Heterostructured Photocatalysts Associating ZnO Nanorods and Ag-In-Zn-S Quantum Dots for the Visible Light-Driven Photocatalytic Degradation of the Acid Orange 7 Dye

Abstract

Heterostructured photocatalysts associating ZnO nanorods (NRs) sensitized by quaternary Ag-In-Zn-S (AIZS) quantum dots (QDs) were prepared by depositing AIZS QDs at the surface of ZnO NRs followed by thermal treatment at 300 °C. The ZnO/AIZS catalysts were characterized by X-ray diffraction, electron microscopy, UV-vis diffuse spectroscopy and by photoelectrochemical measurements. Their photocatalytic activity was evaluated for the bleaching of the Acid Orange 7 (AO7) dye under visible light irradiation. Results show that the association of ZnO NRs with 10 wt% AIZS QDs affords the photocatalyst the highest activity due to the enhanced visible light absorption combined with the improved charge separation. The ZnO/AIZS(10) photocatalyst degrades 98% AO7 in 90 min under visible light illumination, while ZnO NRs can only decompose 11% of the dye. The ZnO/AIZS(10) photocatalyst was also found to be stable and can be reused up to eight times without significant alteration of its activity. This work demonstrates the high potential of AIZS QDs for the development of visible light active photocatalysts.

1. Introduction

Recently, semiconductor photocatalysis has received a large amount of attention due to its potential to solve both environmental pollution and energy shortage issues. Semiconductor photocatalysis was demonstrated to be effective for the degradation of persistent pollutants in wastewater or for air treatment using inexhaustible solar energy [1,2,3,4,5]. The n-type ZnO semiconductor has been widely applied in photocatalytic degradation due to its low cost, high availability, low toxicity and good chemical stability [6,7,8]. However, due to its wide bandgap (3.3 eV), ZnO can only be photo-activated by UV light, which accounts only for ca. 4% of the total energy emitted by sunlight. Moreover, the electron-hole pair generated after photo-activation easily recombines, which markedly decreases the photocatalytic efficiency of ZnO.

Among the most promising strategies developed to improve the photoactivity of ZnO is the association with low-bandgap semiconductors or with graphene derivatives to form a heterojunction that allows the reduction in or even inhibition of the recombination of photo-generated electron-hole pairs, and an increase in the optical absorption [9,10,11,12]. However, the energy levels of ZnO and of the low bandgap semiconductor must match each other to form the staggered heterojunction.

Quantum dots (QDs) are semiconductor nanocrystals whose bandgap can easily be controlled by varying their size, their composition or their surface chemistry [13,14]. QDs such as CdS, CdSe, PbS and carbon or graphene QDs have been demonstrated to be efficient photosensitizers of ZnO and the heterostructures formed markedly enhance the photocatalytic activity under visible light illumination compared to pure ZnO [15,16,17,18]. I-III-VI₂ ternary QDs, such as CuInS₂ or AgInS₂, are valuable alternatives to the binary QDs previously mentioned and to carbon-based QDs due to their large molar absorption coefficient, their energy bandgap, which matches well with the solar spectrum, and because they contain no toxic element [19,20]. AgInS₂ (AIS) QDs and the quaternary Ag-In-Zn-S (AIZS) alloyed QDs obtained after depositing a ZnS shell at their surface were successfully used as photosensitizers and as an electron transport layer to enhance the visible light photoresponse of ZnO or TiO₂ in solar cells [21,22,23,24,25]. Hybrid nanostructured photocatalysts or photoelectrocatalysts composed of AIS nanocrystals or nanosheets and of TiO₂ are also documented in the literature and were successfully used to degrade dyes such as Rhodamine B (RhB), drugs (Docycline, Norfloxacin) and dichlorobenzene, and were also successfully used for bacterial decontamination [26,27,28,29,30,31]. The use of I-III-VI₂ QDs as photosensitizers of ZnO is far less described. CuInSe₂ or CuInS₂ QDs were associated to ZnO to engineer visible-light active photocatalysts which were used to decompose dyes such as RhB or AO7 [32,33,34]. Only one recent report describes the preparation of heterostructured photocatalysts associating ZnO and AIS QDs for phenol degradation and hydrogen photoproduction [35]. Surprisingly, the photosensitizing properties of AIZS QDs to engineer visible light active photocatalysts have never been investigated to date, despite the fact that the energy bandgap of these nanocrystals can more easily be controlled than that of AIS QDs by tuning the Ag/In/Zn/S molar ratio, and the fact that AIZS QDs exhibit a higher photostability than AIS QDs [19,20].

