Pharmaceutically Active Compound (PhAC) Degradation by Means of Cold Plasma Jet Treatment

Abstract

The occurrence of emerging micropollutants of pharmaceutically active compounds (PhACs) in the environment poses a public health concern. Due to PhAC persistence and toxicity even at low concentrations, advanced oxidation processes (AOPs) have gained interest as effective treatment methods. In this context, the present study focuses on the application of a dielectric barrier discharge (DBD)-based plasma jet to Diclofenac (DCF) and Sulfamethoxazole (SMX) degradation in aqueous media. Plasma is sustained by continuous-wave sinusoidal high-voltage of audio frequencies, and negligible total harmonic distortion, in a helium–air mixture. The target pharmaceuticals are chosen based on anticipation of their occurrence due to rehabilitation center (DCF) and hospital (SMX) effluents in sewage systems. The degradation rates are determined by Liquid Chromatography Triple-Quadrupole Mass Spectroscopy (LC-MS/MS). Removal efficiency close to 100%, after 20 min of plasma treatment in the case of DCF at an initial concentration of 50 ppb, is achieved. The post-treatment action of the plasma-induced reactants on PhAC degradation over a day-scale period is studied. The results provide an insight into the dynamic degradation (kinetics) of both DCF and SMX, and they overall highlight the potentiality of the process under consideration for sewage remediation.

1. Introduction

Excess industrial growth and human-related activities lead to the development of a broad variety of new chemical compounds, with little-known environmental and public health risks [1], and no discharge limit for them has been established either in Europe or the US [2,3]. Personal care products, pharmaceuticals, and endocrine disruptors are emerging contaminants that represent a recently discovered broad group of synthetic or natural compound families [4,5,6,7]. Pharmaceuticals’ occurrence in aquatic environments is a globally emerging issue as they are released from the human body without being entirely metabolized [8] and they are designed, by their purpose, to be highly effective at a trace amount [9]. The seemingly ubiquitous presence of pharmaceuticals in municipal wastewater, surface water, and groundwater worldwide [8,10] has been reported in the last decade or so, and more than 80 pharmaceuticals and their metabolites at low concentrations from low ng L⁻¹ to high μg L⁻¹ have been detected [11]. Termed as micropollutants, the omnipresence of these compounds has shifted the focus of the scientific community across the globe, due to their potential to bioaccumulate [12,13,14,15]. Pharmaceutically active compounds (PhACs) enter into the environment mainly via municipal and hospital wastewater treatment plants (WWTPs) [16,17], since the latter are not originally designed for the complete removal of PhACs, and hence are not equipped with advanced technologies to eliminate emerging contaminants [18,19].

As is known, conventional treatment methods for WWTPs include biological, physical, and chemical procedures. Most WWTPs operate under physical and biological treatment steps, while only a few incorporate a tertiary or advanced sewage treatment method such as ultrafiltration, flocculation, ozonation, or osmosis [20]. On a laboratory scale, advanced oxidation processes (AOPs) have been gaining interest as effectual treatment approaches for wastewater, and mainly include photochemical, electrochemical, and catalytic oxidation, ozonation, Fenton or photo-Fenton systems, and sonolysis [21,22,23]. There are certain advantages and disadvantages of each of the processes [24], namely, their non-toxic nature, universal viability, and acceptability, but, on the other hand, high operation costs and the need for pre- or even post-treatment of the matrix. In addition to the above, the long duration of the process and aging effects have not yet been solved. Additionally, numerous studies have demonstrated that cold atmospheric-pressure plasmas belong to the category of AOPs, showing a good opportunity for environmental clean-up applications [25]. They are viewed as an advantageous and promising remediation technique for water and sludge derived from WWTPs, bearing benefits such as high degradation rate, energy efficiency, and relatively short treatment times [26,27].

