Diclofenac Degradation in Aqueous Solution Using Electron Beam Irradiation and Combined with Nanobubbling

Abstract

Diclofenac (DCF) degradation in aqueous solution under electron beam (EB) irradiation after nanobubbling treatment was studied and compared with treatments using nanobubbling or EB irradiation alone. It was found that the removal efficiency of DCF increased by increasing the adsorbed dose, and it depended on the initial concentration of DCF in solution, being higher for the lower concentrations. Furthermore, when using the nanobubbling treatment alone, about 16% of the DCF was removed from the aqueous solution due to the OH radicals generated during the process. On the other hand, using EB treatment at 0.5 kGy, the degradation of DCF increased from 36% to 51% when adding a nanobubbling pretreatment before the EB radiation. At higher doses (5 kGy), the degradation of DCF was 96% using EB radiation and 99% using nanobubbling before EB radiation, indicating that the nanobubbling effect was not synergistic. With an increase in the adsorbed doses, EB radiation seemed to play a more important role on the degradation of DCF, probably due to the reactive species generated. Moreover, the solutions treated with nanobubbling and EB radiation presented higher COD values and radiolytic by-products with aromatic rings with chlorine. This work can support the development of innovative strategies to treat municipal wastewaters using ionizing radiation technologies.

1. Introduction

Diclofenac (DCF) is a non-steroidal anti-inflammatory drug (NSAID) which is widely used in the treatment and management of acute and chronic pain associated with inflammatory conditions [1]. Diclofenac has been detected in the samples of surface water and effluents of wastewater treatment plants (WWTP), which was reviewed by Shamsudin et al. and Yang et al. [2,3]. The presence of antibiotics, DCF and other pharmaceuticals in wastewater has led to an increased concern about their effects on aquatic ecosystems and human health. The release of untreated or partially treated pharmaceutical wastewater can increase environmental toxicity to aquatic life resulting in mutagenic and genotoxic effects [2] and leading to the development of antibiotic-resistant bacteria, which poses a significant public health risk. In fact, diclofenac accumulation in the environment can cause toxic effects, particularly in aquatic animals such as algae, rotifers and crustaceans [4], fish [5], mussels [6], and also in terrestrial animals, such as vultures, due to the consumption of carcasses containing residues of DCF [7].

The conventional methods of treating pharmaceutical wastewater, such as biological treatment and chemical coagulation, have limitations in their effectiveness and efficiency, especially for the removal of persistent organic compounds and pharmaceuticals. Therefore, the development of advanced treatment methods is critical for the efficient and effective removal of antibiotics and other pharmaceuticals from wastewater. The use of advanced treatment methods can lead to a more sustainable and safe approach to managing pharmaceutical wastewater and protecting public health and the environment. The degradation of DCF has been studied intensively in recent years [8]. Advanced oxidation processes (AOPs), such as photolytic processes with UV or solar irradiation [9,10], photocatalytic processes with TiO₂ as the catalyst [11,12], ozonation with/without H₂O₂ addition [13,14], sonolysis [15], photo-Fenton processes [16] and H₂O₂ oxidation [17], were applied. The selection of AOPs for wastewater treatment depends on several factors, including the type of pharmaceuticals present, the characteristics of the wastewater, and the required level of treatment. Among the various AOPs, electron-beam (EB) radiation is emerging as a promising technology due to its ability to efficiently degrade a wide range of contaminants without the need for chemical additives and can be easily scaled up. Radiolysis of DCF with gamma ray irradiation has been explored at the laboratory scale [18,19,20], and DCF degradation using electron-beam (EB) irradiation was also studied in recent years [21,22].

In recent years, nanobubbling was considered as a potential technology to be applied for water purification to remove organic compounds. It has advantages such as a small size, large specific surface area, long residence time in the water, high mass transfer power, high zeta interface potential and the capacity to produce hydroxyl radicals [23,24,25]. The methods to generate nanobubbles include pressure variation [26], hydrodynamic cavitation [27], ultrasonic cavitation [28], electric fields [29] and porous membranes [30]. Michailidi et al. [27] produced air and oxygen bulk nanobubbles in a water solution using hydrodynamic cavitation with an optimum processing time of 30 min. The produced nanobubbles had sizes between 190 and 680 nm and long-term stability up to three months. The molecular dynamics simulations used by Ghaani et al. [29] found that the origin of the nanobubbles’ movement was due to dielectrophoresis, while the substantial nanobubble stabilization was attributed to surface-polarization interactions. On the other hand, Ma et al. [30] developed porous alumina films with straight nanoholes to generate bulk nanobubbles, and observed that the minimum size of the nanobubbles depended on the solubility of the encapsulated gases in water, but it was less than 100 nm. There are studies on the use of nanobubbling for antibiotic degradation [24], but to the best of our knowledge, there is no study on a DCF treatment method using this technology.

