0.176 nm are observed.

In the meantime, Figure c,d shows the TEM images of AgInS2–TiO2 composites. The presence of asymmetric particles identified as AgInS2 in TiO2 is observed. However, the existence of the nanoparticles demonstrates that both AgInS2 result in intimate integration with the TiO2 nanoparticles, and thus result in the formation of the heterostructure.[47,86] High-resolution TEM (HRTEM) images of the composites are further provided in Figure .

Figure 16

(a) and (b) HRTEM images of AgInS2–TiO2.

Photocatalytic Applications

Photodegradation

Pharmaceutical effluent has grown to be a major contaminant source in rivers and groundwater. The rise in contamination shall result in the growth of antibiotic-resistant bacteria in the water bodies. Hence, effective secondary wastewater treatment has resulted to be an imperative solution to this crisis. Photocatalysis is an excellent alternative, and hence in this study, doxycycline (DC) is used as a modal antibiotic to assess the photocatalytic efficiency of AgInS2–TiO2. The photocatalytic trials were done under dark and solar light illumination correspondingly.[87]Figure demonstrates the degradation pattern of AgInS2–TiO2 materials by presenting the variation in DC amount (in concentration) with time [C/C0 % vs time]. As illustrated in Figure , the pristine parent and the composite sample displayed no adsorption activity. Moreover, the DC solution (blank sample) exposed under the same condition did not report any change in the profile and thus nullified any possibility of photodecomposition.

Figure 17

Concentration change of DC with time with TiO2, AgInS2, and AgInS2–TiO2 composites.

Figure 18

(a) Cyclic study and (b) results of species trapping experiment of 0.5 wt % AgInS2–TiO2.

Under visible light irradiation, the AgInS2–TiO2 composites exhibit remarkable photocatalytic efficacy as compared to the TiO2 and AgInS2 exposed under the same condition. The pristine TiO2 and AgInS2 presented the least activity compared to the composite samples. The AgInS2, even though having a lower band gap value, is not sufficient to display enhanced results. This might be ascribed to the higher recombination of the photogenerated excitons (e–h pairs). Although all of the composites unveiled remarkable degradation profiles, the 0.5 wt % material showed the best and the 5 wt % of AgInS2–TiO2 composite presented the lowest activity within 180 min of irradiation. Henceforth, it could be said that the composites display the best outcomes simply at lesser addant ratios. These effects might be additionally confirmed by the hydrogen generation outcomes as conferred in the later section.

So, as to assess the reusability competence of the materials, the used sample was washed, dried, and reused after every single photocatalytic study. The C/C0 % vs time graph of 0.5 wt % AgInS2–TiO2 is illustrated in Figure a. The degradation profile remained unaltered even after 4 cyclic runs, and this illustrates the efficiency and the stability of the catalyst prepared. The scavenging experiment was also completed to clarify the contribution of the potential reactive oxygen species associated with the degradation phenomenon. Scavenger chemicals, AgNO3, triethanolamine, benzoquinone (BQ), and isopropyl alcohol (IPA), were added to the degradation system. As observed in Figure b, it compares the degradation profiles of 0.5 wt % AgInS2–TiO2 with the 4 sets of systems in the reaction mixture on irradiated, under the same irradiation condition. The decrease in the activity was considerably altered on the introduction of silver nitrate and BQ. Thus, electrons and holes are observed to be the major active entities in the reaction mixture. Moreover, the IPA-added reaction mixture also exhibited a nominal decrease, which outwardly demonstrates that hydroxyl radicals are not the key players in the photocatalytic activity. Therefore, electron and holes might be established as the key reactive oxygen elements for AgInS2–TiO2. Moreover, Liu et al. and Zhang et al. studied AgInS2–TiO2 for various applications. They verified the results of the scavenging experiments by the electron paramagnetic resonance (EPR) technique. The composite samples did not display any signal under dark, while on light irradiation, the AgInS2–TiO2 displayed four peak intensities for DMPO-O2*. The DMPO-OH– exhibits a very weak signal intensity, hence substantiating to the results observed in the scavenging experiments.[47,88] Nevertheless, earlier results of EPR on TiO2 and AgInS2 parent samples indicate the formation of superoxide radicals. He et al. studied the major reactive species generated on visible light irradiation on AgInS2 nanoparticles.[89] Weak signals of hydroxyl and singlet oxygen were observed; however, superoxide radicals exhibited strong signals, which concluded it as the major species in the study. Similarly, anatase TiO2 illustrates weak signals for hydroxyl and singlet oxygen for UV light irradiation, and strong peaks were observed for superoxide radicals. The visible light illumination also left weak signals for all of the three major species.[90,91]