Apart from traditional electrical discharge regimes, e.g., corona, arc, etc., the unique features of dielectric barrier discharges (DBDs), as well as DBD-based plasma jets, have been exploited with respect to the above-mentioned applications. Those unique features include restriction to arc transition, relatively low gas temperature, high yield of radicals, engineerability, cost efficiency, scalability, customizability in terms of electrode configuration, electric power delivery, gas feeding, etc. Accordingly, soil remediation [28,29], sewage biosolid sanitization [30,31], and water treatment [32,33] either above or within liquid media are some examples involving DBD-based processes, just to name a few.

The present work investigates the degradation of Diclofenac (DCF) and Sulfamethoxazole (SMX) in aqueous media using a DBD-based plasma jet. The target compounds were selected due to their frequency of occurrence in wastewater and environmental samples, which derives from their extensive use and poor biodegradability. DCF is a non-steroidal anti-inflammatory drug (NSAID) widely used to treat mild-to-moderate pain, as well as in rehabilitation centers to relieve joint, muscle, and bone pain. Almost 1500.00 tons of DCF is used annually, based on the Intercontinental Marketing Services (IMS) health data, which serves 82% of the global population without consumption for veterinary use [34], and it is resistant to biological degradation [35]. Previous studies have brought to the fore the interaction of this compound with other molecules in ecosystems, leading to the production of other toxic contaminants [36]. On the other hand, SMX is a commonly prescribed antibiotic in both human and veterinary medicine to cure a variety of infections of the urinary, respiratory, and gastrointestinal tracts. Since the human body can metabolize almost half (46%) of its intake, there is a high excretion rate, which can be detected in environmental samples in short time intervals, and it is often assessed as an antibiotic pollution indicator in natural habitats [37]. The perseverance of SMX in aquatic environments raises antibiotic resistance in microorganisms in the long term [38]. Due to their recalcitrant properties, DCF and SMX, which are slowly biodegradable or non-biodegradable in WWTPs, remain unaffected and persist even after treatment [39]. Owing to their hydrophobic nature, these PhACs are adsorbed to solid particles during treatment processes in WWTPs, resulting in their accumulation in sludge [40,41].

Several studies on the degradation of these PhACs using AOPs have been reported with different yields. Pizzichetti et al. employed photolysis to treat 20 ppm of DCF, and 100% elimination was achieved after 55 min [42]. The same study explored a UV/H₂O₂ process to minimize the treatment time, achieving the same rate in 30 min [42]. Mirzaei et al. [43] adopted g-N-TiO₂ photocatalysis and achieved 85% degradation of SMX after 5 h of treatment. Zhu et al. applied photochemical oxidation to 100 ppb of SMX, and complete degradation was achieved after 23 min of treatment by UV/H₂O₂ [44]. Liu et al. reported 87% removal of SMX after 40 min of a Fenton-like process [45].

In the context of plasma, Banaschik et al. [46] developed a corona plasma reactor with coaxial geometry, reporting 95% degradation of 500 ppb DCF after 33 min of treatment directly in water. Liu et al. [47] devised a bubble column plasma reactor to create large-volume discharges directly in water, and 88% degradation of DCF was achieved after 45 min of treatment. Dobrin et al. [48] found that DCF in solution was completely removed after 15 min treatment of pulsed corona discharge generated above the liquid surface. In addition, according to Wang et al. [49], enhancement with persulfate and activation with pulsed DBD plasma led to 87% degradation of 50 ppm SMX after 30 min of treatment.

Regarding works where plasma jets have been tested for DCF and SMX degradation, they are practically nonexistent. Kumar et al. [50] considered an atmospheric-pressure RF-driven (0.332 MHz) plasma jet of a pin-type electrode to treat 5 mL DCF, and a removal efficiency of 50 ppm was reported after 1 min of treatment. The present work applies a DBD-based plasma jet, driven by a high sinusoidal voltage (10 kHz), for the efficient removal of DCF and SMX from aqueous media. In parallel, the status of the treated aqueous media within the post-treatment period (i.e., after the plasma action) is studied over two days. The results have important implications for the advancement of WWTPs, especially at the phase of activated sludge processing.

