Photocatalytic degradation of the antidepressant drug Paroxetine using TiO2 P-25 under lab and pilot scales in aqueous substrates
0 Abstract
Pharmaceuticals in water bodies are a significant threat to aquatic life and human health, often persisting due to incomplete degradation in conventional wastewater treatment plants (WWTPs). Heterogeneous photocatalysis is a promising method for efficiently treating wastewater (WW). TiO2 P-25, a well-known photocatalyst, has been widely used to remove persistent organic pollutants (POPs) from aquatic media, primarily on a laboratory scale. In this study, the photocatalytic removal of Paroxetine (PXT), an antidepressant drug, is investigated at the lab scale and pilot scale using a compound parabolic collector (CPC) reactor (85 L) working in batch recirculating mode and secondary treated hospital wastewater (HWW) as the substrate using different catalyst concentrations (200, 300, and 500 mg/L). The lab experiments achieved the fastest PXT degradation with 500 mg/L, while the pilot tests found 200 mg/L to be optimal. Thirteen transformation products (TPs) were identified using liquid chromatography-high-resolution mass spectrometry (LC-HR-MS-Orbitrap), and their ecotoxicity was assessed with ECOSAR software, indicating they were less toxic than PXT. The T.E.S.T. software showed most TPs were not mutagenic but displayed developmental toxicity. Toxicity assessments from the pilot scale using the Microtox bioassay demonstrated that toxicity was eliminated by the end of the photocatalytic treatment. In conclusion, the study provides an integrative approach to photocatalytic degradation of PXT, integrating lab-scale and pilot-scale experiments, pure water and real WW matrices, environmentally relevant concentrations, TPs’ identification along with toxicity assessment of both in-silico and in-vitro, which is not followed in most previous studies dealing with the photocatalytic degradation of pollutants.
Keywords
INTRODUCTION
Recent years’ research on aquatic pollutants has focused on emerging contaminants (ECs) monitoring and decomposition into water ecosystems because it is a new group of chemical substances that are released continuously into the environment, mainly through WW, and many of them are considered persistent and carcinogenic, causing adverse effects on humans and animals’ health[1,2].
A large share of aquatic ECs is “owned” by pharmaceutical active compounds (PhACs), as the pollution from these compounds has increased significantly over the years due to their high consumption[3,4]. It is estimated that 3,000 compounds are used as pharmaceutical products. However, more than 50% of all PhACs worldwide are not prescribed or dispensed legally, and many times, their intake is not indicated[4,5]. Therefore, the increasing use of pharmaceutical products implies the introduction of even tons of these compounds into the WW[6]. Such situations lead to the presence of PhACs in surface water and WW at concentrations ranging from a few ng/L to µg/L, considering them as “pseudo-persistent” pollutants even though they have relatively short half-lives in the environment[7,8]. Even at these trace concentrations, PhACs could be toxic to the environment and aquatic organisms[9].
Among PhACs, selective serotonin reuptake inhibitors (SSRIs) are the most prescribed worldwide and usually are the first choice for mental disorders treatment (e.g., depression)[10,11], but also for various medical purposes, including sleep and eating disorders, alcohol and drug abuse, panic, and post-traumatic stress disorders[12]. SSRIs should be chemically stable to induce a specific pharmacological response in the human body. However, this stability could potentially affect their removal from WW during treatment in conventional wastewater treatment plants (WWTPs). As a result, the appearance of the parent compounds or their transformation products (TPs) in treated WW and, consequently, in the natural aquatic environment is due to incomplete degradation[11].
Paroxetine (PXT) (CAS: 61869-08-7) [Supplementary Figure 1] is an SSRI approved for clinical use and has been available since the early 1990s. It is used for the treatment of clinical depression, obsessive-compulsive disorder, and panic disorder[10,13,14]. As a medicinal product, it is prescribed as hydrochloride salt[15].
In particular, PXT has been detected in several monitoring studies concerning the occurrence of PhACs in influents and effluents of WWTPs. Its removal in WWTPs varies from 30% to approximately 75%, indicating that it is not totally degraded[5,6,11,16-20]. PXT also has a tendency to adsorb onto activated sludge or other organic sediments during biological treatment in a conventional WWTP, thereby limiting its biodegradation[21]. PXT could be considered a “pseudo-persistent” contaminant as other antidepressant pharmaceuticals due to its potential presence in increased concentrations and its frequent appearance in the environment. Generally, antidepressant pharmaceuticals are grouped as pseudo-persistent pollutants of emerging concern[22]. The incomplete degradation of PXT in WWTPs has been noticed to induce spawning in zebra mussels at deficient concentrations of 10-7 M, equivalent to about 30 µg/L, according to toxicological studies[19], indicating potential adverse effects in the aquatic food chain, thus demonstrating the need for more efficient methods such as advanced oxidation processes (AOPs).
AOPs are characterized by their recent application in water decontamination without causing secondary pollution, while at the same time, they are promising, effective, environmentally friendly, and economically feasible techniques[23,24]. The common idea of AOPs is the in situ production of highly reactive species, for example, the hydroxyl radical (•OH) and the anionic superoxide radical (•O2-), which then non-selectively “attack” the organic pollutants to degrade them first into products of lower molecular weight and finally to mineralize them into non-toxic products such as H2O, CO2, and inorganic acids[25-27].
Heterogeneous photocatalysis has been proven a very promising AOP for organic pollutants’ degradation because of its high performance, simplicity, reproducible results, and easy handling[28]. Plenty of semiconductors have been utilized as photocatalysts, but TiO2 remains one the catalysts commonly used in “real-world” applications due to its excellent catalytic performance, low cost, excellent hydrophilicity, remarkable photostability, and non-toxicity[23,29-32]. Another aspect of TiO2 is that it can be used in dispersed (slurry) or in thin film form (immobilized catalytic layer). When the catalyst is applied as a suspension, it has enough advantages; the most significant is the high photocatalytic performance[24,33].
Compound parabolic collectors (CPCs) are the photocatalytic reactor systems that prevail today. Some of their advantages are the collectors’ parabolic geometry, which collects both direct and diffuse sunlight, reflecting it on the tube, a condition that enables CPCs to function during cloudy days, the absence of tracking systems, and the increased quantum yield when the photons’ concentration is high[3,34-37]. Although TiO2 has been widely used in pilot-scale applications for the degradation of organic pollutants, the studies focus on either the use of artificial light sources (e.g., LED lamps), which increase the operating costs, or on small WW volumes operated in batch mode, or specific water matrices, such as municipal WW. Additionally, most studies did not consider the reuse of the catalyst in the process in terms of its impact on performance[34,37-40].
Based on all the above, this work reveals new insights into the photocatalytic removal of PXT in water (lab scale) and secondary hospital wastewater (HWW) effluent (pilot scale) using a CPC photoreactor. TiO2 P-25 catalyst is applied in three different concentrations to identify the optimal concentration for the fastest drug removal using simulated solar radiation for laboratory-scale experiments and natural sunlight for pilot-scale experiments. Furthermore, the study examined the extent of PXT’s mineralization and the identification of its TPs while proposing a transformation pathway with this photocatalyst. The potential ecotoxicity and mutagenic properties of PXT and its TPs are determined by in-silico tools such as ECOSAR and toxicity estimation software tool (TEST) software. Compared to other studies on photocatalytic degradation of organic pollutants in an aqueous matrix, the present study provides an integrative approach, providing experimental data at both the lab scale and pilot scale. It examines degradation in pure water but also in real wastewaters (WW), at environmentally relevant concentrations, studying mineralization and TPs’ formation with toxicity assessment both in-silico and in-vitro.
EXPERIMENTAL
Reagents, solvents, and materials
High-purity PXT (TCI, Zwijndrecht, Belgium) and commercial pharmaceutical product Seroxat (tablet containing 20 mg PXT, GlaxoSmithKline Pharmaceuticals S.A., Poznan, Poland) were used for lab-scale and pilot-scale experiments, respectively. Titanium dioxide (TiO2) P-25 (average particle size: 30 nm, 80% anatase and 20% rutile, specific surface area 50 m2·g-1) was supplied by Evonik (Essen, Germany). Freeze-dried Vibrio fischeri bacterium in solid form (Acute Reagent) and a reconstitution solution used for its activation were supplied by Modern Water (New Castle, DE, USA). For the lab-prepared solutions, ultra-pure water was used by an Evoqua apparatus (Pittsburgh, PA, USA), while for pilot-scale experiments, secondary effluent from the University Hospital of Ioannina City was used as a substrate, whose physicochemical parameters are listed in Supplementary Table 1 and studies about other ECs have already been conducted[5,41]. Further information is provided in the Supplementary Materials.
