Abstract

Using a simple, low cost, and excellent efficient approach, carbon dots (CDs) were fabricated via a one-pot hydrothermal process of coffee waste. Amazingly, the combination of 2% CDs with ZnBi2O4 to form a new and excellent heterogeneous photocatalyst enabled the complete decomposition of 2, 4-dichlorophenoxyacetic acid (2, 4-D) into CO2 and H2O. The findings of this study provide a new perspective on the utilization of agricultural waste for creating products of scientific and practical significance. More than 91% of 2, 4-D (initial concentration of 30 mg/L) was completely decomposed and dechlorinated using 1.0 g/L of CDs (2%)-ZnBi2O4 at pH 4.0 after 120 min of exposure to visible light (with k = 0.0178 min−1), and more than 86% of the decomposed 2, 4-D was mineralized into CO2 and H2O. There was no sign of catalyst deactivation after four cycles of reuse, demonstrating the durability and efficiency of CDs (2%)-ZnBi2O4. The significant improvement in the photocatalytic efficiency of CDs (2%)-ZnBi2O4 compared with that of bare CDs or ZnBi2O4 is due to the formation of defects at the interfaces of the heterojunction; therefore, the movement of photogenerated electrons at the interface between the two components is rapid. The mineralization of 2, 4-D by CDs (2%)-ZnBi2O4 upon exposure to visible light is induced mainly by the photogenerated holes, followed by O2•−, and finally OH radicals.

1. Introduction

Water pollution by agricultural chemicals has become an important environmental problem due to agricultural activities. Therefore, it is important to find solutions to selectively remove toxic compounds. Recently, various techniques have been used for removal of toxic organic compounds such as fenton, photocatalysis, and ozonation. [1]. As an emerging and effective green solution, semiconductor-based photocatalysis has attracted considerable attention from scientists because organic pollutants can be easily removed using freely available solar energy. In addition, the photocatalytic method can decompose hazardous organic pollutants into nontoxic molecules (CO2, H2O, and mineralized inorganic ions) without generating secondary pollutants. Photocatalysis is a technology in which the resulting chemical energy is derived from photonic energy conversion. Through light harvesting by a semiconductor, an excited electron ()-hole () pair is produced. Because of their activated state, and undergo chemical reduction and oxidation to produce strong oxidizers, such as OH and O2•−, in the system, which can then remove organic contaminants. A number of strategies have been employed in designing photocatalysts with good light-harvesting properties, good chemical stability, and suitable bandgaps. One approach involves coupling semiconductors to achieve a new tunable bandgap semiconductor. Various types of catalysts have been combined in composites to achieve improved photocatalytic activity such as TiO2@MgFe2O4 [2], Ag3PO4-TiO2 [3], Mn-ZnO-graphene [4], TiO2-zeolite [5], Bi@Fe3O4 [6], Eu-doped Bi2WO6 [7], Bi2Sn2O7-reduced graphene oxide [8], and Bi2S3–Bi2Sn2O7​ [9].

Nowadays, biomass waste has caused serious environmental issues. Biomass waste, including agricultural waste and food waste, is an abundant and important carbon source to produce nanocarbon materials because of its renewable ability, sustainability, easy availability, nontoxicity, and low cost [10]. As new classes of nanocarbon, carbon dots (CDs) have recently received attention for development. CDs are generally individual quasi-spherical nanoparticles with small sizes (<10 nm) [11] and are prepared from broad-spectrum carbon precursors via bottom-up or top-down pathways via relatively easy processes. Because they are the latest member in the family of quantum dots, CDs also have the potential alternatives to the traditional toxic metal-based quantum dots in use today. Owing to their highly hydrophilic surfaces, easy surface modification, no chemical reaction, low toxicity, high biocompatibility, easy synthesis, excellent stability, water dispersibility, and prospective applications, CDs have attracted the attention of researchers [1214]. CDs have been widely applied in several fields, including sensing transformation, biological imaging, catalysis, and energy conversion/storage [15]. CDs are considered promising nanomaterials for the design of photocatalysts [14, 16, 17]. The key factor affecting the optical performance of CDs is the huge number of functional groups on the surface, such as -OH, C=O, -COOH, and -NH2. CDs in conjunction with various semiconductors can afford improved light harvesting and electron transport and thus enhanced photocatalytic properties [15, 18]. These include CDs/TiO2 [17, 19], CDs/NiFe-LDH/BiVO4 [20], CDs/BiVO4 [21, 22], CDs/Fe2O3 [15], and CDs/Bi2MoO6 [23]. In these photocatalytic systems, the key role of CDs is as an electronic mediator to effectively delay the recombination of the photoinduced pairs, with excellent charge transfer rates.