2. Pharmaceutical Compounds

Analytical-grade DCF and SMX were procured from Sigma-Aldrich (St. Louis, MO, USA). Molecular formulas and weights, Chemical Abstract Service (CAS) numbers, and other relevant properties are summarized in

Table 1. Chemical structure and properties of the PhACs under investigation.

Structure	Formula	Molecular Weight (g/mol)	pKa	CAS
Diclofenac (DCF)	C14H11C12NO2	296.2	4.1	15307-86-5
Sulfamethoxazole (SMX)	C10H11N3O3S	253.3	5.6–6.7	723-46-6

. Liquid chromatography-grade methanol was purchased from Chem-Lab (Torhout, Belgium). Water was purified by a Milli-Q system (Millipore S.A., Barcelona, Spain). Individual stock solutions were prepared by dissolving each compound in methanol at concentrations of 10 ppm and then stored in amber-colored bottles at –18 °C. From these stock solutions, working standard solutions of lower concentrations were freshly prepared before treatment, by diluting with deionized ultrapure water. The concentration of 50 ppb was preferred due to the detection records of these PhACs in waste [48,51]. For each batch, six (6) replicates of each PhAC at a volume of 10 mL were treated. As a sample container, a 10 mL glass beaker was used. To prevent possible contamination, glasswares were cleaned with methanol prior to use. An additional sample was used as the control, which was exposed to all procedures, except plasma. After each treatment session, 1500 μL of sample was taken with an Eppendorf pipette and stored in glass LC-MS/MS vials for analysis, tightly sealed.

A sample from the control was also stored, accordingly. All vials were transferred right after the plasma treatment to the chemical analysis laboratory in thermos containers. During all chromatographic analysis, vials were stored in the LC-MS/MS sample holder at 10 °C and injections proceeded exactly 1, 3, 24, and 48 h after treatment, after being vortexed. The control sample was also measured in each analysis batch.

3. Plasma Setup

The DBD-based plasma jet reactor employed in this work has been presented extensively and the plasma itself has been studied electrically, optically, and thermally, under various operating conditions, in previous works [52,53]. Its efficiency in liposome [54], bacteria [55], and human skin [56] treatment, as well as in producing plasma-activated water [57,58], has been demonstrated. Here, the same reactor is fed with 2 slm high-purity helium gas (99.999% He; Air Liquide) and driven by a high sinusoidal voltage (10 kHz; 15 kV peak-to-peak; total harmonic distortion < 1%; designed by PlasmaHTec^®, Preveza, Greece).

Briefly, as illustrated in Figure 1, the reactor consists of a capillary alumina tube where a thin tungsten wire is inserted (the driven electrode) and an outer hollow brass cylinder (the grounded electrode). The gas is fed through a mass flow controller (Aalborg Instruments & Controls, Inc., New York, NY, USA) and flows through the dielectric tube. Thus, a coaxial DBD is formed along the “driven wire surface/helium gas/alumina wall/grounded cylinder” configuration. At the same time, a plasma jet projects vertically. In this work, a Pyrex^TM beaker (one at a time) containing a 10 mL specimen is placed on a flat grounded plate and aligned with the plasma jet symmetry axis by a specially designed insulating holder. The distance between the reactor nozzle and the specimen–liquid surface is maintained constant at 14 mm. The plasma jet is thus partially immersed in the liquid and stirs it. In this way, apart from the above-mentioned coaxial DBD, a second vertical DBD is formed along the “driven wire tip/plasma jet channel/liquid volume/beaker wall/grounded plate” configuration.