Lab-scale photocatalytic experiments
The lab-scale photocatalytic degradation of PXT took place in aqueous solution (10 mg/L, 200 mL) under simulated solar irradiation using a SUNTEST CPS+ apparatus (Atlas, Linsengericht, Germany). This simulator is equipped with a 1,500 W Xenon arc lamp and a glass filter permitting solar radiation within 300 to 800 nm. The lamp intensity was adjusted to 500 W/m2, corresponding to a radiation dose of 150 kJ per
Pilot-scale photocatalytic experiments
The photocatalytic experiments for PXT degradation at the pilot scale were carried out with a CPC reactor
The experimental process began with filling the feed tank with secondary WW effluent from the WWTP of the University Hospital of Ioannina. The homogenization of the WW was carried out inside the feed tank with mechanical agitation and airflow from the air blower, which is located at the bottom of the tank. During homogenization, WW was spiked with an aqueous solution of the commercial antidepressant drug Seroxat. The spiked concentration of PXT was 20 μg/L. After spiking, homogenization continued for
The decrease in PXT concentration (C/C0) in each photocatalytic experiment was observed as a function of a normalized illumination time (t30W, min) and accumulated UV energy (QUV, kJ/L), as reported in previous studies[35,36,40], according to the following equations:
where tn is the experimental time (min) of treatment, Vi is the irradiated volume (L) of WW, Vt is the total volume (L) of WW, IUV,n is the average value of solar UV radiation (W/m2) (λ = 300-400 nm) measured in the interval Δt, and Ar is the reactor’s irradiated surface (m2).
Analytical determinations
Determination of PXT residual concentration with HPLC
PXT concentration in the samples obtained from lab-scale experiments was determined by a Shimadzu (Kyoto, Japan) HPLC system consisting of a solvent delivery pump (LC-10AD), a degasser unit (DGU-14A), an auto-sampler (SIL-20A), a photodiode array detector (SPD-M10A), and a column oven (CTO-10A). The chromatographic column was a Discovery C18 (15 cm × 4.6 mm, 5 μm particle size) purchased from Supelco (Bellefonte, PA, USA). The injection volume was configured to 20 μL, utilizing a solvent mixture of 70:30 water and acetonitrile with 0.1% v/v Formic Acid as the mobile phase (isocratic elution). The mobile phase flow rate was set to 1 mL/min, with the column oven temperature maintained at 40 °C. The chromatograms were recorded using LabSolutions software. The retention time (Rt) of PXT was observed at 6.30 min.
Determination of mineralization degree by ion chromatography and total organic carbon analyzer
The concentration of F-, NO2-, and NO3- anions in the photocatalytically treated samples was determined by a Shimadzu (Kyoto, Japan) ion chromatography system consisting of a solvent pump (LC-20Ai), a conductivity detector (CDD-10A), and a column oven (CTO-40C). The system is connected to a PC through the communication unit (CBM-40). The anion chromatography column was Shodex IC SI-35 4D (15 cm × 4.6 mm, 3.5 μm particle size) purchased from Showa Denko America (New York, USA). The injection volume was 20 μL, while the mobile phase, an aqueous Na2CO3 solution (3.6 mM), adjusted its flow to 0.6 mL/min. The column oven was set at 40 °C. The acquisition of the chromatograms and the processing of the results were carried out with the LabSolutions software.
A TOC-L automatic analyzer was utilized with an 8-place OCT-L autosampler from Shimadzu (Kyoto, Japan). First, every sample’s total carbon load (TC) was determined through catalytic oxidation at 680 °C in an oxygen-rich environment. Afterward, the sample’s inorganic carbon (IC) was determined, oxidating the carbonate ions contained in the sample in the presence of HCl (1 M). The CO2 produced from both cases was measured through an NDIR detector. Finally, the total organic carbon (TOC) value is obtained by the equation: TOC = TC - IC. The results were obtained via PC-installed TOC Control-L software provided by Shimadzu.
Solid-phase extraction
The samples collected during the lab-scale photocatalytic experiments were subjected to solid-phase extraction (SPE) to pre-concentrate the TPs resulting from PXT degradation. The SPE process was carried out with the 4-position automatic SPE device from LabTech (Bergamo, Italy) connected to an air compressor from JUN-AIR (Norresundby, Denmark) and a laboratory nitrogen gas generator Alliance ALIZE-6/0 of Innovative Gas System Company (Evry, France). CHROMATific HLB+ micro-cartridges
In the case of the pilot scale sample extraction, a protocol that has been previously reported[5,35] was followed. Before the SPE process, 100 mL of each sample was prepared by adjusting the pH to 7 and adding 2 mL of an aqueous Na2EDTA 5% w/v solution. The addition of Na2EDTA solution is because it complexes with metal ions, which are soluble in water and tend to form complexes with pharmaceutical compounds, reducing the extraction yield. The SPE was carried out using a Supelco (Bellefonte, PA, USA) Visiprep DL 12-place extraction apparatus connected to a Laboport (Oxfordshire, United Kingdom) vacuum pump. OASIS HLB cartridges (200 mg/6 mL) from Waters Corporation (Milford, MA, USA.) were used. The extraction procedure reported above was also followed.
Detection and identification of PXT and TPs by UHPLC-LTQ-Orbitrap
A UHPLC Accela LC system, connected with a hybrid LTQ-FT Orbitrap XL 2.5.5 SP1 mass spectrometer equipped with an electrospray ionization (ESI) source (Thermo Fisher Scientific, Inc., GmbH, Bremen, Germany) was used for the detection of PXT in pilot scale experiments and TPs in the treated samples. For the analysis of TPs, a full scan in positive ionization (PI) mode was applied within a mass range of 50-800 Da and a mass resolving power of 60,000 FWHM. In addition, a data-dependent acquisition (full MS/dd-MS2) based on collision induced dissociation (CID) was carried out. More information about the operational parameters of the LTQ-FT Orbitrap instrumentation is listed in Supplementary Table 2. The chromatographic column used was a Hypersil Gold (100 mm × 2.1 mm, 1.9 μm particle size) from Thermo Fisher Scientific (Bremen, Germany). A mixture of water (solvent A) and methanol (solvent B) containing 0.1% v/v formic acid was used as a mobile phase. The elution gradient started with 98% solvent A (initial conditions) and remained for 2 min. Then, it progressed to 70% in 5 min and 50% in 10 min. The percentage of solvent A reached 2% after 15 min, returned to the initial conditions 10 s later, and remained like this until the end of the run. The total run time was 20 min. The injection volume was 10 μL, the flow of the mobile phase was adjusted to 350 µL/min, and the column’s oven temperature was set at 35 °C. The chromatograms, mass spectra, and results processing were acquired with Xcalibur 2.1 software (Thermo Electron, San Jose, CA, USA).
Determination of secondary treated HWW physicochemical parameters
Conductivity was measured by an LF 325 portable conductometer purchased from WTW (Weilheim, Germany). Chemical oxygen demand (COD) was measured using the standard COD Cell Tests: C3/25 (10-150 mg/L) and C4/25 (25-1,500 mg/L), a thermoreactor CR 3200 and a WTW photo Flex photometer, all purchased from WTW (Weilheim, Germany). Five-day biochemical oxygen demand (BOD5) was measured following a standard protocol from WTW (Weilheim, Germany) and using a WTW TS 606-G/2-i incubation cabinet and a WTW portable OxiTop OC 110 meter. The phenolic content of the samples was determined by the Folin-Ciocalteau method. The procedure started with mixing 5 mL of the sample and
In-silico ecotoxicity assessment and prediction of ecotoxicological values (mutagenicity, bioconcentration factor, and developmental toxicity)
The ecotoxicity of PXT and its TPs was assessed using the ECOSAR software (version 2.2). ECOSAR uses a large dataset (130 structural classes) of ecotoxicity studies from the ECOTOX database, following USEPA Office of Chemical Safety and Pollution Prevention guidelines. This software assesses the acute and chronic toxicity of chemical compounds to aquatic organisms using structure-activity relationship (SAR) models. The acute ecotoxicity assessment was calculated at three trophic levels: fish, daphnia (LC50 value), and green algae (EC50 value). Additionally, the chronic toxicity (ChV) values for the chemical compounds studied were calculated using the ECOSAR software.
The TEST software (version 5.1.2) was used to evaluate the potential for mutagenicity, developmental toxicity, and bioconcentration factors (BCFs) of PXT and its TPs. This software, like ECOSAR, was developed by the USEPA and uses QSAR models to make predictions. The consensus method was employed, which provides the most accurate predictions according to the user’s guide.