The gas chromatography–mass spectrometry and high-performance liquid chromatography–mass spectrometry are utilized to identify the polar and nonpolar intermediates generated in the photocatalytic degradation process.[92] The chromatograms observed aid to define the possible degradation pathway of the pollutant. Zhu et al. define a plausible degradation pathway, as illustrated in Figure .[92] The holes generated in the process accept electrons from the pollutant molecules and result in the formation of intermediate products. In the very initiation of the process, DC molecules change from a m/z ratio of 445 to m/z of 431 by losing an N-methyl group and resulting in formation of anionic species. Additional degradation leads to generation of intermediates at m/z 400. A continuous light exposure results in further mineralization resulting in the complete disintegration of the DC molecule. The resulting disintegrated smaller parts of the complex structure undergo additional oxidation and lead to generation of smaller organic molecules such as acids and alcohols.[92−94]

Figure 19

Graphic image of a possible photocatalytic degradation trail of doxycycline.

Photocatalytic H2 Production

With the aim of evaluating the photocatalytic competence, the synthesized composites of AgInS2 were investigated for H2 generation. Visible light-induced water splitting and photoreforming of organic species in aqueous solution have been considered as crucial processes to convert solar to chemical energy. The use of sacrificial organic agents helps to obtain higher photoefficiency due to the oxidation of organic species by photogenerated positive holes, with the simultaneous production of H+ ions. Photogenerated electrons on the surface of the catalyst encounter with hydrogen ions to produce hydrogen gas. Therefore, merging organic wastewater supervision and H2 generation via sunlight might be an encouraging sustainable approach to tackle environmental and energy issues.

The H2 generation rate of AgInS2–TiO2-synthesized materials together with their bare samples (AgInS2 and TiO2) was measured. Methanol ([CH3OH] = 10 vol %) as a sacrificial agent was utilized in this experimental process. As shown in Figure a, the catalysts tested show appreciable hydrogen production for 0.5, 1, and 2 wt % for all over the photocatalytic runs. The composite at low dopant % exhibited H2 output as high as 310 μmol/min. On the contrary, the parent pristine samples (AgInS2 and TiO2) showed 0 μmol/min. The composite structure exhibits efficient output at lower wt % (up to 2); on increasing the dopant value to 5 wt %, the photocatalytic efficiency decreases. With respect to the starting pH values of the solution, no noteworthy variations have been logged for the AgInS2–TiO2 composite materials during the illumination duration (Figure b).

Figure 20

(a) Rate of hydrogen generation and (b) influence of the pH of the solution at different doping percentage of AgInS2–TiO2. [CH3OH] = 10 vol %; sample added = 500 mg/L at RTP.

The illumination measures recorded in the ultraviolet and visible regions of the electromagnetic spectrum on the exterior surface of the apparatus for various AgInS2–TiO2 samples tested are reported in Figure a,b, respectively.

Figure 21

Illumination recorded at the exterior of the reactor walls amid (a) 315–400 nm and (b) 400–1100 nm at different doping ratios of AgInS2–TiO2 samples. [CH3OH] = 10 vol %; sample added = 500 mg/L at RTP.