In this way, the aim of this work was to study the effect of the combined treatment of electron-beam (EB) radiation with nanobubbling pretreatment on the removal of DCF from aqueous solutions, in order to find the conditions for optimal energy usage. Previously, other authors investigated the use of nanobubbling aeration to enhance ozone diffusion in an electro-peroxone AOP for the removal of pharmaceutical contaminants from hospital wastewater [31], which obtained individual removal efficiencies higher than 85%. Nevertheless, as far as the authors know, this study represents the first application of the combination of EB and nanobubbling to degrade DCF. After describing the decomposition studies with different experimental conditions, a discussion on the intermediate products will be carried out.

2. Materials and Methods

2.1. Reagents and Materials

Diclofenac sodium (purity ≥ 98%), formic acid (98–100%), acetonitrile (HPLC grade, >99.9%) and L-alanine (purity > 99%) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Na₂CO₃ (purity ≥ 99%), methanesulfonic acid (purity ≥ 99%) and NaHCO₃ (purity ≥ 99.8%) were acquired from Fluka Chemie GmbH (Buchs, Switzerland). Ferrous ammonium sulfate (FAS) was obtained from Carlo Erba (Val de Reuil, France) and potassium dichromate was obtained from JMGS (Odivelas, Portugal). Silver sulfate was purchased from VWR (Radnor, PA, USA). Deionized Milli-Q water was used for preparing solutions.

2.2. Samples Preparation

To study the degradation of DCF, solutions with different concentrations (10, 125 and 250 mg/L) were prepared by dissolving DCF in Milli-Q water, without pH adjustment. To study the effect of the combination of electron-beam radiation with nanobubbling, solutions of DCF at 125 mg/L were prepared. The concentration used in each experiment is described in the corresponding section.

2.3. Electron-Beam (EB) Irradiation and Dosimetry

The irradiation experiments were performed using the ILU-6 accelerator (2 MeV, 20 kW) located in Institute of Nuclear Chemistry and Technology (INCT, Poland), with an energy of 1.7 MeV. A 0.2 M L-alanine solution was used for dosimetry [32]. The dose rate was adjusted by changing the flow rate of the aqueous solution and electron accelerator parameters, such as the beam current and frequency of the beam pulse.

2.4. Nanobubble Generation and Nanobubble and EB Process

The experimental equipment including a nanobubble treatment unit and electron-beam irradiation unit was built in the laboratory (Figure 1). The installation comprised two stages. The first stage is responsible for nanobubble generation. A CASHIDO pump (CAHIDO, Type MIB 25X UP, Taiwan, China) was used to generate bulk microbubbles. A nozzle was installed inside the stainless-steel container where 2000 mL of the liquid solution with DCF was kept. No additional air was added during the generation of bulk microbubbles. Microbubbles was produced by hydrodynamic cavitation. “White water” was seen inside the container after the pump was started. Dynamic Light Scattering was used to measure the nanobubble size; however, no nanobubbles were observed in this work. The examined solution was circulated between the nanobubble generator (Cashido, Taiwan, China) and the container. As nanobubble generator operation causes solution heating, which may affect the concentration of diclofenac, a cooling system was implemented. It contained a cooling unit equipped with a pump (Julabo, Seelbach, Germany) and cooler in the form of a stainless-steel tube coil immersed in the container with the diclofenac solution. Distilled water, used as a cooling liquid, was circulated between the cooler and cooling unit. The second stage was structured to allow for the irradiation of liquids in flow mode. The main element was the irradiation chamber, which was made of silicone tubes: one tube with inner diameter of 3.5 mm connected with a tee to two tubes with an inner diameter of 1.5 mm and wall thickness of 0.75 mm. It divided the liquid stream into two separate streams to increase the radiation efficiency with a lower overall flow rate. The exits of the smaller tubes were connected again with another tee to merge both streams back to one, so that the irradiated solution could be collected in the container. The tubes forming the irradiation chamber were attached to a wooden base to keep them straight and placed under an electron accelerator exit window for irradiation. To directly examine the solution through this chamber, a peristaltic pump (Aqua-trend, Poland) with adjustable and remotely controlled rotation speed was employed. The dose rate was adjusted by changing the peristaltic pump rotation speed and thus the flow rate or electron accelerator parameters (beam current and frequency).