RESULTS AND DISCUSSION
Photocatalytic degradation of PXT in lab-scale experiments
Studying the photocatalytic degradation of PXT (10 mg/L) in the laboratory showed that its complete degradation is possible within 45 min of irradiation with all three photocatalyst concentrations following pseudo-first-order kinetics [Figure 1A]. As can be observed from Supplementary Table 3, the higher the photocatalyst concentration, the faster the degradation of the pharmaceutical compound. The degree of PXT mineralization was also studied by determining the TOC concentration and F-, NO2-, and NO3- anions. The determination coefficients obtained for each case (Supplementary Table 4, R2 > 0.9) confirm that the pseudo-first-order kinetics model is also followed to a satisfactory degree for TOC decrease. Moreover, as seen in Figure 1B, increasing the catalyst’s concentration leads to TOC’s faster decrease. Specifically, in the case of applying 500 mg/L TiO2, the TOC has decreased by approximately 95% after 360 min of irradiation; in the case of 300 mg/L TiO2 by 86%, while in the case of 200 mg/L TiO2, it has been decreased by 65%. Additionally, it is observed that in the first 60 min of each photocatalytic process (where PXT is not detected after 45 min), the concentration of TOC decreases relatively slowly, concluding that, during PXT’s degradation, TPs are formed, which are further degraded, implying a decrease in TOC.
Figure 1. (A) Degradation kinetics of PXT (C0 = 10 mg/L) and (B) TOC decrease kinetics with TiO2 P-25 catalyst (200, 300, and
Secondly, the mineralization degree was studied by monitoring the F-, NO2-, and NO3- concentrations. The maximum concentrations that can be stoichiometrically derived from the complete oxidation of PXT
Comparing the evolution kinetics of fluorine anions for the three catalyst concentrations applied [Figure 2A], a rapid increase in their concentration is observed in the first 45 min of each photocatalytic process. Accordingly, PXT has been completely removed at this time, which causes the rapid release of fluorine ions. From 45 to 360 min, the concentration remains almost unchanged. Worthy of attention is the case of using 200 mg/L TiO2, where the concentration of fluorine ions formed is higher than those derived from the other catalyst concentrations studied. Theoretically, it would be expected that the higher the catalyst concentration, the higher the release rate of fluorine ions in the solution due to the advanced and faster degradation of the parent compound containing fluorine atoms in its structure. However, after
Figure 2. Evolution kinetics of the produced F- (A) and NO3- (B) anions during the photocatalytic degradation of PXT with TiO2. PXT: Paroxetine.
Regarding the transformation of piperidine-nitrogen heteroatom, no NO2- ions were detected in any samples, regardless of the catalyst concentration. This could be due to their immediate oxidation to NO3- after their production because of the abundance of hydroxyl radicals in the solution. The evolution kinetics of NO3-, as shown in Figure 2B, is “driven” by the concentration of the catalyst. That is, the greater the concentration of the catalyst, the greater the concentration of nitrate ions that appear at any time. The evolution kinetics for nitrate ions appears to have a similar pattern in all three catalyst concentration cases. Specifically, after a sharp increase within the first 15 min of irradiation, the concentration increases at a slower rate of up to 90 min, followed by a second step of higher formation slope up to 180 min (more intense at catalyst concentrations of 200 and 300 mg/L), and then a further gentle increase up to 360 min.
This increase in NO3- concentration in two distinct steps can be associated with the formation of intermediate compounds, which can undergo further reactions, leading to the stepwise release of nitrogen. The first release step (0-15 min) in Figure 2B is most likely due to the degradation of the parent PXT molecule by oxidant species produced by TiO2 photocatalysts. Subsequently, the second release step (60-
Thus, at the end of the photocatalytic process, for the concentration of 200 mg/L TiO2, the concentration of NO3- corresponds to approximately 35% of the stoichiometrically available nitrogen, for the concentration of 300 mg/L, it corresponds to 40%, and for the concentration of 500 mg/L, it is approximately 48%.
Detection and identification of the photocatalytically produced TPs with TiO2 P-25
Thirteen TPs were identified during the photocatalytic processes [Figure 3], and the data obtained from the processing of the chromatograms are shown in Table 1. The primary TP observed was TP-210 with a molecular formula of C12H17ONF. TP-210 appeared at 5.44 min and was formed by removing the benzodioxol part from the PXT molecule, with a mass difference of 120.0203 Da. Its structure was determined by analyzing MS2 fragments with m/z values of 192.1178, 163.0912, 123.0599, and 109.0444, characteristic of the parent compound’s molecule. Additionally, it should be noted this particular TP has been documented previously in the photolytic degradation of PXT[13,14].
Figure 3. Proposed pathway for the photocatalytic transformation of PXT with TiO2 P-25. PXT: Paroxetine.
High-resolution mass data of PXT and its TPs formed during the photolytic and photocatalytic degradation experiments
| Compound | Rt (min) | [M+H]+ | Molecular formula | Δ (ppm) | RDB | MS2 [M+H]+ | Molecular formula | Δ (ppm) | RDB |
| PXT | 11.29 | 330.1491 | C19H21O3NF | -2.993 | 9.5 | 313.1228 192.1178 178.1022 163.0914 151.0386 123.0599 109.0443 | C19H18O3F C12H15NF C11H13NF C11H12F C8H7O3 C8H8F C7H6F | -2.073 -2.416 -2.550 -2.300 -2.189 -4.917 -4.631 | 10.5 5.5 5.5 5.5 5.5 4.5 4.5 |
| TP-208 | 1.58 | 208.1328 | C12H18O2N | -1.515 | 4.5 | 190.1222 161.0958 147.0799 121.0641 107.0482 | C12H16ON C11H13O C10H11O C8H9O C7H7O | -2.370 -1.872 -3.478 -5.712 -9.075 | 5.5 5.5 5.5 4.5 4.5 |
| TP-210 | 5.44 | 210.1288 | C12H17ONF | -0.994 | 4.5 | 192.1178 163.0912 135.0595 123.0599 109.0444 | C12H15NF C11H12F C9H8F C8H8F C7H6F | -2.469 -3.220 -6.923 -4.673 -3.897 | 5.5 5.5 5.5 4.5 4.5 |
| TP-226 | 4.69 | 226.1231 | C12H17O2NF | -2.979 | 4.5 | 208.1129 191.0862 179.0862 177.0706 166.0661 165.0707 163.0552 139.0549 125.0393 | C12H15ONF C12H12OF C11H12OF C11H10OF C9H9ONF C10H10OF C10H8OF C8H8OF C7H6OF | -1.292 -2.301 -2.791 -2.370 -0.895 -2.240 -0.918 -3.377 -3.675 | 5.5 6.5 5.5 6.5 5.5 5.5 6.5 4.5 4.5 |
| TP-238 | 9.55 | 238.1229 | C13H17O2NF | -3.458 | 5.5 | - | - | - | - |
| TP-242 | 4.10 | 242.1181 | C12H17O3NF | -3.420 | 4.5 | - | - | - | - |
| TP-296 | 6.45 | 296.1284 | C15H19O4NF | -2.542 | 6.5 | 278.1181 210.1287 192.1180 190.1019 163.0911 123.0600 | C15H17O3NF C12H17ONF C12H15NF C12H13NF C11H12F C8H8F | -2.258 -0.899 -1.740 -3.809 -3.833 -4.023 | 7.5 4.5 5.5 6.5 5.5 4.5 |
| TP-328 | 7.84 | 328.1537 | C19H22O4N | -3.456 | 9.5 | 311.1272 298.1424 281.1167 234.1120 217.0854 206.1169 190.1222 173.0957 151.0386 | C19H19O4 C18H20O3N C18H17O3 C13H16O3N C13H13O3 C12H16O2N C12H16ON C12H13O C8H7O3 | -1.786 -4.729 -1.853 -1.794 -2.537 -3.034 -2.370 -2.147 -2.321 | 10.5 9.5 10.5 6.5 7.5 5.5 5.5 6.5 5.5 |
| TP-344 | 7.00 | 344.1485 | C19H22O5N | -2.177 | 9.5 | 327.1220 314.1362 234.1121 217.0853 206.1172 189.0908 177.0907 163.0750 151.0387 123.0435 | C19H19O5 C18H20O4N C13H13O3N C13H13O3 C12H16O2N C12H13O2 C11H13O2 C10H11O2 C8H7O3 C7H7O2 | -2.202 -7.814 -1.409 -2.675 -1.481 -1.302 -1.729 -2.306 -1.924 -4.194 | 10.5 9.5 6.5 7.5 5.5 6.5 5.5 5.5 5.5 4.5 |
| TP1-346 | 10.31 | 346.1441 | C19H21O4NF | -2.608 | 9.5 | 329.1179 234.1121 208.1129 194.0968 179.0863 151.0387 139.0550 125.0393 | C19H18O4F C13H16O3N C12H15ONF C11H13ONF C11H12OF C8H7O3 C8H8OF C7H6OF | -1.439 -1.623 -1.388 -3.961 -1.897 -1.725 -2.730 -3.116 | 10.5 6.5 5.5 5.5 5.5 5.5 4.5 4.5 |
| TP2-346 | 12.06 | 346.1440 | C19H21O4NF | -2.146 | 9.5 | 328.1339 208.1129 178.1024 163.0915 151.0386 109.0440 | C19H19O3NF C12H15ONF C11H13NF C11H12OF C8H7O3 C7H6F | -1.427 -1.292 -1.427 -1.897 -2.189 -7.199 | 10.5 5.5 5.5 5.5 5.5 4.5 |
| TP-350 | 8.36 | 350.1392 | C18H21O5NF | -0.078 | 8.5 | 332.1286 306.1492 290.1184 210.1287 192.1180 190.1025 178.1027 163.0916 135.0597 123.0601 | C18H19O4NF C17H21O3NF C16H17O3NF C12H17ONF C12H15NF C12H13NF C11H13NF C11H12F C9H8F C8H8F | -2.086 -2.607 -0.993 -0.613 -1.844 -1.022 0.145 -1.074 -5.516 -3.048 | 9.5 7.5 8.5 4.5 5.5 6.5 5.5 5.5 5.5 4.5 |
| TP-362 | 9.98 | 362.1388 | C19H21O5NF | 1.194 | 9.5 | - | - | - | - |
| TP-378 | 8.06 | 378.1337 | C19H21O6NF | -3.734 | 7.5 | 348.1234 334.1442 210.1284 192.1178 | C18H19O5NF C18H21O4NF C12H17ONF C12H15NF | -2.377 -2.223 -2.089 -2.729 | 9.5 8.5 4.5 5.5 |
TP-208 differs from TP-210 by 1.9960 Da and has the molecular formula C12H18O2N and can be rationalized by the substitution of the fluorine atom in the benzene ring with a hydroxyl group. This is supported by the fragments with m/z = 121.0641 and m/z = 107.0482, which differ by approximately 2 Da from the corresponding fragments of TP-210 (m/z = 123.0599 and m/z = 109.0444), as well as their molecular formulas. Thus, it has been verified that the hydroxyl group is positioned in the benzene ring, substituting fluorine.