Figure 2 provides representative waveforms of the driving voltage, the axial DBD current, and the vertical DBD (plasma jet) current, all recorded in situ, i.e., during PhAC treatment. The waveforms were recorded on a digital oscilloscope (LeCroy Ltd., New York, NY, USA; WaveRunner 44Xi-A) by means of a high-voltage probe (Tektronix; P6015A) and a current monitor (Pearson Electronics Inc., Palo Alto, CA, USA; Model 6585). The total mean electric power consumed by both DBDs is in the order of 10 W.

A portable spectrometer (Avantes, Apeldoorn, The Netherlands; ULS4096CL-EVO-UA-10) was employed to explore the wide-scan optical emission spectra of the plasma during PhAC treatment. It is equipped with a 300 grooves/mm grating (200–1100 nm; blaze 300 nm) and a CMOS photodetector (4096 pixels). The light is collected from a zone of about 6 mm downstream of the reactor orifice. It is efficiently led to the spectrometer by means of a fused silica collimating lens (200–2500 nm; confocal length 8.7 mm) and a fused silica optical fiber (200–2500 nm; NA 0.22).

At the same time, the aqueous media temperature is recorded in real time by means of a non-invasive technique, i.e., a GaAs crystal-based optical thermometer (LumaSense Technologies Inc., Santa Clara, CA, USA). The sensor is mounted close to the beaker bottom. The data were acquired by a datalogger, with recommended standard 232 communication and 0.5 s time resolution.

An evaluation of the mean energy yield (Y), in terms of the PhAC quantity (mg) removed per kWh consumed, may be obtained by using the following equation [27]:

$$\mathrm{Y}\left(\mathrm{m}\mathrm{g}/\mathrm{k}\mathrm{W}\mathrm{h}\right)={\displaystyle \frac{{\mathrm{C}}_0{\mathrm{R}}_{\mathrm{e}}\mathrm{V}}{{\mathrm{P}}_{\mathrm{w}}t}}\times {\displaystyle \frac{1}{100}}$$

(1)

where ${\mathrm{C}}_0$ stands for the initial concentration (mg/L), ${\mathrm{R}}_{\mathrm{e}}$ for the removal efficiency (degradation; %), V for the sample volume (L), ${\mathrm{P}}_{\mathrm{w}}$ for the plasma power (kW), and t for the treatment time (h).

4. LC-MS/MS Analysis

The quantification of the PhACs was performed by a High-Performance Liquid Chromatography Triple-Quadrupole Mass Spectrometer (Shimadzu LC-8050; Kyoto, Japan). The separation was conducted using a Shim-pack Scepter C 18-120 column (4.6 × 100 mm; 5 μm) thermostated at 40 °C. The injection volume was set at 10 μL and the flow rate was 0.2 mL/min. The mobile phase of DCF and SMX was prepared by using ultrapure water with 0.1% Formic Acid and methanol, with isocratic elution at a ratio of 60:40 (v:v) and 90:10 (v:v), respectively. In MS analysis, the ionization was operated with electrospray ionization (ESI) in positive ion mode under multiple reaction monitoring (MRM), detecting the transitions for identification and quantification. Precursor and product ion masses, along with optimized MRM conditions, are shown in

Table 2. Mass spectrometer parameters used to identify and quantify PhACs.

Compound Name	Precursor Ion (m/z)	Product Ion (m/z)	Collision Energy (V)	Retention Time (min)	Repeatability (RSD %)
DCF	296.00	214.20 (quantifier)	−12	9.192	5.55
		249.90	−31	9.191	6.07
		156.07	−15	6.071	4.87
SMX	254.06	108.04	−23	6.074	6.19
		91.90 (quantifier)	−27	6.073	2.5

. For quantitative analysis, the transitions with high intensity are reliable to consider, while the other transitions are used for confirmation of the pharmaceuticals. Figure 3 captures exemplary chromatograms obtained of DCF and SMX via LabSolutions software.

