US-assisted catalytic degradation of paraquat using ZnO/Fe₃O₄ recoverable composite: Performance, toxicity bioassay test and degradation mechanism

Abstract

In this study, the ZnO/Fe₃O₄ catalyst was used as an active catalyst for the oxidation of Paraquat (PQ) herbicide in aqueous solution under ultrasonic (US) waves. FTIR, XRD, FE-SEM, and VSM analyses were performed to characterize the synthesized catalyst. Studies on the effect of radical scavengers were also carried out and the amount of organic matter degradation was determined by measuring the TOC. Under the optimized conditions (catalyst concentration = 0.75 g/L, herbicide concentration = 10 ppm, US power = 70w), the degradation and mineralization rates of the herbicide were acquired as 96.1% and 68% within 60 min, respectively. The quenching tests showed that the hydroxyl (oOH) radical was the most effective oxidant agent in the degradation process of the PQ under ZnO/Fe₃O₄/US system. The toxicity of treated effluent assayed by Daphnia Magna was decreased from %73.16 in raw samples to %7.2 in the treated samples, during 96 h. Finally, it can be concluded that ZnO/Fe₃O₄/US process can be successfully performed as an effective process to herbicides in aqueous solutions, due to the high efficiency and excellent catalytic activity.

Introduction

Herbicides, as a group of organic compounds, have been widely used and disposed into the environment by farmers, establishments, and the general public as consumers have led to severe environmental pollutions. There are above 1400 active commercial components of herbicides in worldwide. Therefore, it is important to prohibit the release of these compounds into the environment (1). Paraquat (PQ) is one of the most widely used herbicides in the world. This compound interferes with the photosynthetic activity of the green parts of the plant through absorption by the leaves, which more absorption is happened in more light intensity and moisture (2). According to the results of studies on the acute toxicity of this herbicide on laboratory animals, this compound is highly toxic from respiration and is classified as class 1 (the highest toxicity class). This chemical compound is classified as carcinogenic class E (no evidence of carcinogenicity). PQ is highly immobilized in soil and is neither hydrolyzed nor photolyzed in aqueous solutions. Moreover, the pollutant is categorized as a refractory organic matter which resist under aerobic and anaerobic degradation (3). According to the European Union standard (98/83/EC) and United States Environmental Protection Agency (USEPA), the maximum permissible concentration of PQ in drinking water is reported to be 0.1 and 3 μg/L, respectively [1–3]. Ultrasonic (US), ozonation, fenton, and photocatalytic processes are among the advanced oxidation processes (AOPs). These processes are based on the generation of very reactive hydroxyl radicals (^oOH) with a considerable potential in the oxidation organic pollutants [4]. The sonolysis process, as an AOP process, has recently received much attention. Besides other AOPs, this process is an environmental-friendly treatment method applied to degrade organic compounds [5, 6]. The AOP using US waves is originally caused by acoustic cavitation producing micro-bubbles in the water at temperatures over 5,000 °K and minimum atmospheric pressure of 1,000 Pa. These conditions finally lead to the formation of reactive radicals, which are widely used to eliminate and degrade the organic pollutants [7]. Contrary to many advantages of the sonolysis process, a long reaction time followed by consuming high amounts of electrical energy is required to complete mineralization of considered pollutants. To overcome these problems, using a suitable semiconductor as a sonocatalyst could be effective to accelerate the reaction alongside the sonolysis process [5, 8]. Sono-luminescence, as the main result of sonocatalytic reaction which is based on the US cavitation, can lead to a wide wavelength range of light. These lights can stimulate semiconductor sonocatalytic compounds to perform a sonocatalytic reaction like the photocatalytic reaction. Therefore, the most important issue is finding an effective and stable semiconductor sonocatalyst due to the relation of sonocatalytic processes to light [9]. By applying heterogeneous catalysts which can generate additional nuclear sites, cavitation bubbles increase is occurred. Among the catalytic compounds, zinc oxide (ZnO) has received much attention because of its proper physico-chemical and optical properties like high potential of ultraviolet (UV) absorption, wide bandgap (3.37 eV), and low cost. Besides, several nanoparticle compounds that are widely used in water and wastewater treatment, ZnO along with magnetic iron (Fe₃O₄) have been the subject of intense attention [10]. ZnO is known as a major semiconductor and piezoelectric substance for optoelectronic devices [11], solar cells [12], and sensor-related applications, or as a photocatalyst. Additionally, ZnO is biologically compatible and safe [13]. Fe₃O₄ represents a large family of U-shaped ferrites. This compound, as one of the most important magnetic materials, has unique magnetic properties such as low toxicity and biological adaptability. Therefore, it has a high potential in various applications including pharmaceuticals, cancer hyperthermia treatment, magnetic resonance imaging (MRI), and catalysis [14]. ZnO nanoparticles (NPs) have a high photocatalytic removal efficiency via adsorption and decomposition of organic matter. Also, super magnetic Fe₃O₄ NPs have a high capacity to adsorb pollutants with very good magnetic properties, leading to easy and fast separation of the adsorbent or degradable material from the solution by the magnetic field [15, 16]. Ineffective methods of catalyst separation can waste catalysts. Today, catalysts immobilization on the surface of magnetic NPs, such as Fe₃O₄, is an effective method to solve the above problem as well as easy and quick magnetic separation of the used catalyst [14]. Thus, combining Fe₃O₄ NPs with ZnO NPs leads the production of a specific functionalized magnetic semiconductor nanocomposite, benefiting from the advantages of both iron oxide and ZnO [17]. Aghaei et al. (2020) investigated the removal of parathion by the sonocatalytic process using ZnO/CuO nanocatalyst under optimum conditions (molar ratio of ZnO to CuO = 90:1, pollutant concentration = 20 mg/L, catalyst loading = 1 g/L, solution pH = 8, and US power = 60 W) which achieved the 100% removal efficiency of the pollutant [6]. The results of the Dikmen et al. (2019) on 2,4-D degradation by sonocatalytic reactions in the presence of titanium dioxide catalyst showed 100% removal efficiency under optimum experimental conditions (pH = 2, catalyst concentration = 0.5 g/L, and pollutant concentration = 75 mg/L) [5]. Zhang et al. (2019) used the Pd-BaZrO₃@WO₃ composite in the removal of Diazinon which showed that the Diazinon sonocatalytic degradation rate in US/Pd-BaZrO₃@WO₃ process at concentrations of 0.5, 1, and 1.5 g/L of Pd-BaZrO₃@WO₃ (3:1) catalyst was 53.4%, 73.9%, and 75.9%, respectively (after 150 min of US radiation) [9]. Considering the mentioned issues and due to the lack of any sonocatalytic study on PQ degradation, especially using ZnO/Fe₃O₄ nanocatalyst, this study aimed to synthesize ZnO/Fe₃O₄ nanocatalyst and evaluate its performance on PQ removal from aqueous solutions, in the presence of sonocatalytic process. Also, the effect of various operating parameters, such as catalyst concentration, solution pH, PQ concentration, and US power was assessed.

