1. Introduction
In the ever-evolving landscape of environmental preservation and pollution control, scientists are continuously exploring innovative technologies to address emerging challenges. One such imperative concern is the removal of pharmaceutical compounds from wastewater. Recent studies reported that pharmaceutical substances such as antibiotics are refractory to biological treatment and have accumulation tendencies in humans and animals through food chain, presenting an ecological and public health hazard [1].
Tetracycline (TC) is among the extensively used antibiotics worldwide in human and veterinary medicine, livestock, and aquaculture to control bacterial infections and promote growth [2]. TC’s complex molecular structure makes it an effective antibiotic by enabling it to bind to the ribosome of bacteria, thereby inhibiting protein synthesis [4]. This selective binding prevents RNA-ribosome attachment, halting peptide chain growth. The structure also allows for efficient penetration of bacterial cells and resistance to degradation, ensuring broad-spectrum activity against various bacteria [4]. TC is an efficient drug thanks to its complexity and stability; however, the over-consumption or/and improper disposal of this antibiotic results in their release into soil and aquatic environment becoming a serious threat [5].
The occurrence of TC in the ecosystems promotes the emergence and dissemination of antibiotic resistance genes in bacteria, disrupting drinking and irrigation water resources, and endangering human and animals’ health [6]. Several technologies such as membrane filtration [7], electrolysis [8], adsorption [9], coagulation-flocculation, flotation [10], and photocatalysis [11] were developed to remove TC and other antibiotics from wastewater. These conventional methods are disadvantageous due to low removal efficiency, high expenses, and the generation of secondary pollution [12]. Hence, the current necessity to find economical and high-efficient technology to remove these antibiotics.
Atmospheric Non-Thermal Plasma (NTP) technology is considered as an attractive, sustainable, and eco-friendly oxidation process for the treatment of organic contaminants. Recently, different types of NTP techniques, namely Pulsed corona discharge (PCD), Dielectric Barrier Discharge (DBD), and Gliding Arc Discharge (GAD) have been used for antibiotics degradation, leading to considerable removal efficiency [13,14]. However, while NTPs have many advantages, they also have some limitations, including energy consumption and the formation of byproducts of unknown toxicity [15,16]. To combat this issue, the optimization of plasma technology performance enhanced with catalysis or coupled with other oxidation processes have emerged as a promising avenue for increasing the efficiency of contaminants removal. Studies have reported that the simultaneous activation of catalysts with reactive species produced by plasma such as hydrogen peroxide (H2O2) and radical hydroxyl (•OH) will exert a synergetic effect on the degradation process to reduce the drawbacks of the plasma treatment alone. Farzinfar et al. [17] found that the combination of DBD plasma with classic photocatalysts TiO2 and ZnO nanoparticles significantly enhanced p-nitrophenol degradation due to photocatalysts activation induced by the UV light emitted in the plasma discharge. The results elucidated that the addition of an optimized quantity of ZnO nanoparticles (250 mg/L) in the plasma reactor could increase the degradation of p-nitrophenol from 49% to 91%. Luu et al, [18] investigated the enhancement of NO degradation in visible light through plasma-treated photocatalytic substrates featuring TiO2@g-C3N4 Z-scheme structure, these nanoparticles exhibit remarkable stability with visible light irradiation. The results show substantial efficiency enhancements, reaching 84.04%. Lou et al, [19] applied DBD plasma coupled with biochar to remove tetracycline hydrochloride. This system showed a better removal performance, shortened treatment time, and an improved energy yield compared to the plasma treatment alone. Hirami et al. [20] reported that plasma-meditated Fenton process can effectively degrade ibuprofen in aqueous solution where a pulsed nonthermal plasma reactor was optimized to achieve faster degradation and high mineralization. The removal of TC by different NTPs has also been investigated in the current literature. For example, Fang et al. presented a unique approach generating plasma by the DBD discharge as the gas was re-circulated into the solution. Multiple treatment cycles were conducted to increase the removal rate and reduce the biotoxicity of the residual products [21]. Moreover, GAD plasma was employed for the removal of TC treated under different feed gases in our previous work that showed successful degradation and TOC abatement especially in oxygen plasma [22]. Although these studies have proven the efficiency of plasma processes with or without catalysis for the elimination of TC and other antibiotics, little attention has been paid to the coupling of different types of plasma with different types of oxidants or catalysts. Ferrate oxidation for instance is a promising green technology capable of degrading recalcitrant organic compounds [23,24] known for its high valence and strong oxidizing power. However, it is reduced quickly to Fe3+ via self-decomposition and it usually requires a high amount to be able to attack heavy organic matter [25,26].
