Tunable triangular and honeycomb plasma structures in dielectric barrier discharge with mesh-liquid electrodes

1.   Introduction

As a third-generation quinolone antibiotic, ciprofloxacin (CIP) has been frequently detected in surface water, wastewater and even drinking water in recent years [1]. CIP that enters the aquatic environment eventually takes the form of a prototype or a metabolite, posing a significant threat to human health and aquatic ecosystems even at trace levels due to its high toxicity and persistence [2–4].

As antibiotics cannot be oxidized effectively through various conventional methods, advanced oxidation technologies (AOTs) have been developed to remove these pollutants. Among the various types of AOT, non-thermal plasma technologies have attracted extensive attention as there are multiple physical (UV irradiation and shock wave) and chemical processes [various reactive oxygen species (ROS)] involved in pollutant degradation [5–8]. However, using discharge plasma for wastewater treatment has the problem of a relatively low mineralization efficiency for organic pollutants. Although efficient removal of organic pollutants could be achieved by enhancing the discharge power and prolonging the treatment time, the energy consumption would also increase significantly. Therefore, setting up synergistic processes based on the non-thermal plasma technologies is an effective solution to overcome the problem.

Persulfate (PS) activation technology has attracted a great deal of attention for pollutant removal in the past few years due to the formation of sulfate radicals (${\mathrm{SO}}_4^{-}$·), which have a strong oxidation capacity (2.5–3.1 V). Studies have proved that ${\mathrm{SO}}_4^{-}$· and a series of ROS produced by PS activation can dissociate various organic pollutants [9]. Many methods, including those using heat, light and chemicals (transition metal ions, zerovalent iron or activated carbon), have been proven to activate PS to successfully form ${\mathrm{SO}}_4^{-}$· [10–12]. Recently, non-thermal plasma has shown excellent PS activation performance and consequently improved the mineralization efficiency of pollutant degradation [13–15]. On one hand, the heat, UV light and high-energy electrons (e*) produced in the plasma system activate the PS to form various radicals which then strengthen the degradation; on the other hand, the main species of O₃ formed during the discharge process provide another route of PS activation. The extra ·OH and ${\mathrm{SO}}_4^{-}$· results in further enhancement of organic compound degradation [16].

Based on the above analysis, a synergistic system of dielectric barrier discharge (DBD) plasma and PS was set up in this work to treat CIP wastewater. A water falling film DBD plasma system, which was considered having more O₃ formation [17, 18], was used to activate the PS. The combining activation of the PS induced by the radiation and O₃ formed concurrently in the DBD process was utilized to promote the more radicals' formation and then the better degradation of the CIP. The reactive species formation and influencing factors analysis were carried out to verify the synergistic effect of the DBD plasma and the PS activation. Other detection techniques, such as chemical oxygen demand (COD) and total organic carbon (TOC) removal, UV–vis spectroscopy, three-dimensional fluorescence spectrum analysis, degradation pathways, and the corresponding toxicity analysis, were also implemented to illustrate the CIP degradation in the synergistic system.

2.   Experiment

2.1   Chemical materials

Sodium persulfate (Na₂S₂O₈, ≥98%), sodium chloride (NaCl, ≥99.5%), sodium hydrogen carbonate (NaHCO₃, ≥99.5%), ethanol (EtOH, ≥99.8%), tert-butyl alcohol (TBA, ≥99%), triethylenediamine (TEDA, ≥96%), sodium dihydrogen phosphate (DHP, ≥98%), p-benzoquinone (BQ, ≥98%), sodium hydroxide (NaOH, ≥96%), titanyl sulfate and 30% hydrogen peroxide were purchased from the China National Pharmaceutical Group Corporation. CIP (≥98%), 5, 5-dimethyl-1-pyrroline N-oxide (DMPO, ≥97%), 2, 2, 6, 6-tetramethylpiperidine (TEMP, 98%), ferrous sulfate (FeSO₄, ≥98%) and copper sulfate (CuSO₄, ≥99%) were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. Acetonitrile (C₂H₃N, ≥99.9%) was purchased from Shanghai Macklin Biochemical Technology Co., Ltd. All reagents were used as purchased without further purification.