Materials and methods

Chemicals

PQ herbicide with the purity of 98% was purchased from Golsam Gorgan Company, Iran. All other chemicals with analytical grade purchased from Samchun, N. Korea. Deionized water was used to prepare all solutions.

Synthesis of zinc oxide/iron oxide nanocatalyst (ZnO/Fe₃O₄)

Synthesis of Fe₃O₄ NPs

Fe₃O₄ NPs were synthesized by co-precipitation method, which was previously described [8, 18]. Briefly, about 200 mL of both ferrous ammonium sulfate ((NH₄)₂Fe(SO₄)₂. 6 H₂O) and iron chloride (FeCl₃) solutions (in the molar ratio of 1:2 = Fe⁺²:Fe⁺³) were combined in a 1 L beaker and stirred at 80 °C at a constant rate of 200 rpm for 15 min. Then, an ammonia solution (50 mL-25%) was added slowly and drop-wisely to the prepared solution. Magnetic iron was formed by converting metallic salts to the metal hydroxides and transformation the hydroxides to ferrite compounds. The resulting black solution was stirred for 1.5 h at 100 °C. Finally, the synthesized Fe₃O₄ NPs were washed with deionized water several times and dried at 103 °C for 3 h.

Stabilizing ZnO NPs on Fe₃O₄ NPs

Zinc chloride (ZnCl₂) was used as the reaction-initiating agent and the sodium hydroxide (NaOH) was used as the precipitating agent. Firstly, 0.1 M ZnCl₂ solution and 0.2 M NaOH solution were prepared. The synthesized Fe₃O₄ from the previous step was added to the ZnCl₂ solution with a volumetric ratio of 1:1. The reaction was continued by stirring for 7 h. Then, the final catalyst was centrifuged at 4,000 rpm, washed with deionized water several times, and finally dried at 100 °C for 3 h [8, 15].

Characterization of ZnO/Fe₃O₄ nanocatalyst

The functional groups of the synthesized catalyst’s surface (ZnO/Fe₃O₄) were identified using Fourier-transform infrared spectroscopy (FT-IR) analysis (Perkin Elmer Frontier-wavelength range from 400 to 4000 cm⁻¹). The X-ray diffraction (XRD) analysis was applied in order to specify the crystal structure of prepared catalyst by using Ni-filtered Cu-Kα radiation (Shimadzu 7000 diffractometer device at 25 °C), in the scan range from 10° to 80° (0.05° step width). The Field Emission Scanning Electron Microscope (FE-SEM) equipped by energy dispersive X-ray spectrometer (EDS) was applied to determine the morphology and combination of the surface of synthesized ZnO/Fe₃O₄ catalyst (Carl-Zeiss SUPRA 55). The magnetic properties of the catalyst were determined at room temperature (25 ± 1 °C) with a Vibrating Sample Magnetometer (VSM), (7400, Lakeshore, USA model) in the field sweeping from -15 to + 15 kOe.