Theoretically, ferrate is considered a great candidate in a plasma hybrid process for many reasons: it reacts quickly with the target compound then totally decomposes to an ion that can transform to Fe2+ species [27]. On the other hand, plasmas are known for their ability to decrease the solution pH and the production of H2O2 creating the perfect conditions for Fenton reaction to take place [20]. UV radiation can also contribute to further improving the removal rates during plasma discharge. On this basis, we have hypothesized that the plasma-ferrate hybrid system could improve the degradation performance and show a better synergetic effect compared to the direct addition of ferric and ferrous ions for the purpose of inducing Fenton reaction. The novelty of this research lies in the innovative combination of GAD plasma with ferrate oxidation, the optimization and comparison with traditional methods, the detailed analysis of degradation pathways and the assessment of byproduct toxicity. Additionally, and to the best of our knowledge, there are no literature reports available on the removal of TC by GAD plasma combined with ferrate oxidation. Therefore, we aim to explore the degradation efficiency of TC using a homogenous catalysis-based plasma by the combination of GAD with Ferrate Fe(VI). And in order to investigate its potential compared to traditional Fenton reagents, we investigated GAD plasma coupled with Fe(II) and Fe(III). The optimization of the plasma reactor and the quantification of reactive oxygen and nitrogen species RONS such as •OH, H2O2, O3, nitrite and nitrate have been studied in previous works [22,28]. Therefore, we shed light on the combination parameters only.
In this study, we focused on the degradation and mineralization of TC under the hybrid systems, the effects of TC initial concentration and catalyst doses, and the effect of radical scavengers on the removal rates. Furthermore, we identified the degradation products by Liquid Chromatography-Mass Spectrometry LC-MS-MS analysis, and we proposed TC degradation pathway. Finally, we evaluated the toxicities of the intermediates in various mediums by prediction software. The findings of this study will not only provide insights into the novel approach of plasma-ferrate combination but also serve as a valuable reference point for future research about wastewater containing antibiotics generally and TC specifically.
2. Materials and Methods
2.1. Chemical Analysis
TC quantification was performed on a High-Performance Liquid Chromatography (HPLC) equipped with a diode array detector (Shimadzu, Japan). The mobile phase used was a mixture of acetonitrile and 0.05% phosphoric acid buffer (90:10, v/v) kept on isocratic mode at a flow rate of 1.0 mL/min and oven temperature of 40°C. All samples and solvents were filtered through a 0.45 um PTFE filter prior to use. TC peak detection was set peaks of 358 nm. Retention time and injection volume were 10 min and 20 μL, respectively. Fresh TC solutions were prepared at different concentrations right before experiments and kept from direct light to avoid the photodegradation of TC molecule. The degradation intermediates analyzed by liquid chromatography with tandem mass spectrometry (LS/MS/MS) system UPLC/Xevo TQ-S micro (Waters, USA) equipped with column C18 Column (2.1 x 100 mm, 1.7 μm). The mass spectrum was scanned in the range of 100 to 600 m/z. All operating conditions are stated in supplementary materials Table S1. Shimadzu Total Organic Carbon (TOC-L) analyzer (Japan) was used under NPOC method for the determination of TC mineralization throughout plasma treatments. All experiments have been conducted at least three times. Detailed information on the reagents used in the present work are listed in supplementary materials Text S1.
2.2. GAD Plasma Reactor
A laboratory-scale GAD reactor system illustrated in Fig. 1. was constructed for this study. The reactor comprised a 400 mL cylindrical glass vessel and two stainless steel electrodes, each 4 mm thick and attached and positioned divergently on the vessel lid. These electrodes were connected to an AC neo-transformer (DAEHANTRANS CO, LTD), which supplied an electric voltage of 15kV. During the electric discharge process, an arc was generated at the narrowest gap (approximately 2 mm) between the electrodes. This arc was then propelled vertically towards the liquid interface by the flow of gas through a 1.5 mm diameter nozzle, operating at atmospheric pressure. The gas flow facilitated the movement of the arc along the electrodes, causing an extension of the arc’s length until it dispersed into a plasma plume. More details about this reactor can be found in our previous publication [28].