Layered double hydroxides (LDHs) are excellent semiconductor photocatalysts because they possess suitable bandgap energies and excellent visible light-absorbing ability [2426]. As an LDH-derived p-type semiconductor with outstanding advantages such as high stability and a low conductivity band edge, ZnBi2O4 has generated a lot of interest in the field of photocatalysis. However, the rapid recombination of photogenerated pairs repressed the photocatalytic activity of ZnBi2O4. It would be a better choice to combine ZnBi2O4 with other functional materials to improve the separation of photogenerated pairs and enhance the surface charge transfer efficiency. ZnBi2O4 has been combined with many different semiconductors such as g-C3N4 [27, 28], Bi2S3 [29], graphite [30], ZnO [31], reduced graphene oxide [32], ZnS [33], and hydrochar [34] to remove toxic organic contaminants. Although these strategies improve the separation efficiency of pairs to some extent, these catalysts are still insufficient for potential industrial applications.

In order to continue the ongoing program of converting biomass waste into valuable products and improving the activity of existing ZnBi2O4 photocatalyst, new catalysts with interesting properties and low cost are being developed with special emphasis. In this paper, the green carbon source, coffee waste powder was chosen to produce CDs by a hydrothermal method; then, CDs-ZnBi2O4 was synthesized using a simple coprecipitation method. To the best of our knowledge, no research has been carried out on developing new photocatalytic CDs-ZnBi2O4 with a favorable energy conduction band that absorbs visible light and testing its catalytic activity for organic pollutant degradation reaction. The intimate association between ZnBi2O4 and CDs exhibited enhanced photocatalytic efficiency. The influence of different operating parameters such as CD percentage in photocatalyst, decomposition time, pH, and 2, 4-D concentration was studied in detail and discussed. The key role of CDs in enhancing photocatalytic activity was clarified. The kinetics of photocatalytic degradation was presented, and the possible mechanism of 2, 4-D photodegradation by CDs-ZnBi2O4 was also proposed.

2. Materials and Methods

2.1. Materials

All chemicals used were of analytical grade and were obtained from Sigma Aldrich including HNO3, NaOH, KOH, Bi(NO3)3·5H2O, Zn(NO3)2·6H2O, tert-butanol, p-benzoquinone, Na2EDTA, and 2, 4-D. Coffee waste powder was obtained from ground robusta coffee (medium roast), which was purchased from Trung Nguyen Company (Vietnam).

2.2. Measurements

A Rigaku Ultima IV X-ray diffractometer (Japan) with CuKα radiation (λ = 1.54051 Å) was used to record the X-ray diffraction (XRD) patterns of the samples. Fourier-transform infrared (FTIR) spectra were obtained using a Thermo Fisher Nicolet Magna-560 FTIR spectrometer (U.S.A.). Scanning electron microscopy (SEM) was obtained on Hitachi S-4800 (Japan) with an operating voltage of 20 keV. A Jasco V770 spectrophotometer (Japan) was used to record UV–Vis diffuse reflectance spectrum absorption. X-ray photoelectron spectroscopy (XPS) measurement was carried out using a Thermo Fisher ESCALAB 250 Xi device (U.S.A.). An IviumStat potentiostat (Ivium Technologies, Netherlands) was used for photoelectrochemical measurement. The silver/silver chloride reference electrode, the Pt mesh counter electrode, and the 0.5 M Na2SO4 electrolyte solution were employed for all electrochemical measurements. A Horiba FL3-22 spectrofluorometer (Japan) was used to record photoluminescence (PL) spectrum. Shimadzu TOC-SSM5000A and a Shimadzu TOC-VCPH analyzer (Japan) were employed to quantify solid and liquid total organic carbon (TOC) of samples, respectively. A PerkinElmer Lamda XLS + spectrophotometer (USA) was employed to quantify the concentration of 2, 4-D remaining in solution.