2.2   Experimental system and methods

The water falling film DBD plasma system used in the study as well as the detailed parameters of the system were described in our earlier research [19]. A schematic diagram is also shown in figure S1 (available online at stacks.iop.org/PST/25/025505/mmedia). The ice bath cooling system and the fan shown in the figure were used to control the solution temperature. The solution temperature change during the reaction period is from 19 ℃ to 34 ℃. An oscilloscope (Tektronix DPO 2014b) and probes (Tektronix, P6015A and TPP0200) were combined together to record the voltage and current waveforms as well as the Lissajous figure for the discharge power calculation. In each experiment, the discharge power was set at 0.4 kW, and 3 l CIP solution or deionized water was circulated in the experimental system using a peristaltic pump (China, Baoding Lead Fluid, BT600/YZ15) at a flow rate of 1.5 l min^-1. A magnetic stirrer (China, Xinrui, DJ-1) was used to ensure uniform mixing of the solution. The total reaction time was 40 min and samples were taken every 10 min for the following analysis.

2.3   Analytic methods

High-performance liquid chromatography apparatus (China, Lunan Ruihong, P1010) assembled with a UV detector and a Symmetry C18 column (Φ4.6 × 250 mm) was used to measure the CIP concentration at a wavelength of 277 nm. An acetonitrile and water (0.1% formic acid) 13:87 v/v solution at a flow rate of 1 ml min^-1 was used as the mobile phase. The CIP degradation efficiency (DE) was calculated according to formula (1), and the energy yield (EY) of CIP degradation [G₅₀, mg (kWh)^-1] was calculated using formula (2). The synergetic factor (SF) was also calculated (formula (3)) to illustrate the synergetic effect of the DBD plasma and the PS [20].

Here, C₀ (mg l^-1) is the initial concentration of CIP; C_t (mg l^-1) is the CIP concentration obtained after t min reaction; V (L) is the solution volume (3 l in this study); f (Hz) is the applied discharge frequency; C (F) is the internal sampling capacitance of the power supply; S is the calculated area according to the obtained Lissajous figure; t₅₀ (h) is the time taken for 50% CIP degradation (obtained from the kinetic analysis results of CIP degradation). k_DBD+PS is the kinetic constant of CIP degradation in the DBD–PS system; k_DBD is the kinetic constant of CIP degradation in the DBD-only system; k_PS is the kinetic constant of CIP removal by only the addition of PS.

The initial pH value of solution was adjusted using 0.1 M NaOH and 0.1 M HCl solutions. During the experiments, no more than two drops of HCl solution were used to adjust the solution pH value, and the corresponding Cl^- concentration was about 0.03 mM, so the effect of the Cl^- on the collection of radicals was overlooked. A pH meter (China, Shanghai INESA, PHS-3E) was used to measure the pH values of the different solutions, and the solution conductivity was monitored by a conductivity analyzer (China, Shanghai INESA, DDS-307A). Full-band scanning of the CIP solutions during degradation at different reaction times was recorded with a UV-visible spectrophotometer (Japan, Shimadzu, UV-2600). Solution COD and TOC values were detected by a COD analysis meter (China, Lianhua, 5B-3A) and a TOC-measuring instrument (German, German element, Vario TOC select) respectively. Molecular fluorescence properties of the degradation intermediates were characterized by a three-dimensional fluorescence spectrophotometer (Hitachi, F-4600). An electron spin resonance instrument (ESR; Germany, Bruker, Micro ESR) was used to determine the relative amount of HO· and ${\mathrm{SO}}_4^{-}$· radicals. H₂O₂ concentration was determined by the titanyl sulfate method [7] with a UV spectrophotometer (China, Shanghai Yidian, N4S). O₃ concentration in the liquid phase was measured using an ozone detector (China, Shanghai Haiheng, CY-1A). The byproducts formed during the CIP degradation process were identified via liquid chromatography–mass spectrometry (LC-MS; Thermo LXQ). All experiments were repeated three times and the results are shown with an error bar.

