Evaluation of Three Sample Preparation Methods for LC-HRMS Suspect Screening of Contaminants of Emerging Concern in Effluent Wastewater

Evaluation of Three Sample Preparation Methods for LC-HRMS Suspect Screening of Contaminants of Emerging Concern in Effluent Wastewater Dana Orlando-Véliz,*Rocío Inés Bonansea, Manuel García-Vara, Varvara Nikolopoulou, and Miren López de Alda* Cite This: Anal. Chem. 2025, 97, 25946−25957 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information ABSTRACT: Suspect screening is a valuable tool for studying the pollution footprint of environmental samples, but the sample preparation (SP) method significantly affects the results. In this study, three SP methods�lyophilization, direct injection, and online solid-phase extraction (SPE)�were evaluated for suspect screening of contaminants of emerging concern (CECs) in effluent wastewater using liquid chromatography-high resolution mass spectrometry (LCHRMS). LC-HRMS was performed with a Q-ToF mass analyzer in both positive and negative electrospray ionization modes, and pollutants were identified using the NORMAN SusDat database. A total of 119 CECs were identified with high confidence (all above level 2a). Of these, 115 were detected using lyophilization, 37 with direct injection, and 49 with online SPE. Principal component analysis revealed distinct patterns for each SP method. In all cases, the pollution profile was dominated by pharmaceuticals, followed by industrial chemicals. Articaine, a local anesthetic previously unreported in the aquatic environment, was the most abundant compound found. Industrial chemicals included phthalates, flame retardants, benzotriazoles, and perfluoroalkyl substances, among others. The remaining identified CECs were personal care products, pesticides, and drugs of abuse. Compound prioritization, based on abundance, persistence, mobility, bioaccumulation potential, and toxicity, identified the antibiotics clindamycin and tiamulin, the analgesic flufenamic acid, the musk fragrance metabolite galaxolidone, and the antidepressant venlafaxine metabolite, O-desmethyl venlafaxine, as top priority. These results highlight the importance of applying an appropriate SP method to obtain a comprehensive CEC footprint in suspect screening analysis and demonstrate the suitability of lyophilization for wastewater characterization. ■INTRODUCTION The exponential population growth and its associated economic activities have increased the water shortage globally, forcing authorities to implement various measures to preserve water quality. 1 In Europe, wastewater reuse has become one of the main strategies to face this problem. Today, reclaimed water is present in many areas such as agricultural irrigation, aquifer recharge, landscaping, municipal sanitation, and even indirect potable water purification for human consumption. 2−4 The minimum quality parameters required based on the intended use of treated water are compiled in the Regulation EU 2020/741. 5 Although some wastewater treatment plants (WWTP) employ advanced remediation processes, they do not guarantee the complete removal of all organic micropollutants and this is strongly evident in the case of conventional treatment plants. 6−8 As a result, these effluents become a source of pollutants emission into the environment. These micropollutants are referred to as contaminants of emerging concern (CECs), chemical substances whose environmental fate and impact are not yet well characterized. The increasing environmental monitoring programmes worldwide and the improvement of the analytical methodologies and instrumentation in recent years have broadened the CECs chemical space. 4,8 These compounds exhibit a wide range of physical-chemical properties (polarity, structure, solubility, etc.) and uses such as pharmaceuticals, 9−11 personal care products, 12 disinfection byproducts, pesticides, 13 flame retardants, 14,15 industrial chemicals, 16 their derivatives, and other classes. 17,18 To minimize the impact of CECs in the aquatic environment and ensure water sustainability, the EU Directive 2013/39 19 established a list of priority substances to control chemical pollution in surface waters. Additionally, in 2015, it was introduced the concept of Watch List, which includes substances that may pose a significant risk to or via the aquatic Received: March 19, 2025 Revised: October 30, 2025 Accepted: November 5, 2025 Published: November 19, 2025 Articlepubs.acs.org/ac © 2025 The Authors. Published by American Chemical Society 25946 https://doi.org/10.1021/acs.analchem.5c01659 Anal. Chem. 2025, 97, 25946−25957 This article is licensed under CC-BY 4.0 Downloaded via CSIC on December 3, 2025 at 07:35:27 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. environment. 