Degradation efficiency is quantified by calculating the spectrum area under the considered peak and correlating it with the corresponding area of the spectrum of untreated samples as follows:

$$\mathrm{D}\mathrm{e}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\left(\mathrm{\%}\right)={\displaystyle \frac{{(\mathrm{A}\mathrm{r}\mathrm{e}\mathrm{a}\_\mathrm{P}\mathrm{e}\mathrm{a}\mathrm{k})}_0-{(\mathrm{A}\mathrm{r}\mathrm{e}\mathrm{a}\_\mathrm{P}\mathrm{e}\mathrm{a}\mathrm{k})}_{\mathsf{\tau }}}{{(\mathrm{A}\mathrm{r}\mathrm{e}\mathrm{a}\_\mathrm{P}\mathrm{e}\mathrm{a}\mathrm{k})}_0}}\times 100$$

(2)

where ${(\mathrm{A}\mathrm{r}\mathrm{e}\mathrm{a}\_\mathrm{P}\mathrm{e}\mathrm{a}\mathrm{k})}_0$ and ${(\mathrm{A}\mathrm{r}\mathrm{e}\mathrm{a}\_\mathrm{P}\mathrm{e}\mathrm{a}\mathrm{k})}_{\mathrm{t}}$ are the areas of the spectrum under the considered peak for untreated samples (t = 0) and the sample treated over a time of τ (t = τ), respectively.

5. Results

Figure 4 summarizes the results obtained after the treatment of the two pharmaceutical compounds for different plasma exposure durations. Namely, Figure 4a shows the dynamic evolution of DCF over the first 20 min of the treatment, unveiling two different reaction rates, i.e., a relatively fast one during the first 10 min and a relatively slower one during the last 10 min. Notably, complete degradation of the DCF compound (50 ppb) is implied due to the reaction of the aqueous medium with the plasma jet species. Similar dynamic behavior is revealed for the SMX compound (Figure 4b). However, in this case, both degradation rates are lower, while the degradation is not full within the considered time interval, i.e., 20 min of treatment. Despite that, a mean removal of about 80 ± 10% is clearly achieved for the SMX compound (50 ppb).

It is underlined that the above-mentioned results refer to analyses conducted at the time point of exactly 1 h after the plasma treatment for each of the durations under consideration (5, 10, 15, and 20 min). This time of 1 h is necessary due to the sample transport, handling, and insertion into the analysis system. It is thus meaningful to call this interval of 1 h the post-treatment time. Therefore, the stability of the treatment and the potential effect of the post-treatment time on the samples (due, e.g., to the accumulative action of the plasma-induced species) are studied systematically. Post-treatment times up to 48 h are considered for both PhACs and for all the durations of the actual plasma action. The data are plotted in Figure 5.

According to the mean values plotted in Figure 5, one could judge that the species induced by the plasma in the aqueous medium may continue to degrade the PhACs (see 5 min case). A speculative mechanism of the potential PhAC–plasma interaction, with respect to the degradation of the compounds, points to the activation of the aqueous medium where PhACs are diluted, due to the formation of long-lived radicals. However, this post-treatment action of the plasma becomes debatable when the data deviations among different series of experiments (six in our case) are considered. Thus, the precedent curves of Figure 4 present a reliable profile of the PhAC degradation dynamics, even if the samples are analyzed 1 h after the plasma action.

Apart from the high degree of degradation achieved by means of the plasma jet, another promising feature of this process is the limited increase in the temperature of the aqueous media. Figure 6 presents representative traces of the temperature increase during the treatment of both PhACs, recorded in real time. A monotonous rise over the first 5 min, approximately, is followed by a quasi-saturation phase. This saturation corresponds to a value of about 45 °C. Thus, it is rational to claim that any degradation shown in Figure 4 and Figure 5 can be attributed to chemically rather thermally triggered reactions. This claim is further supported if one considers the reactive species that are formed within the plasma/liquid interface. Consequently, Figure 7 demonstrates the notable modification of the wide-scan optical emission spectrum of the free-running plasma jet when the aqueous media are placed downstream of the plasma jet.