3.   Results and discussion

3.1   Reactive species formation

The concentrations or relative concentrations of critical reactive species formation, including O₃, H₂O₂, ·OH, ${\mathrm{SO}}_4^{-}$· and ¹O₂, in the DBD-only system and the DBD–PS system with deionized water were first analyzed to verify the synergetic effect of the DBD plasma and the PS. Figure 1(a) reveals that the O₃ concentration in the DBD–PS system was higher than that in DBD-only system. This might be due to the higher solubility of O₃ in the solution with a lower pH value, which was caused by the addition of PS [18]. In addition, during the first 30 min reaction time, O₃ accumulated and the concentration increased, and the concentrations decreased at 40 min. This might be due to the slight increase of solution temperature during the reaction, which led to O₃ decomposition. Figure 1(b) lists the change of H₂O₂ concentrations in the DBD-only system as well as the DBD–PS system during the reaction process. Evidently, H₂O₂ concentration increased gradually by prolonging the reaction time in the DBD-only system, which was due to the accumulation of the active substances accompanying the reaction. However, the H₂O₂ concentration in the DBD–PS system decreased with the extension of reaction time and the concentrations were lower than those in the DBD-only system. The results might be attributed to the consumption of H₂O₂ induced by the PS activation (reactions (4) and (5)) [21].

Figure  1.  Reactive species formation: (a) O₃, (b) H₂O₂, (c) ESR (DBD-only), (d) ESR (DBD–PS), (e) ESR (DBD–PS). Experimental conditions: P = 0.4 kW, deionized water, PS concentration is 4 mM.

DownLoad: Full-Size Img PowerPoint

ESR/DMPO and ESR/TEMP spin-trapping experiments were carried out to investigate the relative concentrations of ·OH, ${\mathrm{SO}}_4^{-}$· and ¹O₂ in the different reaction systems and the results are described in figures 1(c)–(e). Characteristic signals of DMPO–·OH adducts (with hyperfine splitting constants of α_N = α_H = 14.9 G (Gauss)) and DMPO– ${\mathrm{SO}}_4^{-}$· adducts (with hyperfine splitting constants of α_N = 13.2 G, α_H = 9.6 G, α_H = 1.48 G and α_H = 0.78 G) were observed [22, 23]. The signals shown in figure 1(d) verify the formation of ${\mathrm{SO}}_4^{-}$· in the DBD–PS system, and there were stronger signals of DMPO–·OH adducts in the DBD–PS system. The results confirmed the synergistic effect of the DBD plasma and the PS activation, not only for the new reactive species generation (${\mathrm{SO}}_4^{-}$·) (reaction (6)), but also for the strengthening of the ·OH production (reactions (7) and (8)). The signal intensity of TEMP–¹O₂ adducts shown in figure 1(e) shows that the ¹O₂ formed in the DBD–PS system were significantly lower than those in the DBD-only system. This might be due to the reaction between the ·HO₂ and the S₂${\mathrm{O}}_8^{2-},$ which hinders the ¹O₂ generation (reactions (9)–(12)) [19].

3.2   Influencing factor analysis

3.2.1   PS concentration

The effect of PS concentration on CIP degradation in the DBD–PS system was investigated in the study and the results are summarized in figure 2, including degradation efficiency, kinetic analysis, energy yield and SF values. The reviewed PS concentrations were 1, 2, 3, 4 and 5 mM, respectively.

Figure  2.  Effect of the PS concentration. (a) Degradation efficiency, (b) kinetic analysis, (c) energy yield, (d) SF value. Experimental conditions: P = 0.4 kW, C₀ = 10 mg l^-1.

DownLoad: Full-Size Img PowerPoint

As an increase of PS concentration could lead to more radical formation and accordingly more efficient degradation of pollutants [24], figure 2(a) shows that the degradation efficiency of CIP was enhanced by the gradual increase of PS concentration and the enhancement was dramatically evident compared with those in the DBD-only and PS-only systems. The degradation efficiency reached 97.73% after 40 min treatment in the DBD–PS system with 4 mM PS addition. The results confirmed the synergistic effect of the DBD plasma and the PS on CIP degradation, which was mainly due to the PS activation induced by the physical and chemical effects formed in the plasma system. In the activation process, ${\mathrm{SO}}_4^{-}$· and the extra ·OH (as shown in figure 1) were formed through a series of reactions and then contributed to the further degradation of the organic compounds. Comparing the CIP degradation in the DBD–PS system with different additive concentrations of PS, it can be seen that the increase of CIP degradation was more obvious in the 1–4 mM PS addition range than with 5 mM PS addition. The persulfate anion reacted with the formed ${\mathrm{SO}}_4^{-}$· when excessive PS was added into the reaction system (reaction (13)) [25], which accordingly resulted in a moderate increase of CIP degradation when the additive concentration of the PS exceeded 4 mM.

