《1. Introduction》

1. Introduction

Metronidazole (1-hydroxyethyl-2-methyl-5-nitroimidazole; MNZ), one of the most common nitroimidazole-based antibiotic drugs, is widely used to treat infections by anaerobic bacteria and protozoa, such as Trichomonas and Giardia lamblia [1,2]. MNZ is also added to poultry and fish feed to eliminate parasites. More importantly, MNZ possesses potential biological carcinogenicity and mutagenicity [3–5]. As a pharmaceutical substance and a potential carcinogen, MNZ tends to accumulate in aquatic environments, resulting in the pollution of surface water and groundwater [6–8]. Due to the toxicity and high water solubility of MNZ [9], the elimination of this compound from the environment has become a significant issue. Hence, it is imperative to exploit new techniques to remove MNZ pollution effectively.

Various technologies have been reported to degrade MNZ, including ultraviolet (UV) irradiation [10], Fenton/UV [11], heterogeneous photocatalysis [12], nanoscale zero-valent metal reduction [13], and adsorption/bio-adsorption on activated carbon [14]. Aside from these methods, non-thermal plasma generated by electrical discharge in liquid or at the gas–liquid interface, inducing a variety of chemical and physical effects in situ, has attracted a considerable amount of attention [15]. These effects include intense UV radiation, shockwaves, high-energy electrons, and, especially, the formation of various reactive oxygen species (i.e., ozone, hydrogen peroxide, hydroxyl radicals, and perhydroxyl radicals) [16]. Therefore, plasma discharge has been widely used for the removal of different antibiotics from aqueous solutions, such as pentoxifylline [17], three β-lactam antibiotics [18], and carbamazepine [19]. However, there are few reports on the degradation of MNZ by plasma discharge, even though this technology may be efficiently utilized as an environmentally friendly technology for MNZ degradation.

Diverse plasma-discharge modes have been studied with different electrode geometries (e.g., needle–needle, needle–plate, and wire–plate) for degrading different types of organic compounds [20]. Recently, a plasma-assisted rotating disk reactor (plasmaRDR), which involves the coupled fields of plasma and high gravity, was innovatively developed to enhance the degradation efficiency of organic dye compounds [21]. The main advantages of the plasma-RDR are as follows: ① Because of the strong centrifugal force, a micron-liquid film is obtained, which favors the renewal of the plasma–liquid interface and thus increases the utilization of the hydroxyl radicals; and ② all the involved energies produced by plasma discharge can act directly on the liquid film.

Titanium dioxide (TiO2) is regarded as an ideal photocatalyst due to its high charge separation efficiency and its effective utilization of UV light radiation [22,23]. During photocatalytic processes, reactive oxygen species—which are key active sources in the photocatalytic degradation of antibiotics—can be generated [24–26]. Thus, a high degradation efficiency of MNZ can be expected in a plasma-RDR coupled with TiO2.

This work studied the enhancement of the degradation efficiency of MNZ. The effects of the operating conditions on the energy yield were evaluated, and a correlation between the energy yield and the operating conditions in the plasma-RDR was obtained. A plasma-RDR and a TiO2 photocatalyst were coupled in order to enhance the degradation efficiency of MNZ. The structure and morphology, catalytic performance, and reusability of the TiO2 before and after discharge were systematically characterized. Three-dimensional (3D) excitation emission fluorescence matrix (EEFM) spectroscopy was selected to qualitatively investigate the concentration of MNZ. Major degradation intermediates of MNZ were identified, and a possible degradation pathway for MNZ was then proposed.

《2. Experimental method and procedure》

2. Experimental method and procedure

《2.1. Plasma generator》

2.1. Plasma generator

The input voltage of the plasma generator (Beijing Ruiant Technology Co., Ltd., China) was an alternating current (AC) pulse signal. The voltage and current in the plasma-RDR were measured by a Rigol PVP2150 high-voltage probe (RIGOL Technologies Co., Ltd., USA) and a Cybertek CP0030A current probe (Shenzhen Zhiyong Electronics Co., Ltd., China) connected to a Tektronix TDS3032C oscilloscope (Tektronix, Inc., USA). The average power of the plasma-RDR, as determined by the voltage and current waveforms, can be calculated using Eq. (1):

where P is the average power, U is the instantaneous voltage, I is the instantaneous current, T is the pulse duration, and f is the pulse repetition frequency. Considering previous reports on discharge characteristics for the degradation of rhodamine B, the peak voltage (Vpp) and electrode gap (d) were set as 44 kV and 8 mm, respectively [21]. 