Figure 7a presents a typical pattern of the optical emission spectrum when the plasma jet runs free, i.e., it does not impinge on the liquid target. On the other hand, Figure 7b presents the corresponding spectrum when the aqueous medium is placed downstream of the plasma jet. In this figure, by conserving the same vertical scale of OES intensity, the noteworthy enrichment of the plasma/liquid interface in reactive species and radicals is unveiled. Not only new species are detectable, but the intensity of most species also increases considerably with respect to the free-running plasma case (compare Figure 7a,c), in agreement with the results of other groups in analogous cases [59]. In Figure 7c, the identity of the detected species is marked, including excited neutrals of nitrogen, positive ions of nitrogen, hydroxyl, nitric oxide, atomic oxygen, helium atoms, and hydrogen atoms. The species detailed identification is given in

Table A1. Identification of the species shown in the spectra of Figure 7.

Species	Wavelength (nm)	Vibrational Transition v′ − v″ (Δv)
${\mathrm{N}\mathrm{O}}_{\mathsf{\gamma }}:{\mathrm{A}}^2{\mathsf{\Sigma }}^{+}-{\mathsf{{\rm X}}}^2\mathsf{\Pi }$	215.49	1-0 (+1)
	226.94	0-0 (0)
	237.02	0-1 (−1)
	247.87	0-2 (−2)
	259.57	0-3 (−3)
	272.22	0-4 (−4)
$\mathsf{O}\mathsf{H}:{\mathrm{A}}^2{\mathsf{\Sigma }}^{+}-{\mathsf{{\rm X}}}^2\mathsf{\Pi }$	282.90	1-0 (+1)
	289.27	2-1 (+1)
	296.24	3-2 (+1)
	308.9	0-0 (0)
${\mathsf{{\rm N}}}_2\left(\mathrm{S}\mathrm{P}\mathrm{S}\right):{\mathrm{C}}_3{\mathsf{\Pi }}_{\mathrm{u}}-{\mathrm{B}}_3{\mathsf{\Pi }}_{\mathrm{g}}$	296.20	3-1 (+2)
	297.68	2-0 (+2)
	315.93	1-0 (+1)
	326.81	4-4 (0)
	328.53	3-3 (0)
	330.90	2-2 (0)
	337.17	0-0 (0)
	350.05	2-3 (−1)
	353.67	1-2 (−1)
	367.19	3-5 (−2)
	371.05	2-4 (−2)
	375.54	1-3 (−2)
	380.49	0-2 (−2)
	394.3	2-5 (−3)
	399.84	1-4 (−3)
	405.94	0-3 (−3)
	409.48	4-8 (−4)
	414.18	3-7 (−4)
	420.05	2-6 (−4)
	426.97	1-5 (−4)
	434.36	0-4 (-4)
	441.67	3-8 (−5)
	449.02	2-7 (−5)
	457.43	1-6 (−5)
	464.94	4-10 (−6)
	481.47	2-8 (−6)
	491.68	1-7 (−6)
${\mathsf{{\rm N}}}_2\left(\mathrm{F}\mathrm{P}\mathrm{S}\right):{\mathrm{B}}^3{\mathsf{\Pi }}_{\mathrm{g}}-{\mathrm{A}}^3{\mathsf{\Sigma }}_{\mathrm{u}}^{+}$	559.29	6-1 (+5)
	575.52	12-8 (+4)
	601.36	7-3 (+4)
	606.97	6-2 (+4)
	639.47	9-6 (+3)
	646.85	8-5 (+3)
	654.48	7-4 (+3)
	662.36	6-3 (+3)
	678.86	4-1 (+3)
	687.50	3-0 (+3)
	727.33	6-4 (+2)
	738.66	5-3 (+2)
	750.39	4-2 (+2)
	762.62	3-1 (+2)
	775.32	2-0 (+2)
	789.64	7-6 (+1)
	804.74	6-5 (+1)
	854.18	3-2 (+1)
	872.23	2-1 (+1)
	891.19	1-0 (+1)
${\mathrm{N}}_2^{+}\left(\mathrm{F}\mathrm{N}\mathrm{S}\right):{\mathsf{{\rm B}}}^2{\mathsf{\Sigma }}_{\mathrm{u}}^{+}{-\mathrm{X}}^2{\mathsf{\Sigma }}_{\mathrm{g}}^{+}$	385.79	2-2 (0)
	388.43	1-1 (0)
	391.44	0-0 (0)
	427.81	0-1 (−1)
	470.92	0-2 (−2)
${\mathrm{H}}_{\mathsf{\beta }}:\mathrm{n}=4-\mathrm{n}=2$	486.1	$-$
${\mathrm{H}}_{\mathsf{\alpha }}:\mathrm{n}=3-\mathrm{n}=2$	656.3	$-$
$\mathrm{H}\mathrm{e} \mathrm{I}$	447.15	$-$
	492.19	$-$
	501.57	$-$
	587.56	$-$
	667.82	$-$
	706.57	$-$
	728.13	$-$
O I	777	$-$
	844.6	$-$
	926	$-$