Combining the results of the degradation efficiency and energy yield of the CIP degradation under different amounts of PS addition, 4 mM was chosen as the optimal PS additive concentration for investigating synergistic effects in the subsequent experiments.

3.2.2   CIP concentration

Figure 3 shows the CIP degradation in both the DBD–PS system and the DBD-only system when changing the initial concentration of the CIP. Apparently, significant removal of CIP was observed in the DBD–PS system and higher degradation could be obtained at a lower initial CIP concentration in the DBD–PS system as well as the DBD-only system (figure 3(a)). Furthermore, it can be seen from the results of the corresponding kinetic analysis (figure 3(b)) that the kinetic constants increased by 12.53 times, 10.68 times and 3.30 times, when the initial concentrations were 5 mg l^-1, 10 mg l^-1 and 20 mg l^-1, respectively. The results were due to the relatively stable reactive species formation in both the DBD–PS system and the DBD-only system as the input power was constant. The limited reactive species could only oxidize a certain amount of target molecules, which led to a decrease of degradation percentage in accordance with the increase of the initial concentration of CIP. The slightly higher energy yield in the case of 10 mg l^-1 CIP solution (figure 3(c)) was caused by the relatively higher collision probability between the reactive species and the CIP molecules under the condition of a higher solution concentration.

Figure  3.  Effect of initial concentration of the CIP. (a) Degradation efficiency, (b) kinetic analysis, (c) energy yield. Experimental conditions: P = 0.4 kW, PS concentration is 4 mM.

DownLoad: Full-Size Img PowerPoint

3.2.3   Initial pH value

The effect of the initial solution pH on the CIP degradation in the DBD–PS system was reviewed by adjusting the solution pH value from 2.45 to 11.03. The results are shown in figure 4. It can be obtained from the results that the degradation efficiency, the k value as well as the energy yield were relatively higher under the condition of initial pH of 2.45. It can also be seen that there was a synergistic effect of the DBD plasma and the PS under different solution pH values and the G₅₀ was highest under the condition of an initial pH of 2.45. Under the condition of acidic solution, PS could be decomposed into ${\mathrm{SO}}_4^{-}$· (reactions (14) and (15)) and then attack organic pollutants (such as CIP) rapidly [26]. Under neutral or alkaline pH conditions, ${\mathrm{SO}}_4^{-}$· could undergo a quenching reaction, which reduced the formation of active substances and slowed down the CIP degradation (reactions (16)–(19)). Although ${\mathrm{SO}}_4^{-}$· can be converted to ·OH, the presence of various ions (such as ${\mathrm{SO}}_4^{2-}$ and ${\mathrm{CO}}_3^{2-}$) in the solution might play an inhibitory role in the reactivity of ·OH or ${\mathrm{SO}}_4^{-}$· [26–28].

Figure  4.  Effect of the initial pH value. (a) Degradation efficiency, (b) kinetic analysis, (c) energy yield. Experimental conditions: P = 0.4 kW, C₀ = 10 mg l^-1, PS concentration is 4 mM.

DownLoad: Full-Size Img PowerPoint

3.2.4   Inorganic ion addition

To broaden the possible application of the synergistic system in practical wastewater treatment, the effects of adding different metal ions (Fe²⁺, Cu²⁺) and inorganic anions (Cl^-, HC${\mathrm{O}}_3^{-}$) on CIP degradation were also studied and the results are described in figure 5. The corresponding chemicals representing the ions were FeSO₄, CuSO₄, NaCl and NaHCO₃ (2 mM).

Figure  5.  Effect of inorganic ion addition. Experimental conditions: P = 0.4 kW, C₀ = 10 mg l^-1, PS concentration is 4 mM.