Herein, the first photosensitization of ZnO using AIZS QDs is described. ZnO nanorods (NRs) were prepared via a solvothermal process and decorated with water-dispersible 3-mercaptopropionic acid (MPA)-capped AIZS QDs to engineer visible-light active ZnO/AIZS heterostructured photocatalysts. ZnO NRs were used in this study because one-dimensional (1-D) ZnO nanostructures exhibit an improved photocatalytic activity compared to conventional spherical particles, not only due to their large specific surface area but also because the charge separation is improved in elongated 1-D nanostructures [9,36,37]. The activity of the ZnO/AIZS heterojunctions was evaluated in the photodegradation of the AO7 diazo dye under visible light irradiation. Results show that the association of 10 wt% AIZS QDs with ZnO NRs affords the photocatalyst the highest activity due to the increased optical absorption and enhanced charge separation. Based on the experimental results, the ZnO/AIZS photocatalyst behaves as a type II heterostructure under visible light irradiation [2].

2. Results and Discussion

2.1. ZnO/AIZS Heterojunction Synthesis and Characterization

The synthetic strategy used to assemble ZnO NRs with AIZS QDs is depicted in Scheme 1. Due to the negative charge of MPA-capped AIZS QDs (Zeta potential of −44 ± 3.2 mV at pH 7), the dots strongly bind to the surface of ZnO NRs (Zeta potential of +27.6 ± 1.2 mV at pH 7) via electrostatic attraction, which leads to an intimate contact between ZnO NRs and AIZS QDs. Next, the MPA ligand is decomposed by heating at 300 °C for 5 h, which allows the building of a heterojunction between ZnO and AIZS QDs. The point of zero charge (pzc) of pure ZnO NRs is ca. 9, a value in good accordance with previous reports [38]. A slight decrease in the pzc value is observed after association with AIZS QDs (8.8 and 8.7 for ZnO/AIZS(5) and ZnO/AIZS(10) catalysts, respectively), suggesting the presence of traces of the MPA ligand after thermal treatment at 300 °C for 5 h (Figure S1).

Representative SEM images show that ZnO NRs have an average length up to 500 nm and that their average diameter is ca 17 ± 5 nm (Figure 1a). Upon association with AIZS QDs and heating at 300 °C, an association of these rods in micrometric and more compact structures is observed (Figure 1b). TEM images show that AIZS QDs are well deposited on the surface of ZnO NRs, forming a heterostructure (Figure 1c). Isolated AIZS QDs non-bound to ZnO NRs could not be detected. It is also noteworthy that the size of AIZs QDs increased from ca. 2 nm to ca. 13 nm after the thermal treatment as a result of the decomposition of the MPA capping ligand followed by the coalescence of the nanocrystals. The enlarged HR-TEM image shows lattice spacings of 0.31 and 0.26 nm, belonging to the (111) and (002) crystal planes of cubic AIZS and wurtzite ZnO, respectively, further demonstrating that the heterostructured catalyst was formed (Figure 1d).

The HAADF-STEM image of the ZnO/AIZS(10) photocatalyst and the corresponding elemental mapping images of Zn, O, S, Ag, and In elements further demonstrate the formation of the heterostructured photocalyst and confirm that AIZS QDs are bound to the surface of ZnO NRs (Figure 2).

The photocatalysts were further characterized by X-ray diffraction (Figure 3). AIZS QDs exhibit a cubic structure (JCPDS No. 03-065-9585) and the wide diffraction peaks confirm the small size of the nanocrystals [20,38]. Sharp and intense diffraction peaks can be observed for ZnO NRs, indicating their highly crystalline nature. All diffraction peaks are in good agreement with the hexagonal wurtzite structure of ZnO (JCPDS No. 00-036-1451) and no crystalline impurity can be detected [39]. These results are in good agreement with HR-TEM observations. All reflections observed for ZnO/AIZS catalysts correspond to ZnO. AIZS QDs were not detected by XRD likely due to the weak loading in QDs, as well as their lower crystallinity compared to ZnO. Finally, no impurities were detected, which shows that the ZnO/AIZS heterostructures were successfully prepared.