TP-226 was observed eluting at 4.69 min. Its molecular formula is C12H17O2NF, which differs from the molecular formula of TP-210 by 15.9943 Da, suggesting a hydroxylated derivative of TP-210. The additional hydroxyl group is positioned in the benzene ring, as indicated by MS2 fragments with m/z = 139.0549 and m/z = 125.0393 and molecular formulas C8H8OF and C7H6OF, respectively. The corresponding fragments without the addition of oxygen are also present in the parent compound with m/z = 123.0599 and m/z = 109.0443 and molecular formulas C8H8F and C7H6F, respectively.
TP-296 was identified as having the molecular formula C15H19O4NF. The MS2 fragments characteristic of PXT consist of m/z = 192.1180, m/z = 163.0911, and m/z = 123.0600. The proposed molecular formula for TP-210 matches the fragment with m/z = 210.1287. For the fragment with m/z = 278.1181, the molecular formula C15H17O3NF was suggested. Despite obtaining sufficient information from the MS2 fragments, a chemical formula for TP-296 cannot be determined, and it does not come to an agreement with the literature suggesting C19H19ONF as the molecular formula of TP-296[13].
TP-328, with m/z = 328.1537, corresponds to a molecular formula of C19H22O4N. The TP appears to result from replacing a fluorine atom in the PXT molecule with a hydroxyl group, similar to when TP-208 was created from TP-210. The fragment with m/z = 190.1222 and a molecular formula of C12H16ON is the same as that reported for TP-208, indicating the hydroxyl group is located in the benzene ring. A fragment with m/z = 151.0386 and a molecular formula of C8H7O3, which contains the benzodioxol moiety also found in the PXT molecule, was also detected (could be called a methoxy-benzodioxol structure). This fragment originated from the ether bond cleavage and includes three of the four oxygen atoms present in the TP. The substitution of fluorine in the benzene ring was confirmed by fragments with m/z = 234.1120 and m/z = 217.0854, resulting from the cleavage of the bond between the benzene and piperidine ring. These fragment ions contain three oxygen atoms, suggesting that the fourth oxygen atom will be found in the benzene ring.
TP-344 results from the addition of an oxygen atom to the structure of TP-328 as shown by the mass difference of 15.9948 Da between them, but also by the shorter retention time of TP-344 (7.00 min) than TP-328 (7.84 min), a fact showing that the former is a more polar molecule than the latter. Molecular structure identification was based on MS2 fragment ions. Specifically, four common fragments with TP-328 are identified, with m/z values of 234.1121, 217.0853, 206.1172, and 151.0387. The first two (234.1121 and 217.0853) come from the cleavage of the bond between the benzene ring and the piperidine ring. From the piperidine ring, the nitrogen atom has also been removed as ammonia by the opening of the ring and its rearrangement to form a double bond. Thus, the difference of 17.0268 Da between the two fragments is also justified. Furthermore, a fundamental observation is that three of the five oxygen atoms in the molecular formula of the given TP are located in the molecular formula of these two fragments, which means that the remaining two oxygen atoms are located in the structure corresponding to TP-210. This observation is supported by the common fragment with m/z = 151.0387 having a methoxy-benzodioxol moiety (the three oxygen atoms are detected). It has resulted from the cleavage of the ether bond between the oxygen and the methyl carbon of the piperidine ring. Although the fragment with m/z = 206.1172 has the same molecular formula as the corresponding fragment of TP-328, it has a different structure. This concludes from the fragment with m/z = 123.0435, which indicates the existence of two hydroxyl groups on the benzene ring, which is connected to the piperidine ring. Therefore, since one hydroxyl group has replaced the fluorine atom in the PXT molecule, the second one will be located either in the o-position or in the m-position, with the o-position having more chances due to the activation of the benzene ring by the hydroxyl group present.
TP-346 is detected at two retention times in the chromatogram, at 10.31 and 12.06 min, considering that this particular TP presents two isomers. It was assigned the molecular formula C19H21O4NF, confirming the addition of an oxygen atom to the PXT molecule, as indicated by a difference of 15.9950 Da. The structure for each isomer was identified by MS2 data, aiming to find the location of the added hydroxyl group on the PXT molecule. In both TPs, the fragment with m/z = 151.0387 and molecular formula C8H7O3 is detected, which has also been detected in the MS2 spectrum for TP-328 and is the methoxy-benzodioxol moiety contained in the PXT molecule. As mentioned, three of the four oxygen atoms that the two isomers of TP-346 have are located in this structure. Therefore, the added hydroxyl group could be in the piperidine or benzene rings. From the fragment ions obtained for the first isomer (TP1-346) and, more specifically, the one with m/z = 125.0393 and a molecular formula C7H6OF, it is indicated that the hydroxyl group is located in the benzene ring with the most likely addition positions being o- position to fluorine.
The same finding has been made in the TP-226. On the contrary, for the second isomer (TP2-346), a fragment with m/z = 109.0440 and a molecular formula of C7H6F was obtained, meaning that the oxygen atom is not located in the benzene ring. Still, the moiety of the molecule where the hydroxyl group can be located is in the piperidine ring, following the previous finding for the fragment with m/z = 151.0837.
TP-350 with m/z = 350.1392 corresponds to the molecular formula C18H21O5NF. This compound appears to be created by bi-hydroxylation in the PXT molecule and the five-membered ring’s concurrent opening. The hydroxylated character is also denoted by the characteristic loss of H2O (18.0106 Da), presenting a fragment ion with m/z equal to 332.1286. A fragment ion with an m/z value of 306.1492 is formed due to the loss of a CO2 molecule (43.9900 Da)[13]. When trying to identify the additional hydroxyl groups on the molecule, characteristic fragment ions found in PXT were observed (m/z = 192.1180, m/z = 187.1027, m/z = 163.0916, and m/z = 123.0601). Moreover, the fragment ion with m/z = 210.1287 has the same structure as the TP-210, indicating that one of the five oxygen atoms in the molecule is located in this structure. Consequently, it is suggested that the two additional hydroxyl groups are positioned in the benzene ring of the benzodioxol moiety. However, due to steric effects, it is challenging to pinpoint both added hydroxyl groups in the mentioned ring, as their simultaneous inclusion likely led to the ring’s opening.