20 The purpose of this list is to monitor these substances and assess their potential risks in order to support future prioritization exercises. Most CECs are not regulated or even known and, therefore, their discharge into the environment is not controlled. The inherent hazard associated with these pollutants is determined by their intrinsic physical-chemical properties, which influence their environmental risk and fate. 21 CECs can be persistent in the environment, mobile in aqueous media, and exhibit toxicity or bioaccumulation in aquatic organisms. CECs can subsequently result in ecosystem deterioration and enter the human trophic chain becoming a human health risk. Monitoring CECs is a valuable tool to control their environmental occurrence. However, the successful detection of some of these compounds can be highly challenging due to their diverse chemical properties and the complexity of environmental matrices. Targeted monitoring programs handle a very narrow range of CECs and, therefore, do not provide a complete characterization of the CEC footprint in the analyzed samples. In the last decades, liquid chromatography coupled to high resolution mass spectrometry (LC-HRMS) has led to the development of suspect screening (SS) and nontarget screening (NTS) methodologies, which have significantly widened CEC surveillance. In both techniques, the range of compounds covered by the method is not known a priori. However, suspect-type analyses allow the investigation of socalled “known unknowns” by comparing acquired spectral features against suspect lists from databases. 22 Depending on the parameters contrasted, i.e., exact mass match, isotopic pattern fit, coincidence and match of fragmentation profile, etc. the confidence level of identification achieved will be different. 23 In this type of analyses, sample treatment becomes a critical step, as selective extraction and cleanup steps can result in the loss of many CECs while more universal techniques can be affected by high matrix effects and/or low sensitivity. In this context, optimizing sample preparation is essential for a holistic characterization of CECs. Therefore, the objectives of this work were (i) to evaluate the efficiency of three different sample preparation approaches for CEC analysis in effluent wastewater samples using LCHRMS suspect screening, (ii) to characterize the CECs present in the samples under study, and (iii) to prioritize the identified compounds based on their occurrence and potential environmental risk. ■EXPERIMENTAL SECTION Reagents and Chemicals. Information on the isotopically labeled compounds used as internal standards (IS) for data quality control is provided in Table S1 in Supporting Information. All solvents used were LC-HRMS grade. Acetonitrile (ACN), methanol (MeOH) and water used for LC-HRMS analysis were purchased from Thermo Fisher Scientific Inc. (Waltham, MA). MeOH, ethyl acetate (EtOAc) and water (H2O) used for sample preparation, and formic acid (purity >98%) and ammonium acetate used for chromatographic separation were purchased from Merck (Darmstadt, Germany). Case Study Area and Sample Collection. The WWTP under study is provided with conventional secondary treatment based on activated sludge. It is located in an industrial area in Barcelona province (Catalonia, Spain). This WWTP receives 40,000 m3/day of, mainly, industrial and urban wastewater. It has capacity for 300,000 inhabitants. From this WWTP, flowproportional 24h composite effluent wastewater samples were collected in three different days (Wednesday, Friday and Sunday) within a week in January 2022. Two fractions of 400 mL of the 24h composite effluent samples were transferred to glass amber bottles and stored at −20 °C until analysis. Sample Preparation Methods. Three different sample preparation methods were evaluated for their efficiency in the nonselective extraction of CECs from wastewater effluent samples for subsequent wide scope suspect screening analysis. In all cases, before any further processing, samples (400 mL) were spiked with a mixture of isotopically labeled standard compounds at a concentration of 1 μg/L to monitor analytical performance and instrumental sensitivity. Sample preparation method 1 (SP-1) is a modified version of an approach previously developed by our group. 6 It involves freeze-drying the sample, followed by redissolution of the residue in a series of solvents with different polarities, which is, in principle, a simple and cost-effective sample preconcentration method. 13,24−26 Specifically, in this method, 200 mL of the previously frozen sample was lyophilized using a LyoAlfa freeze-dryer (Telstar) at a final condenser temperature of −64 °C and a vacuum pressure of 0.031 mbar. This process requires 2−4 days to complete a batch of six samples. The residue was then reconstituted with 15 mL of MeOH, followed by 15 mL of EtOAc. The extracts were transferred to a glass centrifuge tube and centrifuged at 4000 rpm for 10 min (Eppendorf Centrifuge 5810R). The supernatant was then evaporated with a stream of N2at 10 psi to a final volume of 0.2 mL and reconstituted to 1 mL with HPLC-grade water. Finally, the obtained extracts were centrifuged at 10,000 rpm for 10 min and transferred to HPLC vials for subsequent analysis by LCHRMS. In sample preparation method 2 (SP-2), 2 mL of the sample was centrifuged under the same conditions described above, and the supernatant was transferred to an HPLC vial for direct injection (100 μL) into the LC-HRMS system. In sample preparation method 3 (SP-3), 2 mL of the sample was centrifuged and transferred to an HPLC vial as above and subjected to a subsequent online solid-phase extraction (online SPE) process. This process was performed using an Elute OLE (Online Extraction) system coupled to the LC-HRMS system and an ISOLUTE ENV+ online SPE cartridge (30 mm ×2.1 mm with a particle size of 40 μm) from Biotage (Uppsala, Sweden). The cartridge was first conditioned with 1 mL of ACN followed by 1 mL of H2O, both with a flow rate of 0.4 mL/min. Then, the water sample was loaded with 1.7 mL of a mixture of H2O/ACN at initial chromatographic conditions (95/5, v/v) at a flow rate of 0.3 mL min−1. A