The performance of US irradiation on PQ degradation

This study was conducted in the laboratory scale. The degradation of PQ was evaluated by considering affecting parameters as follows: solution pH = 6, ZnO/Fe₃O₄ concentration (0.1–0.75 g/L), PQ concentration (10–50 mg/L), and US power (10–70 W, 20 kHz) in the reaction time (0–60 min), which was defined based on the results of our previous study [8]. Experiments were performed based on the one factor at a time method. The kinetics of the degradation of PQ were examined considering the effect of US, US/ZnO, US/Fe₃O₄, US/ZnO/Fe₃O₄, and at different time intervals (0–60 min), under the optimum conditions of the parameters. The residual PQ concentrations in the solution were measured by the high-performance liquid chromatography (HPLC-Agilent 1260) at a wavelength of 258 nm. In this regard, an ultimate Zorbax C18 column (4.6 mm × 100 mm) was used as a stationary phase and phosphate buffer solution (0.1 M) and methyl cyanides (88:12, v/v) was prepared as the mobile phase (adjusted pH = 3.0). The flow rate of sample eluent was fixed at 1 mL/min.

The collected data were analyzed by Microsoft Excel, and finally the mean of two replicates was recorded and reported.

Results and discussion

Characteristics of the synthesized nanocatalyst

Fe-SEM and EDS analyses

The results of FE-SEM analysis of ZnO (a), Fe₃O₄ (b), and ZnO/Fe₃O₄ (c) nanocatalysts and the EDS spectra for ZnO/Fe₃O₄ (d) are shown in Fig. 1. As can be seen, the ZnO surface (Fig. 1A) was more poriferous than ZnO/ Fe₃O₄ (Fig. 1C) which may occurred due to the accumulation of Fe₃O₄ NPs and filling some pores with Fe₃O₄ during the magnetization process. Figure 1B demonstrates the Fe₃O₄ NPs accumulation which shows that pure Fe₃O₄ NPs have a spherical-like shape. The aggregated Fe₃O₄ NPs were ultrafine and connected tightly to one another to form groups. Some particles showed a nearly spherical morphology with a rough surface, while some others were regrouped to form larger aggregates. Figure 1C shows that Fe₃O₄ NPs were uniformly distributed on ZnO, in comparison with the pure Fe₃O₄ (Fig. 1B). According to the EDS analysis (Fig. 1D), which exhibits the elemental composition of the ZnO/Fe₃O₄ nanocatalyst, all peaks were related to Zn, O, and Fe elements and no unanticipated peaks were identified [19] that proves the high purity of ZnO/ Fe₃O₄ catalyst.

Fig. 1.

FE-SEM analysis of (A) ZnO (B) Fe₃O₄ and (C) ZnO/ Fe₃O₄ nanocatalysts and (D) EDS spectra for ZnO/Fe₃O₄

X-ray diffraction analysis

The X-ray diffraction analysis of as-synthesized catalysts proved the formation of the crystalline phases of cubic magnetite and hexagonal ZnO wurtzite phases (Fig. 2). The broadness of the diffraction peaks demonstrates the formation of nanosized phases. Results showed that the XRD pattern of ZnO and Fe₃O₄ samples were completely comparable to the XRD spectra of standard pure ZnO and Fe₃O₄, respectively. The identified diffraction peaks in the XRD pattern in Fig. 1a belonged to the wurtzite phase ZnO (JCPDS No. 36–1451), indicating that ZnO had a wurtzite structure. For the XRD pattern of Fe₃O₄, all identified peaks were similar to standard magnetite, which was in agreement with JCPDS No. 19–0629 [19]. In XRD pattern of ZnO/Fe₃O₄ nanocatalyst, all of peaks were related to both ZnO and Fe₃O₄ and there were no other peaks found in the ZnO/Fe₃O₄ nanocatalyst structure, indicating the purity of the synthesized material. According to Scherrer’s equation, the average crystallite size for Fe₃O₄ and ZnO/Fe₃O₄ nanocatalysts was 33.04 and 40.17 nm, respectively.

Fig. 2.

X-ray diffraction spectra of (A) ZnO, (B) Fe₃O₄ and (C) ZnO/Fe₃O₄ samples

Magnetic properties and FTIR analysis of nanocatalyst

Figure 3A displays the magnetic properties of Fe₃O₄ and ZnO/Fe₃O₄ at a magnetic field of -15 kOe to 15 kOe. As can be seen, the magnetization saturation value was decreased as Fe₃O₄ coupled with ZnO NPs. According to this analysis, the magnetization saturation values was equal to 79 and 26.2 emu/g for pure Fe₃O₄ and ZnO/Fe₃O₄, respectively. The significant difference between the values of these magnetization saturation may be attributed to (i) the minor content of Fe₃O₄ in the ZnO/Fe₃O₄ catalyst and (ii) the presence of diamagnetic ZnO in the catalyst structure. These results also showed that both of catalysts had great magnetic properties, so they could be easily collected from the solution, leading to a remarkable decrease in the operational costs in real environments. FT-IR analysis was used (400–4,000 cm⁻¹) to determine the functional groups on the surface of synthesized catalysts (Fig. 3B).

Fig. 3.