(Appendix A). Molecular transitions are identified according to [60] and atomic transitions according to [61].

6. Discussion

It becomes clear that the advanced oxidation process proposed here for the degradation of pharmaceutically active compounds is promising in terms of degradation efficiency, and it is related to the activation of the aqueous media by the plasma. Plasma-activated water has been extensively produced and studied in the past [62]. Based on such results, including former ones obtained by the present authors [57,58], it is known that aqueous media are chemically activated due to the interactions between the plasma-induced reactive species (charged and neutrals). At the same time, the medium’s temperature remains low. Chemical activation refers to the formation of hydrogen peroxide, superoxide, hydroxyl, singlet oxygen, nitrite, nitrate, etc., with a consequent pH decrease and redox potential increase.

Thus, the activated medium can initiate and propagate chain reactions, leading to the rapid and efficient degradation of PhACs. Radicals such as hydroxyl (•OH) can attack and break down their molecular structures through a series of oxidative reactions. Secondary radicals generated by primary radicals can interact with the water molecules and even with the PhACs themselves, leading to further oxidation reactions. Thus, the interaction of radicals with the target compounds forms transformation products. Continuous exposure to radicals can further degrade these intermediates, potentially leading to complete mineralization [49], which is pivotal for reducing the toxicity of micropollutants in the environment. The role of hydrated electrons, produced in aqueous samples through the interaction of high-energy electrons from the helium plasma with water molecules, should be mentioned separately because they are among the strongest reductants known. They can act synergistically with other reactive species such as hydrogen radicals, improving the breakdown of contaminants [63,64].

There are multiple respected studies in the literature dedicated exclusively to the intermediates produced during PhAC degradation, though the identification and quantification of transformation products are not in the framework of this study. Nevertheless, Figure 8 illustrates potential transformational pathways of DCF and SMX based on literature findings and the knowledge obtained from our previous studies. Based on the literature [34,48,49,50,65], degradation occurs due to the mechanisms of dehydration, hydroxylation, dechlorination, and decarboxylation, leading to ring opening or possible C–N, S–N, and C–S bond cleavage, creating transformation products.