The dosimetry of the experimental system was estimated with a 0.2 M L-alanine solution according to the method of Kovacs et al. [32] using conductivity measurements. The prepared 0.2 M L-alanine solution was calibrated against alanine pellets (Aerial, Illkirc, France). Solution samples (50 mL) were placed under a 10 MeV Elektronika 10/10 electron accelerator (Siberia, Russia) and irradiated with alanine pellet dosimeters. The conductivity of the irradiated alanine solution samples was measured at 20 °C using a conductivity meter (Elmetron CX 461, Warsaw, Poland) and the alanine pellet absorbed dose was measured using an EPR spectrometer MagnetTech MS 5000 (Freiberg/Bruker, Berlin, Germany). Conductivity values were correlated with absorbed dose values, which allowed for the creation of a calibration curve and equation. Later, the same solution was entered into the experimental setup, nanobubbled, and then irradiated. The nanobubbling of alanine solution before irradiation was conducted to keep the same irradiation conditions for both the dosimetric solution and diclofenac solution. The absorbed dose was estimated by conductivity measurements performed under the same conditions as those given above and the results were correlated to electron accelerator work parameters and peristaltic pump adjustment. This allowed us to determine the calibration curve and equation correlating the absorbed dose and peristaltic pump adjustment fitted to the specified electron accelerator parameters.

2.5. Analytical Methods

The DCF concentration was analyzed using reversed-phase High Performance Liquid Chromatography (HPLC) (Shimadzu LC-10 AT, Shimadzu, Kyoto, Japan) with a diode array UV–Vis detector (model SPD-M10AVP, Kyoto, Japan). The analytical column was a Phenomenex Luna C18 (2) (250 × 4.6 mm, 5 μm, Torrance, CA, USA) and the detection was made at 280 nm. The mobile phase was a mixture of a 0.2 M aqueous solution of formic acid and acetonitrile (40:60) with an injected sample volume of 20 μL.

The removal efficiency (R%) of DCF was is calculated according to Equation (1) [33]:

$$\mathrm{R}\mathrm{\%}=\frac{\left({\mathrm{C}}_0-{\mathrm{C}}_{\mathrm{t}}\right)}{{\mathrm{C}}_{\mathrm{t}}}\times 100$$

(1)

where C₀ is the initial concentration of DCF and C_t is the concentration of the treated DCF.

The pH measurements were performed using an Elmetron CX-461 multimeter (Warsaw, Poland) designed for accurate measurements of pH, which had been previously calibrated with buffered solutions.

Dissolved oxygen was measured using a Mettler Toledo DO meter purchased from Sigma Aldrich (Poznań, Poland).

Total carbon (TC) and inorganic carbon (IC) were analyzed using a TOC analyzer multi N/C 3100 (Analytik Jena GmbH+Co. KG, Jena, Germany), with a combustion temperature at 820 °C.

A Dionex ICS-5000 SP chromatograph (ThermoFisher Scientific, Frederick, MD, USA) was applied to analyze cations and anions. A Dionex Ion PacTM AS23 + AG23 (4 × 250 mm) column and Dionex Ion PacTM CS12A + CG12A (4 × 250 mm) were used to examine anions and cations, respectively. Dionex seven anion standard II and Dionex six cation standard II were used to make the calibration curves. Mobile phase 4.5 mM Na₂CO₃ + 0.8 mM NaHCO₃ and mobile phase 20 mM methanesulfonic acid at a flow rate of 1 mL/min were used for the analysis of anions and cations, respectively, with a sample injection volume of 25 µL.

Chemical oxygen demand (COD) was determined according to Method 5220 C of the Standard Methods for the Examination of Water and Wastewater using the Titrimetric Method [34]. COD measures the oxygen equivalent to the organic matter content of a sample that is susceptible to chemical oxidation by potassium dichromate.