2.2. Optical Properties

The optical properties of ZnO NRs, AIZS QDs and of ZnO/AIZS catalysts were investigated by DRS and the bandgap energies were determined using the Kubelka-Munk equation (αhγ)ⁿ = A (hγ-Eg), where α is the absorption coefficient, h is the Planck’s constant, γ is the light frequency, A is a constant and n = 2 for the direct bandgap ZnO semiconductor (Figure 4). Pure ZnO NRs do not absorb in the visible region due to their wide bandgap of 3.19 eV. After thermal treatment at 300 °C, the bandgap of AIZS QDs is 1.99 eV, indicating that the dots may absorb visible light until ca. 623 nm. The heterostructured ZnO/AIZS photocatalysts show enhanced visible light absorption, resulting from the low bandgap of AIZS QDs and from their high visible light absorption capability. The bandgaps determined are 3.17, 3.15, 3.09 and 3.06 eV for ZnO/AIZS(1), ZnO/AIZS(2), ZnO/AIZS(5) and ZnO/AIZS(10) composites, respectively. The improved optical absorption of ZnO/AIZS catalysts should favor the production of electron-hole pairs, and thus improve the photocatalytic performance under visible light irradiation, as described below.

2.3. Photocatalytic Degradation of OA7 and Kinetic Study

The influence of the ZnO/AIZS photocatalysts’ composition on their photocatalytic activity was assessed in the degradation of the OA7 diazo dye (concentration of 10 mg/L) selected as a target pollutant. Experiments were conducted under visible light irradiation (λ > 420 nm, irradiance of 20 mW/cm²). As can be seen from Figure 5a, ca. 15% of the dye is adsorbed at the surface of ZnO and ZnO/AIZS catalysts after 30 min of stirring in the dark, while no adsorption could be observed on AIZS QDs. This likely originates from the electrostatic repulsion between the negatively charged OA7 dye at pH 7 and ZnO NRs [40,41], and from residues of the MPA ligand still present at the surface of the AIZS QDs after the thermal treatment at 300 °C. Upon irradiation, a weak photocatalytic activity is observed for AIZS QDs, ZnO NRs, ZnO/AIZS(1) and ZnO/AIZS(2) composites and the degradation does not exceed 11%, likely due to the poor separation ability of electron-hole pairs and to the wide bandgap of these catalysts (see Figure 4). Much better results were obtained when increasing the loading of AIZS QDs at the surface of ZnO NRs, and the degradation reached 49 and 98% for ZnO/AIZS(5) and ZnO/AIZS(10) catalysts, respectively, after 90 min irradiation. These results show that the amount of AIZS QDs associated to ZnO NRs plays a major role in the efficiency of the photogenerated electron-hole separation, which is induced by a sufficient and intimate heterogeneous interface between AIZS QDs and ZnO NRs. Figure 5b shows the evolution of the UV-visible absorption spectra of OA7 during the photocatalytic treatment using the ZnO/AIZS(10) composite, which further confirms the decomposition of the dye.

The photodegradation kinetics were determined using the pseudo-first-order model ln(C₀/C_t) = kt, where C₀ and C_t are the initial concentration and after a time t of OA7, and k is the kinetic constant. The k values determined for AIZS QDs, ZnO NRs, ZnO/AIZS(1), ZnO/AIZS(2), ZnO/AIZS(5) and ZnO/AIZS(10) composites are 0.61x 10⁻³, 1.17 × 10⁻³, 0.26 × 10⁻³, 1.40 × 10⁻³, 5.99 × 10⁻³ and 36.27 × 10⁻³ min⁻¹, respectively, and confirm that the ZnO/AIZS(10) catalyst displays the highest activity (Figure S2).

2.4. Reuse and Stability of the Photocatalyst

The reusability and the stability of the ZnO/AIZS(10) photocatalyst were investigated for the degradation of OA7. As shown in Figure 6a, only a small decrease in the photodegradation yield from 98 to 91% is observed after eight cycles, indicating that the catalyst exhibits a high photostability. The XRD patterns of the starting and of the reused ZnO/AIZS(10) catalysts (Figure 6b) and the corresponding SEM images (Figure S3) show no change in the crystallinity and in the morphology, further confirming the high stability of the catalyst and its potential for practical photocatalytic applications.