sample volume of 1.8 mL was loaded into the OLE system to fill the entire loop and 1 mL (full loop) of the sample was transferred to the chromatographic system. The ISOLUTE ENV+ online SPE cartridge is a polymeric cartridge that contains a hyper crosslinked polystyrene polymeric sorbent with a hydroxylated surface, designed for the extraction of compounds with a wide range of polarities from aqueous samples. Instrumental. LC-HRMS analysis was performed using an Elute UHPLC system coupled to an Impact II Q-TOF mass spectrometer (Bruker Daltonics, Billerica, MA) provided with a Vacuum Insulated Probe Heated Electrospray Ionization (VIP-HESI). For instrumental details, see Supporting Information SI 1. For data quality control, procedural blanks (HPLC water spiked with the IS mix) were prepared in the Analytical Chemistry pubs.acs.org/ac Article https://doi.org/10.1021/acs.analchem.5c01659 Anal. Chem. 2025, 97, 25946−25957 25947 Table 1. List of Identified Compounds in Effluent Samples Including Their CAS Number, Retention Time (RT), Molecular Formula, Ionization Mode, m/z, Chemical Category, Confidence Level, and Detected Method a N°compound CAS N° RT (min) molecular formula ESI m/zcategory confidence level detected method (*) 1 1,2,3-benzotriazole 95-14-7 5.8 C6H5N3 ±120.0561/118.0405 industrial chemicals 1 A, B, C 2 1,8-diazabicyclo [5.4.0]undec-7ene 6674-22-2 3.6 C9H16N2 + 153.1392 industrial chemicals 2a A, C 3 10,11-dihydro-10,11dihydroxycarbamazepine 35079-97-1 6.7 C15H14N2O3 + 271.1083 pharmaceuticals 2a A 4 10-hydroxycarbazepine 29331-92-8 7.3 C15H14N2O2 + 255.1134 pharmaceuticals 2a A 5 1-naphthol 90-15-3 6.6 C10H8O −145.0653 pharmaceuticals/ pesticides 2a A, B 6 2,4-diaminotoluene 95-80-7 2.5 C7H10N2 + 123.0922 industrial chemicals 2a A, C 7 2-amino-4-cresol 95-84-1 2.4 C7H9NO + 124.0762 industrial chemicals 2a A, B, C 8 2-amino-6methylmercaptopurine 1198-47-6 4.2 C6H7N5S + 182.0500 pharmaceuticals 2a A, B 9 2-aminophenol/3-Hydroxy-2methylpyridine/Nicotinyl alcohol 95-55-6 2.2 C6H7NO + 110.0606 industrial chemicals 2a A, C 10 2-methoxy-5-methylaniline 120-71-8 3.6 C8H11NO + 138.0919 industrial chemicals 1 A, B, C 11 2-methyl-S-benzothiazole 615-22-5 11.7 C8H7NS2 + 182.0098 pesticide/ industrial chemicals 2a A, C 12 2-phenylbenzimidazole-5sulfonic acid (Ensulizole) 27503-81-7 4.8 C13H10N2O3S + 275.0490 industrial chemicals 2a A 13 3,5-ditert-Butyl-4hydroxybenzoic acid 1421-49-4 12.8 C15H22O3 + 251.1647 industrial chemicals 2a A 14 3-[(hexanoyloxy)ethanimidoyl]- 1H-pyrrole - 8.1 C12H18N2O2 + 223.1447 other 2a A 15 4,4′-dihydroxybiphenyl - 9.4 C12H10O2 −187.0759 industrial chemicals 2a A, C 16 4-acetamidoantipyrine 83-15-8 5.3 C13H15N3O2 + 246.1243 pharmaceuticals 2a A, B, C 17 4-formylaminoantipyrine 1672-58-8 5.2 C12H13N3O2 + 232.1086 pharmaceuticals 2a A 18 4-methylbenzotriazole 29878-31-7 7.0 C7H7N3 ±134.0718/132.0561 industrial Chemicals 2a A, B, C 19 5-methyl-1H-benzotriazole 136-85-6 7.1 C7H7N3 ±134.0718/132.0562 industrial Chemicals 1 A, B, C 20 6-methoxyquinoline 5263-87-6 4.1 C10H9NO + 160.0762 industrial Chemicals 2a A 21 6-methyl-2-pyridinemethanol 1122-71-0 3.6 C7H9NO + 124.0762 industrial Chemicals 2a A, B, C 22 acetaminophen 103-90-2 5.9 C8H9NO2 −152.0711 pharmaceuticals 2a A 23 amantadine 768-94-5 5.3 C10H17N + 152.1439 pharmaceuticals 2a A 24 amisulpride 71675-85-9 5.7 C17H27N3O4S + 370.1801 pharmaceuticals 2a A, C 25 amphetamine 300-62-9 3.6 C9H13N + 136.1126 drugs of abuse 2a A 26 ampyrone 83-07-8 4.1 C11H13N3O + 204.1137 pharmaceuticals 2a A 27 articaine 23964-58-1 6.0 C13H20N2O3S + 285.1273 pharmaceuticals 2a A, B, C 28 atenolol 29122-68-7 4.0 C14H22N2O3 + 267.1709 pharmaceuticals 2a A, B, C 29 atenolol acid (metoprolol acid) 56392-14-4 5.1 C14H21NO4 + 268.1549 pharmaceuticals 2a A, B, C 30 benzydamine 642-72-8 8.8 C19H23N3O + 310.1919 pharmaceuticals 2a A 31 betahistine 5638-76-6 3.9 C8H12N2 + 137.1079 pharmaceuticals 2a B, C 32 bis(2-ethylhexyl) amine 106-20-7 10.4 C16H35N + 242.2848 industrial Chemicals 2a A, C 33 bisoprolol 66722-44-9 7.2 C18H31NO4 + 326.2331 pharmaceuticals 2a A, B, C 34 bupropion 34911-55-2 7.2 C13H18ClNO + 240.1155 pharmaceuticals 2a A 35 caffeine 58-08-2 5.2 C8H10N4O2 + 195.0882 other 2a A 36 caprolactam 105-60-2 4.6 C6H11NO + 114.0919 industrial chemicals 1 A 37 carbamazepine 298-46-4 9.3 C15H12N2O + 237.1028 pharmaceuticals 2a A, C 38 chlorthiazide 58-94-6 4.8 C7H6ClN3O4S2 −295.9567 pharmaceuticals 2a A 39 citalopram 59729-33-8 8.4 C20H21FN2O + 325.1716 pharmaceuticals 1 A, B, C 40 clarithromycin 81103-11-9 9.3 C38H69NO13 + 748.4847 pharmaceuticals 1 A 41 clindamycin 18323-44-9 7.0 C18H33ClN2O5S + 425.1877 pharmaceuticals 2a A, B, C 42 clopidogrel carboxylic acid 144457-28-3 6.8 C15H14ClNO2S + 308.0512 pharmaceuticals 2a A 43 cordycepin 73-03-0 2.3 C10H13N5O3 + 252.1097 pharmaceuticals 2a A Analytical Chemistry pubs.acs.org/ac Article https://doi.org/10.1021/acs.analchem.5c01659 Anal. Chem. 2025, 97, 25946−25957 25948 Table 1. continued N°compound CAS N° RT (min) molecular formula ESI m/zcategory confidence level detected method (*) 44 darunavir 206361-99-1 11.3 C27H37N3O7S + 548.2431 pharmaceuticals 2a A 45 decanamide 2319-29-1 11.4 C10H21NO + 172.1701 other 2a A 46 decanophenone 6048-82-4 13.3 C16H24O + 233.1905 other 2a A, B, C 47 DEET 134-62-3 10.3 C12H17NO + 192.1388 pesticides 1 A, C 48 desacetyl diltiazem 42399-40-6 7.7 C20H24N2O3S + 373.1586 pharmaceuticals 2a A, C 49 desmethylcitalopram 62498-67-3 8.3 C19H19FN2O + 311.1560 pharmaceuticals 2a A, C 50 