2.3. Potassium Ferrate Synthesis
Potassium Ferrate utilized in this work was prepared through wet oxidation method as described in [29,30]. 30 g of KOH was added to 100 mL of NaOCl solution and stirred until completely dissolved. Then, the solution is cooled and filtered with GF/C filter to remove NaCl particles. The filtration process is repeated three to four times to ensure the complete removal of NaCl particles. Next, the filtered solution should be kept in the refrigerator to maintain cold temperature. Next step is to add 11.1 g of Fe(NO3)3·9H2O to the solution and stir for 40 minutes to synthesize FeO42−. Up until this stage it is still in the liquid state. In order to transform it to solid 16.7 g of KOH is added while continuously stirring for 40 minutes more, then stored for about 30 minutes. After cooling down, the solution is passed through a G4 glass filter (pore size: 10–16 μm), the filtrate is immediately administered to a saturated KOH solution (11M) and refrigerated again. A repeated step of filtration with GF/C filter paper is necessary before Ferrate (VI) crystals are obtained [30]. The associated chemical reactions are listed in Eq. (1)., Eq. (2)., and Eq. (3):
The Fe(VI) particles that remain on the filter paper are washed with a mix of different solvents namely 25 ml of n-hexane, 10 ml methanol, and 10 ml diethyl ether four times. The particles are then collected and put in a vacuum dryer chamber. The purity of the final Ferrate obtained was measured with a DR 5000 UV-spectrometer (HACH) using spectrophotometric analysis. The experiments were carried out when the purity achieved was above 93%. Specific information about purity measurement can be found in supplementary materials Text S2.
3. Results and Discussions
3.1. Influence of Catalysts and Combination Process on Plasma Performance
To investigate the effect of Fe(II), Fe(III) and Fe(VI) on GAD plasma treatment of TC experimental tests were conducted both with and without the catalysts. The experimental parameters were as follows: power supply of 15 kV, air as the gas feed, and air flow of 13 L/min. The results of TC degradation (%) and mineralization (% TOC removed) are presented in Fig. 2. The results show that the impact of catalyst presence was significant, as the hybrid processes exhibited improved results compared to plasma treatment alone. In the absence of catalysts, and after 30 minutes of reaction, the degradation and mineralization of TC only reached 30.75% and 25.67%, respectively. However, when catalysts were used, all the degradation and mineralization values increased. Particularly noteworthy were the combinations of plasma with Fe(VI) and Fe(II). Total removal of TC was achieved by the plasma-Fe(VI) system, and 97.28% removal rate was constated for the plasma-Fe(II) system by the end of the treatment, demonstrating an improvement of nearly 70.45% proving the synergistic potential of the combination processes. In the plasma-Fe(III) system, there was a modest enhancement in degradation of 10.23% and in mineralization of 3.67%. These results can be explained by the oxidation-reduction occurring between ROS generated through plasma and Fenton-like reactions taking place in TC solution thanks to H2O2 presence and the acidification of plasma treated solution (pH solution reaches 3 after 15 minutes of plasma discharge).
The main reactions in plasma-Fe(II) and plasma-Fe(III) systems that are responsible for the enhancement of degradation and mineralization of TC are stated in the equations blow:
The reaction presented in Eq.(6). is similar to the process taking place in Fenton and electro-Fenton processes [31]. The in-situ generation of an additional amount of •OH through the oxidation of Fe ions by H2O2 produced by GAD plasma will only add more oxidation power in the aqueous solutions. Consequently, •OH will react directly with TC in water during plasma/catalysis treatment and also with the generated by-products leading to an increase of mineralization. Thus, a decrease in the content of organic carbon is expected in the presence of different catalysts. Song et al. investigated Enrofloxacin drug degradation by DBD plasma in presence of peroxydisulfate catalyst. The results show that TOC removal rates improved by double after 25 minutes of treatment compared to DBD treatment alone [32]. Similar results were reported by Hirami et al. where a combination of plasma DBD system with Fe ions improved the degradation significantly. They constated an enhancement in degradation from 75.9% in plasma alone to 88.3% when combined with an iron-based catalyst. As for mineralization, TOC decreased from 32.5 mg/L to reach 15.7 mg/L after 6 hours of reaction in the plasma-Fe based system [20]. The plasma-Fe(VI) process seems to be the most effective hybrid system according to the results obtained in this study. However, it should be noted that Fe(VI) has a strong oxidative power on its own and that in plasma-Fe(VI) system it is more of a hybrid of two AOPs than just a plasma catalysis process. The effect of catalysts alone on TC removal is investigated in the next section.