2.5. Photocatalytic Mechanism

To investigate the separation, the migration and the recombination of photogenerated electron-hole pairs in ZnO/AIZS catalysts, photoelectrochemical measurements were conducted by depositing the catalysts on FTO glasses and using a bias potential of 0.5 V vs. Ag/AgCl [10]. The photocurrent response of ZnO and ZnO/AIZS photocatalysts was determined during repeated on-off cycles of visible light illumination (Figure 7a). Note that no photocurrent response was detected for AIZS QDs. All ZnO and ZnO/AIZS catalysts show fast photocurrent responses in each light cycle and the response decreases quickly when the light was turned off during the ten cycles of light irradiation. The photocurrent produced by pure ZnO is weak but increases with the loading of AIZS QDs on the surface of ZnO NRs, the highest value being measured for the ZnO/AIZS(10) (ca. 6.8 times higher than that of pure ZnO), which is in good agreement with photocatalytic experiments.

The electrochemical impedance spectroscopy (EIS) measurements performed on ZnO and ZnO/AIZS catalysts are described in Figure 7b. The diameter of the semicircular Nyquist plot is the smallest for the ZnO/AIZS(10) composite, which is in good accordance with photocurrent measurements, and further confirms that loading 10 wt% of AIZS QDs onto ZnO NRs allows a decrease in the transfer resistance and thus shows that the charge carriers’ recombination in the ZnO/AIZS(10) catalyst is limited compared to the other samples. From the regression, the charge transfer resistance R_CT was determined (see Table S1). As expected by the decrease in the diameter of the semi-circles on Nyquist plots, the decrease in R_CT observed with the increase in the AIZS QDs loading confirms that the charge transferability is the highest in the ZnO/AIZS(10) catalyst. This indicates that QDs deposited at the surface of ZnO NRs provide a charge-transfer pathway, via interparticular charge delocalization, for photogenerated electrons.

Scavenging experiments of the active species involved in the photodecomposition of AO7 were also conducted to explore the possible mechanism. For that purpose, t-BuOH, benzoquinone (BQ), DMSO, EDTA, NaN₃ and the catalase enzyme were used as scavengers of hydroxyl (^•OH) radicals, superoxide (O₂^•−) radicals, electrons (e⁻), holes (h⁺), singlet oxygen (¹O₂) and hydrogen peroxide (H₂O₂), respectively [8,10]. Figure 8 shows that species playing a key role in the degradation of AO7 are holes, hydrogen peroxide and ^•OH radicals. When adding EDTA, catalase and t-BuOH during photocatalytic experiments, the photodegradation falls to 3, 30 and 45%, respectively, after 90 min irradiation. Singlet oxygen and O₂^•− radicals are also involved in the decomposition of the dye but to a lesser extent (64 and 67% degradation after 90 min), while the influence of electrons is very low (91% degradation after 90 min).

Using UV-visible DRS spectra, the valence band (VB) and the conduction band (CB) potentials of ZnO NRs and AIZS QDs were determined using Formulas (1) and (2):

E_VB = χ − E^e + 0.5 Eg

(1)

E_CB = E_VB − Eg

(2)

where E_VB, E_CB and Eg are the VB potential, the CB potential and Eg is the bandgap (2.47 eV for the ZnO/AIZS(10) composite according to DRS results), respectively, and E^e is the energy of free electrons on the hydrogen scale (ca. 4.5 eV) [42]. The χ values for ZnO and AIZS QDs are 5.89 and 4.95, respectively. The values determined for E_VB(ZnO) and E_CB(ZnO) are 2.88 and −0.3 eV, respectively, and E_VB(AIZS) and E_CB(AIZS) are 1.44 and −0.54 eV, respectively. The energy band structure of the ZnO/AIZS catalyst is described in Figure 9a.