dextromethorphan 125-71-3 7.9 C18H25NO + 272.2014 pharmaceuticals 2a A 51 dibutyl adipate 105-99-7 13.5 C14H26O4 + 259.1909 industrial chemicals 2a A 52 dibutyl hydrogen phosphate 107-66-4 13.8 C8H19PO4 + 211.1099 industrial chemicals 2a C 53 dibutyl phthalate 84-74-2 15.2 C16H22O4 + 279.1596 industrial chemicals 2a A, C 54 diclofenac 15307-86-5 12.7 C14H11Cl2NO2 + 296.0245 pharmaceuticals 2a A 55 diethyl phthalate 84-66-2 11.7 C12H14O4 + 223.0970 industrial chemicals 2a A 56 diltiazem 56209-45-1 8.6 C22H26N2O4S + 415.1692 pharmaceuticals 2a A, C 57 diphenyl phosphate 838-85-7 6.5 C12H11O4P −251.0473 industrial chemicals 2a A 58 dodecyl sulfate 151-41-7 10.0 C12H26O4S −267.1630 industrial chemicals 2a A 59 dodecylbenzenesulfonic acid 121-65-3 11.1 C18H30O3S −327.1994 industrial chemicals 2a A 60 doxylamine 76210-47-4 4.9 C17H22N2O + 271.1810 pharmaceuticals 2a A, C 61 erucamide 112-84-5 20.3 C22H43NO + 338.3423 industrial chemicals 2a A 62 esomeprazole 119141-88-7 6.8 C17H19N3O3S + 346.1225 pharmaceuticals 2a A 63 fenofibric acid 42017-89-0 12.7 C17H15ClO4 + 319.0737 pharmaceuticals 2a A 64 fexofenadine 83799-24-0 9.3 C32H39NO4 + 502.2957 pharmaceuticals 2a A 65 flecainide 54143-55-4 8.6 C17H20F6N2O3 + 415.1456 pharmaceuticals 1 A, B 66 florfenicol 73231-34-2 7.8 C12H14Cl2FNO4S −358.0083 pharmaceuticals 2a A 67 fluconazole 86386-73-4 6.6 C13H12F2N6O + 307.1119 pharmaceuticals 2a A 68 flufenamic acid 530-78-9 13.6 C14H10F3NO2 ±282.0741/280.0585 pharmaceuticals 2a A, B, C 69 gabapentin 60142-96-3 4.5 C9H17NO2 + 172.1338 pharmaceuticals 2a A 70 galaxolidone 507442-49-1 15.3 C18H24O2 + 273.1855 other 2a A, B, C 71 gemcitabine 95058-81-4 1.9 C9H11F2N3O4 + 264.0796 pharmaceuticals 2a A 72 hexamethylenetetramine 100-97-0 1.6 C6H12N4 + 141.1140 pharmaceuticals 2a B 73 hydrochlorothiazide 58-93-5 5.6 C7H8ClN3O4S2 −297.9723 pharmaceuticals 2a A, B 74 irbesartan 138402-11-6 9.5 C25H28N6O + 429.2403 pharmaceuticals 2a A, C 75 ketamine 6740-88-1 5.8 C13H16ClNO + 238.0999 pharmaceuticals 1 A, B 76 ketoprofen 22071-15-4 11.0 C16H14O3 + 255.1021 pharmaceuticals 2a A 77 lacosamide 175481-36-4 6.6 C13H18N2O3 + 251.1396 pharmaceuticals 2a A 78 lamotrigine 84057-84-1 6.2 C9H7Cl2N5 + 256.0157 pharmaceuticals 2a A, B, C 79 levamisole 14769-73-4 4.9 C11H12N2S + 205.0800 pharmaceuticals 2a A 80 lidocaine 137-58-6 5.7 C14H22N2O + 235.1810 pharmaceuticals 2a A 81 lincomycin 154-21-2 3.5 C18H34N2O6S + 407.2216 pharmaceuticals 2a A 82 losartan 114798-26-4 10.0 C22H23ClN6O + 423.1700 pharmaceuticals 2a A, C 83 melamine 108-78-1 1.7 C3H6N6 + 127.0732 industrial chemicals 1 A 84 mepivacaine 96-88-8 5.7 C15H22N2O + 247.1810 pharmaceuticals 2a A 85 metformin 657-24-9 1.6 C4H11N5 + 130.1093 pharmaceuticals 2a A, B, C 86 methocarbamol 532-03-6 6.9 C11H15NO5 + 242.1029 pharmaceuticals 2a A 87 mirtazapine 85650-52-8 6.0 C17H19N3 + 266.1657 pharmaceuticals 2a A, B 88 m-Xylene-4-sulfonic acid (2,4Dimethylbenzenesulfonic acid) 88-61-9 5.0 C8H10O3S −187.0429 other 2a A, B 89 N,N′-diphenylguanidine (DPG) 102-06-7 6.0 C13H13N3 + 212.1188 industrial chemicals 1 A, C 90 N-[(S)-(+)-1-ethoxycarbonyl-3phenylpropyl]-L-alanine 82717-96-2 7.1 C15H21NO4 + 280.1549 pharmaceuticals 2a A 91 N-butylbenzenesulfonamide 3622-84-2 11.2 C10H15NO2S + 214.0902 industrial chemicals 2a A 92 N-ethylaniline 103-69-5 3.5 C8H11N + 122.0970 industrial chemicals 2a C 93 N-ethyl-N-methylcathinone 1157739-24-6 3.6 C12H17NO + 192.1388 drugs of abuse 2a A Analytical Chemistry pubs.acs.org/ac Article https://doi.org/10.1021/acs.analchem.5c01659 Anal. Chem. 2025, 97, 25946−25957 25949 same way as samples, and analyzed together with the samples, solvent blanks, and standards. Postacquisition Data Processing and Confirmation Procedure. The retrospective analyses of the samples involved comparison with the suspect list mzCloud (S19, https://www.norman-network.com/nds/SLE/), which contains approximately 8,000 CECs from various applications. This database is available at the NORMAN Substance Database. For more information about the data-dependent acquisition processing see SI 2, and for confirmation with references standards, see SI 3. Relative Peak Areas (RPA) were calculated by dividing the absolute chromatographic peak areas of each analyte by the peak area of the internal standard metconazole-d6, selected for its intermediate RT, and then taking the average of these ratios. Exploratory Analysis. Differences between sampling days and sample preparation methods were studied using Principal Component Analysis (PCA). 27,28 For this purpose, the peak areas of the identified compounds were extracted for each sample and evaluated using the Statistical Analysis (one factor) program through the MetaboAnalyst 6.0 platform. In this analysis, the three daily samples were treated as replicates of each type of sample treatment and the data were also grouped according to sample treatment type. Environmental Risk Assessment and Prioritization. To evaluate the persistence, the probability of aerobic and anaerobic biodegradation of organic compounds was estimated using QSAR models via EPI Suite version 4.1 employing BIOWIN 2, 3, and 6 models. These models were utilized to classify a substance as “Potentially Persistent (P) or very Persistent (vP)” according to the two options indicated in Table S2. Regarding bioaccumulation and mobility, an estimation of the intrinsic physical-chemical properties, such as logarithmic octanol−water (log Kow) and organic carbon− water (log Koc) partition coefficients, was