TP-378 eluted at 8.06 min and was assigned to the molecular formula C19H21O6NF. It is considered adding three oxygen atoms (3 × 15.9949 Da) to the PXT molecule, most likely in hydroxyl groups. Two characteristic fragments were detected in the MS2 spectrum, which is also detected for PXT having m/z = 210.1284 and m/z = 192.1178 and molecular formulas C12H17ONF and C12H15NF, respectively. Two other fragment ions were detected with m/z = 348.1234 (molecular formula: C18H19O5NF) and with m/z = 334.1442 (molecular formula: C18H21O4NF), which cannot indicate the positions of the hydroxyl groups on the molecule’s structure. Therefore, the only conclusion that can be drawn, and indeed not with absolute certainty, is that the specific TP is a tris-hydroxylated PXT molecule.
Finally, MS2 data were not obtained for TP-238, TP-242, and TP-362 with molecular ions m/z = 238.1229, m/z = 242.1181, and m/z = 362.1388, respectively. However, their identification was possible considering their molecular type, retention time, and evolutionary profiles, indicating that these chemical structures did not pre-exist but were formed during the degradation of the parent compound; their concentration reached a maximum, and over time, they were also degraded. In particular, TP-238 with a molecular formula of
TP-242 with molecular formula C12H17O3NF probably originates from the addition of an oxygen atom to TP-226 (the difference between them is equal to 15.9951 Da). Additionally, TP-242 elutes earlier (4.10 min) than TP-226 (4.69 min), an observation that confirms the more polar character of TP-242. However, the absence of fragment ions does not allow its precise identification. Accordingly, TP-362 (molecular formula C19H21O5NF) differs from TP-346 by 15.9947 Da, and by extension, an oxygen atom has been added to the molecule of TP-346. TP-362, as a more polar molecule, elutes earlier than TP-346.
In summary, various primary and secondary hydroxylated products were consistently formed with the well-known potential of TiO2 for hydroxyl radical generation, suggesting hydroxylation pathways as the major ones for PXT transformation.
The evolutionary profiles of the TPs are presented in Figure 4, showing that after 90 min of irradiation, regardless of the concentration of the photocatalyst, most of them have either been completely degraded or are presented in small amounts, such as TP-208, TP-210, and TP-226 until their complete degradation. This observation agrees with the TOC results, where after 90 min, the rate of its decrease becomes faster. TP-210 appears to be the most “resistant” to degradation, as regardless of the catalyst’s concentration, it is no longer detected after 360 min. Finally, the maximum concentration of each TP (at 15 or 30 min of irradiation), especially in the case of the most hydroxylated products (TP-208, TP-226, TP-242, TP-362, TP-378), occurred when the catalyst concentration was the highest due to the higher production of hydroxyl radicals.
Figure 4. Evolutionary profiles of each TP detected in photocatalysis with TiO2 P-25 in concentrations 200, 300, and 500 mg/L. TP: Transformation product.
In-silico ecotoxicity assessment (lab-scale experiments)
Table 2 shows the theoretical acute (LC50, EC50) and chronic toxicity (ChV) values of PXT and its TPs for fish, daphnids, and green algae. The Globally Harmonized System of Classification and Labelling of Chemicals (GHS) was used. For TPs with undetermined functional groups, the average toxicity, mutagenicity, and BCF values were calculated by placing the functional group at each possible position on the molecule. Furthermore, no definitive conclusion can be drawn about the ecotoxicity of TP-296 due to the inability to suggest its molecular formula and the TEST software’s incapability to calculate mutagenicity and BCF values for TP-238.
In-silico assessment of acute and chronic toxicity of PXT and its TPs with ECOSAR v2.2 software (chemical category: aliphatic amins)
| Compound | Acute toxicity (LC50/EC50) | Chronic toxicity (ChV) | ||||
| Fish LC50 (mg/L) | Daphnid LC50 (mg/L) | Green algae EC50 (mg/L) | Fish ChV (mg/L) | Daphnid ChV (mg/L) | Green algae ChV (mg/L) | |
| PXT | 3.29[b] | 0.493[a] | 0.259[a] | 0.096[a] | 0.050[a] | 0.101[a] |
| TP-208 | 116[d] | 12.7[c] | 12.3[c] | 8.64[b] | 0.963[a] | 3.86[b] |
| TP-210 | 41.7[c] | 4.96[b] | 4.11[b] | 2.45[b] | 0.406[a] | 1.36[b] |
| TP-226 | 92.8[c] | 10.4[c] | 9.67[b] | 6.46[b] | 0.810[a] | 3.08[b] |
| TP-238 | 23.6[c] | 2.96[b] | 2.21[b] | 1.18[b] | 0.256[a] | 0.762[a] |
| TP-242 | 1,355.4[d] | 121.6[d] | 176.3[d] | 190[d] | 7.61[b] | 47.6[c] |
| TP-328 | 2.77[b] | 0.420[a] | 0.216[a] | 0.078[a] | 0.044[a] | 0.085[a] |
| TP-344 | 6.01[b] | 0.862[a] | 0.494[a] | 0.200[a] | 0.085[a] | 0.187[a] |
| TP1-346 | 2.16[b] | 0.335[a] | 0.164[a] | 0.056[a] | 0.036[a] | 0.066[a] |
| TP2-346 | 10.7[c] | 1.47[b] | 0.918[a] | 0.407[a] | 0.138[a] | 0.337[a] |
| TP-350 | 16.5[c] | 2.19[b] | 1.46[b] | 0.692[a] | 0.200[a] | 0.524[a] |
| TP-362 | 28.0[c] | 3.50[b] | 2.63[b] | 1.42[b] | 0.303[a] | 0.901[a] |
| TP-378 | 11.14[c] | 5.39[b] | 4.53[b] | 2.45[b] | 0.453[a] | 1.44[b] |
The summarized data show that PXT is highly toxic to daphnids and green algae. Regarding fish, PXT is toxic after brief exposure and highly toxic in chronic exposure. For the TPs, TP-208 and TP-242 were identified as non-harmful to fish in short exposure, and TP-242 continued to be non-harmful during chronic exposure. In general, the majority of TPs detected are either highly toxic or toxic to the aquatic organisms mentioned above during both short and long-term exposure.
The high toxicity of PXT has been confirmed by in-vitro studies on various microalgae species, including Scenedesmus rubescens, Chlorococcum sp., and Dunaliella tertiolecta algae, which play an essential role as aquatic producers. PXT notably impacts algal growth within the first 24 h, with some instances showing complete inhibition of growth rates at 100%. Possibly, the acute ecotoxicity of PXT on algae relates to its chemical structure, particularly the aromatic ring and amino group, which may interact with the biological membranes of microorganisms[43]. Additionally, other studies examined the toxicity of PXT metabolites excreted from the human body, but these products were not detected in the present study[44,45]. Thus, the available studies in the literature do not address the potential impact of the detected PXT’s TPs on different organisms. It is known that in aquatic environments, particularly at lower trophic levels (e.g., microalgae), population reductions caused by toxic compounds directly affect the subsequent trophic level (e.g., green microalgae are consumed by daphnids). Therefore, the presence of PXT’s TPs may have potential effects on aquatic food chains.
Determining toxicological risk due to potential harm and carcinogenicity is crucial to assessing mutagenicity. QSAR models are utilized to predict the mutagenicity of new or data-lacking chemicals[46,47]. It has been established that PXT does not possess mutagenic potential. From Figure 5, it has been observed that TPs that could cause mutagenic reactions are TP-210 and TP-226. To address the formation of these specific TPs during photoinduced processes, their evolutionary profiles are considered, where complete degradation is observed, producing non-mutagenic compounds.
Figure 5. In silico predicted values for (A) mutagenicity, (B) BCF, and (C) developmental toxicity of PXT and its TPs formed during photocatalysis with TiO2 P-25. BCF: Bioconcentration factor; PXT: Paroxetine; TPs: transformation products.
When assessing chemical pollutants, it is crucial to also consider their BCF, which indicates their potential to accumulate in organisms in aquatic environments. Modeling BCF is often done due to challenges in conducting experiments for many pollutants[48,49]. Predicted BCF values suggest that PXT’s TPs are less likely to accumulate in aquatic organisms than the parent compound, indicating a lower potential for harmful effects.
Regarding developmental toxicity, the TEST software classified both PXT and all the TPs formed as “developmental toxicant” as the predicted values for all compounds are > 0.5. TP1-346, TP2-346, and TP-362 were much more “developmental toxicant” than PXT. Although this may be concerning, considering the evolutionary profiles of the corresponding TPs, it is evident that after 60 min of the photocatalytic process, regardless of the photocatalyst concentration, the TPs are completely degraded.
Photocatalytic degradation of PXT in pilot-scale experiments and environmentally relevant concentration levels
The photocatalytic degradation of PXT (20 μg/L) was studied with a CPC pilot reactor under natural solar irradiation and TiO2 photocatalyst. Figure 6A presents the data from the photocatalytic degradation of PXT as a function of the accumulated UV energy per liter of WW (QUV), while Figure 6B presents the same data as a function of a normalized illumination time (t30W). PXT’s degradation kinetics follows a pseudo-first-order model, as in lab-scale experiments. From the data in Table 3, PXT’s removal from WW was satisfying, ranging from 76% to 92%. Also noteworthy is that in the case of the 200 mg/L photocatalyst concentration, removal was achieved faster (k = 0.05 L/kJ) than the other two photocatalyst concentrations and the highest removal was achieved (92%).