The degradation products of DCF were identified using Liquid Chromatography-Mass Spectrometry (LC-MS), using a C18 reverse-phase column. The detailed experimental conditions are given in the Appendix A. The chemical identity of each identified degradation product was confirmed by accurate mass determination on an ImpactII ESI-QqTOF mass spectrometer (details are given in the Appendix B).

2.6. Statistical Analysis

Origin software version 7.5 was used for data analysis. The results are represented as the mean of at least three independent experiments.

3. Results

3.1. Influence of Dose on DCF Degradation under EB Irradiation

The effect of the adsorbed dose on DCF degradation under EB irradiation was studied using initial concentrations of DCF varying from 10 mg/L to 250 mg/L (Figure 2). The removal efficiency of DCF increased with increasing adsorbed dose, and it depended on initial concentration of DCF in the solution, being higher for the lowest concentration. It was 99%, 45% and 33% at the 0.5 kGy absorbed dose for initial concentrations of DCF of 10, 125 and 250 mg/L, respectively. As far as the authors know, this is the first study evaluating the influence of the initial concentration on the degradation of diclofenac by EB irradiation; however, the same tendency was also reported in studies on the decomposition of diclofenac using EB irradiation [21] and gamma radiation [33,35] and other pharmaceutical compounds using gamma radiation [36,37,38].

The degradation of DCF in aqueous solution by EB irradiation was previously studied by other authors with or without the use of additives [21,22,39]. Tominaga et al. [22] reported the complete decomposition of DCF (50 mg/L) in aqueous solutions at 5 kGy, while He et al. [21] indicated that the removal efficiency of DCF (10–40 mg/L) was almost 100% for all the samples at 0.5 kGy. Zhuan and Wang [20] observed that 86% of DCF (30 mg/L) was degraded after gamma radiation at 1 kGy.

DCF decomposition was initiated by the primary products from water radiolysis, mainly H, ${\mathrm{e}}_{\mathrm{a}\mathrm{q}}^{-}$ and OH radicals, which formed according to the following reaction (2) [40]:

$${\mathrm{H}}_2\mathrm{O}\stackrel{\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{i}\mathrm{o}\mathrm{l}\mathrm{y}\mathrm{s}\mathrm{i}\mathrm{s}}{\to }\left[0.28\right]{}^{\bullet }\mathrm{OH}+\left[0.27\right]{}^{\bullet }\mathrm{H}+\left[0.07\right]{\mathrm{H}}_2{\mathrm{O}}_2+\left[0.05\right]{\mathrm{H}}_2+\left[0.27\right]{\mathrm{H}}^{+}$$

(2)

The numbers in brackets represent the radiation chemical yield (G-value, in µM/J).

The main species generated from water radiolysis are ${}^{\bullet }\mathrm{OH}$, ${}^{\bullet }\mathrm{H}$ and ${\mathrm{e}}_{\mathrm{a}\mathrm{q}}^{-}$.

DCF degradation can be written as

$$\frac{-\mathrm{d}\left[\mathrm{D}\mathrm{C}\mathrm{F}\right]}{\mathrm{d}\mathrm{t}}={\mathrm{k}}_{\mathrm{O}\mathrm{H}}\left[{}^{\bullet }\mathrm{OH}\right]\left[\mathrm{D}\mathrm{C}\mathrm{F}\right]+{\mathrm{k}}_{\mathrm{H}}\left[{}^{\bullet }\mathrm{H}\right]\left[\mathrm{D}\mathrm{C}\mathrm{F}\right]+{\mathrm{k}}_{{\mathrm{a}\mathrm{q}}^{-}}\left[{\mathrm{a}\mathrm{q}}^{-}\right]\left[\mathrm{D}\mathrm{C}\mathrm{F}\right]$$

(3)

The initial pH of the DCF solution was 6, meaning that the ^●H concentration was low and its contribution to DCF degradation can be neglected under this condition. The hydroxyl radical is a strong oxidant species with a standard reduction potential compared to a normal hydrogen electrode (NHE) of E⁰ (^●OH/OH⁻) = 1.9 V, whereas a hydrated electron is a reductant, with a standard reduction potential (vs. NHE) of E⁰ (${\mathrm{e}}_{\mathrm{a}\mathrm{q}}^{-}$) = −2.9.