2.3. Fabrication of CDs-ZnBi2O4 Heterogeneous Photocatalyst

First, the CDs were prepared using a one-step hydrothermal method. Coffee waste powder (2.0 g) was mixed with 20 mL of 1.0 M KOH. Next, the mixture was placed into a 50-mL Teflon-lined stainless-steel autoclave which was heated to 180°C and held for 6 h. After gradual cooling to room temperature, the fluorescent CDs were obtained by centrifugation at 15,000 rpm for 20 min to remove large particles. The supernatant was dialyzed for 24 h to remove micron-sized particles. The eluate was then lyophilized to obtain solid CDs.

Second, a salt solution of Zn2+ and Bi3+ (molar ratio 3 : 1) in HNO3 (5%) was slowly added to aqueous NaOH (1.0 M) and ultrasonicated at 100 W. The suspension was kept at pH 10 throughout the synthesis. The suspension was then ultrasonicated at 100 W for 24 h. The solid material (ZnBi2O4) was collected by centrifugation, then washed with distilled water, dried at 70°C for 10 h, and annealed at 450°C for 3 h.

Third, ZnBi2O4 was further dispersed in solutions with different CD percentages (0%, 1%, 2%, and 5%). The suspension was ultrasonicated at 100 W for 6 h. The resulting material was separated using centrifugation, dried at 70°C for 24 h, and finally labeled as bare ZnBi2O4, CDs (1%)-ZnBi2O4, CDs (2%)-ZnBi2O4, and CDs (5%)-ZnBi2O4, respectively.

2.4. Photocatalysis Experiment

The performance of the catalyst was evaluated based on the decomposition of 2, 4-D (30 mg/L) after exposure to visible light emitted from a halogen lamp (300 W) through a UV cutoff filter (λ > 420 nm). Magnetic stirring was performed in the dark for 60 min before turning on the halogen lamp to achieve adsorption-desorption equilibrium of 2, 4-D on the catalyst. During 120 min of irradiation, 5 mL of suspension was taken every 15 min, and catalysts were removed by centrifugation. A PerkinElmer Lambda XLS + spectrometer was used to quantify 2, 4-D in solution. All experiments were repeated thrice (n = 3). The photodegradation rate of 2, 4-D was determined by applying the pseudo-first-order kinetic equation given by , where k represents the photodegradation rate constant (min−1) and and represent the concentration of 2, 4-D (mg/L) before irradiation with visible light and at irradiation time t, respectively.

3. Results and Discussion

3.1. Material Characterization

Figure 1(a) presents the XRD patterns of bare ZnBi2O4 and the samples of ZnBi2O4 incorporating different amounts of CDs (CDs-ZnBi2O4). Strong, relatively sharp, and typical peaks of tetragonal structure of zinc bismuth oxide (JCPDS No. 043-0449), hexagonal ZnO (JCPDS No. 079-0207), and bismuth hydroxide (JCPDS No. 001-0898) were observed in the XRD pattern of bare ZnBi2O4. The new peak at 23.8° in the pattern of the CDs-ZnBi2O4 samples confirmed the presence of CDs in the lattice of ZnBi2O4. The peak at 23.8° closely matched the graphite nature with a (002) lattice plane [35]. The interlayer spacing value () is 3.72 Å, which is larger than that of bulk graphite, indicating that the graphite core is disordered [36]. Most of the XRD pattern of CDs-ZnBi2O4 samples had similar diffraction peaks but slightly shifted compared to the peaks of bare ZnBi2O4. The shift of the diffraction peaks of CDs-ZnBi2O4 samples may be due to the formation of intimate contact between CDs and ZnBi2O4. This also suggests that the intimate contact between CDs and ZnBi2O4 facilitated the movement of photogenerated pairs at the interface within CDs-ZnBi2O4.