DownLoad: Full-Size Img PowerPoint

Figure 5 shows that the CIP degradation performance was significantly improved within the first 10 min reaction when Fe²⁺ was added into the DBD–PS system. The concurrent addition of Fe²⁺ and PS in the DBD plasma system could contribute to increased production of ${\mathrm{SO}}_4^{-}$· and ·OH through PS activation by the Fe²⁺ and the Fenton reaction (reactions (20) and (21)) [14], and then lead to an rapid increase of CIP degradation in the primary stage. CIP removal was also enhanced by addition of Cu²⁺, the CIP degradation percentage reached 100% after 40 min reaction. The results might be due to the PS activation induced by the Cu²⁺ and the catalysis of Cu²⁺ for the H₂O₂ decomposition (reactions (22)–(24)) [29]. Figure 5(c) illustrates the CIP degradation in the DBD–PS system with Cl^- addition. It can be seen from the figure that the degradation efficiency was reduced by 13.13% after 40 min reaction when 2 mM Cl^- was added into the synergistic system, which was mainly due to the formation of Cl· by the reactions between the Cl^- and ${\mathrm{SO}}_4^{-}$· or ·OH (reactions (25) and (26)) [30]. Furthermore, the CIP degradation efficiency decreased with a further extension of reaction time when HC${\mathrm{O}}_3^{-}$ was added, which was due to the inhibition reactions of HC${\mathrm{O}}_3^{-}$ with ${\mathrm{SO}}_4^{-}$· and ·OH (reactions (27) and (28)) [31, 32].

3.2.5   Radical scavenger addition

In the DBD–PS system, ·OH and ${\mathrm{SO}}_4^{-}$· were thought to be the two most important reactive species for organic compound degradation. To figure out the contributions of ·OH and ${\mathrm{SO}}_4^{-}$· to the degradation, quenching tests (figure 6(a)) were performed by adding different radical scavengers (TBA for ·OH and EtOH for ${\mathrm{SO}}_4^{-}$·). TBA can react 1000-fold faster with ·OH (k = (3.8–7.6)×10⁸ M^-1 s^-1) than with ${\mathrm{SO}}_4^{-}$· (k = (4.0–9.1)×10⁵ M^-1 s^-1) (reactions (29) and (31)) [28]. EtOH can react with ${\mathrm{SO}}_4^{-}$· (k = (1.2–2.8)×10⁹ M^-1 s^-1) at a faster reaction rate constant than with ·OH (k = (1.6–7.7)×10⁷ M^-1 s^-1) (reactions (30) and (32)) [18]. Meanwhile, BQ, DHP and TEDA were used to quench ${\mathrm{O}}_2^{-}$·, e^* and ¹O₂, respectively (figure 6(b)).

Figure  6.  Effect of radical scavenger addition. (a) Quenching tests for ·OH and ${\mathrm{SO}}_4^{-}$· (200 mM), (b) quenching tests for ${\mathrm{O}}_2^{-}$·, e^* and ¹O₂ (5 mM). Experimental conditions: P = 0.4 kW, C₀ = 10 mg l^-1, PS concentration is 4 mM.

DownLoad: Full-Size Img PowerPoint

It can be observed from figure 6(a) that the CIP degradation exhibited an obvious decline from 97.73% (no scavenger) to 41.31% (TBA addition) and 61.48% (EtOH addition), respectively, indicating that the CIP degradation obeyed the following inhibition order: TBA > EtOH. Combining the different reaction rate constants of TBA and EtOH with ·OH and ${\mathrm{SO}}_4^{-}$·, it could be concluded that there was more ·OH in the DBD–PS system, which led to the stronger CIP degradation. The addition of BQ, DHP and TEDA also had negative effects on CIP degradation (shown in figure 6(b)), which indicated the positive effects of ${\mathrm{O}}_2^{-}$·, e^* and ¹O₂ on CIP oxidation.

3.3   Degradation of the CIP

3.3.1   UV–vis analysis

Figure 7 illustrates the UV–vis spectra of the CIP solutions at different reaction times in the DBD-only and DBD–PS systems. It can be seen from the figure that the characteristic peak of CIP appears at 277 nm, and the intensity of the characteristic peak decreases with the prolonged reaction time in the DBD system with or without PS addition. The decrease is more significant in the DBD–PS system, which confirms the synergetic effect of DBD plasma and PS, and beyond that, the byproducts formed by CIP degradation were responsible for the gradual increase of the absorption intensity around 200–250 nm. The byproducts with a conjugated double bond, benzene ring and an unsaturated carbonyl group had absorption peaks at 190–230 nm [33], and further oxidation of these byproducts would cause the continuous accumulation of substances with absorption peaks of 210–230 nm.