Subsequently, eliminating the ultraviolet A filter, a prompt surge to the UV illuminance measure has been logged. The higher illuminance values herein logged for the AgInS2 material indicating a lesser suspension absorption, thus resulting in the absence of photocatalytic hydrogen generation. The illumination measures obtained from the outer walls of the reactor did not exhibit any prominent alteration at the visible region. On comparing the irradiance values with hydrogen production rates in the visible region for samples with different composition in weight percentage, it can be stated that higher radiation absorption values (i.e., lower irradiance values recorded on the photoreactor walls) result in slightly improved hydrogen generation rates. In accordance with pecking order of irradiance values in the visible region, the photocatalytic activity had the following sequence: 2% AgInS2–TiO2 > 1% AgInS2–TiO2 ≈ 0.5% AgInS2–TiO2 ≫ 5% AgInS2–TiO2.

Photocatalytic Antimicrobial Studies

The photocatalytic efficacy of AgInS2–TiO2 samples with their pristine parent samples was also studied by light-induced microbial disinfection using a suspension test. A sample mixture with the bacterial inoculate and catalyst was placed under both dark and light environments, respectively.

Additionally, bare materials (only with microbial strains) were too illuminated devoid of the presence of any sample. On irradiating bacterial cells under visible light by catalysts, these displayed higher photocatalytic proficiency by the AgInS2–TiO2 samples compared to their parent materials. More than a 5-log decrease was achieved in 30 min of illumination for AgInS2–TiO2 (Figure ).

Figure 22

Log reduction of microbial strains with AgInS2, titania, and AgInS2–TiO2 on 30 min of light irradiation.

The development of the 2 strains was constant in light and dark environments devoid of the presence of any photocatalysts. The bacterial progression of the AgInS2 parent sample displayed a complete antibacterial efficiency only after 80 min. In the case of pristine TiO2, the total bacterial inactivation for both the strains was reached only after 60 min of light illumination.

The pictures of agar plates with bacteriological clusters at different periods for AgInS2–TiO2 are given in Figure . All of the composite samples were more active than their parent samples. This could be attributed to the enhanced photocatalytic efficiency of the composite sample. None of the test samples showed any bacterial disinfection property under dark conditions.

Figure 23

Photographic illustration of reduction of (a) Escherichia coli colonies and (b) Staphylococcus aureus colonies on agar plates at different periods using AgInS2–TiO2.

Photocatalytic Mechanism

The composite design of AgInS2 using TiO2 displayed remarkable photocatalytic performance when equated to their bare materials. The computational results of the electronic and optical properties of AgInS2 and TiO2 helped in corroborating with the experimentally obtained orthorhombic and tetragonal phases, respectively. Additionally, the band structure and PDOS investigation illustrate that the Ti 3d orbital of TiO2 dominates the valence band, while the 2p orbital of oxygen is dominated by the conduction band. Similarly, the In–S interaction dominates the conduction band minima of AgInS2, while the Ag–S interaction determines the valence band maxima. Hence, on solar light illumination on the composite structure, the electrons migrate from the 4d orbital of Ag and 3p orbitals of sulfur toward the 5s orbital of In of AgInS2 and latter proceed to the 2p orbital of oxygen in titania.[47,95]

In accordance with the theoretical study, trapping experiments, and the band gap values from UV-DRS, a likely photocatalytic and interfacial electron-transfer mechanism for AgInS2–TiO2 heterojunction is illustrated in Figure . The conduction band (CB) and valence band (VB) potentials of the heterostructure components (AgInS2 and TiO2) are vital to comprehend the migration of photoexcited charge carriers above the nanocomposites. The relevant calculations and the summarized explanation of the process are given in the Supporting Information.

Figure 24

Graphical illustration of the photocatalytic mechanism for AgInS2–TiO2 (a) before and (b) after heterojunction formation.