Figure 1(b) displays the FTIR spectra of the CDs, ZnBi2O4, and CDs (2%)-ZnBi2O4. For bare ZnBi2O4, the peaks appearing at 1630 and 3450 cm−1 correspond to the bending and stretching vibration peaks of O-H, respectively [37]. The absorption peaks appear at wavenumbers of about 1381 and 847 cm−1 confirm the presence of the Bi-O bonds [31, 38] and Bi-O-Bi bonds [6, 39, 40], respectively. As seen in Figure 1(b), the bands in the range of 3600–3200 cm−1 in the bare CD spectrum are the typical bands of the stretching vibration modes for hydroxyl (O-H) and N-H groups. The significant absorption peak appearing at 1634 cm−1 is attributed to the C=O bond in the carboxyl (COO) group [41, 42]. The weak band at 1552 cm−1 is compatible with the N-H bending, demonstrating the presence of amino-containing functional groups. The band at 1400 cm−1 is consistent with the C=C bond of the aromatic ring, confirming aromatization. The appearance of the C=C bond (Figure 1(b)) indicates that the CDs have a predominantly graphitic structure [41], whereas the O-H, N-H, and C=O bond vibrations indicate that these characteristic groups are attached to the surface of the CDs. For CDs-ZnBi2O4 with different CDs contents, the characteristic bands at 1634 and 1387 cm−1 demonstrate the appearance of C=O and C=C bond vibrations. The 3 cm−1 shift of the C=C bonds for the CDs-ZnBi2O4 samples compared to that of the bare CDs is owing to the electrostatic interaction of CDs with and ZnBi2O4​ [43].

As illustrated in Figure 2(a), the TEM image indicates that the CDs were mainly monodisperse and spherical. The HRTEM image shows the lattice fringes with a d-spacing of 0.21 nm, corresponding to the (100) plane lattice of graphite [44]. The size distribution histogram also indicated particle sizes of CDs less than 7.5 nm with an average size of 5.1 nm. Furthermore, the SEM-EDX spectrum of CDs confirmed the presence of C (38.73%), N (12.10%), and O (39.85%) elements in the sample. The SEM micrographs of the samples in Figure 2(b) show that bare ZnBi2O4 consisted of individual plates that were different sizes and stacked on top of each other. For the CDs-ZnBi2O4 samples, the CDs were relatively uniformly dispersed, with some amount of aggregation on the surface of ZnBi2O4. From the EDX spectrum of CDs (2%)-ZnBi2O4, the proportions of the elements in CDs (2%)-ZnBi2O4 were as follows: Zn (43.75%), Bi (37.59%), O (16.85%), N (0.41%), and C (1.38%).

3.2. Performance of 2, 4-D Photodegradation

Experiments evaluating the influence of the amount of CDs in CDs-ZnBi2O4 on 2, 4-D decomposition efficiency were conducted at 30 mg/L 2, 4-D, a solution pH of 4.2, and using different catalysts at a dosage of 1.0 g/L. As displayed in Figure 3(a), the visible-light photolysis assay confirmed the photochemical stability of 2, 4-D. Approximately 12.9% and 46.1% of 2, 4-D were degraded by the bare CDs and ZnBi2O4 after 120 minutes of exposure to visible radiation, respectively. The photocatalytic performance of the heterojunction CDs-ZnBi2O4 was superior to that of the bare CDs and ZnBi2O4 components, although the photocatalytic activity of CDs (5%)-ZnBi2O4 was comparable that of bare ZnBi2O4. Among the photocatalysts, CDs (2%)-ZnBi2O4 displayed outstanding adsorption and photocatalytic properties. With the participation of CDs (2%)-ZnBi2O4, the efficiency of 2, 4-D removal reached 82.7% after adsorption for 60 min and visible radiation for 120 min. However, Figure 3(a) shows that the activity of CDs (5%)-ZnBi2O4 was clearly reduced; therefore, the excessive introducing of CDs onto ZnBi2O4 may adversely affect the activity of photocatalysts. The improvement in catalytic performance of CDs (2%)-ZnBi2O4 is indexed to the efficiency of decoupling and delaying recombination of pairs via the heterostructure. However, with excess CDs, as in the case of CDs (5%)-ZnBi2O4, the CDs can perform as intermediates, favoring photoinduced recoupling. These results demonstrate the formation of the heterogeneous junction between the CDs and ZnBi2O4 semiconductors through intimate surface contact.