Based on the experimental results described above, the following photodegradation mechanism can be proposed (Figure 9b). For the ZnO/AIZS photocatalyst, a typical type-II band alignment is observed, in which the positions of the VB and CB of AIZS QDs are higher than those of ZnO [2]. It is noteworthy that the difference of chemical potential between AIZS QDs and ZnO generates an interfacial band bending. The built-in field thus generated drives photogenerated electrons located in the CB of AIZS QDs after visible irradiation to move to the CB of ZnO and thus enhances the spatial separation and the lifetime of charge carriers. The electrons located in the CB of ZnO reduce O₂ into O₂^•− radicals, which were shown to play an important role in the bleaching of the dye. The holes in the VB of AIZS QDs play a major role in the decomposition of OA7 and may either directly oxidize the dye (AO7 → AO7⁺) and/or reduce O₂^•− radicals into ³O₂ and ¹O₂, the latter species also being involved in the photodegradation.

Hydrogen peroxide is likely produced by the protonation of O₂^•− radicals followed by their reduction (reactions (3) and (4)).

O₂^•− + H₂O → HO₂^• + HO⁻

(3)

HO₂^• + H₂O → H₂O₂ + ^•OH

(4)

The key role played by H₂O₂ in the oxidation of the dye was demonstrated by the addition of the catalase enzyme which decomposes H₂O₂ and thus strongly inhibits the photodegradation (reaction (5)) [43].

$$2 {\mathrm{H}}_2{\mathrm{O}}_2\stackrel{\mathrm{Catalase}}{\to }2{ \mathrm{H}}_2\mathrm{O}+{\mathrm{O}}_2$$

(5)

The VB potential of AIZS QDs (+1.44 eV vs. NHE) is higher than the potential of the H₂O/^•OH couple (+2.40 eV vs. NHE), indicating that holes cannot oxidize H₂O and produce ^•OH radicals, which are the species primarily involved in photodegradation. We assume that these radicals may be produced by a reaction of the HO₂^• radical with H₂O (reaction (4)), reduction of H₂O₂ (reaction (6)), reaction of H₂O₂ with O₂^•− radicals (reaction (7)) or by the decomposition of H₂O₂ under visible light irradiation (reaction (8)).

H₂O₂ + e⁻ → ^•OH + HO⁻

(6)

H₂O₂ + O₂^•− → ^•OH + HO⁻ + O₂

(7)

H₂O₂ → 2 ^•OH

(8)

3. Materials and Methods

3.1. Materials and Instruments

Zn(OAc)₂•2H₂O (Sigma), NaOH (Sigma), Acid Orange 7 (AO7, Sigma), tert-butanol (Sigma), benzoquinone (Sigma), ethylenediaminetetraacetic acid disodium salt dihydrate (Fisher), sodium azide (Sigma), catalase (Sigma), DMSO (Fisher) and ethanol (VWR) were of analytical grade and used as obtained. MPA-capped AIZS QDs with a Ag/In molar ratio of 1/5 were prepared in aqueous phase according the synthetic protocol recently reported [20].

The structure of the photocatalysts was investigated by powder X-ray diffraction (XRD) using a Panalytical X’Pert pro MPD diffractometer with Cu Kα radiation. The morphology of the photocatalysts was studied using scanning electron microscopy (SEM) performed on JEOL JSM-6490 LV and JEOL JSM IT800 instruments. Transmission electron microscopy (TEM) images were taken using Philips CM200 equipment operating at an acceleration voltage of 200 kV. A high-resolution TEM (HR-TEM, JEOL ARM200F) was used to acquire the high angle annular dark field-scanning transmission electron microscopy (HAADF-STEM) image and the EDS mapping of the samples (a 68–174.5 mrad inner-outer collection angle and a 0.1 nm probe size were used).

The Zeta potential of the catalysts was measured using a Malvern Zetasizer Nano ZS equipment.

UV-visible absorption spectra were recorded on a Thermo Scientific Evolution 220 spectrometer. UV-visible spectra diffuse reflectance spectra (DRS) of powders were measured using a Shimadzu 2600–2700 spectrometer.