Magnetization curves of (A) Fe₃O₄ and (B) ZnO/Fe₃O₄ and FTIR spectra of ZnO, Fe₃O₄, and ZnO/Fe₃O₄ nanocatalysts

The IR bands at 470, 733, 935, and 3,480 cm⁻¹ are related to ZnO nanocatalyst. The IR spectrum of ZnO appeared in the absorption peaks between 400–500 cm⁻¹. A weak peak at 1600 cm⁻¹, related to H–O-H bond, was detected during the FT-IR sample disks preparation that was due to moisture adsorption of the sample. In the ZnO spectrum, the peak observed at 3500 cm⁻¹ was return to the traction vibration of O–H. The Fe–O bond was detected through a strength peaks of 560 cm⁻¹ [15]. A similar FT-IR spectra of ZnO/Fe₃O₄, ZnO and Fe₃O₄ nanocatalysts confirmed the proper ZnO-Fe₃O₄ hybridization. Moreover, Zn–O and Fe–O bonds were detected at peaks of 470 and 560 cm⁻¹ through the broad adsorption peak of 500 cm⁻¹ which were return to ZnO and Fe₃O₄ FTIR spectra, respectively. Finally, –CH–,–CO–, and O–H bonds were identified in the absorption peaks at 1379, 1637, and 3400 cm⁻¹ respectively, in the Fe₃O₄ and ZnO/ Fe₃O₄ FT-IR analysis [20, 21].

Investigating the effect of the studied variables on the process of PQ sonocatalytic degradation

Investigating the effect of nanocatalyst concentration, initial PQ concentration, and US power on PQ degradation in ZnO/Fe₃O₄/US system

The effect of various ZnO/Fe₃O₄ nanocatalyst concentrations on the efficiency of the ZnO/Fe₃O₄/US process in PQ degradation was investigated (Fig. 4A). As shown, by increasing the nanocatalyst concentration from 0.1 to 0.75 g/L, the PQ removal efficiency increased from 45 to 73% (after 60 min). This increase could be related to the increasing the number of active sites on the surface of the catalyst at higher concentrations, leading to the production of additional nuclei to form cavitation bubbles [19, 22]. Furthermore, high catalyst concentrations provided more surface area for the pollutant adsorption and consequently, the contact between the PQ and the active species produced on the catalyst was increased; which led to an increase in the degradation efficiency. Therefore, the concentration of 0.75 g/L was applied as the optimum catalyst concentration for the next steps of experiments. Figure 4B illustrates the effect of different concentrations of PQ on its removal process in the above system. According to Fig. 4B, by increasing PQ concentration from 10 to 50 mg/L, the degradation rate was decreased from 87 to 52.5%. As a result, the highest removal efficiency (87%) was observed at the lowest PQ concentration (10 mg/L). The reasons for the decreased removal efficiency with increasing pollutant concentration were threefold: (1) increasing competition between intermediate products and major pollutant for the reaction with radical species; (2) decreasing the production rate of oxidizing radicals due to the occupation of catalytic surface active surface by PQ molecules and decreasing reactions between US radiation and sonocatalyst; and (3) at the constant production rate of oxidizing radicals, all pollutant molecules couldn't be decomposed [23]. Similar results were reported in sonocatalytic degradation of different pollutants by ZnO, ZnO–biosilica, and Co Fe₂O₄@ZnS catalysts [24, 25]. Figure 4C shows the effect of US powers on the sonocatalytic degradation of PQ by the ZnO/Fe₃O₄/US system under optimum conditions. The US power is a significant parameter regarding its effect on acoustic cavitation. According to Fig. 4c, by increasing the US power from 10 to 70 W, the PQ degradation efficiency was increased. This increase could be due to the greater energy transfer to the system, leading to the increased pulsation and consequently the collapse of more cavitation bubbles and therefore, production of more free radicals [26, 27]. Thus, the power of 70 W was the most suitable US power for further experiments. The above findings were consistent with the results of previous studies reported by other researchers [25, 28]. pH is one of the most important factors in photocatalytic activity is due to: (1) the effect on the degree of ionization of the catalyst and the pollutant, (2) hydrolysis of the pollutant, (3) the change in structure and surface properties of the catalyst, and (4) the effect on solubility and activity of oxidants and free radicals. All experiments were performed at optimum pH equal to 6. This value was based on the pre-test experiments conducted to determine the effect of sonocatalytic removal efficiency of PQ. The results of the pre-tests showed that in acidic pHs, the removal efficiency of PQ was higher compared to the alkaline condition. On the other hand, according to pH_pzc measured in our previous study [8] which was equal to 6.2, the slightly less solution pH than pH_pzc could led to the adsorption of PQ molecules onto the surface through the methyl group. So, the degradation of adsorbed species was done with ^oOH radicals produced by catalyst surface.

Fig. 4.

Effect of (A) catalyst concentration, (B) PQ concentration, and (C) US power on PQ degradation in ZnO/Fe3O4/US process (pH = 6, catalyst conc. = 0.75 g/L, PQ conc. = 30 ppm (for section (C), PQ conc. = 10 ppm), US power = 40 w)

Studies on the kinetic of PQ degradation in the presence of different processes

Evaluation of the kinetic of PQ degradation under US processes, ZnO/US, Fe₃O₄/US, and ZnO/Fe₃O₄/US revealed that the degradation of PQ has the highest consistency within the pseudo first order kinetic model, as shown in Fig. 5 [29–31]. The pseudo-first-order kinetic equation is as follows:

$$\ln (\frac{C_t}{{\mathrm{c}}_0})=-{\mathrm{k}}_{\mathrm{obs}}\mathrm{t}$$ 1

where C_t and C₀ were the remaining and initial concentrations of PQ, respectively, and k_obs was the observed reaction rate. The PQ degradation rate by sonocatalytic processes was greater than alone sonolysis process which the values are 0.044, 0.061, 0.076, 0.0965 (min⁻¹) for the US alone, Fe₃O₄, ZnO/US, ZnO/Fe₃O₄/US, in that order, and the increased k_obs may happened because of the synergistic impact among ZnO and Fe₃O₄ and the radical species production [8].

