Multiresidue analytical method for pharmaceuticals and personal care products in sewage and sewage sludge by online direct immersion SPME on-fiber derivatization – GCMS

1 Multiresidue analytical method for pharmaceuticals and personal care products in 1 sewage and sewage sludge by online direct immersion SPME on-fiber derivatization – 2 GCMS 3 4 Rebeca López-Serna () • David Marín-de-Jesús • Rubén Irusta-Mata • Pedro Antonio 5 García-Encina • Raquel Lebrero • María Fdez-Polanco • Raúl Muñoz 6 7 Department of Chemical Engineering and Environmental Technology, University of 8 Valladolid, C/ Dr. Mergelina, s/n, 47011 Valladolid, Spain 9 10 () Corresponding author: 11 Rebeca López-Serna 12 e-mail: rebeca.lopezs[email protected]a.es 13 14 Abstract 15 The work here presented aimed at developing an analytical method for the 16 simultaneous determination of 22 pharmaceuticals and personal care products, 17 including 3 transformation products, in sewage and sludge. A meticulous method 18 optimization, involving an experimental design, was carried out. The developed 19 method was fully automated and consisted of the online extraction of 17 mL of water 20 sample by Direct Immersion Solid Phase MicroExtraction followed by On-fiber 21 Derivatization coupled to Gas Chromatography - Mass Spectrometry (DI-SPME – On-22 fiber Derivatization – GC - MS). This methodology was validated for 12 of the initial 23 compounds as a reliable (relative recoveries above 90% for sewage and 70% for 24 sludge; repeatability as %RSD below 10% in all cases), sensitive (LODs below 20 ng L-1 25 in sewage and 10 ng g-1 in sludge), versatile (sewage and sewage-sludge samples up to 26 15,000 ng L-1 and 900 ng g-1, respectively) and green analytical alternative for many 27 medium-tech routine laboratories around the world to keep up with both current and 28 forecast environmental regulations requirements. The remaining 10 analytes initially 29 considered showed insufficient suitability to be included in the final method. The 30 methodology was successfully applied to real samples generated in a pilot scale 31 sewage treatment reactor. 32 2 33 Keywords: DI-SPME • GC-MS • On-Fiber Derivatization • PPCPs • Sewage sludge • 34 Wastewater 35 36 1 Introduction 37 The development of analytical methodologies for the determination of 38 pharmaceuticals and personal care products (PPCPs) in environmental matrices has 39 boomed in the past years. In this context, Zwiener and Frimmel [1] reported that the 40 analysis of PPCPs has been traditionally dominated by Liquid Chromatography 41 detected by tandem Mass Spectrometric (LC-MS/MS) techniques. Fischer et al. [2] 42 recently observed major trends in the use of Ultra High Performance Liquid 43 Chromatography (UHPLC) [3] and High Resolution Mass Spectrometry (HRMS) [4-6] like 44 Time Of Flight (TOF) and Orbitrap [7] analyzers. However, these techniques require 45 costly instrumentation not affordable by many laboratories worldwide. In contrast, 46 Gas Chromatography coupled to single quadrupole Mass Spectrometry (GC-MS) is an 47 analytical configuration far more common in routine analysis laboratories around the 48 world, including developing countries. Despite PPCPs are mainly polar compounds and 49 not readily analyzable by GC, Lopez-Serna et al. [8] recently showed how GC-MS is a 50 valid instrumental technique for the analysis of emerging contaminants in 51 environmental matrices like sewage, when a derivatization step is included in the 52 method. In terms of sample preparation, Solid-Phase Extraction (SPE) represents 53 nowadays the most popular technique for the extraction of pollutants from 54 environmental aqueous samples, and recent developments in this field have mainly 55 focused on SPE automation [9]. In addition, a great effort has been lately made to 56 develop new analytical methodologies able to perform direct analyses using 57 miniaturized equipment, thereby achieving high enrichment factors, minimizing 58 solvent consumption and reducing waste [7, 10] in accordance to the requirements of 59 green analytical chemistry. Solid-Phase MicroExtraction (SPME) was firstly developed 60 in the 1990s by Pawliszyn and coworkers [11]. Since then many configurations have 61 been successfully implemented, which can be classified into static and dynamic 62 techniques [12]. Static procedures are typically carried out in stirred samples, including 63 fiber SPME, and constitute the most common format for this technique. Fiber SPME 64 3 utilizes a sorbent coating on the outer surface of a fused silica fiber to extract the 65 analyte(s) from the sample matrix in a process that occurs through direct immersion 66 (DI-SPME) or from the sample headspace in a closed container (HS-SPME) [10]. Thus, 67 analytes that exhibit a high vapor pressure can be extracted either by immersing the 68 fiber into the aqueous sample or by sampling its headspace. In contrast, analytes that 69 exhibit a low vapor pressure could only be extracted by immersion. Fiber SPME has 70 become a very popular technique, especially for volatile compounds, due