performed using the same software. Compounds with log Kow > 4.5 were assigned as potentially bioaccumulative (B) or very bioaccumulative (vB) in aquatic biota, whereas for evaluating the mobility of compounds in soil, generic thresholds of log Koc < 4 and log Koc < 3 were used to assign a compound as mobile (M) or very mobile (vM), respectively (Table S2). Substance toxicity was assessed on the basis of the predicted no-effect concentration (PNEC) of the compound in freshwater. For substances included in the EU Watch Lists, the corresponding maximum acceptable method detection or quantification limit was taken as PNEC. Meanwhile, for priority substances listed in the Directive 2013/39/EU, 19 the PNEC corresponded to the lowest Environmental Quality Standards (EQS) in surface water. The other values were either obtained through experimental data or predicted by QSAR models from the NORMAN Ecotoxicology Database (https://www.norman-network.com/nds/ecotox/). Compounds were then prioritized taken into account all these parameters and their abundance. To this end, percentiles Table 1. continued N°compound CAS N° RT (min) molecular formula ESI m/zcategory confidence level detected method (*) 94 noramidopyrine 519-98-2 4.1 C12H15N3O + 218.1293 pharmaceuticals 2a A, B 95 O-desmethyl venlafaxine 93413-62-8 5.8 C16H25NO2 + 264.1964 pharmaceuticals 1 A 96 ofloxacin/levofloxacin 100986-85-4 5.7 C18H20FN3O4 + 362.1516 pharmaceuticals 2a A 97 omeprazole sulfone 88546-55-8 8.1 C17H19N3O4S + 362.1175 pharmaceuticals 2a A 98 oxcarbazepine 28721-07-5 8.4 C15H12N2O2 + 253.0977 pharmaceuticals 2a A, C 99 perfluorobutanesulfonic acid (PFBS) 375-73-5 8.2 C4H1F9O3S1 −300.9580 industrial chemicals 2a A, B 100 perfluorohexanoic acid (PFHxA) 307-24-4 8.0 C6H1F11O2 −314.9878 industrial chemicals 2a A 101 phenazone 60-80-0 6.2 C11H12N2O + 189.1028 pharmaceuticals 2a A, B, C 102 pilocarpine 92-13-7 5.7 C11H16N2O2 + 209.1290 pharmaceuticals 2a A 103 pirlimycin 79548-73-5 7.1 C17H31ClN2O5S + 411.1721 pharmaceuticals 2a A, B, C 104 propranolol 525-66-6 7.8 C16H21NO2 + 260.1651 pharmaceuticals 2a A, C 105 pyrimethanil 53112-28-0 8.9 C12H13N3 + 200.1188 pesticides 2a A 106 pyroquilon 57369-32-1 3.8 C11H11NO + 174.0919 pesticides 2a A 107 sitagliptin 486460-32-6 6.9 C16H15F6N5O + 408.1259 pharmaceuticals 1 A, B, C 108 sulfamethoxazole 144930-01-8 7.7 C10H11N3O3S + 254.0599 pharmaceuticals 1 A 109 sulpiride 15676-16-1 4.5 C15H23N3O4S + 342.1488 pharmaceuticals 1 A, B, C 110 tapentadol 175591-23-8 6.5 C14H23NO + 222.1858 pharmaceuticals 2a A, B, C 111 terbutryn 886-50-0 9.6 C10H19N5S + 242.1439 pesticides 1 A, C 112 tiamulin 55297-95-5 9.2 C28H47NO4S + 494.3304 pharmaceuticals 2a A, B, C 113 trazodone 19794-93-5 7.5 C19H22ClN5O + 372.1591 pharmaceuticals 2a A 114 tritbutyl phosphate 126-73-8 14.0 C12H27O4P + 267.1725 industrial chemicals 1 A, B, C 115 tributylamine 102-82-9 7.8 C12H27N + 186.2222 industrial chemicals 1 A 116 triethyl phosphate 78-40-0 7.7 C6H15O4P + 183.0786 industrial chemicals 1 A 117 trimethoprim 738-70-5 5.4 C14H18N4O3 + 291.1457 pharmaceuticals 1 A 118 tris(2-butoxyethyl) phosphate 78-51-3 14.7 C18H39O7P + 399.2512 industrial chemicals 1 A, B, C 119 venlafaxine 93413-69-5 7.2 C17H27NO2 + 278.2120 pharmaceuticals 1 A a (*)A: Lyophilization, B: Direct injection, C: Online SPE Analytical Chemistry pubs.acs.org/ac Article https://doi.org/10.1021/acs.analchem.5c01659 Anal. Chem. 2025, 97, 25946−25957 25950 (20th, 40th, 60th, and 80th) were calculated for the PNEC and the relative peak area parameters and were assigned the scores described in Table S3. ■RESULTS AND DISCUSSION Identified Compounds. After peak picking, 2012, 1015, and 1749 features were annotated with SP methods 1, 2, and 3, respectively, and library matching reduced afterward these numbers to 115, 37, and 49 identifications. In total, 119 compounds were found in the effluent wastewater samples analyzed. Among them, 22 compounds were further confirmed with a confidence level of 1, “Confirmed structure”, after measuring the corresponding reference standard. The confirmation of these compounds was done using the retention time, exact mass, and the mSigma and MS2scores obtained through the MetaboScape software. The remaining 97 compounds were tentatively identified with a level 2a, “Probable structure”, according to Schymanski et al. scale. 23 This involved the matching of the exact mass and MS2 fragments from the spectral library. It is worth noting that for some structural isomers, an unambiguous assignment could not be made due to their identical fragmentation patterns. This was the case for compounds such as ofloxacin/levofloxacin, and 2-aminophenol/3-hydroxy-2-methylpyridine/nicotinyl alcohol. The identification details, along with the corresponding compound categories and PNECs, are provided in Table 1. Additionally, MS2fragments matching the mass spectrometry database are available in Table S4, while the chromatographic peak areas obtained for each compound in each sample treatment are reported in Table S5 (SP method-1), Table S6 (SP method-2), and Table S7 (SP method-3). The identified compounds were predominantly pharmaceuticals, industrial chemicals, pesticides, and drugs of abuse. Pharmaceuticals were the most abundant (in terms of peak areas) and frequently detected category of compounds, accounting for 61.3% of detections in effluent wastewater samples (73 compounds and metabolites) (Figure 1,Figures S1 and S2). This category included antibiotics, analgesics, antihypertensive agents, and antidepressants, among others. The highest RPA was obtained for articaine, a local anesthetic used in dental interventions, whose presence, to our knowledge, has not been previously reported in environmental fate studies. In second place, lamotrigine, an antiepileptic agent commonly found in conventional wastewater effluents due to its high persistence and low biodegradation in activated sludge treatments, was also detected at elevated concentrations. 