Fig. 5.

Degradation kinetics of PQ by sonolysis and sonocatalysis processes in optimum condition (pH = 6, catalyst conc. = 0.75 g/L, PQ conc. = 10 ppm, US power = 70 w)

Comparing the efficiency of various processes in the removal of PQ herbicide under optimum conditions

In order to determine the contribution of adsorption or oxidation processes on the PQ removal, the effectiveness of various procedures was examined under the equal experimental conditions [8]. As demonstrated in Fig. 6, the degradation of PQ by Fe₃O₄, ZnO, ZnO/Fe₃O_4, US, Fe₃O₄/US, ZnO/US, ZnO/Fe₃O₄/ US under optimum conditions (after 60 min) was 17%, 21.5%, 26%, 41.5%, 56.5%, 73.6%, and 96.1%, respectively. The adsorption process played the main role in the PQ removal by ZnO, Fe₃O₄, and ZnO/Fe₃O₄ catalysts. The higher removal efficiency in ZnO/Fe₃O₄ process in compare to ZnO and Fe₃O₄ showed the synergistic effect between ZnO and Fe₃O₄. Moreover, it demonstrated that the PQ removal rate by sonocatalytic processes (Fe₃O₄/US, ZnO/US, ZnO/Fe₃O₄/US) was higher than sonolysis alone, which could be related to the generation of more oxidizing radicals in sonocatalytic processes compared to sonolysis alone [32]. In the Fe₃O₄/US process, the removal efficiency was 56.5%, which could be due to the combined role of adsorption and degradation processes in the removal of the pollutant [33]. The removal efficiency in the presence of ZnO/US was increased and reached to 73.6% which was due to the electron–hole pair formation after electrons excitation at the surface of the ZnO semiconductor catalyst, followed by US irradiation. Finally, the highest removal efficiency (96.1%) was observed in ZnO/Fe₃O₄/US, which was owing to the synergistic effect between the effective factors on the removal of pollutant via the production of reactive radicals. Additionally, the high degradation efficiency of PQ by ZnO/ Fe₃O₄/US process, in contrast with other methods, indicated that the ZnO/Fe₃O₄ nanocatalyst (in the presence of US waves) had considerable catalytic activity in producing of oxidizing species. Therefore, ZnO/ Fe₃O₄ nanocatalyst could play a main role in the removal of organic pollutants with the production of oxidizing radicals [8]. In order to better compare the obtained results about of the efficiency of the ZnO/Fe₃O₄/US process within the removal efficiency of PQ in other previous studies, Table 1 was prepared.

Fig. 6.

The effect of various processes in the removal of PQ under optimum conditions (pH = 6, catalyst conc. = 0.75 g/L, PQ conc. = 10 ppm, US power = 70 w, contact time = 60 min)

Table 1.

Comparison of PQ degradation of the current study with the literatures (33–40)

No	Catalyst	Ultrasonic frequency (kHz)	Pollutant type and concentration (mg L−1)	Catalyst dosage (g L−1)	Degradation efficiency (%)	Reaction rate constant (min)	Ref
1	ZnO/Fe3O4	20	Amoxicilin (10)	0.8	90.0	0.026	[8]
2	TiO2	20	Atrazine (43)	0.2	93.0	0.034	[33]
3	Sono-electro-Fenton (SEF)	37	Diazinon (25)	0.03	98.9	0.045	[34]
4	ZnO	20	Flonicamid (75)	1.0	60.5	7.8 × 10 − 3	[35]
5	TiO2/CdS	20	Reactive Black 5 (50)	0.05	97.3	0.418	[36]
6	MgO/Z	20	Textile (2600)	0.7	78.0	9.2 × 10 − 5	[37]
7	S2O8	35	Tetracycline (25)	0.95	94.2	2.29 × 10 − 5	[38]
8	Na2SO4	24	2,4-dichlorophenoxyacetic acid (20)	7	98.2	0.028	[39]
9	Fe3O4/H2O2	24	Reactive Orange (50)	0.8	99.2	0.106	[40]
10	ZnO/Fe3O4	20	Paraquat (10)	0.75	96.1	0.096	This study

Determining the radical scavengers and main water anions in sonocatalytic removal of PQ herbicide in ZnO/Fe₃O₄/US system under optimum conditions