The rate constants for (k) 2, 4-D decomposition were computed using the pseudo-first-order kinetic equation and are presented in Figure 3(b). Remarkably, the rate constant for photodegradation of 2, 4-D over the heterojunction elevated noticeably compared to that obtained with the individual catalyst components. The heterojunction CDs (2%)-ZnBi2O4 afforded the outstanding activity, with k = 0.0139 min−1, and the decomposition efficiency was about 12.6 and 2.7-fold greater than those of the CDs (k = 0.0011 min−1) and ZnBi2O4 (k = 0.0051 min−1), respectively. Consequently, the formation of the heterojunction improved the photoactivity of the bare CDs and ZnBi2O4. The factors affecting the photodecomposition efficiency of 2, 4-D were investigated by using the heterojunction CDs (2%)-ZnBi2O4, which had the best visible-light catalytic activity.

The best catalytic activity of CDs (2%)-ZnBi2O4 was also confirmed by XPS, UV-vis-DRS, photoluminescence (PL), and electrochemical impedance spectroscopy (EIS) analyses.

XPS was employed to determine the elemental compositions of the materials. As shown in Figure 4(a), the deconvoluted C 1s XPS profile of CDs (2%)-ZnBi2O4 shows a number of peaks at binding energies of 284.5, 286.1, and 288.1 eV, corresponding to various types of carbon bonds: sp2 C=C, C-O, and C=O, respectively [45]. Binding energy peaks of the Zn 2p1/2 and Zn 2p3/2 states were observed at 1044.7 and 1021.5 eV for bare ZnBi2O4 (Figure 4(b)), indicating that Zn is in the divalent state [46]. However, the two strong peaks of Zn 2p in the spectrum of CDs (2%)-ZnBi2O4 shifted towards lower binding energy (0.5–0.6 eV) compared with that of the bare ZnBi2O4 sample; this may be due to the strong interaction between the CDs and ZnBi2O4, which is favorable for electron movement between the CDs and ZnBi2O4. Binding energy peaks of the Bi 4f7/2 and Bi 4f5/2 were located at 159.3 and 164.6 eV for ZnBi2O4 (Figure 4(c)), indicating that Bi exists as Bi3+ [46]. However, the Bi 4f peak shifted by approximately 0.2 eV towards lower binding energy for CDs (2%)-ZnBi2O4 compared with bare ZnBi2O4, due to the sustained interaction between the CDs and ZnBi2O4. The O 1s XPS profiles of ZnBi2O4 show broad peaks at 530.6 eV (Figure 4(d)), which is assigned to lattice oxygen [47]. Furthermore, the O 1s XPS profile of the CDs (2%)-ZnBi2O4 displays two peaks at 531.3 eV and 529.7 eV, indexed to the chemically adsorbed oxygen on the surface (Ochem) and O2− ions, respectively [48]. The two peaks in the deconvoluted oxygen profile of CDs (2%)-ZnBi2O4 compared to that of bare ZnBi2O4 may be because of the chemical bonding between ZnBi2O4 and CDs.

As illustrated in the UV-vis DRS profile of CDs (2%)-ZnBi2O4 (Figure 5(a)), the range of light absorption was broadened towards the visible region, indicating intimate association between the CDs and ZnBi2O4. Figure 5(a) (inset) displays the typical Tauc plot; the bandgap energy (Eg) of ZnBi2O4 was estimated to be 3.10 eV; however, the Eg of the CDs (2%)-ZnBi2O4 decreased to 2.93 eV. Indeed, the intimate association of the CDs with ZnBi2O4 narrowed the bandgap, thereby increasing the visible-light catalytic activity.