3.2. Effect of Catalysts on TC Without Plasma Discharge
It is crucial to explore the degradation potential of the catalysts studied in the work in order to set the stage for talking about the mechanism and effects of plasma catalysis. Initially, we examined the impact of Fe(II), Fe(III), and Fe(VI) on the removal of TC of 50 mg/L concentration and a catalyst dose set at 50 mg/L, without the interference of plasma. TC solutions in these experiments were introduced in the same plasma reactor without turning the power on nor the feed gas, just stirring. TC samples were taken for HPLC analysis after each time of treatment. Results are presented in Fig. S1. As shown in the figure, unlike Fe(VI), the other two catalysts alone were unable to oxidize TC, resulting in a zero-removal rate even after 30 minutes of reaction. Fe(VI), being a strong oxidant, achieved up to 60.84% TC removal within the first 5 minutes. Beyond this point, no further change was observed, as Fe(VI) loses its oxidation power and decomposes after reacting with all available organic matter molecules [25].
3.3. Effect of TC Initial Concentration
Fig. 3 illustrates the effect of various TC concentrations (20, 50, and 100 mg/L) on the degradation of TC in sole NTP and plasma-catalysis systems ([catalyst]= 50 mg/L). The results show that in all systems lower initial concentration is in favor of more efficient removal, achieving higher degradation rates at shorter reaction time. Plasma-Fe(VI) unsurprisingly topped the results by achieving complete degradation for all TC initial concentrations, only the treatment time varies (5, 15, 20 minutes of reaction for TC initial concentration of 20, 50, 100 mg/L, respectively). The second most efficient system was plasma-Fe(II) with degradation rates of 97.89%, 92.55%, 85.63% for 20, 50, 100 mg/L respectively after 30 minutes of reaction time. The plasma-Fe(III) system has shown a weaker removal potential reaching 70.25% for the smallest TC initial concentration. The main reason behind the observed results could be attributed to the fact that under fixed conditions of applied voltage and catalysts dosage, a limited amount of ROS is produced that is capable of attacking the heavy organic structure of TC. With a high initial concentration, the amount of organic matter outnumbers ROS resulting in decreasing degradation rates. Furthermore, as the mineralization rate increases, the formation of degradation intermediates that compete with TC molecules to react with the active radicles will take place. However, in the presence of catalysts, the removal efficiency improves the decomposition of TC due to the oxidizing effect of Fe(VI) in the case of plasma-Fe(VI) system and the production of additional amount of radical species in the solution by all three systems.
3.4. Effect of Catalyst Dose
The degradation efficiencies of TC after 30 min of plasma treatment with catalysts at different concentrations were compared with that of sole plasma and the results are shown in Fig. 4. For these experiments, initial TC concentration was set at 50 mg/L, and the voltage at 15 kV. TC degradation was significantly higher as the catalysts dose increased. Notably, 30 mg/L of Fe(VI) coupled with plasma achieved total degradation while a concentration of 100 mg/L of Fe(II) lead to the same results. This showcases the ability of Fe(VI) to reach full decomposition at smaller dose and shorter time implying that the physical and chemical effects of NTP process are more efficient for TC removal in combination with Fe(VI). Fe(III) coupled with plasma showed a timid elimination capacity even at high dose of 100 mg/L (80.42%). In the field of wastewater treatment, it is more interesting to use lower dosages of reagents and shortened treatment times to cut costs. The plasma-Fe(VI) hybrid system had shown a huge enhancement of degradation rate even at lowest dose and shortest reaction time. These results demonstrate the potential benefits of using plasma-Fe(VI) coupling in water treatment.