Figure  7.  UV–vis analysis. (a) DBD-only system, (b) DBD–PS system. Experimental conditions: P = 0.4 kW, C₀ = 10 mg l^-1, PS concentration is 4 mM.

DownLoad: Full-Size Img PowerPoint

3.3.2   Three-dimensional fluorescence spectrum analysis

The conjugated heterocyclic structures in organic compounds show different fluorescence intensities in three-dimensional fluorescence spectra [34]. The fluorescence in different regions also represents different components, and the corresponding change of intensities can reflect the relative concentration change of the organic compounds [35]. Therefore, the evolution process of the CIP solution during the degradation process was analyzed by three-dimensional fluorescence spectroscopy, as shown in figure 8 (obtained at 0 min, DBD-only 40 min and DBD–PS 40 min, respectively). Figure 8(a) presents that the fluorescent peaks were located in the range of E_x/E_m = (376–550 nm)/(200–375 nm), which belong to humic acid-like and fulvic acid-like substances. With the extension of reaction time, the fluorescence intensities within the above range decreased (figure 8(b)), which indicated that the conjugated heterocyclic structures of the CIP were destroyed and the heterocyclic structure intermediates were oxidized. After 40 min DBD–PS treatment, the fluorescence in the position significantly reduced (figure 8(c)), which illustrated that the conjugate heterocyclic structures of the CIP were largely destroyed. The results verified the more effective CIP degradation in the DBD–PS system.

Figure  8.  Three-dimensional fluorescence spectrum analysis. (a) 0 min, (b) DBD-only 40 min, (c) DBD–PS 40 min. Experimental conditions: P = 0.4 kW, C₀ = 10 mg l^-1, PS concentration is 4 mM.

DownLoad: Full-Size Img PowerPoint

3.3.3   Variation of solution pH and conductivity

Figure 9 displays the variation of the CIP solution pH and conductivity during reactions in the DBD-only system and the DBD–PS system. Figure 9(a) presents that the solution pH decreased from the initial value of pH 3.83 to 2.25 during the 40 min reaction time in the DBD-only system and the corresponding solution conductivity increased from 0.0099 to 1.596 mS cm^-1, while the solution pH value decreased from pH 3.47 to 2.00 and the solution conductivity increased from 0.607 to 3.407 mS cm^-1 in the DBD–PS system (figure 9(b)). With the mineralization of CIP, some byproducts, such as organic acids and inorganic ions (i.e. HC${\mathrm{O}}_3^{-},$${\mathrm{CO}}_3^{2-},$ etc), were produced in the system [4], which resulted in the decrease of solution pH and the increase of solution conductivity.

Figure  9.  Variation of the solution pH and conductivity during the reaction. (a) DBD-only system; (b) DBD–PS system. Experimental conditions: P = 0.4 kW, C₀ = 10 mg l^-1, PS concentration is 4 mM.

DownLoad: Full-Size Img PowerPoint

Furthermore, nitric acid and carbonic acid could also be produced by the gas ionization and then cause a change of solution properties [9]. The lower solution pH and higher conductivity in the DBD–PS system after the reaction illustrated the deeper mineralization of the CIP by the synergistic effect of the DBD plasma and the PS activation.

3.3.4   TOC and COD removal

In order to evaluate the mineralization and degradation degree of the CIP solution in the DBD plasma system with or without PS addition, TOC and COD removal was examined. It can be observed from figure 10(a) that the TOC removal in the DBD plasma system was only 13.81% and the value was enhanced to 25.75% after adding PS. In addition, the calculated energy yield of the TOC reduction in the DBD–PS system was 136.71 mg (kWh)^-1, which was 58.81 mg (kWh)^-1 higher than that in the DBD-only system (77.9 mg (kWh)^-1). The results indicate an accelerated mineralization process with the addition of PS. Meanwhile, the COD removal results shown in figure 10(b) display that COD removal was 29.59% in the DBD-only plasma system, while it rose to 49.40% after adding the PS. The results indicated that DBD plasma with added PS could relieve water pollution more effectively.

Figure  10.  TOC and COD removal. Experimental conditions: P = 0.4 kW, C₀ = 10 mg l^-1, PS concentration is 4 mM.