Energy yield, was adopted to illustrate the energy consumption required to degrade MNZ, as calculated by Eq. (2) [27]:

where Y is the energy yield (g∙(kW∙h)–1 ), C0 is the initial concentration of MNZ (g∙L–1 ), V is the volume of MNZ solution (L), η is the degradation efficiency, and t is the treatment time (h).

《2.2. Experimental procedure》

2.2. Experimental procedure

Fig. 1 shows the experimental setup for the degradation of MNZ. Deionized water containing MNZ and oxygen was fed into the plasma-RDR by gas–liquid co-current flow. In our previous research, we provided details on plasma-RDR with gas–liquid cocurrent flow [28]. Oxygen with a flow rate of 0.5 m3 ∙h–1 was fed into the plasma-RDR for 30 min before the start of the experiments in order to generate more reactive oxygen species. Then deionized water containing MNZ was introduced into the plasma-RDR, flowed radially outward through the rotating low-voltage electrode, was collected in a fluid reservoir below, and was circulated into the plasma-RDR by means of a pump. The MNZ solution was sampled every 20 min at the liquid outlet.

《Fig. 1》

Fig. 1. Schematic diagram of the experimental setup.

《2.3. Analytical methods》

2.3. Analytical methods

The concentration of MNZ was determined by measuring the absorption intensity of MNZ at the wavelength of 318 nm with a UV spectrophotometer (U-2800, Hitachi, Japan) [7,29]. The typical absorption spectra of MNZ is shown in Fig. 2. The efficiency with which the MNZ solution was degraded (herein referred to as the degradation efficiency of MNZ) was calculated based on the following equation:

where C is the final concentrations of MNZ.

《Fig. 2》

Fig. 2. Absorption spectra of MNZ solution before and after treatment. N: rotational speed.

The crystal structure of TiO2 was analyzed by X-ray diffraction (XRD; D/max 2550, Rigaku Co., Japan) and the angular range of 2θ was from 10° to 80°. The processes were carried out at 40 kV and 50 mA.

Intermediates of MNZ degradation during the plasma treatment process were identified through liquid chromatography–mass spectrometry (LC-MS; Xevo G2 QTof, Waters Co., USA).

《3. Results and discussion》

3. Results and discussion

《3.1. Enhanced degradation performance of plasma-RDR》

3.1. Enhanced degradation performance of plasma-RDR

As shown in Fig. 3(a), the effect of rotational speed (N) on the degradation efficiency of MNZ was evaluated. The role of rotation on MNZ degradation was found to be significant. The degradation efficiency of MNZ was merely 9.5% at 0 r∙min1 and increased by 24% at 400 r∙min–1 . When the rotational speed was increased to 500 r∙min–1 , the degradation efficiency was as high as 50.2%. The degradation efficiency of MNZ changed slightly when the rotational speed exceeded 500 r∙min–1 . Due to the strong centrifugal force, the plasma–liquid interface renewal was clearly enhanced, followed by improvement in the utilization of the hydroxyl radicals. As shown in Fig. 3(b), the concentration of hydroxyl radicals generated in the plasma-RDR increased by 2.954 mg∙L–1 when the rotational speed was increased from 0 to 500 r∙min–1 .

《Fig. 3》

Fig. 3. Enhanced (a) degradation efficiency of MNZ and (b) production of hydroxyl radicals in the plasma-RDR. L: liquid flow rate.