3.5. Role of Reactive Species in TC Removal
The quantification of RONS generated in this work have been studied thoroughly in our previous publications [22,28]. To evaluate the contribution of various kinds of reactive species on TC degradation, we investigated the effect of chemical scavengers, including Dimethyl Sulfoxide (DMSO) and tert-butyl (TBA) (used as •OH scavengers), p-benzoquinone (p-BQ, as a superoxide O2 •− scavenger) and salicylic acid (as free electrons scavenger) on the removal efficiency in plasma alone and plasma-catalysts systems. Since •OH is the most important reactive species in plasma processes, we opted to use different scavengers to assess its role in TC oxidation. Both DMSO and TBA are extremely reactive with •OH and induce a strong inhibition effect [33,34]. p-BQ was selected as scavenger for superoxide anion radical (O2 •−) (k = 0.9–1.0 × 109 M−1 s−1). Due to the importance of free electrons in the solution on the major plasma interactions, we added salicylic acid as a scavenger for free electrons [35]. The results of DMSO effect on TC degradation are shown in Fig. 5. A decrease in TC degradation efficiency was observed in the presence of DMSO for all systems. The degradation rates decreased from 20.69% to 9.84% in plasma alone, meaning that a 10% of degradation potential was inhibited by DMSO acting as a radical scavenger. However, during plasma catalysis application, it is observed that the gap in this decrease is reduced to 4% and 20% in plasma/Fe(II) and plasma/Fe(III) systems respectively, indicating that despite the presence of radical inhibitors, the catalyst-plasma combination process will achieve higher degradation rates compared to plasma treatment alone. plasma/Fe(VI) system in the other hand, have exhibited an opposite outcome in the presence of DMSO compared to other systems. The degradation efficiency in presence of DMSO was actually higher than its absence. This can be contributed to the complex nature of the treated solution. As Fe(VI) decomposition in the solution produces more •OH their consumption by DMSO seems to have little effect on the degradation of TC in this case. Chen et al. also have reported that DMSO is degraded by the Fe based oxidation, further concluding that the presence of DMSO led to higher TC removal rate [36]. In the presence of Fe(VI) and Fe(III)– which is the decomposition product of ferrate–more •OH will be produced throughout the treatment. The accumulated concentration of these radicals can be quickly consumed by all the different species present in the aqueous solution [37].
The effect of TBA, p-BQ, and salicylic acid on TC removal was investigated and presented in Fig. 6 and Fig. S2. The experiments were carried out using an initial TC concentration of 50 mg/L and optimal operational conditions. Each test was conducted separately with the addition of only one scavenger. The concentration of each scavenger was set as 100 mM. The results indicate that TBA effect was the most noticeable as the removal rate dropped from 30% to 10% in presence of TBA, further solidifying the role of •OH in the plasma degradation. Superoxide radicals also managed to decrease the removal efficiency in their presence indicating that indeed it engages with TC oxidizing reactions. The figure also shows that electrons are involved in this oxidation since the removal rate dropped to 20% after 30 minutes of treatment. According to the results, all three scavengers cause a decrease in TC removal efficiency. The reduction in TC residual concentration was influenced the following order: salicylic acid < p-BQ < TBA, implying that •OH had the highest contribution to overall TC decomposition followed by superoxide and then free electrons. By the end of treatment, TBA, p-BQ and salicylic acid reduced the TC removal from 31.11% to 9.8%, 15.93%, and 22.78%, respectively. The comparison of these findings reveals that •OH and superoxide have the most significant role in the degradation of TC as their scavengers showed the most inhibition in removal efficiency compared to electron scavenger.