The separation of charge carriers and the charge transfer efficiency can be determined from the PL and EIS data. The PL emission peak of the CDs (2%)-ZnBi2O4 (Figure 5(b)) was in a similar position to that of bare ZnBi2O4 but with significantly lower intensity, indicating that separation of the photoexcited charge carriers is maximal in CDs (2%)-ZnBi2O4, which is very useful for achieving efficient photocatalytic performance. Electrochemical impedance spectroscopy (EIS) is a powerful technique for measuring the charge transfer efficiency of a material by constructing Nyquist plots. The Nyquist plot of CDs (2%)-ZnBi2O4 (Figure 5(c)) has a smaller arc radius than that of bare ZnBi2O4, suggesting that CDs (2%)-ZnBi2O4 exhibits excellent charge transfer efficiency with low resistance. The lower impedance of CDs (2%)-ZnBi2O4 may be related to extension of the visible light absorption range over a wider region in comparison to that of ZnBi2O4, leading to a greater carrier density and faster interfacial charge transfer in CDs (2%)-ZnBi2O4. Indeed, the intimate association of ZnBi2O4 with the CDs leads to better charge separation and more efficient electron transfer at the boundary between ZnBi2O4 and the CDs.

Because the pH adjusts the surface charge of the photocatalyst, different toxic contaminants must be decomposed at different pH levels. Therefore, pH is a major parameter influencing photocatalytic decomposition. Experiments to evaluate the influence of pH on the photodegradation efficiency of 30 mg/L 2, 4-D by 1.0 g/L CDs (2%)-ZnBi2O4 catalyst were performed by altering the pH of the reaction solution from 2.0 to 7.0. The concentration of 2, 4-D decreased significantly after 120 minutes of exposure to visible radiation at pH 4.0 (Figure 6(a)). Approximately 91.1% of 2, 4-D was degraded over the CDs (2%)-ZnBi2O4 catalyst at pH 4.0 (with k = 0.0178 min−1) after 120 minutes of exposure to visible radiation, whereas the decomposition of 2, 4-D reached ∼78.6% (k = 0.0110 min−1) for pH 2.0, 84.3% (k = 0.0134 min−1) for pH 3.0, 85.7% (k = 0.0139 min−1) for pH 4.2, and 71.4% (k = 0.0082 min−1) for pH 7.0. The decrease in 2, 4-D decomposition efficiency at pH values less than or greater than 4.0 could be interpreted as follows: The point of zero charge (pHPZC) and pKa of 2, 4-D have a crucial effect on the catalyst activity and protonation of carboxyl groups of 2, 4-D, respectively. As shown in Figure 6(b), the pHPZC of CDs (2%)-ZnBi2O4 was found to be ∼6.78. When pHsolution < pHPZC, the surface of CDs (2%)-ZnBi2O4 becomes positively charged; however, when pHsolution = 2.0, less than pKa of 2, 4-D (pKa = 2.74) hinders protonation of the 2, 4-D molecule; therefore, unprotonated 2, 4-D is the main species in the solution, resulting in less 2, 4-D degradation. In comparison, at pHsolution = 7.0 (>pHPZC), the surface of CDs (2%)-ZnBi2O4 becomes negatively charged. Electrostatic repulsion prevents the 2, 4-D anion from contacting with the negatively charged surface of the CDs (2%)-ZnBi2O4 catalyst, resulting in less 2, 4-D degradation. The optimal decomposition of 2, 4-D at pH 4.0 is attributed to the availability of the −COOH groups of 2, 4-D for protonation; therefore, it is very convenient for the 2, 4-D anions to approach the positively charged surface of the CDs (2%)-ZnBi2O4 catalyst, resulting in significant 2, 4-D degradation.

The effect of catalyst dosage on the photodegradation of 2, 4-D was evaluated by altering the amount of CDs (2%)-ZnBi2O4 in the range of 0.5–2.0 g/L, while the initial concentrations of 2, 4-D and pH were carried on with 30 mg/L and 4.0, respectively. As illustrated in Figure 6(c), the CDs (2%)-ZnBi2O4 dosage of 1.0 g/L yielded the optimal 2, 4-D degradation efficiency. Catalyst dosages greater or less than 1.0 g/L did not yield the optimal degradation efficiency of 2, 4-D. These findings can be put forward for interpretation as follows: With a reasonable ratio between the catalyst dosage, 2, 4-D concentration, and the number of photons occupied on the catalyst surface, the photodecomposition rate of 2, 4-D increased. However, a very low CDs (2%)-ZnBi2O4 dosage led to insufficient generation of oxidizing radicals for degrading 2, 4-D. Likewise, an excessive dose of CDs (2%)-ZnBi2O4 increased turbidity such that light could not reach the active sites of the CDs (2%)-ZnBi2O4 catalyst, resulting in reduced formation of photoinduced pairs, leading to insufficient oxidizing radicals; thus, the rate of 2, 4-D photodegradation decreased.