The radical scavengers were used to determine the effect of different radical species produced in the ZnO/Fe₃O₄/US process in the removal of PQ. These compounds included tert-Butanol as an ^oOH scavenger, benzoquinone as a superoxide (^oO₂) radical scavenger, silver nitrate as an electron scavenger, and ethylenediaminetetraacetic acid as hole (^oH) scavenger (2 mM). As can be observed in Fig. 7, by the presence of radical scavenger compounds, the removal efficiency of the pollutant was reduced due to the occupation of the catalyst surface by these compounds [8]. The presence of tert-Butanol caused the observation of lowest removal efficiency (52.35%) of PQ. Therefore, it could be concluded that ^oOH radicals could be considered as the most effective factor in the sonocatalytic removal process of this herbicide in the ZnO/Fe₃O₄/US system (Eqs. 2–20). According to Kumar et al. (2018), the most effective oxidizing species in PQ sonocatalytic degradation were ^oOH, $O_2^{-o}$, and $H^o$ radicals in the order of their appearance. In this study, the removal efficiency in the presence of tert-Butanol and benzoquinone was reduced from 99.3% to 20.1% and 41.3%, respectively [41].

$$ZnO/{\mathit{Fe}}_3{O}_4+)))\to h_{V{B}^{+}}+e_{C{B}^{-}}$$ 2

$$ZnO/{\mathit{Fe}}_3{O}_4(h_{V{B}^{+}})+H_{2}O\to {}^{o}OH{}_{(ads)}+H^{+}$$ 3

$$ZnO/{\mathit{Fe}}_3{O}_4(e_{C{B}^{-}})+O_2\to O{}_2+O_2^{-o}$$ 4

$$ZnO/{\mathit{Fe}}_3{O}_4(e_{C{B}^{-}})+O_2^{-o}+2H^{+}\to H_2{O}_2$$ 5

$$O_2^{-o}+O_2^{-o}+2H^{+}\to H_2{O}_2+O_2$$ 6

$$O_2^{-o}+2H^{+}\to H{O}_2^o$$ 7

$$H{O}_2^o+H{O}_2^o\to H_2{O}_2+O_2$$ 8

$$O_2^{-o}+H_{2}O\to {}^{o}OH{}_{(ads)}+{\mathit{OH}}^{-}$$ 9

$$O_2^{-o}+H_{2}O\to H{O}_2^o+{\mathit{OH}}^{-}$$ 10

$$O_2^{-o}+H_2{O}_2\to {}^{o}OH{}_{(ads)}+{\mathit{OH}}^{-}+O_2$$ 11

$$H_{2}O+)))\to {}^{o}OH+H^o$$ 12

$$H_{2}O+H^o\to {}^{o}OH+H_2$$ 13

$$O_2+H^o\to H{O}_2^o$$ 14

$$2{}^{o}OH+{2H}^o+)))\to H_2{O}_2+H_2$$ 15

$${}^{o}OH+{}^{o}OH\to H_2{O}_2$$ 16

$$H_2{O}_2+)))\to 2{}^{o}OH$$ 17

$$H_2{O}_2+e_{C{B}^{-}}\to {}^{o}OH+{\mathit{OH}}^{-}$$ 18

$$PQ+{}^{o}OH\to (OH){\mathit{PQ}}^o$$ 19

$$\left(O,H\right){\mathit{PQ}}^o\to products\to {\mathit{CO}}_2+H_{2}O$$ 20

Fig. 7.

Efficacy of (A) various radical scavengers and (B) water anions on the removal of PQ in the ZnO/Fe₃O₄/US process under optimum conditions (pH = 6, catalyst conc. = 0.75 g/L, PQ conc. = 10 ppm, and US power = 70 w)

The presence of water anions (carbonate, chloride, nitrate, phosphate and sulphate) could affect the oxidation process, in real conditions, which was due to their role as a radical scavengers. Hence, the degradation of PQ by ZnO/Fe₃O₄/US process, under the optimum conditions, was conducted in the presence of some common water anions including sodium bicarbonate (NaHCO₃), sodium sulfate (Na₂SO₄), sodium chloride (NaCl), sodium carbonate (Na₂Co₃), sodium nitrate (NaNO₃), and sodium phosphate (NaH₂PO₄) (at 0.1 mM concentrations). The results (Fig. 7B) showed that PQ removal efficiency was decreased in the presence of all these anions, but this efficiency reduction was greater in the presence of phosphate (PO₄⁻³), nitrate (NO₃⁻), and bicarbonate (HCO₃⁻) ions, respectively. This decrease in the degradation efficiency of PQ could be due to the possession of the catalytic surfaces holes by these ions, which may reduce the adsorption of the pollutant and consequently decreased the rate of ^oOH radical’s production [42]. Mussavi et al. (2014) also reported the decrease in photocatalytic removal efficiency of Diazinon in the presence of several anions such as NO₃⁻, HCO₃⁻ and PO4⁻³ which also was due to the reaction of the aforementioned anions with the H^o and ^oOH radical to produce other anionic radicals with lower oxidation potential [42]. According to Eqs. 21–24, ^oHCO₃⁻ and ^oSO₄²⁻ radicals were produced during the reaction between h⁺/^oOH radicals within HCO₃⁻ and SO₄²⁻ anions, respectively. Although the production of ^oHCO₃⁻ and ^oSO₄²⁻ radicals and their relative potential of oxidizing organic compounds were occurred, however ^oOH radicals had much more oxidation potential that led to a decrease in the removal efficiency of organic pollutants [29]:

$${\mathrm{HCO}}_3^{-}+{\mathrm{h}}^{+}\to {{}^o\mathrm{HCO}}_3^{-}$$ 21

$${4\mathrm{HCO}}_3^{-}+{}^{o}OH\to {{}^o\mathrm{HCO}}_3^{-}+{\mathrm{OH}}^{-}$$ 22

$${\mathrm{SO}}_4^{2-}+{\mathrm{h}}^{+}\to {{}^o\mathrm{SO}}_4^{2-}$$ 23

$${\mathrm{SO}}_4^{2-}+{}^{o}OH\to {{}^o\mathrm{SO}}_4^{2-}+{\mathrm{OH}}^{-}$$ 24