《3.2. Investigation of the degradation efficiency and energy yield of the plasma-RDR》

3.2. Investigation of the degradation efficiency and energy yield of the plasma-RDR

3.2.1. Effects of operating conditions on degradation efficiency

Fig. 4 illustrates the effects of the operating conditions on the degradation efficiency. The effects of the initial concentration of MNZ, pulse repetition frequency, and volume of solution were evaluated when the rotational speed and liquid flow rate were respectively set as 500 r∙min–1 and 80 L∙h–1 .

《Fig. 4》

Fig. 4. Effects of (a) initial concentration of MNZ, (b) pulse repetition frequency, and (c) volume of solution on the degradation efficiency of MNZ.

As shown in Fig. 4(a), the degradation efficiency of MNZ decreased with the increase of the initial concentration of MNZ. When there was a lower concentration of MNZ, the initial degradation efficiency was higher. The final degradation efficiency reached 73.1% and 68.1% for MNZ concentrations of 5.0 mg∙L–1 and 7.5 mg∙L–1 , respectively. However, the degradation efficiency was only 54.8% for an MNZ concentration of 15 mg∙L–1 . Thus, it can be reasonably speculated that the degradation of high concentrations of MNZ requires more energy and a longer treatment time. Hence, a concentration of 7.5 mg∙L–1 was selected as the experimental condition for subsequent experiments using only plasmaRDR.

Pulse repetition frequency was selected to change the average power. Fig. 4(b) shows that the degradation efficiency of MNZ increased with an increase in pulse repetition frequency. The degradation efficiency increased slightly when the pulse repetition frequency increased from 200 to 240 Hz. The experimental results can be explained as follows: At higher pulse repetition frequencies, the electrical condenser had insufficient time to fully recharge; therefore, the energy injected to the plasma-RDR decreased [30].

The volume of treated solution affected the capacity to degrade MNZ. As shown in Fig. 4(c), the degradation efficiency reached 73% with a solution volume of 1.75 L and then decreased slightly for volumes ranging from 1.75 to 2.25 L. The degradation efficiency decreased by 20.4% with an increase in the solution volume from 2.25 to 2.50 L. When the volume of the solution exceeded 2.25 L, the capacity to treat MNZ was seriously restricted. Based on these experimental results, a solution volume of 2.00 L was selected for the subsequent experiments.

3.2.2. Effects of operating conditions on energy yield and a correlation between energy yield and operating conditions

Degradation efficiency can be illustrated by the quantity of MNZ decomposed per unit of energy (i.e., yield). The energy yield relies on the type of plasma reactor, the initial concentration, and the characteristics of the compound [31]. Approximately 70% degradation was obtained within 120 min using the starting parameters mentioned earlier. Fig. 5 presents detailed data on the energy yield under different operating conditions. It was noted that the energy yield as a function of time decreased and changed with the pulse repetition frequency, the volume of solution, and the initial concentration of MNZ. When the pulse repetition frequency was increased from 120 to 240 Hz, the energy yield increased from 0.19 to 0.38 g∙(kW∙h)–1 for an MNZ concentration of 7.5 mg∙L–1 within 60 min. Similarly, with an increase in the initial concentration of MNZ from 5.0 to 15.0 mg∙L–1 , the energy yield increased from 0.28 to 0.44 g∙(kW∙h)–1 within 60 min. When the volume of solution was increased from 1.75 to 2.50 L, the energy yield decreased from 0.38 to 0.20 g∙(kW∙h)–1 for a concentration of 7.5 mg∙L–1 within 60 min.

《Fig. 5》

Fig. 5. Effects of operating conditions on the energy yield of the plasma-RDR.

Based on the above experimental data, a correlation between the energy yield, Y, and the other operating conditions in a plasma-RDR was proposed, as shown in Eq. (4):

It can be seen that the pulse repetition frequency and the volume of the solution have a greater impact on energy yield, whereas the treatment time has the smallest effect among the operating conditions. Fig. 6 shows the values of energy yield obtained experimentally versus the values predicted by Eq. (4), which exhibits a deviation of ±15%.

《Fig. 6》

Fig. 6. Comparison of the calculated and experimental energy yield, Y.