to its 71 simplicity, relatively short extraction time, solvent-free nature, full automation 72 potential and easy coupling with chromatography [12]. These advantages eventually 73 reduce the contamination of the original sample and the loss of analytes. In addition, 74 SPME can also be used for onsite sample extraction and is able to obtain good results 75 even for trace analytes in complex matrices [12]. However, its application to the 76 environmental analysis of polar compounds has been poorly explored, especially when 77 this sample pretreatment is coupled to GC. This application implies the addition of a 78 derivatization step, which is essential for the analysis of non-volatile and/or 79 thermolabile compounds by GC. Today, two approaches are commonly used to carry 80 out derivatization when SPME is the pretreatment technique. The first one, namely in-81 situ derivatization, is based on the addition of the derivatizing agent directly to the 82 sample and the collection of the derived volatile analytes by SPME in the headspace of 83 a closed vial. In the second approach, namely on-fiber derivatization, analyte 84 extraction occurs via direct fiber immersion in the sample combined with a headspace 85 derivatization by exposing the analytes-loaded fiber to the vapors of the derivatizing 86 agent. This second approach is environmentally and economically preferred, because 87 the derivatizing agent can be reused for a large number of analyses (with the 88 subsequent decrease of reagent consumption). 89 This study aimed at developing and optimizing a fully automated method 90 consisting of Online DI-SPME - On-Fiber Derivatization - GC-MS for the analysis of 19 91 PPCPs and 3 of their Transformation Products (TPs) in sewage (SW) and sludge (SS) 92 using statistical experimental design. To the authors’ knowledge, there are only two 93 other publications [13, 14] proposing the use of this technique for the analysis of 94 PPCPs in sewage and none for sludge. However, none of them included the level of 95 4 automation here presented. Finally, the analytical limitations encountered during the 96 application of this innovative methodology were also discussed. 97 98 2 Material and methods 99 2.1 Chemicals 100 The standards for all PPCPs and their TPs, provided in Table S1 as 101 supplementary data, were of high purity grade (>95%). They were purchased from 102 Sigma-Aldrich (Tres Cantos, Madrid, Spain) as neutral non-solvated molecules, except 103 for amoxicillin (acquired as trihydrate), atorvastatin (acquired as calcium salt) and 104 diclofenac (acquired as sodium salt). The isotopically labelled compounds Diclofenac-105 d4, Ibuprofen-d3, Salicylic acid-d4, Naproxen-d3, Propylparaben-d7 and Triclosan-d3 106 were obtained from TRC Canada (Toronto, ON, Canada). 107 Individual stock solutions at 1 g L-1 for both PPCPs standards and isotopically-108 labelled-internal-standards were prepared on a weight basis in methanol (MeOH), 109 except for the fluroquinolones (ciprofloxacin, levofloxacin and norfloxacin), which 110 were dissolved in a water-methanol (H2O/MeOH) mixture (1:1) containing 0.2% v/v 111 hydrochloric acid (HCl) due to their low solubility in pure MeOH [15]. From them, a 112 stock solution with all the analytes was then prepared in MeOH at 20 mg L-1. Serial 113 aqueous dilutions were subsequently prepared from it. A separate mixture of 114 isotopically labelled internal standards and further dilutions were also prepared. After 115 preparation, all stock solutions were stored at -20 °C in darkness. 116 High purity solvents, i.e., SupraSolv® GC-MS grade MeOH by Merck Millipore 117 (Madrid, Spain), LC-MS Chromasolv® grade Ethyl Acetate (EA) by Fluka (Madrid, Spain), 118 Sodium chloride (NaCl) and 37% HCl were supplied by Panreac (Barcelona, Spain). 119 Acetone, 99% pure, was supplied by Cofarcas (Burgos, Spain). N-tert-120 Butyldimethylsilyl-N-methyltrifluoroacetamide, with a purity >99%, (MTBSTFA), was 121 obtained from Regis Technologies Inc. (Morton Grove, IL, USA). SPME fibers were 122 purchased from Supelco (Tres Cantos, Madrid, Spain). Milli-Q® grade water was in-123 house produced. Helium 99.999% (He) was purchased from Abelló Linde S.A. (Alcalá de 124 Henares, Madrid, Spain). 125 5 126 2.2 Sewage analytical methodology 127 The development of the analytical method, further explained in Sections SD.1.1 128 and SD.1.2 within the Supplementary data (SD), was carried out in Milli-Q® water and 129 validated for sewage as detailed in Section 3.2.1. In addition, the optimized method 130 based on Online DI-SPME – On-Fiber Derivatization – GC – MS was applied to the 131 analysis of raw and treated wastewater from a pilot scale activated sludge reactor, and 132 the results are presented in Section 3.2.2. 133 134 2.2.1 Online DI-SPME – On-Fiber Derivatization 135 Water samples (100 mL) were supplemented with NaCl at 30 % (wt./vol.). After 136 stirring for 20 min to assure complete dissolution, the resulting water sample pH was 137 adjusted to 3 by adding as few drops of diluted solutions of HCl (1%, 0.1% and/or 138 0.01%) as needed. A volume of 17 mL of the resulting solution was placed in a 20-mL 139 SPME vial along with 200 µL of an aqueous mixture of the isotopically labelled internal 140 standards at 0.5 mg L-1. 