The rate constants of DCF with ^●OH and hydrated electrons are given in Equations (4) and (5), respectively [19,39]:

$${}^{\bullet }\mathrm{OH}+\mathrm{D}\mathrm{C}\mathrm{F}=\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}\mathrm{s}(0.93~1.24)\times {10}^{10} {\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$

(4)

$${\mathrm{e}}_{\mathrm{a}\mathrm{q}}^{-}+\mathrm{D}\mathrm{C}\mathrm{F}={\mathrm{C}\mathrm{l}}^{-}+\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}(1.53~3.1)\times {10}^9 {\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$

(5)

The reaction rate of ${}^{\bullet }\mathrm{OH}$with DCF is faster than that with ${\mathrm{e}}_{\mathrm{a}\mathrm{q}}^{-}$. The reaction of the OH radical with DCF involves an addition reaction to a benzene ring (4) and ${\mathrm{e}}_{\mathrm{a}\mathrm{q}}^{-}$ reacts with DCF by a dissociative electron capture reaction and releases Cl⁻ ions (5).

3.2. Nanobubbling Effect

The nanobubbling effect on DCF removal was studied in two cases. In the first case, DCF was diluting using distilled water after bubbling, and it was found that there were no changes in DCF concentration after 48 h. The second case involved bubbling the DCF solution; it was observing that the DCF concentration was about 10% lower than its initial concentration (10 mg/L) after 30 min of nanobubbling (Figure 3). The dissolved oxygen concentration was reduced with increased bubbling time, probably caused by the coalescence of the numerous bulk microbubbles [27]. The reduction in DCF concentration during the nanobubbling process was caused by OH radicals (Equation (3)), which were generated by the cleavage of the entrapped water molecular due to collapse of nanobubbles [25].

3.3. DCF Degradation in Aqueous Solution by EB Irradiation after Nanobubbling Process

Low concentrations of DCF were easily decomposed under EB irradiation (Figure 2). Almost 100% of the DCF was decomposed at 0.5 kGy when the DCF initial concentration was 10 mg/L; thus, a higher initial concentration of DCF was selected to determine the effect of nanobubbling. The degradation of DCF was studied after nanobubbling pretreatment followed by EB irradiation for a solution with an initial DCF concentration of 125 mg/L. The DCF solution was first nanobubbled for 30 min, followed by EB irradiation. After 30 min of nanobubbling, about 16% of the DCF was removed from the aqueous solution due to the reaction of OH radicals, which were generated by with nanobubbling process, with DCF (Equation (4)). With an increasing adsorbed dose, the concentrations of ${\mathrm{e}}_{\mathrm{a}\mathrm{q}}^{-}$ and OH radicals also increased. The removal efficiency of DCF was improved under EB irradiation with or without nanobubbling pretreatment. At a 5.0 kGy absorbed dose, the concentration of OH radicals generated by EB irradiation was much higher than that generated through the nanobubbling process. Approximately 99% and 96% of the DCF were efficiently removed under EB irradiation with and without the nanobubbling pretreatment, respectively (Figure 4).

The inorganic ions products (Cl⁻, NH₄⁺) from the DCF degradation in the aqueous solution are presented in Figure 5 and Figure 6, respectively. The concentrations of Cl⁻ and NH₄⁺ (shown as Cl and N, respectively) increased with increasing dose, i.e., with the decrease in DCF concentration. Cl⁻ was generated due to the dissociated electron capture reaction of hydrated electron with DCF (Equation (4)). These results are in agreement with those reported by He et al. [21]. Based on the calculation of mass balance, at a 5 kGy absorbed dose, about 53% and 49% of Cl (equivalent to 2 Cl atoms) were released from DCF degradation under EB irradiation with and without the nanobubbling pretreatment (

Table 1. DCF degradation (%) and formation of its main inorganic products (%) in water under electron-beam irradiation at 5 kGy, with (Nano + EB) and without (EB) nanobubble pretreatment ([DCF]₀ = 125 mg/L).

	EB (5 kGy)	Nano + EB (5 kGy)
DCF	96%	99%
Cl−	53%	49%
${\mathrm{N}\mathrm{H}}_4^{+}$	29%	35%
CO2	<1%	<1%

).

For both processes, CO₂ formation was less than 1% and NH₄⁺ formation was about 29% and 35%, respectively, under EB irradiation alone and EB irradiation together with the nano-bubbling pretreatment, respectively (Table 1).