《3.3. Degradation of MNZ in a plasma-RDR coupled with TiO2》

3.3. Degradation of MNZ in a plasma-RDR coupled with TiO2

3.3.1. Effect of TiO2 additive amount

The effect of the amount of TiO2 added to the solution in the plasma-RDR on the degradation efficiency of MNZ is exhibited in Fig. 7(a). The experiments were conducted with a liquid flow rate of 80 L∙h–1 , an initial MNZ concentration of 15.0 mg∙L–1 , and a pulse repetition frequency of 200 Hz. The experimental results revealed that 0.2 g∙L–1 was the optimal TiO2 additive amount for a higher degradation efficiency of MNZ. When the additive amount of TiO2 exceeded this value, the degradation efficiency induced by the plasma discharge decreased. It can be understood that an increase in TiO2 particles improved the number of photons that were absorbed. When the concentration of TiO2 particles exceeded 0.2 g∙L–1 , light scattering and screening effects ensued. Excessive opacity of the suspension restricted the TiO2 nanoparticles that were furthest in from being irradiated [32]. Due to the scattering and screening effects, the catalytic activity of the TiO2 particles was reduced [33]. Particle aggregation always occurs at high concentrations [34]. The optimum amount of TiO2 for degrading MNZ was deemed to be 0.2 g∙L–1 . Furthermore, the synergistic effect of the plasma-RDR plus TiO2 was significant in generating hydroxyl radicals. As shown in Fig. 7(b), the concentration of hydroxyl radicals increased by 0.671 mg∙L–1 within 120 min with the aid of the TiO2 photocatalyst.

《Fig. 7》

Fig. 7. Effects of TiO2 dosage on (a) degradation efficiency of MNZ and (b) production of hydroxyl radicals.

3.3.2. Characteristics of TiO2 before and after plasma discharge

To examine the morphology and structural features of the TiO2 particles before and after discharge in the plasma-RDR, scanning electron microscope (SEM) and transmission electron microscope (TEM) patterns of TiO2 were obtained for comparison, as shown in Figs. 8(a)–(f). Figs. 8(a) and (c) respectively show the SEM and TEM images of the original TiO2 particles, from which it can be clearly concluded that many of the TiO2 particles were irregularly spherical. After discharge for the treatment of MNZ, the surface morphology of the TiO2 particles (Figs. 8(b) and (d)) was found to be almost identical to that of the original TiO2 particles. Figs. 8(e) and (f) indicate nearly the same plane spacing, which was approximately 0.350 nm both before and after discharge, corresponding to the spacing of the (101) plane of anatase TiO2 [35,36].

Fig. 8(g) also shows the XRD patterns of anatase TiO2 before and after discharge. After plasma discharge, the TiO2 particles retained the same anatase-phase diffraction peaks as the original TiO2 particles, which verified that the TiO2 particles were not destroyed during the plasma discharge.

In the process of catalytic degradation, it was important to study the stability of the catalysts in long-term reactions, in order to evaluate the catalysts’ applicability for water treatment. Fig. 8(h) shows the degradation efficiency of MNZ with used TiO2 that was recovered by filtration and thoroughly dried. The degradation efficiency of MNZ dropped slightly after four cycles, which indicates that the plasma discharge has little influence on the photocatalytic property of TiO2.

《Fig. 8》

Fig. 8. SEM images of TiO2 (a) before and (b) after discharge; TEM images of TiO2 (c) before and (d) after discharge; high-resolution transmission electron microscope (HRTEM) images of TiO2 (e) before and (f) after discharge; (g) XRD patterns of TiO2 before and after discharge; (h) cycling tests of TiO2 particles.

《3.4. 3D EEFMs analysis, degradation intermediates, and a possible degradation pathway》

3.4. 3D EEFMs analysis, degradation intermediates, and a possible degradation pathway

3D EEFMs were adopted to analyze the fluorescent information of MNZ based on the conjugated heterocyclic structures in the organic compounds [37]. Fluorescence in various regions indicates different components of the molecules. The fluorescence intensity qualitatively reflects the concentration of organic compounds. Therefore, 3D EEFM analysis was implemented to illustrate changes in the molecular structure and concentration of MNZ. As shown in Fig. 9, the fluorescent peaks were situated in the following ranges for the excitation over emission wavelengths (EX/EM, EX denotes excitation wavelength, EM denotes emission wavelength): (200–500 nm)/(300–600 nm). As exhibited in Figs. 9(b) and (c), the fluorescent peaks disappeared gradually, suggesting that the conjugated heterocyclic structures in MNZ were destroyed. The experimental results show that the plasma-RDR/TiO2 system can be used for MNZ removal.