Therefore, by means of the calculated values, as given in the Supporting Information, a likely photocatalytic mechanism is proposed below. The CB and VB edge values of AgInS2 and titania, prior and later the AgInS2–TiO2 heterojunction creation, are illustrated in Figure . Figure a displays the Fermi level positions of AgInS2 and titania at 0.27 and −0.1 eV correspondingly. The formation of the heterostructure results in the Fermi levels of AgInS2 and titania to an extent at a uniform level (Figure b). Since the dopant level of AgInS2 is extremely low, thus the band position of AgInS2 is altered, and hence the Fermi level of AgInS2 reaches −0.1 eV.[47] The refreshed band edge locations for AgInS2 are found to be at +0.825 and −0.625 eV, as both the VB and CB shift, as shown in Figure b. Charged excitons are produced on the surface of titania and AgInS2 on being irradiated with solar light. The holes on the VB of TiO2 and the holes on the VB of AgInS2 participate in the reaction process individually,[96] while the electrons from the CB of AgInS2 migrate to the conduction band of titania. The electrons participate along with the adsorbed O2 in the surface of the photocatalyst to form anO2* (superoxide) radical. As observed in the result of trapping experiments, the electrons, superoxide radicals, and the holes played the role of predominant species. These radicals were further engaged in photocatalytic reactions.

Conclusions

In summary, the current study reported the fabrication of ternary metal chalcogenide heterostructure nanocomposites of titania. The composite material showcased a dramatic improvement in visible light absorption. The increase in the dopant percent in the composite ensued in the narrowing of the band gap. Furthermore, the shift in the peak signals of XPS suggested an alteration in the chemical surroundings of the heterostructure. Importantly, the degradation outcomes illustrated the enhanced efficiency, and comparable results were observed in the case of the light-induced hydrogen generation. The composite structures displayed 300 times enhanced hydrogen production output when matched with their bare sample materials. A log-5 microbial inactivation was accomplished in 30 min of light irradiance. A possible interfacial electron transfer and the photocatalytic mechanism are offered based on the computational analysis, species trapping experiment, and also the band gap value estimation from the absorbance plot. Therefore, the current investigation details the positive composite preparation and highlights the multifunctional photocatalytic applications.

Materials and Method

Photocatalyst Preparation

Pristine anatase titania was synthesized by the sol–gel method, using titanium isopropoxide (TTIP) as the precursor. In this experiment, 20 mL of TTIP was suspended in 100 mL of isopropanol. Further on continuous stirring, distilled water was added slowly to the solution of TTIP in isopropanol. The volume of water and isopropanol was equal, and the resultant sol was left for gel formation for 120 min at room temperature. The white colloid gel that formed was allowed to dry for 12 h at 100 °C in an oven. The dried xerogel was ground and further calcined using a muffle furnace at 500 °C, with a gradient rate of 600 °C per h for 2 h.

The one-pot solvothermal method was used to prepare AgInS2–TiO2 heterostructure composites. A certain quantity of AgNO3, In(NO3)3·xH2O, and thiourea (1:1:2) was vigorously stirred in 30 mL of ethanolamine for 30 min. The prepared TiO2 was added to the mixture and was sonicated for the next half an hour. The above mixture was transferred into a 50 mL steel Teflon-lined vessel and heated up to 180 °C for 12 h. After being cooled to room temperature, the resultant solid product was washed with water and ethanol multiple times to remove the residue. The washed powder was dried at 60 °C for 12 h and denoted as AgInS2–TiO2, and it was used for further characterization and applications. The parent sample, AgInS2, was synthesized using the same method without the addition of TiO2.[47]

Photocatalytic Activity

To evaluate the photocatalytic efficiency of synthesized AgInS2–TiO2 samples, photocatalytic degradation of doxycycline (DC) under solar light radiance by means of a photoreactor unit was done. In this experiment, 0.10 g of the photocatalyst was added into 100 mL of DC (10–5 M) solution (1 g/L) in water and was stirred for 1 h in dark prior to irradiation to attain absorption–desorption equilibrium. During the illumination period, aliquots were withdrawn at an interval of 15 min. A UV–vis spectrophotometer was used to measure the degradation profiles of the doxycycline with respect to time of the sample aliquots. The reduction in the peak intensity (357 nm) of doxycycline was analyzed. Similarly, the degradation profiles of bare samples (TiO2 and AgInS2) were performed for a comparative evaluation. To assess the recyclability of the samples, the photocatalysts were subjected to centrifugation, washing, drying, and these were reused after each experiment.