The photoelectrochemical measurements were performed in a 0.1 M Na₂SO₄ aqueous solution on a SP150 Bio-Logic potentiostat connected to a three-electrode cell (with a Pt wire as counter electrode, a saturated Ag/AgCl reference electrode and FTO glass substrates (1 cm²) coated with ZnO or ZnO/AIZS particles as the working electrode). The photocurrent density was evaluated from chronoamperometric measurements by applying a positive potential of +0.5 V vs. Ag/AgCl. Electrochemical impedance spectroscopy (EIS) was performed at open circuit with a magnitude of the modulation signal of 10 mV in a frequency range of 500 kHz to 10 mHz. The impedance spectra were fitted by a Randles equivalent circuit including a solution resistance R_S, a constant phase element (CPE), charge transfer resistance R_CT and a Warburg impedance W.

3.2. Solvothermal Synthesis of ZnO NRs

ZnO rods were prepared by hydrolysis of Zn(OAc)₂ under solvothermal conditions. Briefly, Zn(OAc)₂•2H₂O (511 mg) was added to 35 mL ethanol and the mixture was stirred until complete dissolution. A solution of NaOH (466 mg) in 35 mL ethanol was dropwise added to Zn(OAc)₂ and the mixture was allowed to react for 30 min at room temperature to form the ZnO nuclei. Next, the solution was poured into a 100 mL Teflon-sealed autoclave and heated at 160 °C for 24 h. After the solution was naturally cooled to room temperature, ZnO rods were recovered by centrifugation, washed with water (2 × 20 mL), with ethanol (2 × 20 mL), and finally dried at 50 °C overnight (ca. 150 mg of ZnO NRs are obtained).

3.3. Synthesis of ZnO/AIZS Photocatalysts

A representative synthesis of ZnO/AIZS(5) is described. To MPA-capped AIZS QDs (5 mg) dispersed in water at pH 7 were added 95 mg of ZnO NRs and the solution was stirred for 1 h at 20 °C before being sonicated for 10 min. Water was then removed by heating at 70 °C for 24 h. Finally, the powder was calcinated in air at 300 °C for 5 h with a heating rate of 2 °C/min, and the heterostructured ZnO/AIZS catalyst obtained was used without further treatment. Other ZnO/AIZS photocatalysts were prepared using the same synthetic protocol but by varying the ZnO to AIZS QDs mass ratio (1, 2, 5 or 10 wt% AIZS QDs relative to ZnO NRs).

3.4. Photocatalytic Tests

The photodegradation of AO7 (concentration of 10 mg/L) was conducted in a batch reactor. An Osram Ultra Vitalux lamp fitted with an UV cutoff filter (>420 nm) was used as light source (the light irradiance at the surface of the OA7 solution was fixed at 20 mW/cm²). In a typical reaction, 30 mg of the catalyst and 30 mL of the OA7 solution were placed in the reactor and the mixture was sonicated for 2 min in the dark. Before irradiation, the mixture was further magnetically stirred for 30 min in the dark to reach the adsorption/desorption equilibrium. At given intervals (15 or 30 min), aliquots (2 mL) were removed, centrifuged (15,000 rpm for 2 min) to separate the catalyst and analyzed by UV-visible spectroscopy. Blank experiments in the absence of catalyst or light irradiation confirmed that the degradation of the dye could only be observed in the presence of the ZnO/AIZS catalyst and under light illumination.

Scavenging experiments were performed to identify the species involved in the photodegradation process. These experiments were conducted using tert-butanol (t-BuOH, 10 mM), benzoquinone (BQ, 10 mM), DMSO (10 mM), ethylenediaminetetraacetic acid disodium salt dihydrate (EDTA, 10 mM), sodium azide (NaN₃, 10 mM) and catalase (28,600 U) as scavengers of ^•OH and O₂^•− free radicals, electrons and holes, singlet oxygen ¹O₂, and H₂O₂, respectively.

4. Conclusions

In summary, novel ZnO/AIZS heterostructured photocatalysts were prepared through the association of ZnO NRs with AIZS QDs under sonication followed by thermal treatment at 300 °C. The ZnO/AIZS(10) catalyst exhibits the highest activity for the AO7 dye photodegradation due to the enhanced visible light absorption and to the type II heterojunction between ZnO NRs and AIZS QDs, which promotes the charge separation upon photo-activation. Our study demonstrates the high potential of AIZS QDs as a photosensitizer of wide bandgap semiconductors such as ZnO, and should not only find applications in emerging pollutant photodegradation but also in other photocatalytic processes linked to energy production.