10,11,17 O-desmethyl venlafaxine also showed high abundance, and greater than that of its parent compound, likely due to its slower biodegradation rate (the apparent elimination half-life is 5±2 h for venlafaxine and 11 ±2 h for their metabolite). 29 The presence of O-desmethyl venlafaxine in effluent wastewater results from both human metabolism and degradation in WWTP processes. 11,30 Other interesting identifications with RPAs greater than the 75th percentile corresponded to the antibiotics clindamycin and tiamulin. These compounds pose a potential risk to the environment due to the emergence of antibiotic resistance. 31 In addition, at least three analgesics, including 4-acetamidoantipyrine, tapentadol, and phenazone, were found. Figure 1. Relative peak area of the most abundant compounds by category (complete figure with all compounds in Figure S1). Analytical Chemistry pubs.acs.org/ac Article https://doi.org/10.1021/acs.analchem.5c01659 Anal. Chem. 2025, 97, 25946−25957 25951 Compounds used in industrial applications were the second most detected group of contaminants (a total of 33 compounds and metabolites), accounting for 27.7% of the total identifications (Figure 1 and Figures S1B and S2). This group included phthalates, flame retardants, benzotriazoles, and perfluoroalkyl compounds, among others. These substances are considered an emerging environmental concern due to their intrinsic physical-chemical properties, such as persistence and mobility. 21 Within this category, many compounds have multiple applications, making it difficult to trace their exposure sources. For example, caprolactam is used to synthesize Nylon 6 fibers but is also employed in film coatings, synthetic leather, vehicle painting, and as a plasticizer. Among the identified industrial chemicals, benzotriazole and its metabolites 4(5)-methyl benzotriazole occupy top positions in the RPA list. The occurrence of these compounds in natural water bodies has been frequently reported in the literature, as they may persist in the environment for long periods due to their low volatility, slow biodegradation, and high polarity, making their removal in conventional WWTPs difficult. 16,21,32 Organophosphorus flame retardants, such as tris(2-butoxyethyl) phosphate (TBEP), tributyl phosphate and dibutyl hydrogen phosphate (DBHP or DBP), were another category detected in effluent samples with high RPAs. These nonchlorinated esters are primarily used as plasticizers, antifoaming agents, and additives. 33,34 Toxicological studies have linked TBEP to toxic effects that pose risks to both the environment and human health. 35 2,4-Diaminotoluene also presented a high pollution load, possibly due to its multiple uses in hair dyes and as an intermediate for dyes, polymers, and other chemicals. It is classified as a human carcinogen by the Environmental Protection Agency (EPA). 36 The last compound within the 75th percentile of the RPA for this category is N,Ndiphenylguanidine, a guanidine-derivative used as a vulcanization accelerator in rubber products such as tires. This substance is classified as a persistent and mobile organic compound (PMOC) with accumulation potential in the aquatic environment. 21,37 In the case of pesticides (4.2% of the total identifications, Figure S2), five compounds used as fungicides, herbicides, and insecticides were detected. Terbutryn, a priority substance regulated in surface waters under Directive 2013/39/EU 19 was identified in water samples. Despite its regulation, these results indicate the difficulty of removing it from the aquatic environment. Since the WWTP is not located in an agricultural area, these compounds may originate from domestic sewage. Finally, the urban origin of the wastewater also resulted in the presence of the drugs of abuse amphetamine and N-ethylN-methylcathinone (1.7% of the total identifications, Figure S2), the stimulant caffeine, and galaxolidone, a metabolite of the personal care product galaxolide. Exploratory Analysis of the Sample Preparation Methods. The normalized peak areas of the identified compounds were used to perform the PCA. Figure 2 shows the PCA score plot (PC1 vs PC2) of the two principal components explaining the largest amount of variance (58.2% and 28.9%, respectively). This graph shows three clusters of samples corresponding to the different sample treatments applied (SP-1, SP-2 and SP-3). Specifically, PC1 clearly separates between lyophilized water samples (SP-1) and those that are only centrifuged before injection in the (online SPE) LC-HRMS system (SP-2 and SP-3). On the other hand, PC2 differentiates between the other two sample preparation methods, SP-2 and SP-3. These aggregations indicate that the pattern of CECs, in terms of presence or abundance, exhibits differences within samples subjected to distinct SP Figure 2. Score diagram of PC1 vs PC2, depicting three distinct clustering of samples corresponding with each sample treatment (SP1_lyophilization “Lyo”, SP-2_direct injection “DI”, and