Species Trapping Experiment

Different scavenger tests were performed to analyze the ROS (reactive oxygen species) formed during the catalytic reaction in the mixture. Radical scavengers like isopropanol (10 mM), AgNO3 (6 mM), benzoquinone (6 mM), and triethanolamine (10 mM) were added into separate photocatalytic systems to quench the formed hydroxyl radicals (OH–), electron, superoxide radicals (O2•), and holes, respectively. The concentration of catalysts was kept as same as for degradation studies (0.08 g in 80 mL of aqueous DC). These solutions were kept in dark condition for 60 min, and further, they were allowed to expose to light irradiation for another 180 min. Aliquots were taken at an interval of 15 min and tested for DC concentration for the photocatalytic experiments as explained above.

Photocatalytic Hydrogen Generation

The photocatalytic hydrogen generation experiments were performed in a 300 mL annular glass batch reactor, which was concealed with an aluminum foil (Figure ).

Figure 25

Schematic diagram of the photocatalytic reactor.

A 125 W high-pressure mercury vapor lamp from Helios Italquartz was used as the light source. During the experiments, the temperature of the reactor was maintained at 25 °C using a thermostatic bath (Falc GTR 90). The reactants and gaseous nitrogen were fed into the reactor through an inlet, and gaseous sample aliquots at different times were taken through an outlet.

In a typical H2 experiment, a fixed amount of the photocatalytic material and CH3OH solution was added into 300 mL of aqueous solution, and the subsequent solution was fed into the reactor under magnetic stirring. To prevent the unwanted reactions of dissolved oxygen with photogenerated electrons, the solution was flushed with nitrogen gas for 30 min before the photocatalytic runs. Furthermore, nitrogen was continuously fed to the reactor with a flow rate (QN) of 0.3 L/min throughout the experiments.

The pH of the solution was observed with an Orion 420Ap pH-meter (Thermo) during the experiments. The collected liquid aliquots at different time intervals were quickly filtered using regenerated cellulose filters (pore diameter 0.20 μm, Scharlau), and the pH was measured using the filtrate.

The Tedlar gas sampling bags were used to collect the gaseous samples recovered from the outlet, and then the hydrogen concentration is evaluated. A gas chromatograph (Agilent 7820A) equipped with an HP–PLOT Molesieve 5A column (Agilent) and a TCD detector with argon as the carrier gas were used to measure the concentration of hydrogen in the samples.

The radiation that emerged from the reactor was evaluated on the exterior walls of the reactor with radiometers in the wavelength ranges from 315 to 400 and to 400 to 1100 nm during the experiments. 1 M sodium nitrite solution was introduced inside the cooling jacket for maintaining 25 °C temperature during the experiment and to cut-off UVA radiation emitted from the lamp (⌊ < 400 nm) for a fixed reaction time of 60 min in all of the same experiments.

Photocatalytic Antimicrobial Activity

Bacterial disinfection efficiency of AgInS2–TiO2 composites was calculated with Gram-negative (E. coli) and Gram-positive (S. aureus) strains and compared the activity with the parent samples TiO2 and AgInS2. A fixed concentration of 1 g/L of the catalyst in bacterial suspensions (strain in 0.9% NaCl solution) was taken in a cylindrical glass and placed inside a photochamber under visible light irradiation for 3 h. The concentration of bacterial colonies in the buffer solution was maintained as 104 CFU/mL (colony forming units/mL) for all experiments. Aliquots of 1 mL of the cell suspension were taken at fixed time intervals of 15 min. To attain a countable CFU, aliquots were diluted in PBS (phosphate-buffered saline) solution (pH 7), and then 0.1 mL of aliquots was spread onto an agar plate and incubated for 24 h at 37 °C. The viable cells were counted and recorded as CFU/mL. Control experiments were performed with the bare bacterial solution, and all samples were also examined under dark environments. All analyses were conducted in triplicate. Log reduction and N/N0% vs time plot were utilized to calculate the antimicrobial efficiency.