The existence of NO₃⁻ and PO₄⁻³ anions could essentially decrease the removal rate of organic matters due to their scavenging effects on oxidative radicals. [24] (Eqs. 25 and 26). These reactions caused the production of anion radicals which they had lower oxidation power in compare with ^oOH radicals and thus their reaction with pollutant molecules was less effective [30].

$${\mathrm{PO}}_4^{2-}+{\mathrm{OH}}^{-}\to {\mathrm{PO}}_4^{2-}+{\mathrm{OH}}^{-}$$ 25

$${\mathrm{NO}}_3^{-}+{\mathrm{OH}}^{-}\to {\mathrm{NO}}_3^{-}+{\mathrm{OH}}^{-}$$ 26

Therefore, the inhibitory effect of chloride (Cl⁻) anions could be discussed by following reasons: (1) Formation of active chlorine radical species (Cl^o, Cl₂^o−, ClOH^o−) by the reaction of Cl⁻ and surface holes of the catalyst (Eq. 27); (2) Production of Cl^o, Cl₂^o− and ClOH^o− radicals through consumption of ^oOH radicals and oxidation of Cl⁻ with lower oxidation power (Eqs. 28 and 29). The degradation rate of PQ in the presence of Cl⁻ was decreased due to the competition of Cl⁻ with pollutant molecules to react with ^oOH or other reactive species [23]. The other reactions related to chlorine radicals were shown in Eqs. 30–35.

$${\mathrm{Cl}}^{-}+{\mathrm{h}}^{+}\to {\mathrm{Cl}}^{\mathrm{o}}$$ 27

$${\mathrm{Cl}}^{-}+{}^{\mathrm{o}}OH\to {\mathrm{Cl}}^{\mathrm{o}}+{\mathrm{OH}}^{-}$$ 28

$${\mathrm{Cl}}^{-}+{}^{\mathrm{o}}OH\to {\mathrm{ClOH}}^{\mathrm{o}-}$$ 29

$${\mathrm{ClOH}}^{-}+{\mathrm{H}}^{+}\to {\mathrm{Cl}}^{\mathrm{o}}+{\mathrm{H}}_2\mathrm{O}$$ 30

$${\mathrm{Cl}}^{\mathrm{o}}+{\mathrm{Cl}}^{-}\to {\mathrm{Cl}}_2^{\mathrm{o}-}$$ 31

$${\mathrm{Cl}}_2^{\mathrm{o}-}+{\mathrm{Cl}}_2^{\mathrm{o}-}\to {\mathrm{Cl}}_2+{2\mathrm{Cl}}^{-}$$ 32

$${\mathrm{Cl}}^{\mathrm{o}}+{\mathrm{Cl}}^{\mathrm{o}}\to {\mathrm{Cl}}_2$$ 33

$${\mathrm{Cl}}_2+{\mathrm{H}}_2\mathrm{O}\to \mathrm{HOCl}+\mathrm{HCl}$$ 34

$$\mathrm{HOCl}\to {\mathrm{H}}^{+}+{\mathrm{HClO}}^{-}$$ 35

The inhibitory effect of CO₃⁻² was due to the ^oOH scavenging for the formation of carbonate radicals (CO₃^o−) (Eqs. 36–38).

$${\mathrm{CO}}_3^{2-}+{\mathrm{OH}}^{\mathrm{o}}\to {\mathrm{CO}}_3^{\mathrm{o}-}+{\mathrm{OH}}^{-}$$ 36

$${2\mathrm{CO}}_3^{\mathrm{o}-}+{\mathrm{H}}_2\mathrm{O}\to {2\mathrm{CO}}_2+{\mathrm{HO}}_2^{-}+{\mathrm{OH}}^{-}$$ 37

$${\mathrm{OOH}}^{-}+{\mathrm{OH}}^{\mathrm{o}}\to {\mathrm{OH}}^{-}+\mathrm{OOH}$$ 38

Finally, it could be concluded that water-anions could significantly affected the degradation of PQ by contributing the consumption of reactive species. Briefly, the inhibitory effect of these anions was due to (1) usage of reactive radicals in different pathway, (2) transforming of reactive radicals to anion radicals with a lower oxidation potential, and (3) and limited activity of catalyst surface due to adsorption of anions [31]. The results were in accordance with the findings of other researchers in this field [8, 31].