SP-3_online SPE “OLE”). Analytical Chemistry pubs.acs.org/ac Article https://doi.org/10.1021/acs.analchem.5c01659 Anal. Chem. 2025, 97, 25946−25957 25952 methods. However, the amount of variation between replicates within each SP method is barely noticeable. This shows the high reproducibility of the analysis and, on the other hand, the similar contaminant load of the identified CECs in the various tested weekdays. The PCA loading plot in Figure S3 represents the variables (identified compounds) of the model. This diagram indicates that compounds with the highest values in PC1 and/or PC2 are the most important in explaining the greatest variability in the data set, regardless of whether the contribution is positive or negative. Compounds Identified with Each Sample Preparation Method. Sample preparation methodologies play a crucial role in the outcome of the analysis. Certain pretreatment conditions can modify the sensitivity and selectivity of the analytical method, as corroborated by the exploratory analysis in the previous section. To further understand these effects, several aspects of the three sample preparation methods are discussed and compared here. Figure 3 shows the number of compounds within each category of use detected with each of the sample preparation methods tested, and the sum of their corresponding peak areas (the data is available in TS5-S7). Out of the total 119 compounds found in effluent wastewater, 115 were identified through sample lyophilization (SP-1), 37 with direct injection (SP-2), and 49 with online SPE (SP-3). These results demonstrate that lyophilization leads to the detection of a larger number of compounds, in spite of potential matrix effects increasing. In all cases, the predominant categories were, first, pharmaceuticals and, then, industrial chemicals. Pesticides were only detected after lyophilization and online SPE, not by direct injection, and drugs of abuse only after lyophilization, probably because of their low concentration in the treated wastewater. In the case of pharmaceuticals, it is observed that, despite identifying the highest number of these compounds using lyophilization, the sum of the peak areas does not increase proportionally. The same is observed in the case of pesticides when comparing lyophilization and online SPE. This suggests that many of these compounds did not exhibit a high contaminant load in the effluent wastewater and could experiment losses during lyophilization and/or high matrix effects. To evaluate the range of polarity, expressed as n-octanol/ water partition coefficient (log Kow), organic carbon−water partition coefficient (log Koc), and Henry’s law constant of the compounds identified with each sample pretreatment method, these coefficients were estimated with QSAR models using the EPI Suite program. The obtained values are summarized in Table S8 and represented in Figure S4. Out of the three sample pretreatment methods tested, lyophilization covered the broadest range of polarity, with log Kow values ranging from −2.34 to 8.44, followed by online SPE (log Kow between −2.34 and 7.91), and direct injection (log Kow between −4.15 and 5.60), which allowed the detection of more polar compounds than the other two. These polar compounds could have been lost in the lyophilization approach due to the solvents used for extract reconstitution (MeOH and EtOAc, but not H2O) and in the online SPE approach because of the characteristics of the polymeric sorbent used for extraction of the samples. Regarding log Koc, values up to 6.38, corresponding to the antidiabetic agent sitagliptin, were reached with all three methods. However, lyophilization exhibited the broadest range, reaching log Koc values as low as 0.82, corresponding to fexofenadine. Lyophilization was also the method affording the identification of the most volatile compounds. Figure 4 shows in a Venn diagram the number of compounds identified with each sample preparation method (SPM). Twenty-six compounds, with a polarity range between log Kow −2.34 and 5.60, were detected with all three methods. Twenty-three out of these 26 compounds showed high abundance in the samples, based on their chromatographic peak areas (TS5-S7), which explains their detection with all three methods, including the one that is, in principle, least sensitive (direct injection). The remaining three compounds exhibited medium to low polarity, which also facilitates their detection with all three methods. Sixty compounds were exclusively detected with the lyophilization method, one compound (the antibiotic hexamethylenetetramine, with log Kow −4.15) with the direct injection method, and two compounds (the organophosphorus flame retardant dibutyl hydrogen phosphate and the industrial chemical N-ethylaniline, with log Kow 2.29 and 2.16, respectively) with the online Figure 3. Number of CEC identified with each of the sample preparation methods tested, classified according to category of use, and sum of the corresponding peak areas. Figure 4. Van Venn diagrams showing the number of compounds identified with each sample pretreatment procedure and coincidences among them. Analytical Chemistry pubs.acs.org/ac Article https://doi.org/10.1021/acs.analchem.5c01659 Anal. Chem. 2025, 97, 25946−25957 25953 SPE method. Apart from these three compounds, the other compound missed by the lyophilization method and detected with the other two methods was the pharmaceutical betahistine, used to treat the symptoms of Meniere’s