Specifically, in the case of DCF (Figure 8a), the initial reaction with HO• creates two possible pathways, either H-abstraction or the electrophilic addition of HO• in the electron-donating position of the aromatic ring. From the above-mentioned pathways, the production of “A” (C₁₄H₉Cl₂NO₃, m/z: 308.0) [50] is expected. From that point, decarboxylation can occur, forming “B” (C₁₄H₁₅ClNO₂, m/z: 264.0). The chemical formation of “G” (C₁₄H₉Cl₂NO₃, m/z: 308.0) results from possible cyclization followed by water elimination of the parent compound [50]. An alternative pathway is the reductive degradation of DCF due to the existence of hydrated electrons, along with hydrogen radicals, leading to transformation products “E” (C₁₃H₉ClNO₂) and “F” (C₁₄H₉Cl₂NO₃, m/z: 260.0) [63]. Since hydroxylation and dechlorination are two of the major transformation routes, “C” (C₁₄H₁₂ClNO₃, m/z: 276.0) and “D” (C₁₄H₁₃NO₄) are frequently encountered [25,63]. Subsequently, under further cleavage and the formation of smaller molecules, degradation continues in the background. This tentative possible degradation pathway of DCF leads to the final step of mineralization to CO₂, (COOH)₂, and water [25].

As for SMX (Figure 8b), four mechanisms of degradation are proposed, i.e., hydroxylation, oxidation, and subsequent C-N bond cleavage (“D” (C₆H₈N₂O₂S, m/z: 172.2) and “E” (C₄H₇N₂O, m/z: 98.1)), or N-S bond cleavage (“C” (C₆H₈NO₂S, m/z: 158.0) and “D”) [66,67], with the formation of either inorganic ions (namely ammonium, nitrate and sulfate ions) or mineralization products (like CO₂) and water deriving from ring-opening reactions. Other studies have identified hydroxylation derivatives of “A” (C₁₀H₁₁N₃O₄S, m/z: 270.0) and “B” (C₁₀H₁₁N₃O₄S, m/z: 269.3) [66,68] as the major formed products. The transformation product assigned to the nitro-derivative “F” (C₁₀H₁₀N₃O₅S, m/z: 284.0) is a result of simultaneous hydroxylation and N-oxidation mechanisms [69]. It is likely that there are a plethora of other by-products in different concentrations that have resulted from both oxidative and reductive mechanisms, and they may be investigated in dedicated research in the future.

Finally, Equation (1) gives an approximate value of the process yield. By setting C₀ = 5 × 10⁻² mg/L, V = 10 × 10⁻³ L (ignoring any evaporation during treatment), P_W = 10 × 10⁻³ kW, t = 1/12 to 1/3 h, and R_e = mean values of the data in Figure 4a (DCF) and Figure 4b (SMX), the corresponding values of Figure 9 are found.

Following these plots, the mean energy yield of the present process is a decreasing function of the degradation efficiency, since the degradation rate approaches saturation (Figure 4) while the mean power delivered to the plasma remains practically constant. This tendency agrees with previously published results [50]. On the other hand, the yield values are much lower as compared to those obtained with other plasma setups; a difference of up to three orders of magnitude may be found in other reports [48,50,70] for much denser aqueous media, although a direct comparison among various configurations and operating conditions is not straightforward. However, it is speculated that the values of Figure 9 do not mirror the maximal potential efficiency of the present process, due to the very low PhAC concentrations considered here and the complex dependence between the removal/energy efficiency and the initial concentration [48,50]. This speculation remains to be tested by further experiments with much denser aqueous media.

7. Conclusions

To recap, in the present study, a cold plasma helium jet served as a precursor-free advanced oxidation process for the degradation of DCF and SMX in aqueous solutions of low concentration (5 × 10⁻² mg/L). The results indicated a high degradation degree for both compounds (up to 100% and 80%, respectively), under the present discharge regime and experimental conditions (10 mL volume solution, 2 slm He gas flow, 14 mm working distance, 10 W mean power, 20 min treatment time). Overall, it is proposed that the plasma-induced reactive species (charged and neutral) alter the chemical composition of the solutions and PhAC degradation takes place progressively. The role of temperature in the process efficiency was proven to be comparatively negligible. A nonreciprocal relationship between the energy yield and the degradation degree was clearly unveiled. Nonetheless, the former was found to be significantly low in terms of mg/kWh, and denser PhAC solutions must be tested.