《Fig. 9》

Fig. 9. 3D EEFMs of MNZ solution after treatment times of (a) 0 min, (b) 60 min, and (c) 120 min in the plasma-RDR/TiO2 system. EX: excitation wavelength; EM: emission wavelength.

Table 1 lists the major intermediates that were detected by LC-MS. The results revealed that the degradation of MNZ was complex. Heterocyclic intermediates were mainly formed by hydroxylation with denitration (No. 3 and No. 5), oxidation of the lateral N-ethanol to an N-acetic acid group (No. 2 and No. 3), and the loss of these lateral groups (No. 4 and No. 5). It is anticipated that further breakage of the imidazole portion and oxidation of the resulting linear compounds would produce a mixture of short linear aliphatic carboxylic acids. The presence of oxalic, oxamic, formic, and acetic acids was corroborated by highperformance liquid chromatography (HPLC) analysis [38,39]. These aliphatic acids can be mineralized into CO2 and H2O. According to the above experimental results, a possible degradation pathway is proposed, as shown in Fig. 10.

《Table 1》

Table 1 Intermediates identified by LC-MS.

《Fig. 10》

Fig. 10. Possible degradation pathway of MNZ in the plasma-RDR/TiO2 system.

《3.5. Degradation mechanism of MNZ in the plasma-RDR/TiO2 system》

3.5. Degradation mechanism of MNZ in the plasma-RDR/TiO2 system

Based on the experimental results described above, a schematic diagram of the MNZ degradation mechanism by the plasma-RDR/ TiO2 system is exhibited in Fig. 11. UV–visible light generated by the plasma-RDR induced TiO2 particles and caused the separation of electron–hole pairs. The produced h+ was able to react with H2O and OH– to form ∙OH. The generated conduction electrons were trapped by both the O2 and the O3 produced by the discharge, further generating ∙OH. The produced ∙OH by means of photocatalysis, together with the formed reactive oxygen species by plasma discharge, jointly reacted with MNZ to produce intermediates.

《Fig. 11》

Fig. 11. Degradation mechanism of MNZ in the plasma-RDR/TiO2 system. UV-Vis: ultraviolet–visible; VB: valence band; CB: conduction band.

《4. Conclusions》

4. Conclusions

This work studied the degradation of MNZ in aqueous solution. The degradation efficiency of MNZ and the concentration of hydroxyl radicals generated in the plasma-RDR respectively increased by 41% and 2.954 mg∙L–1 with an increase in the rotational speed from 0 to 500 r∙min–1 . A higher pulse repetition frequency favored the degradation efficiency, whereas a higher initial concentration of MNZ and a greater volume of MNZ solution were unfavorable to the degradation efficiency. Based on the experimental data, the energy yield was calculated and a correlation was proposed. The deviation between the predicted values and the experimental data was within ±15%. The enhanced degradation performance of the plasma-RDR plus TiO2 was significant, as the concentration of hydroxyl radicals increased by 0.671 mg∙L–1 within 120 min. 3D EEFMs and LC-MS demonstrated that the conjugated heterocyclic structures of MNZ were destroyed and small molecules were produced. The plasma-RDR/TiO2 system is thus a promising technology for antibiotic wastewater treatment.

《Acknowledgments》

Acknowledgments

This work was supported by the National Natural Science Foundation of China (21725601).

《Compliance with ethics guidelines》

Compliance with ethics guidelines

Yong Cai, Guang-Wen Chu, Yong Luo, Meng-Jun Su, Bao-Ju Wang, Bao-Chang Sun, and Jian-Feng Chen declare that they have no conflict of interest or financial conflicts to disclose.