Figure 6. Degradation kinetics of PXT (C0 = 20 µg/L) with TiO2 P-25 catalyst (200, 300, and 500 mg/L) under natural solar irradiation as a function of QUV (A) and t30W (B). PXT: Paroxetine.
Pseudo-first-order rate constant (kapp), correlation coefficient (R2), and % removal of PXT after pilot-scale photocatalytic treatment
| TiO2 P-25 | PXT (20 μg/L) | ||
| C (mg/L) | kapp (L/kJ) | R2 | % |
| 200 | 0.05 | 0.9213 | 92 |
| 300 | 0.034 | 0.9219 | 79 |
| 500 | 0.039 | 0.9268 | 76 |
In contrast to the lab-scale photocatalytic experiments, it was reported that the higher the photocatalyst concentration, the faster the degradation of the pharmaceutical compound was, whereas this condition was not followed. This can be observed in Figure 6A, where QUV values are greater at the exact sampling times in the case of photocatalysis with 200 mg/L TiO2.
Additionally, the concentration of 200 mg/L is presented as the optimal catalyst concentration for this pharmaceutical’s removal compared to the other two catalyst concentrations (300 and 500 mg/L). It thus ensures the effective absorption of photons from the catalyst, as referred to in other case studies[35,50]. When higher catalyst concentrations (300 and 500 mg/L) are applied, the reaction rate decreases, and unwanted scattering of light and difficulty of its penetration into the solution may occur[33,51]. Of course, other factors such as pH, temperature, differences in the substrate’s organic composition, etc., affect the performance of heterogeneous photocatalysis[25,36,52-54], but it is believed that the intensity of radiation and the concentration of the catalyst effect are the most crucial factors in this case.
The reusability of the TiO2 catalyst [Figure 7A and B] was also studied on a pilot scale for three consecutive days (three catalytic cycles), letting the photocatalyst settle down in the feed tank after the end of each experiment without washing it, and then used the next day (next catalytic cycle). Before the beginning of each experiment, WW was spiked with an appropriate volume of PXT solution, concluding at a concentration of 20 μg/L. The data obtained from this series of experiments are summarized in Table 4. In the first photocatalytic cycle, PXT is removed from WW at a rate of 78%. In the second cycle, it is removed by 66%, while in the third cycle, this percentage drops to 32%, indicating changes on the surface of the catalyst and, by extension, the decrease in its total active sites, resulting in deactivation. This phenomenon is also confirmed by the constant rate values of degradation (kapp) as they gradually decrease from cycle to cycle. Therefore, the separation of the catalyst from the suspension by an economical and efficient technique and, afterward, treatment to remove adsorbed substances should be studied in future studies. To evaluate the effectiveness of photocatalytic treatment in HWW amelioration, the physicochemical parameters Abs254, total phenols, BOD5, and COD were measured before and after substrate treatment [Table 5].
Figure 7. Degradation kinetics of PXT (C0 = 20 µg/L) by reusing TiO2 P-25 catalyst (300 mg/L) under natural solar irradiation as a function of QUV (A) and t30W (B). PXT: Paroxetine.
Pseudo-first-order rate constant (kapp), correlation coefficient (R2), % removal of PXT, and degree of degradation kinetic constant reduction (%Δkapp) for each catalytic cycle
| TiO2 P-25 | PXT (20 μg/L) | |||
| C = 300 mg/L | kapp (L/kJ) | R2 | % | Δk% |
| 1st catalytic cycle | 0.051 | 0.9932 | 78 | - |
| 2nd catalytic cycle | 0.023 | 0.8876 | 66 | 54.9 |
| 3rd catalytic cycle | 0.008 | 0.9276 | 32 | 65.2 |
Abs254, total phenols, BOD5, COD values, and the BOD5/COD ratio of HWW before and after treatment by TiO2 in three consecutive PCC
| TiO2 (mg/L) | Abs254 (nm) | Total phenols (mg/L) | COD (mg/L) | BOD5 (mg/L) | BOD5/COD | |||||||||
| 0 min | 300 min | % variation | 0 min | 300 min | % variation | 0 min | 300 min | % variation | 0 min | 300 min | % variation | 0 min | 300 min | |
| 200 mg/L | 0.1596 | 0.1526 | -4.39 | 1.241 | 1.102 | -11.2 | 142 | 40 | -71.8 | 47.9 | 23.6 | -50.7 | 0.34 | 0.59 |
| 300 mg/L (1st PCC) | 0.1793 | 0.1763 | -1.67 | 1.079 | 0.948 | -12.1 | 23 | 10 | -56.5 | 2.1 | 5.6 | 166.7 | 0.09 | 0.56 |
| 300 mg/L (2nd PCC) | 0.1699 | 0.1669 | -1.77 | 1.468 | 1.281 | -12.7 | 31 | 10 | -67.7 | 8.7 | 1.7 | -80.5 | 0.28 | 0.17 |
| 300 mg/L (3rd PCC) | 0.1652 | 0.1597 | -3.23 | 0.659 | 0.641 | -2.61 | 47 | 20 | -57.4 | 5.6 | 4.2 | -25.0 | 0.12 | 0.21 |
Results showed a reduction in Abs254 and total phenolic content after treatment, as shown in Table 5. The BOD5/COD ratio was calculated to observe better the treated WW’s quality, where a ratio of 1 indicates complete biodegradability. In contrast, the separate analysis of BOD5 and COD alone does not fully explain the substrate’s quality in the organic load. An increase in the BOD5/COD ratio after each photocatalytic process was estimated in most cases, indicating enhanced biodegradability of the produced WW.
In vitro toxicity studies during the photocatalytic process (pilot-scale experiments)
Monitoring the toxicity of samples during photocatalytic processes is crucial for evaluating their effectiveness. The risk of increasing toxicity due to intermediate oxidation products makes it imperative to monitor toxicity levels during the application of AOPs[38,40]. The toxicity of the samples was evaluated by monitoring changes in the light intensity of the naturally luminescent bacterium Vibrio fischeri. The results obtained from samples collected at the beginning and end of each pilot-scale photocatalytic experiment are listed in Table 6. In general, the percentage of bioluminescence inhibition at the beginning of the photocatalytic treatment is generally low, except for the experiment with 200 mg/L of catalysts, where 65.81% bioluminescence inhibition was observed, indicating that this sample was highly toxic. In most cases, toxicity has been considerably decreased at the end of the process, while in some cases, the samples demonstrated the phenomenon of hormesis. Hormesis is the phenomenon in which the sample is not toxic enough to reduce the population of the Vibrio fischeri microorganism; instead, it contributes to its growth and, consequently, the intensity of the emitting light. In conclusion, significant or total detoxification of WW was demonstrated, and the elimination of PXT and its TPs and the amelioration of WW effluent quality with the decomposition of other organic compounds contributed to the initial toxicity observed in each experiment.
Toxicity study of pilot-scale photocatalytic experiments with different catalyst concentrations and catalyst reuse
| Irradiation time (min) | % Bioluminescence inhibition of Vibrio fischeri | |||||
| TiO2 concentration | Cycles with TiO2 (300 mg/L) | |||||
| 200 mg/L | 300 mg/L | 500 mg/L | 1st cycle | 2nd cycle | 3rd cycle | |
| 0 | 65.81 | 17.90 | 21.11 | 13.59 | 20.74 | 15.73 |
| 300 | 30.85 | Hormesis | 3.75 | 6.72 | Hormesis | 4.89 |
CONCLUSIONS
In lab-scale experiments, PXT concentration was degraded within a short time, up to 45 min, with slower removal of TOC and release of fluorine and nitrate anions. Thirteen TPs were identified, denoting hydroxylation, carbonylation, ether bond scission, fluorine atom substitution, and loss of CO2 molecule degradation pathways of the parent compound and its derivatives.
Assessing the in silico ecotoxicity of PXT and its TPs revealed, in most cases, less toxic TPs than the parent compound for the three species of aquatic organisms studied (fish, daphnid, and green algae). TP-328 and TP1-346 were identified as those TPs posing higher toxicity. Based on the TEST software predictions for mutagenicity, BCF, and developmental toxicity values, all TPs were classified as “developmental toxicant”, with only TP-210 and TP-226 classified as mutagenic. The BCF values show that PXT is the only compound that tends to accumulate more in aquatic organisms.