3.6. Degradation Products and Proposed Pathway
To determine the degradation mechanism of TC under plasma-catalysts systems, the degradation products were identified and detected using LC-MS-MS analysis on samples treated for 15 minutes in plasma-Fe(II), plasma-Fe(III), and plasma-Fe(VI) systems. The obtained chromatograms are presented in Fig. S3. Most intermediates were detected in all three combination systems. Therefore, the degradation pathway is general for all the systems. TC structure possesses four aromatic rings that possess a number of different functional groups, including hydroxyl, amino, double bond, and amide groups, and the phenolic hydroxyl and enol groups are attached to them. These groups are acidic and can be easily degraded under GAD plasma [38]. The major decomposition reactions that take place on TC molecules are: hydroxylation, dehydroxylation, dehydration, deamidization, demethylation, deamination, and ring opening [38,39]. The bond cleavage near the position of dense electrons in the molecular structure is reported to cause the ring opening [40,41]. Based on the literature and the results of LC-MS-MS analyses, a possible degradation pathway of TC is proposed in Fig. 7. The decomposition followed three pathways three pathways. First, P-461, P-431, P-459 are formed after demethylation and hydroxylation reaction. P-461 would experience a serious consecutive hydroxylation to form P-477, P-453, and P-413. Secondly, since amino and methyl groups possess electronegative charge compared to other groups, they tend to be more vulnerable to reactive species attacks, leading to the formation of P-431 and P-400 consecutively. The rest of the by-products formed after multiple reactions (loss of amino, amid, and alkyl groups), eventually leading to ring opening and transformation to smaller organic compounds. At last, these intermediates will get further oxidized until they reach mineralization [42]. These findings were in agreement with previous researchers’ findings that provided an insight into the transformation mechanism of TC during the catalytic process [29,31,43].
3.7. Prediction of Degradation Products Toxicities
Toxicities of TC degradation intermediates are indicated by five parameters: toxicity development, mutagenicity, Daphnia magna LC50 (48hr), Oral rat LD50, and bioconcentration factor in aquatic systems. It was evaluated by the Toxicity Evaluation Software Tool (T.E.S.T) program. This software, developed by the United States Environmental Protection Agency (U.S. EPA), can estimate the toxicity values and physical properties of organic compounds based on their molecular structure and has been used widely for evaluating the toxicity of degradation intermediates and, the monitoring of secondary pollution in effluents [43]. The predicted results are illustrated in Fig. 8 and the toxicity values are found in S3 files. Fig. 8.(a) represents the developmental toxicity of TC intermediates. The results show that some by-products like P-238 and P-413 are more toxic than TC. The toxicity of the remaining intermediates ranged from being equally toxic as TC to being less harmful. P-202, P-339, P-158, P-114, P-163, and P-145 have low developmental toxicity and are labeled as completely non-toxic. Furthermore, the mutagenicity prediction shown in Fig. 8.(b) indicated that seven degradation intermediates out of 19 are considered to be mutagenicity negative, however, many by-products are less ‘mutagenicity positive’ than TC indicating that the mutagenicity indeed decreased under plasma-catalysis treatment. The Daphnia magna LC50 (48 hr) and Oral rat LD (50) toxicity results presented in Fig. 8.(c) and Fig. 8.(d) showed various patterns, but remarkably, intermediates of m/z values < 300 had less acute toxicities than TC. The bioconcentration factor calculations are presented in Fig. 8.(e) The results indicate that P-461 has a bioconcentration factor equal to that of TC. P-114, P202, and P-191 have a higher factor while the remaining intermediates all possess a small or even negative bioconcentration factor. These comprehensive results show that the degradation by-products formed by plasma process combined with Fe(II), Fe(III), or Fe(VI) are mostly less harmful than TC in the exception of few cases.
4. Conclusions
The combination of plasma discharge with different iron-based catalysts demonstrates promising potential for treating recalcitrant pollutants such as TC. Notably, the plasma-Fe(VI) system exhibited a remarkable synergetic effect achieving total degradation within 15 minutes of treatment for different TC initial concentrations. TOC analysis confirmed that this combination system effectively decomposes TC degradation by-products, increasing in the mineralization rates. This study identified the roles of reactive species such as •OH, superoxide, and free electrons in TC elimination, with •OH showing the highest oxidation potential. LC/MS/MS analyses of treated solutions identified nineteen degradation products, and a degradation pathway was proposed according on these results and the information found in literature. Various toxicity parameters were assessed based on degradation by-products and DATA from toxicity estimation software. This study provides insights into the application of plasma processes, especially, combined with ferrate or Fe(II) for the treatment of TC. The hybrid process offers significant benefits, including reduced treatment time, minimal catalyst dosage, and complete pollutant removal. These advantages make it a crucial component for future water decontamination applications, as it ensures lower energy consumption and increased cost-efficiency.
PDF Links
PubReader
Full text via DOI
Download Citation 
Supplement
Print 