Experiments to evaluate the effect of initial 2, 4-D concentration on the rate of photodegradation were carried out with a 2, 4-D concentration range of 10–50 mg/L and pH of the solution at 4.0. The optimal CDs (2%)-ZnBi2O4 dosage was 1.0 g/L. High 2, 4-D concentrations resulted in low decomposition efficiency (Figure 6(d)). These findings can be explained as follows: Since the amount of CDs (2%)-ZnBi2O4 and the light intensity remained unchanged, a certain amount of reactive radicals such as pairs, •OH, and O2• was generated while the initial concentration of 2, 4-D increased. Therefore, the number of active radicals attacking 2, 4-D was insufficient, resulting in a decrease in the photodegradation efficiency of 2, 4-D.

The key to practical applications of catalysts is their stability and reusability. To evaluate the reuse efficiency, CDs (2%)-ZnBi2O4 was exposed to 30 mg/L 2, 4-D at pH 4.0, with a ratio of catalyst/2, 4-D solution of 1 : 1. The catalyst was extracted by centrifugation at the end of the decomposition reaction (120 minutes). Before use for the next run, the recovered catalyst was heated at 100°C for 90 min and reused four times in succession. The reuse efficiency data of CDs (2%)-ZnBi2O4 are summarized in Figure 7 and Table 1.

The results in Table 1 indicated that CDs (2%)-ZnBi2O4 exhibited significant photostabilization after 120 min, and the percentage photodegradation of 2, 4-D reached 91.1 ± 1.6% for the first use, 88.9 ± 3.1% for the second reuse, 87.8 ± 2.9% for the third reuse, and 86.4 ± 1.5% for the fourth reuse. Overall, the CDs (2%)-ZnBi2O4 is an excellent catalyst for the efficient removal of toxic organic substances over multiple cycles of exposure to visible light. Most of the decomposed 2, 4-D was dechlorinated, as confirmed by quantifying the amount of inorganic chloride in solution before and after the decomposition reaction.

As summarized in Table 1, the conversion of chloride in the structure of 2, 4-D (toxic organic chloride) to nontoxic inorganic chloride by CDs (2%)-ZnBi2O4 after 120 min of visible light exposure was 90.7 ± 1.2% for the first run, 88.8 ± 1.1 for the second run, 85.3 ± 1.4 for the third run, and 82.2 ± 1.9 for the fourth run. After four consecutive tests, approximately 2.3–6.8% of 30 mg/L 2, 4-D was present as chloride-free organic carbon intermediates. The efficiency of 2, 4-D mineralization by CDs (2%)-ZnBi2O4 catalyst was determined by quantifying organic carbon (TOC) in the solution at the beginning and end of the decomposition reaction. The TOC of the 2, 4-D solution decreased by approximately 86.6 ± 1.6% in the first run (Table 1). Based on the differences between 2, 4-D decomposition, dechlorination efficiency, and 2, 4-D mineralization, it can be confirmed that 2, 4-D was almost completely decomposed to CO2 and H2O after 120 min of visible radiation exposure using CDs (2%)-ZnBi2O4 at pH 4.0.

A number of different catalysts have studied the decomposition efficiency of 2, 4-D using different light sources, and the comparative results are presented in Table 2.