In CPC pilot-scale photocatalytic experiments under environmentally relevant conditions, the most efficient PXT removal was achieved, with the catalyst’s concentration being 200 mg/L. The reusability of the catalysts for three cycles indicated partial deactivation and the need for effective separation and regeneration to maintain catalytic efficiency. Pilot-scale photocatalytic treatment led to less or non-toxic (hormesis phenomenon) WW effluents.
In summary, TiO2 P-25-photocatalysis effectively removes PXT and its TPs while ameliorating the physicochemical characteristics of effluents; however, more practical applications need to be explored, and future research should focus on efficient recovery and reactivation methods for photocatalysts. Additionally, reactor engineering, the synthesis of stable and efficient visible light catalysts, the influence of matrices on the catalysts’ surface, and the estimation of toxicity in mixtures of contaminants or their metabolites are crucial for incorporating heterogeneous photocatalysis into WWTPs, particularly for eliminating PhACs.
The main purpose of employing heterogeneous photocatalysis as an alternative and more efficient method to conventional WW treatment lies in the degradation of toxic compounds, such as PhACs, that are not fully degraded in conventional WWTPs and can subsequently affect humans. Effectively managing WW in an economical and environmentally friendly manner is a significant challenge for society, which aims to eliminate sources of aquifer pollution while safeguarding human health.
DECLARATIONS
Acknowledgments
The authors would like to thank the Unit of Environmental, Organic, and Biochemical high-resolution analysis-ORBITRAP-LC–MS of the University of Ioannina for providing access to the facilities.
Authors’ contributions
Investigation, formal analysis, visualization, writing - original draft and editing: Sioulas, S.
Investigation and formal analysis: Rapti, I.; Kosma, C.
Conceptualization, project administration, methodology, supervision, resources, writing - original draft preparation, and writing - review and editing: Konstantinou, I.
Conceptualization, project administration, methodology, supervision, resources: Albanis, T.
Availability of data and materials
The data presented in this study are available in this article and the associated Supplementary Materials.
Financial support and sponsorship
This research is co-financed by Greece and the European Union (European Social Fund- ESF) through the Operational Programme «Human Resources Development, Education and Lifelong Learning 2014-2020» in the context of the project “Chemistry and Technologies for Pollution Control and Environmental Protection (MIS 6004807)”.
Conflicts of interest
Konstantinou, I. is the Guest Editor of the Special Issue “Recent Progress on Photocatalytic Treatment of Water and Wastewater” and an Editorial Board member of the journal Water Emerging Contaminants & Nanoplastics. Konstantinou, I. was no t involved in any steps of editorial processing, notably including reviewer selection, manuscript handling, and decision making. The other authors declared that there are no conflicts of interest.
Ethical approval and consent to participate
Not applicable.
Consent for publication
Not applicable.
Copyright
© The Author(s) 2025.
Supplementary Materials
REFERENCES
1. Ahmed, S.; Khan, F. S. A.; Mubarak, N. M.; et al. Emerging pollutants and their removal using visible-light responsive photocatalysis - a comprehensive review. J. Environ. Chem. Eng. 2021, 9, 106643.
2. Lykos, C.; Kourkouta, T.; Konstantinou, I. Study on the photocatalytic degradation of metronidazole antibiotic in aqueous media with TiO2 under lab and pilot scale. Sci. Total. Environ. 2023, 870, 161877.
3. Antonopoulou, M.; Kosma, C.; Albanis, T.; Konstantinou, I. An overview of homogeneous and heterogeneous photocatalysis applications for the removal of pharmaceutical compounds from real or synthetic hospital wastewaters under lab or pilot scale. Sci. Total. Environ. 2021, 765, 144163.
4. Kaur, A.; Umar, A.; Kansal, S. K. Heterogeneous photocatalytic studies of analgesic and non-steroidal anti-inflammatory drugs. Appl. Catal. A. Gen. 2016, 510, 134-55.
5. Kosma, C. I.; Kapsi, M. G.; Konstas, P. G.; et al. Assessment of multiclass pharmaceutical active compounds (PhACs) in hospital WWTP influent and effluent samples by UHPLC-Orbitrap MS: temporal variation, removals and environmental risk assessment. Environ. Res. 2020, 191, 110152.
6. Paíga, P.; Correia, M.; Fernandes, M. J.; et al. Assessment of 83 pharmaceuticals in WWTP influent and effluent samples by UHPLC-MS/MS: hourly variation. Sci. Total. Environ. 2019, 648, 582-600.
7. Hernando, M. D.; Mezcua, M.; Fernández-Alba, A. R.; Barceló, D. Environmental risk assessment of pharmaceutical residues in wastewater effluents, surface waters and sediments. Talanta 2006, 69, 334-42.
8. Gopinath, K. P.; Madhav, N. V.; Krishnan, A.; Malolan, R.; Rangarajan, G. Present applications of titanium dioxide for the photocatalytic removal of pollutants from water: a review. J. Environ. Manage. 2020, 270, 110906.
9. Frédéric, O.; Yves, P. Pharmaceuticals in hospital wastewater: their ecotoxicity and contribution to the environmental hazard of the effluent. Chemosphere 2014, 115, 31-9.
10. Hwang, S.; Kim, J. H.; Jo, S. H. Inhibitory effect of the selective serotonin reuptake inhibitor paroxetine on human Kv1.3 channels. Eur. J. Pharmacol. 2021, 912, 174567.
11. Silva, L. J.; Pereira, A. M.; Meisel, L. M.; Lino, C. M.; Pena, A. A one-year follow-up analysis of antidepressants in Portuguese wastewaters: occurrence and fate, seasonal influence, and risk assessment. Sci. Total. Environ. 2014, 490, 279-87.
12. Santoke, H.; Cooper, W. J. Environmental photochemical fate of selected pharmaceutical compounds in natural and reconstituted Suwannee River water: role of reactive species in indirect photolysis. Sci. Total. Environ. 2017, 580, 626-31.
13. Gornik, T.; Carena, L.; Kosjek, T.; Vione, D. Phototransformation study of the antidepressant paroxetine in surface waters. Sci. Total. Environ. 2021, 774, 145380.
14. Kwon, J. W.; Armbrust, K. L. Hydrolysis and photolysis of paroxetine, a selective serotonin reuptake inhibitor, in aqueous solutions. Environ. Toxicol. Chem. 2004, 23, 1394-9.
15. Carvalho, P. S.; de, M. C. C.; Ayala, A. P.; Ellena, J. X-Ray diffraction, spectroscopy and thermochemical characterization of the pharmaceutical paroxetine nitrate salt. J. Mol. Struct. 2016, 1118, 288-92.
16. Gurke, R.; Rößler, M.; Marx, C.; et al. Occurrence and removal of frequently prescribed pharmaceuticals and corresponding metabolites in wastewater of a sewage treatment plant. Sci. Total. Environ. 2015, 532, 762-70.
17. Lajeunesse, A.; Smyth, S. A.; Barclay, K.; Sauvé, S.; Gagnon, C. Distribution of antidepressant residues in wastewater and biosolids following different treatment processes by municipal wastewater treatment plants in Canada. Water. Res. 2012, 46, 5600-12.
18. Metcalfe, C. D.; Chu, S.; Judt, C.; et al. Antidepressants and their metabolites in municipal wastewater, and downstream exposure in an urban watershed. Environ. Toxicol. Chem. 2010, 29, 79-89.
19. Schlüsener, M. P.; Hardenbicker, P.; Nilson, E.; Schulz, M.; Viergutz, C.; Ternes, T. A. Occurrence of venlafaxine, other antidepressants and selected metabolites in the Rhine catchment in the face of climate change. Environ. Pollut. 2015, 196, 247-56.
20. Yuan, S.; Jiang, X.; Xia, X.; Zhang, H.; Zheng, S. Detection, occurrence and fate of 22 psychiatric pharmaceuticals in psychiatric hospital and municipal wastewater treatment plants in Beijing, China. Chemosphere 2013, 90, 2520-5.
21. Duarte, P.; Almeida, C. M. R.; Fernandes, J. P.; et al. Bioremediation of bezafibrate and paroxetine by microorganisms from estuarine sediment and activated sludge of an associated wastewater treatment plant. Sci. Total. Environ. 2019, 655, 796-806.
22. Ebele, A. J.; Abou-Elwafa, A. M.; Harrad, S. Pharmaceuticals and personal care products (PPCPs) in the freshwater aquatic environment. Emerg. Contam. 2017, 3, 1-16.
23. Li, R.; Li, T.; Zhou, Q. Impact of titanium dioxide (TiO2) modification on its application to pollution treatment - a review. Catalysts 2020, 10, 804.
24. Oturan, M. A.; Aaron, J. Advanced oxidation processes in water/wastewater treatment: principles and applications. a review. Crit. Rev. Env. Sci. Tec. 2014, 44, 2577-641.