DownLoad: Full-Size Img PowerPoint

3.4   Degradation byproducts and pathway analysis

3.4.1   Anion formation

The anions formed during the CIP degradation were determined by ion chromatography and the results are shown in figure 11. The concentrations of F^-, N${\mathrm{O}}_2^{-},$ N${\mathrm{O}}_3^{-}$ and ${\mathrm{SO}}_4^{2-}$ gradually increased with the extension of discharge time. After 40 min reaction, the concentrations of F^-, N${\mathrm{O}}_2^{-},$ N${\mathrm{O}}_3^{-}$ and ${\mathrm{SO}}_4^{2-}$ were 0.164, 0.32, 252 and 939 mg l^-1, respectively. F^- was formed through the defluorination reaction induced by the ROS attack. The dissolution of nitrogen oxides formed by N₂ and O₂ in air plasma could generate N${\mathrm{O}}_2^{-}$ and N${\mathrm{O}}_3^{-}.$ In addition, NO· and NO₂· formed by a radical chain reaction might also generate N${\mathrm{O}}_2^{-}$ and N${\mathrm{O}}_3^{-}$ with the dissolved O₂. (reactions (33)–(36)) [36, 37]. ${\mathrm{SO}}_4^{2-}$ in the solution was produced by the self-decomposition of PS (reactions (37) and (38)) [14].

Figure  11.  Anion formation. Experimental conditions: P = 0.4 kW, C₀ = 10 mg l^-1, PS concentration is 4 mM.

DownLoad: Full-Size Img PowerPoint

3.4.2   Degradation pathway analysis

The organic intermediates of CIP degradation were identified by LC-MS and the possible degradation pathways in the DBD–PS system were then proposed. By analyzing the LC-MS spectra obtained at different time points (10, 30, 40 min), the substances with high content were selected to speculate the degradation path. The corresponding spectra are shown in figure S2. The obtained mass charge ratio (m/z), molecular formula and different chemical structures of the byproducts formed during the CIP degradation are summarized in table S1. According to the theoretical calculation of the Fukui functions and frontier electron densities (FEDs) of CIP reported by Liu et al and Xu et al [38, 39], the CIP reaction sites were preferentially attacked by ROS. The O18, C9, C1, N22, O14 and C6 atoms shown in figure 12 were more vulnerable to attack by ·OH and ${\mathrm{O}}_2^{-}$·. The N22, C10 and F24 positions were vulnerable to attack by ¹O₂ and ${\mathrm{SO}}_4^{-}$· radicals, which can preferentially react with organic compounds through the electron transfer mechanism. The N1 and N4 atoms on the piperazine ring contained lone pair electrons [40], which might be attacked by ${\mathrm{SO}}_4^{-}$· and then produce ammonium cation radicals through the electron transfer mechanism, and then the cracking of organic compounds could be finally realized.

Figure  12.  Ball-and-stick model of CIP.

DownLoad: Full-Size Img PowerPoint

According to the above analysis of the possible CIP oxidation reactions, three probable degradation pathways were proposed, as shown in figure 13. Pathway Ⅰ: the piperazine ring of CIP (m/z = 332) suffered from cleavage and oxidation to produce E1 (m/z = 305) due to the attack by strong oxidizing substances (${\mathrm{SO}}_4^{-}$· or ·OH) [40, 43], and the E1 was further transformed to E2 (m/z = 262) by losing the piperazine ring and a decarbonization reaction. E2 then underwent a hydroxyl substitution reaction and a desorption reaction by the ·OH attack to form E3 (m/z = 261). Finally, E4 (m/z = 201) was formed by losing HO- and a quinolone decarboxylation [41]. Pathway Ⅱ: E5 (m/z = 274) could be formed by the attack of ${\mathrm{SO}}_4^{-}$· and ·OH on the piperazine ring and a simultaneous defluorination reaction on the quinolone ring [40]. Then, E6 (m/z = 245) was produced through decarboxylation and hydroxyl substitution reactions [41]. Under the continuous attack of ROS, the ring-side group of the quinolone ring was opened and E6 transformed into E7 (m/z = 177) [42]; then, –COOH groups were cracked to form E8 (m/z = 149). Pathway Ⅲ: the piperazine ring of CIP was oxidized and cracked to produce E9 (m/z = 361). Then, E10 (m/z = 333) was generated via the missing of one carbonyl group from E9. Subsequently, E11 (m/z = 290) was formed by deamination and a carbonylation reaction, and the intermediate product of E12 (m/z = 260) was derived from decarboxylation and a hydroxyl substitution reaction of E11[4]. The further oxidative cleavage of the cyclopropane could result in the loss of -CH₂ and the formation of E13 (m/z = 220) [40]. Finally, E14 (m/z = 181) was generated through the cleavage of the quinolone rings and the loss of N22 [42]. These intermediate products then might undergo a series of reactions and be eventually mineralized to CO₂, H₂O, F^-, N${\mathrm{O}}_2^{-}$ and N${\mathrm{O}}_3^{-}$ [1, 13].