Moreover, the dissolved CO₂ and carbonic salts seemed to increase with the absorbed dose (Figure 7).

From the analysis of the total carbon (TC) and inorganic carbon (IC) present in the samples before and after treatment at different doses (Figure 7 and Figure 8), it was possible to conclude that the total organic carbon (TOC) for both processes, measured by subtracting IC from TC, increased until 2 kGy, with the values remaining relatively stable at higher absorbed doses. The increase in TOC could be related to the higher amounts of organic matter containing carbon in the irradiated aqueous samples. Contrary to these results, Homlok et al. [18] reported 90% mineralization of 0.5 mM DCF using gamma radiation at 40 kGy and Tominaga et al. [22] and He et al. [21] observed TOC removal efficiencies lower than 6.5% at 2 kGy and 7.5 kGy, respectively.

Looking at DCF’s structure (Figure 9), it is possible that Cl was released from position “1” (blue color) of DCF and NH was released from position “2” (green color) of DCF, forming Cl⁻ and NH₄⁺ through related chemical reactions, and CO₂ was formed from bond cleavage in position “3” (red color).

The degradation of DCF under EB irradiation, EB irradiation together with nanobubbling, and the nanobubbling process alone were studied in flow system. Concerning nanobubbling alone, about 20% of DCF was removed after this process. In the case of the nanobubbling + EB process, where the DCF solution was irradiated just after nanobubbling, its removal efficiency was slight lower than that of the EB process alone at absorbed doses below 5 kGy (Figure 10). At 6 kGy, about 91% of the DCF seemed to be removed from the aqueous solution in both the EB and nanobubbling + EB processes (Figure 10).

3.4. Chemical Oxygen Demand (COD)

Chemical oxygen demand (COD) is equivalent to the amount of oxygen required to chemically oxidize the organic matter content in a water sample. The COD of the DCF solution before and after EB irradiation with and without nanobubbling was measured (Figure 11). The DCF decomposition was markedly influenced by the dose up to 3 kGy. In comparison with the control non-treated sample (154 mg O₂/L), the COD value increased after EB irradiation at 3 kGy (461 mg O₂/L) and after nanobubbling and EB irradiation in the batch system (307 mg O₂/L). On the other hand, the nanobubbling + EB irradiation process in the flow system markedly increased the COD value (1382 mg O₂/L). These results are not in agreement with those obtained by He et al. [21], which described a COD removal efficiency of 18% at 2 kGy, or by Homlok at al. [18], who reported that the COD decreased by 20% at 20 kGy. The higher values of COD for these samples could be attributed to the increase in oxidable organic matter in the samples, corroborating the obtained results for TC (Figure 8). The presence of filamentous fungi detected in the samples of the nanobubbling and EB irradiation process might suggest a high biodegradability of this solution.

3.5. Intermediate Products and Degradation Mechanism

The intermediate products of DCF (10 mg/L) in aqueous solution under EB irradiation and EB irradiation after nanobubbling in a flow system were analyzed using LC-MS (

Table 2. DCF and its radiolytic by-products in aqueous solution under EB irradiation (EB) and EB together with nanobubbling in flow system (Nano + EB, flow) (DCF₀ = 10 mg/L). Experimental and calculated isotope profiles are shown in Figure A1.

Sample	Peak	m/z [M + H]+	Molecular Weight	Molecular Formula	Reference(s)
DCF			296	C14H11Cl2NO2
Nano + EB, flow	P1	134	133	C9H11N	[22]
	P2	260	259	C14H10ClNO2	[10,39]
	P3	312	311	C14H11Cl2NO3	[20,39]
	P4	312	311	C14H11Cl2NO3	[20,39]
	P5	294	293	C14H9Cl2NO2	[16]
EB	P*	312	311	C14H11Cl2NO3	[18,20,39]
	P#	152	151	C8H9O2N	[22,39]
	P$	134	133	C9H11N	[22]
	P^	126	125	C6H4ClN	This work
	P’	106	105	C7H7N	[22]

). As an example, Figure 12 represents the chromatogram of a DCF solution after EB irradiation and nanobubbling in a flow system, recorded at 280 nm. DCF and its radiolytic by-products gave strong mass spectrometric signals under positive ionization conditions; thus, the measurements were performed mainly in this mode, where DCF generated a peak at m/z 296 ([DCF + H]⁺). In comparison with the non-irradiated solution, the peak area of DCF indicated a considerable decrease after both treatments, meaning that degradation of the compound was taking place and corroborated the results presented in Figure 2 with a decomposition of 99% at 0.5 kGy. At the same time, other peaks appeared, and they could be attributed to decomposition products of DCF (Figure 12).