disease, with log Kow 0.68. Twenty compounds were detected by both lyophilization and online SPE. These compounds were not highly abundant in the samples and, therefore, were only detectable due to the preconcentration of the water. Screening Environmental Assessment and Prioritization. The risk posed by the presence of these CECs in the environment is highlighted by the extent of emissions, their persistence, and the ease with which they can be transported across natural barriers from their point of release. Moreover, in the aquatic environment, these chemicals may present diverse toxicity and bioaccumulation potential depending on their intrinsic physical-chemical properties and the climatic conditions. To assess the environmental fate of the compounds detected in wastewater effluents in the aquatic environment of the area under study, their persistence, bioaccumulation potential, and mobility were evaluated by classifying them as “Potentially P or vP”, “Potentially B or vB in aquatic organisms“, mobile (M), or very mobile (vM), following the criteria proposed by Neumann et at. 38 and ECHA. 39 This evaluation was performed considering the information compiled in Table S8. The classification of each environmental fate category and the score assigned to each of them for prioritization purposes is provided in Table S3. Regarding persistence, up to 48 compounds were clearly classified as Potentially P or vP, while 23 of the remaining compounds were ambiguous and did not meet the criteria. For these, a definitive assessment of their P/vP properties should be performed based on degradation half-life testing. Concerning substances mobility, 67 and 33 compounds were categorized as vM and M, respectively. Finally, in terms of bioaccumulation, only 10 compounds, namely, bis(2-ethylhexyl) amine, decanophenone, dibutyl phthalate, diclofenac, dodecylbenzenesulfonic acid, erucamide, flufenamic acid, galaxolidone, irbesartan, and tiamulin were considered Potentially B or vB. For PBT assessment, the list established by ECHA in which PMT (persistent, mobile and toxic) or vPvM (very persistent and very mobile) substances are provided, was also considered. Thus, the compounds melamine, perfluorobutanesulfonic acid, and perfluorohexanoic acid, included in this list, coincided with the classification performed, while the compounds 1,2,3benzotriazole and its derivatives, 2-methyl-S-benzothiazole, dibutyl phthalate and bis(2-ethylhexyl) amine did not match with the persistence results obtained in the exercise performed. This shows that an estimation of persistence based on the compound structure is not sufficient since for some degradable substances, other factors must be considered. For instance, continuous delivery to the environment may render them pseudopersistent. This would be the case of the benzotriazole family. 12,40 Prioritization of the tentatively identified CECs in the effluent wastewater samples was based on their abundance Table 2. Classification of the Top-Prioritized Compounds as P (or vP), M (or vM) or B, PNEC, and Total Score for Prioritization a conclusion of classification rank compound CAS N°P (op1) P (op2) M B PNEC freshwater (μg/L) total SCORE 1 clindamycin 18323-44-9 Pot. P or vP Pot. P or vP vM not B and not vB 0.044 6 2 flufenamic acid 530-78-9 Pot. P or vP Pot. P or vP M Pot. B or vB 0.4 6 3 galaxolidone 507442-49-1 not P or vP Pot. P or vP not M Pot. B or vB 0.1 5.5 4 O-desmethyl venlafaxine 93413-62-8 Pot. P or vP Pot. P or vP M not B and not vB 0.006 5.5 5 tiamulin 55297-95-5 Pot. P or vP Pot. P or vP not M Pot. B or vB 0.25 5.5 6 venlafaxine 93413-69-5 Pot. P or vP Pot. P or vP M not B and not vB 0.006 5 7 2,4-diaminotoluene 95-80-7 Pot. P or vP Pot. P or vP vM not B and not vB 12 5 8 2-amino-6-methylmercaptopurine 1198-47-6 Pot. P or vP Pot. P or vP vM not B and not vB 0.49 5 9 bis(2-ethylhexyl) amine 106-20-7 not P or vP not P or vP not M Pot. B or vB 0.52 5 10 decanophenone 6048-82-4 not P or vP not P or vP M Pot. B or vB 0.07 5 11 diclofenac 15307-86-5 Pot. P or vP Pot. P or vP vM Pot. B or vB 0.05 5 12 terbutryn 886-50-0 Pot. P or vP Pot. P or vP vM not B and not vB 0.34 5 13 hexamethylenetetramine 100-97-0 Pot. P or vP Pot. P or vP vM not B and not vB 11 4.5 14 1,2,3-benzotriazole 95-14-7 not P or vP not P or vP vM not B and not vB 19 4.5 15 4-methylbenzotriazole 29878-31-7 not P or vP not P or vP M not B and not vB 5.9 4.5 16 articaine 23964-58-1 not P or vP Pot. P or vP vM not B and not vB 4.02 4.5 17 dibutyl phthalate 84-74-2 not P or vP not P or vP M Pot. B or vB 10 4.5 18 lamotrigine 84057-84-1 Pot. P or vP Pot. P or vP M not B and not vB 8 4.5 19 melamine 108-78-1 Pot. P or vP Pot. P or vP vM not B and not vB 360 4.5 20 N,N′-diphenylguanidine (DPG) 102-06-7 not P or vP Pot. P or vP M not B and not vB 1.05 4.5 21 ofloxacin/levofloxacin 100986-85-4 Pot. P or vP Pot. P or vP vM not B and not vB 0.026 4.5 22 perfluorobutanesulfonic acid (PFBS) 375-73-5 Pot. P or vP Pot. P or vP vM not B and not vB 372 4.5 23 phenazone 60-80-0 not P or vP not P or vP vM not B and not vB 1.1 4.5 24 pirlimycin 79548-73-5 Pot. P or vP Pot. P or vP vM not B and not vB 1.61 4.5 25 2-methyl-S-benzothiazole 615-22-5 not P or vP not P or vP M not B and not vB 0.69 4.5 a Pot.: Potentially, P: Persistent, vP: very Persistent, M: Mobile, vM: very Mobile, B: Bioaccumulative, vB: very Bioaccumulative Analytical Chemistry pubs.acs.org/ac Article https://doi.org/10.1021/acs.analchem.5c01659 Anal. Chem. 2025, 97, 25946−25957 25954