25. Antoniou, M. G.; Zhao, C.; O’shea, K. E.; et al. CHAPTER 1. Photocatalytic degradation of organic contaminants in water: process optimization and degradation pathways. In: Dionysiou DD, Li Puma G, Ye J, Schneider J, Bahnemann D, editors. Photocatalysis. Cambridge: Royal Society of Chemistry; 2016. pp. 1-34.
26. Fotiou, D.; Lykos, C.; Konstantinou, I. Photocatalytic removal of the antidepressant fluoxetine from aqueous media using TiO2 P25 and g-C3N4 catalysts. J. Environ. Chem. Eng. 2024, 12, 111677.
27. Vagi, M. C.; Petsas, A. S. Recent advances on the removal of priority organochlorine and organophosphorus biorecalcitrant pesticides defined by Directive 2013/39/EU from environmental matrices by using advanced oxidation processes: an overview (2007-2018). J. Environ. Chem. Eng. 2020, 8, 102940.
28. Zhu, D.; Zhou, Q. Action and mechanism of semiconductor photocatalysis on degradation of organic pollutants in water treatment: a review. Environ. Nanotechnol. Monit. Manag. 2019, 12, 100255.
29. Suleiman Khasawneh O, Palaniandy P. Removal of organic pollutants from water by Fe2O3/TiO2 based photocatalytic degradation: a review. Environ. Technol. Innov. 2021, 21, 101230.
30. Fujishima, A.; Rao, T. N.; Tryk, D. A. Titanium dioxide photocatalysis. J. Photoch. Photobio. C. 2000, 1, 1-21.
31. Katal, R.; Masudy-Panah, S.; Tanhaei, M.; Farahani, M. H. D. A.; Jiangyong, H. A review on the synthesis of the various types of anatase TiO2 facets and their applications for photocatalysis. Chem. Eng. J. 2020, 384, 123384.
32. Velempini, T.; Prabakaran, E.; Pillay, K. Recent developments in the use of metal oxides for photocatalytic degradation of pharmaceutical pollutants in water - a review. Mater. Today. Chem. 2021, 19, 100380.
33. Gaya, U. I.; Abdullah, A. H. Heterogeneous photocatalytic degradation of organic contaminants over titanium dioxide: a review of fundamentals, progress and problems. J. Photoch. Photobio. C. 2008, 9, 1-12.
34. Mccullagh, C.; Skillen, N.; Adams, M.; Robertson, P. K. Photocatalytic reactors for environmental remediation: a review. J. Chem. Tech. Biotech. 2011, 86, 1002-17.
35. Rapti, I.; Kosma, C.; Albanis, T.; Konstantinou, I. Solar photocatalytic degradation of inherent pharmaceutical residues in real hospital WWTP effluents using titanium dioxide on a CPC pilot scale reactor. Catal. Today. 2023, 423, 113884.
36. Spasiano, D.; Marotta, R.; Malato, S.; Fernandez-Ibañez, P.; Di, S. I. Solar photocatalysis: materials, reactors, some commercial, and pre-industrialized applications. A comprehensive approach. Appl. Catal. B. Environ. 2015, 170-1, 90-123.
37. Sundar, K. P.; Kanmani, S. Progression of photocatalytic reactors and it’s comparison: a review. Chem. Eng. Res. Des. 2020, 154, 135-50.
38. Brienza, M.; Mahdi, A. M.; Escande, A.; et al. Use of solar advanced oxidation processes for wastewater treatment: follow-up on degradation products, acute toxicity, genotoxicity and estrogenicity. Chemosphere 2016, 148, 473-80.
39. Fernández-Ibáñez, P.; Blanco, J.; Malato, S.; de, N. F. J. Application of the colloidal stability of TiO2 particles for recovery and reuse in solar photocatalysis. Water. Res. 2003, 37, 3180-8.
40. Sousa, M.; Gonçalves, C.; Vilar, V. J.; Boaventura, R. A.; Alpendurada, M. Suspended TiO2-assisted photocatalytic degradation of emerging contaminants in a municipal WWTP effluent using a solar pilot plant with CPCs. Chem. Eng. J. 2012, 198-9, 301-9.
41. Konstas, P.; Kosma, C.; Konstantinou, I.; Albanis, T. Photocatalytic treatment of pharmaceuticals in real hospital wastewaters for effluent quality amelioration. Water 2019, 11, 2165.
42. Babaeivelni, K.; Khodadoust, A. P. Adsorption of fluoride onto crystalline titanium dioxide: effect of pH, ionic strength, and co-existing ions. J. Colloid. Interface. Sci. 2013, 394, 419-27.
43. Antonopoulou, M.; Dormousoglou, M.; Spyrou, A.; Dimitroulia, A. A.; Vlastos, D. An overall assessment of the effects of antidepressant paroxetine on aquatic organisms and human cells. Sci. Total. Environ. 2022, 852, 158393.
44. Cunningham, V. L.; Constable, D. J.; Hannah, R. E. Environmental risk assessment of paroxetine. Environ. Sci. Technol. 2004, 38, 3351-9.
45. Dmitriev, A.; Rudik, A.; Filimonov, D.; et al. Integral estimation of xenobiotics’ toxicity with regard to their metabolism in human organism. Pure. Appl. Chem. 2017, 89, 1449-58.
46. Diniz, R. R.; Domingos, T. F. S.; Pinto, G. R.; Cabral, L. M.; de, P. M.; de Souza, A. M. T. Use of in silico and in vitro methods as a potential new approach methodologies (NAMs) for (photo)mutagenicity and phototoxicity risk assessment of agrochemicals. Sci. Total. Environ. 2023, 904, 167320.
47. Toropova, A. P.; Toropov, A. A.; Roncaglioni, A.; Benfenati, E. The enhancement scheme for the predictive ability of QSAR: a case of mutagenicity. Toxicol. In. Vitro. 2023, 91, 105629.
48. Gómez-Regalado, M. D. C.; Martín, J.; Santos, J. L.; Aparicio, I.; Alonso, E.; Zafra-Gómez, A. Bioaccumulation/bioconcentration of pharmaceutical active compounds in aquatic organisms: assessment and factors database. Sci. Total. Environ. 2023, 861, 160638.
49. Khan, K.; Kumar, V.; Colombo, E.; Lombardo, A.; Benfenati, E.; Roy, K. Intelligent consensus predictions of bioconcentration factor of pharmaceuticals using 2D and fragment-based descriptors. Environ. Int. 2022, 170, 107625.
50. Sousa, M.; Gonçalves, C.; Pereira, J. H.; Vilar, V. J.; Boaventura, R. A.; Alpendurada, M. Photolytic and TiO2-assisted photocatalytic oxidation of the anxiolytic drug lorazepam (Lorenin® pills) under artificial UV light and natural sunlight: a comparative and comprehensive study. Sol. Energy. 2013, 87, 219-28.
51. Wetchakun, K.; Wetchakun, N.; Sakulsermsuk, S. An overview of solar/visible light-driven heterogeneous photocatalysis for water purification: TiO2- and ZnO-based photocatalysts used in suspension photoreactors. J. Ind. Eng. Chem. 2019, 71, 19-49.
52. Chen, D.; Cheng, Y.; Zhou, N.; et al. Photocatalytic degradation of organic pollutants using TiO2-based photocatalysts: a review. J. Clean. Prod. 2020, 268, 121725.
53. Chong, M. N.; Jin, B.; Chow, C. W.; Saint, C. Recent developments in photocatalytic water treatment technology: a review. Water. Res. 2010, 44, 2997-3027.
Cite This Article
Open AccessHow to Cite
Sioulas, S.; Rapti, I.; Kosma, C.; Konstantinou, I.; Albanis, T. Photocatalytic degradation of the antidepressant drug Paroxetine using TiO2 P-25 under lab and pilot scales in aqueous substrates. Water Emerg. Contam. Nanoplastics 2025, 4, 5. http://dx.doi.org/10.20517/wecn.2024.73
Download Citation
Export Citation File:
Type of Import
Tips on Downloading Citation
Citation Manager File Format
Type of Import
Direct Import: When the Direct Import option is selected (the default state), a dialogue box will give you the option to Save or Open the downloaded citation data. Choosing Open will either launch your citation manager or give you a choice of applications with which to use the metadata. The Save option saves the file locally for later use.
Indirect Import: When the Indirect Import option is selected, the metadata is displayed and may be copied and pasted as needed.
About This Article
Special Issue
Copyright
Data & Comments
Data
0
Comments
Comments must be written in English. Spam, offensive content, impersonation, and private information will not be permitted. If any comment is reported and identified as inappropriate content by OAE staff, the comment will be removed without notice. If you have any queries or need any help, please contact us at [email protected].