For the solutions treated by EB radiation and EB irradiation after nanobubbling in flow system, five peaks were positively identified, with two similar products (P1 and P$, and P3, P4 and P*) (Table 2).

P3, P4 and P* ([C₁₄H₁₁C_l2NO₃ + H]⁺ at m/z 312) were the products formed from the addition reaction of ^•OH radicals to the aromatic ring of DCF, followed by H atom abstraction from H-C bonds. The specificity of electrophilic aromatic substitution is normally governed by the nature of the substituent. In this context, the amino group is a strong donating group and acts as the ortho-para director, which makes it possible for a hydroxyl radical to be added at two positions of the aromatic ring [39], as can be seen for Nano + EB (Table 2, P3 and P4). These products were also observed in previous studies [18,20,39]. P# ([C₈H₉O₂N + H]⁺ at m/z 152), P$ ([C₉H₁₁N + H]⁺ at m/z 134) and P’ ([C₇H₇N + H]⁺ at m/z 106) could be related to C-N bond cleavage, maintaining a nitrogen-containing group on the aromatic moiety [22,39]. P^ ([C₆H₄ClN + H]⁺ at m/z 126) was not previously detected as a DCF by-product but can be described as a dechlorination product of DCF. P2 ([C₁₄H₁₀ClNO₂ + H]⁺ at m/z 260) and P5 ([C₁₄H₉Cl₂NO₂ + H]⁺ at m/z 294) were only observed as degradation products of DCF when using EB irradiation after nanobubbling in the flow system. P2 seemed to result from the dechlorination-cyclization of DCF [39], as was also reported by Peng et al. [10] using UV-activated peroxymonosulfate. P5 could be generated by the possible cyclization of P3/P4 with the loss of H₂O, as described by Pérez-Estrada et al. [16] for the degradation of DCF under a photo-Fenton treatment.

Taking into consideration the previous results and the identification of radiolytic by-products of DCF, its degradation mechanism under EB irradiation after the nanobubbling process was proposed (Figure 13).

It is important to mention that irradiation doses can influence the generated degradation products of target compounds. Based on this, the identification of other radiolytic by-products of DCF by Zhuan and Wang [20], Homlok et al. [18] and Tominaga et al. [22] could be attributed to the different doses applied in each study. On the other hand, it is also evident that the decomposition of DCF can follow different reaction pathways depending on the degradation treatment used, as can be observed using the photo-Fenton method [16], UV-activated peroxymonosulfate [10] or photocatalysis using TiO₂-based zeolite catalyst [12].

4. Conclusions

The degradation of diclofenac (DCF) in aqueous solution was studied using electron-beam (EB) irradiation after a nanobubbling treatment and compared with the treatments of nanobubbling and EB irradiation alone. The results demonstrated that at lower doses (0.5 kGy), the degradation of DCF was increased by 15%, from 36% to 51%, when using EB radiation after nanobubbling pretreatment in comparison with EB radiation. Nevertheless, at higher doses (5 kGy), the difference between the processes was not so evident, since 96% and 99% of the DCF was degraded using EB treatment and nanobubbling followed by EB radiation, respectively. Thus, the nanobubbling pretreatment seemed to have no significant effect on the degradation of DCF in aqueous solution. Actually, although the inorganic ion formation slightly increased, the results indicated that the COD value markedly increased after the nanobubbling pretreatment in the flow system, which could be related to the amount of organic matter in the sample. Concerning the degradation products, some with an aromatic ring with chlorine were detected and identified in solutions treated by EB radiation after nanobubbling, which can justify the higher COD values on these solutions. Further studies should be performed with increasing radiation doses in order to achieve mineralization of the DCF solutions. The toxicity of the radiolytic by-products should be also assessed to understand the effectiveness of the DCF solution treatments. This work can support the development of strategies to remove pharmaceutical pollutants from wastewaters using ionizing radiation technologies as an alternative process to the conventional ones, which have been demonstrated to be ineffective in the elimination of DCF and other pharmaceuticals.