The hole in the valence band (), hydroxyl radicals (OH), and superoxide radicals (O2•−) are highly oxidizing agents that may be active in photocatalysis for pollutant degradation. To confirm the importance of radicals (, OHO2•−) that can interfere with 2, 4-D photodegradation by CDs (2%)-ZnBi2O4, experiments were conducted with 30 mg/L 2, 4-D at pH 4.0 and a catalyst/solution ratio of 1 : 1 in the participation of radical scavengers. Figure 8(a) indicates that when a scavenger (2.0 mM tert-butanol) was added in the photocatalytic experiment, approximately 58.7% of the 2, 4-D was degraded using CDs (2%)-ZnBi2O4. The resulting degradation rate was ∼0.0066 min−1, which indicates that OH• was not the dominant radical for 2, 4-D degradation over CDs (2%)-ZnBi2O4. By adding 2.0 mM p-benzoquinone to the decomposition system as a scavenger, the efficiency of 2, 4-D decomposition over CDs (2%)-ZnBi2O4 declined. Approximately 43.2% of 2, 4-D was decomposed after 120 minutes of exposure to visible radiation in the participation of p-benzoquinone as a scavenger of O2•−. When 1.0 mM Na2EDTA, an electron donor, was used as the scavenger to remove , the efficiency of 2, 4-D decomposition was significantly reduced. Approximately 16.5% of 2, 4-D was decomposed after 120 minutes of exposure to visible radiation, where the decomposition rate was 0.0010 min−1. The results confirm that is a powerful agent for the degradation of 2, 4-D catalyzed by CDs (2%)-ZnBi2O4, whereas the O2•−and OH radicals had relatively little influence on the degradation of 2, 4-D.

Based on the radical scavengers research, a feasible mechanism for the catalytic decomposition of 2, 4-D over CDs (2%)-ZnBi2O4 is proposed in Figure 8(b). Owing to their upconversion fluorescence properties, CDs can serve as efficient energy-transfer materials in CDs-based composite photocatalysts by converting visible and near-infrared light to shorter wavelength, for instance, an emission of ultraviolet light under visible light excitation [50]. When the CDs (2%)-ZnBi2O4 system was excited with visible radiation, the CDs absorbed long-wave radiation and emitted short-wave radiation owing to their photoluminescence properties [51]. Meanwhile, ZnBi2O4 absorbed both long-wave and short-wave radiation, resulting in being excited and creating more electron-hole pairs. These photoinduced electrons in the conduction band (CB) of ZnBi2O4 tend to be shuttled in the conducting network of the CDs because of the conjugated system of the CDs [52, 53]. Therefore, the CDs in CDs (2%)-ZnBi2O4 act as acceptors for the photoinduced electrons emitted from ZnBi2O4, thereby slowing down recombination of the photoinduced electron-hole pairs at the interface of the composite [50, 52]. The migrated electrons that accumulated on the CDs reacted with adsorbed O2 to form O2•− radicals. The photoinduced holes in the valence band (VB) of ZnBi2O4 react with adsorbed reductants (usually OH) to produce oxidizing species (e.g., OH). As a result, the very strong oxidizing agents (O2•−, OH, and photoinduced ) degraded the 2, 4-D molecules to generate CO2 and H2O.

The following reactions illustrate the degradation of 2, 4-D by CDs (2%)-ZnBi2O4 under visible light:

4. Conclusions

A series of CDs-ZnBi2O4 photocatalysts were successfully synthesized. The photocatalyst containing 2% CDs by weight displayed the highest photocatalytic activity after 120 minutes of exposure to visible radiation. The significantly improved catalytic activity of CDs (2%)-ZnBi2O4 can be assigned to its good ability to capture visible radiation and the movement of photoinduced electrons from the CB of ZnBi2O4 to the CDs, resulting in a prolonged separation time of the photoinduced pairs. CDs (2%)-ZnBi2O4 is relatively stable, and the activity did not decrease over four consecutive reuse cycles. Approximately, 91.1–86.4% of 2, 4-D was decomposed after 120 minutes of exposure to visible radiation in four consecutive cycles, and 86.6–79.9% of decomposed 2, 4-D was mineralized to CO2 and H2O. Photoinduced also plays a key role, whereas the O2•− and OH radicals play a supporting role in 2, 4-D degradation. The findings in this study are evidence of the dual benefits of utilizing biowaste to develop new photocatalysts to simultaneously eliminate the impact of biological waste on the environment and decompose toxic organic pollutants.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that they have no conflicts of interest regarding the publication of this paper.

Acknowledgments

This work was supported by the Vietnam Academy of Science and Technology (VAST) under grant number THTETN.08/23-24.