Determining the mineralization rate of PQ herbicide in Fe₃O₄/ZnO/US process

Figure 8 shows the mineralization rate of PQ herbicide in the ZnO/Fe₃O₄/US process under optimum conditions by calculating organic carbon removal percentage. According to this figure, 68% of the PQ were converted to inorganic compounds, after 60 min of reaction time (carbon dioxide, water, and inorganic acids). The complete mineralization of the PQ by the ZnO/Fe₃O₄/US process needed more reaction time (> 60 min) which may be because of the production of intermediates and their interfering effect on the sonocatalytic degradation of PQ molecules [43]. Macias-Sanchez et al. (2015) and Selvin et al. (2017) showed that in order to reach the complete mineralization of organic pollutants (2,4-D, picloram and rhodamine B) reaction times above 270 min and 180 min were required, respectively [44, 45]. Mirzaie et al. (2017) studied the removal of PQ from aqueous solutions using Fenton and pseudo-Fenton oxidation processes. According to their results, the rate of TOC removal by Fenton and pseudo-Fenton processes under optimum conditions (Fe³⁺ = Fe²⁺ = 0.9 mM, PQ conc. = 0.194 mM, pH = 3 H₂O₂ = 29.4 mM) and 120 min was 52.15% and 45%, respectively [46].

Fig. 8.

Total organic carbon removal rate during PQ degradation in ZnO/Fe₃O₄/US system under optimum conditions (pH = 6, catalyst conc. = 0.75 g/L, PQ conc. = 10 ppm, and US power = 70 w)

Evaluating the biotoxicity effect of PQ herbicide residues before and after sonocatalytic removal process in ZnO/Fe₃O₄/US system on Daphnia magna

Results of the biotoxicity test in different dilutions before and after the sonocatalytic degradation of PQ are given in Table 2. As can be seen, the mortality rate of Daphnia was increased by increasing the time. According to Table 2, after 24, 48, 72, and 96 h, the toxicity unit of raw samples containing PQ herbicide were reduced from 4.52, 15.46, 33.07, and 73.16 to 0.3, 0.53, 3.26, and 7.2, after treatment process by ZnO/Fe₃O₄/US system which showed that the acute toxicity in the treated samples was decreased by 10.16 times over 96 h. These results also showed that the toxicity of pollutants and effluents was decreased over time, probably because of the efficient role of Fe₃O₄/ZnO/US process in pollutant degradation and the formation of intermediate products with lower toxicity. Eventually, the findings of this study confirmed that Fe₃O₄/ZnO/US process could effectively reduce the toxicity of the studied herbicide. Hence, this process can be used as an effective method in detoxifying agricultural runoffs and industrial effluents which contains herbicides. Segura et al. (2008) [47], Malato et al. (2001) [48], and Farzadkia et al. (2014) [49] reported a remarkable decrease in the acute toxicity of Imidacloprid pesticide and phenol on Daphnia magna by photo-Fenton, photocatalytic oxidation, and catalytic ozonation processes.

Table 2.

LC₅₀ and TU values for raw and treated samples of PQ under ZnO/Fe₃O₄/US process during 96 h

Sample	time	LC50	95% Confidence Limits,		TU
	(h)	% (v:v)	Lower Bound	Upper Bound
Raw	24	22.12	13.23	109.01	4.52
	48	6.47	2.78	21.09	15.46
	72	3.02	0.99	12.19	33.07
	96	1.36	0.1	4.65	73.16
Treated	24	192.307	51.43	2110.2	0.3
	48	188.68	20.26	273.07	0.53
	72	30.67	4.38	47.54	3.26
	96	13.88	1.59	26.39	7.2

Conclusion

In this study, synthesizing, characterization and application of magnetic ZnO/Fe₃O₄ nanocatalyst for the degradation of PQ was investigated. The results of this research showed proper ZnO stabilization on the Fe₃O₄ surface. The catalyst also exhibited a considerable magnetic behavior and physical characteristics which provided it with high separation capability and reusability potential. The performance of US irradiation on PQ degradation was improved by adding the catalyst to the reaction. Under optimum conditions (catalyst concentration = 0.75 g/L, herbicide concentration = 10 mg/L, and US power = 70 W), the removal and mineralization rate, during 60 min reaction time, was 96.1% and 68%, respectively. Although many reactive species such as ^oOH, ^oO₂ and ^oH were generated during the ZnO/Fe₃O₄/US process, ^oOH had the major role in the degradation of PQ. The results also proved the reduction of the removal efficiency of PQ in the presence of specified scavengers and water anions. This decrease was due to the use of ^oOH radicals, which has been the most effective factor in the PQ oxidation process in the ZnO/Fe₃O₄/US system. Moreover, the results showed that the PO₄⁻³ and NO₃⁻ anions showed the highest inhibitory effect on the pollutant removal process in the above system, among the common water anions. According to the results of the biotoxicity test by Daphnia magna, the toxicity unit values in samples containing PQ herbicide was dramatically reduced from 73.16 in raw samples to 7.2 in the treated samples by ZnO/Fe₃O₄/US system (after 96 h). Finally, the ZnO/Fe₃O₄/US process showed a proper removal of PQ from aqueous solutions with advantages of great reusability, facility of use, and stable performance.

Declarations

Conflict of interests

The authors declare no conflicts of interest about the publication of submitted work.