Figure  13.  Three probable degradation pathways of CIP in the DBD–PS system. Experimental conditions: P = 0.4 kW, C₀ = 10 mg l^-1, PS concentration is 4 mM.

DownLoad: Full-Size Img PowerPoint

3.5   Toxicity evaluation

The toxicity of the byproducts produced by organic pollutant degradation has important theoretical and practical significance for improving water environment quality. Acute LC₅₀ (LD₅₀) is a statistic of a substance that can be expected to cause death in 50% of the animals when given by a specified route as a single dose and the animals observed for a specified period of time [44]. Developmental toxicity represents the harmful effects of exogenous physical and chemical factors on offspring before they reach adulthood, including structural deformity, growth retardation, dysfunction and death [20]. Mutagenicity refers to the ability of a substance to change the chromosome base sequence of biological (especially human) cells. Therefore, the developmental toxicity and mutagenicity of CIP and its degradation byproducts (formed through the three pathways) on minnow (LC₅₀), daphnia (LC₅₀) and rat (LD₅₀) were analyzed using the Toxicity Estimation Software Tool (TEST) software to evaluate the effectiveness of the DBD–PS oxidation process. The results are presented in figure 14.

Figure  14.  Toxicity evaluation of CIP and its byproducts. (a) Pathway Ⅰ, (b) Pathway Ⅱ, (c) Pathway Ⅲ.

DownLoad: Full-Size Img PowerPoint

As displayed in figure 14(a) (Pathway Ⅰ), the toxicity of E1, E2, E3 and E4 was lower than that of CIP to the three animals. The developmental toxicity and mutagenicity of the four byproducts were also lower than those of CIP. Similar results could also be obtained from figures 14(b) and (c) (Pathway Ⅱ and Pathway Ⅲ). Based on the toxicity analysis results, it can be concluded that CIP could not only be mineralized by the DBD–PS treatment, but also the toxicity of the initial CIP solution decreased, indicating the effectiveness of the synergistic effect of the DBD plasma and the PS activation.

3.6   Economic analysis

To illustrate the economy of the DBD–PS technology, its energy yield for CIP degradation was compared with that of other technologies [8, 20, 45, 46] and the results are summarized in table S2. It can be seen from the table that the energy yield of the DBD–PS process was significantly higher than that of other technologies. The comparison proved the economy of the technology used in the research.

4.   Conclusions

CIP degradation using water falling film DBD plasma in the presence of PS in this work verified that PS could be activated by the DBD plasma, and then improved the degradation and mineralization of the CIP. The chemical and ESR analysis proved the production of reactive species, including O₃, H₂O₂, ·OH, ${\mathrm{SO}}_4^{-}$· and ¹O₂, in the DBD–PS synergistic system. A higher PS additive amount, a lower initial concentration of CIP and an acidic pH solution were favorable for CIP degradation in the DBD–PS system. The presence of Fe²⁺ and Cu²⁺ ions could accelerate the degradation, while the presence of Cl^- and HC${\mathrm{O}}_3^{-}$ had an inhibitory effect. The radical scavenger addition verified the roles of ·OH, ${\mathrm{SO}}_4^{-}$·, ${\mathrm{O}}_2^{-}$·, e^* and ¹O₂ in the CIP degradation. The UV–vis and three-dimensional fluorescence spectrum analysis proved the better degradation by the DBD–PS system. Furthermore, three probable pathways of CIP degradation were proposed according to the results of the detected inorganic ions as well as the organic intermediates. The toxicity evaluation results showed that the organic byproducts usually have lower toxicities than its parent CIP. A laboratory study could provide proof of the synergistic effect of the DBD plasma and the PS for organic compound degradation, and a scaling up of treatment capacity for real wastewater could be conducted in the future based on this study.