141 The resulting vial was placed in the sample rack of a CTC PAL RSI autosampler. A 142 SPME tool held a 2-cm long 50/30-µm thick 143 Divinylbenzene/Carboxen/Polydimethylsiloxane (DVB/CAR/PDMS) StableFlex/SS fiber 144 that was protected inside a 23 Ga needle. The fully automated DI-SPME method 145 included a fiber pre-conditioning for 15 min at 270 °C in the spare GC inlet, followed by 146 120 min sample extraction at a penetration depth of 60 mm, which entailed that the 147 fiber was fully immersed in the sample (DI-SPME). On-fiber derivatization of the 148 analytes absorbed onto the fiber was then carried out by introducing the fiber in 149 another 20-mL SPME vial containing 1 mL of the derivatizing agent MTBSTFA for 48 150 min at a penetration depth of 60 mm. Thus, the fiber was exposed to the vapors of the 151 MTBSTFA in the headspace of the vial. Both the DI-SPME and On-Fiber Derivatization 152 were carried out at a constant temperature of 50 °C under orbital agitation at 500 rpm 153 with a stirring regime of 6s on / 30 s off. The fiber, loaded with the derivatized 154 analytes, was then taken to the GC inlet connected to the GC column for desorption at 155 250 °C for 3 min. Finally, the fiber was post-conditioned for 15 min at 270 °C in the 156 spare GC inlet prior to the next analysis. 157 6 158 2.2.2 GC – MS 159 Chromatographic runs started concomitantly with fiber desorption in a pulsed 160 splitless mode at 250 °C in the split/splitless back inlet. A SPME injection sleeve, 0.75 161 mm i.d., was used as a liner. The tests were performed in an Agilent 7890B GC System 162 coupled to a 5977A MSD. A capillary HP-5MS GC column (30 m length, 0.25 mm i.d., 163 0.25 μm film thickness) was used for the chromatographic separation with He as 164 carrier gas at a constant flow rate of 1.2 mL min-1. Injector temperature was set at 250 165 °C, while the GC oven temperature increased from 70 °C (held for 3 min during fiber 166 desorption) to 120 °C at 20 °C min-1, then to 250 °C at 10 °C min-1 and finally to 300 °C 167 (held for 5 min) at 5 °C min-1. The total analysis time for each GC run was 33.5 min. The 168 multimode front GC inlet was set at 270 °C in split mode to facilitate the elimination of 169 residual compounds during fiber preand post-conditioning. 170 Mass detection was obtained in electron impact ionization mode (70 eV) with 171 selected ion monitoring (SIM) and a filament delay of 12 min. The GC–MS interface, 172 ion source and quadrupole temperatures were set at 280, 230 and 150 °C, 173 respectively. Quadrupole resolution was set at low. Target compounds were recorded 174 in five acquisition windows along the run time. Table 1 shows the primary (in italics) 175 and the two secondary ions monitored per compound. Acquisition stopped at min 26. 176 Instrument control and data acquisition were performed by Agilent Technology Mass 177 Hunter B.07.03.2129 software. 178 179 2.3 Sewage sludge analytical methodology 180 Aerobic sludge was used to develop and validate the methodology further 181 discussed in Sections SD.1.3 and 3.2.1, respectively. The sewage sludge analytical 182 method was designed as follows: 1) One hundred milliliters of fresh sludge sample 183 were freeze-dried. 2) A known amount of dried sludge (~800 mg) was weighed into a 184 20-mL glass vial, along with 200 µL of a mixture of the isotopically labelled internal 185 standards at 20 mg L-1 in acetone. 3) The mixture was thoroughly vortex-stirred and 186 remained overnight to allow solvent evaporation and internal standard fixation. 4) A 187 volume of 12 mL of MilliQ® water at pH 9 was then added to the vial, which was then 188 vigorously vortex-stirred to obtain a homogenous suspension. 5) Then, the vial 189 7 underwent Ultrasound Assisted Extraction (UAE) for 30 min at room temperature in a 190 JP Selecta Univeba ultrasonic bath of 50 W and 60 Hz (Barcelona, Spain). 6) 191 Subsequently, the suspension was centrifuged for 5 min at 2,655 x g in a Fisher 192 Bioblock Scientific Centrifuge 2-16P (Madrid, Spain). 7) The resulting supernatant was 193 then collected with a glass pipette and transferred to a 20-mL glass vial. 8) Steps 4-7 194 were repeated once more and the supernatants were pooled together. 9) The resulting 195 solution was analyzed by Online DI-SPME – On-fiber derivatization – GC-MS using the 196 optimized method described in Section 2.2, except for the addition of internal 197 standards as they were already added in step 2. 198 199 2.4 Experimental design 200 As a first approach, a screening design was carried out. Hence, the key 201 parameters influencing the performance of the Online DI-SPME – On-Fiber 202 Derivatization methodology were identified for the development of the instrumental 203 leg of both sewage and sludge methods. As a result, a total of 18 parameters were 204 sorted out in four categories, depending on the target of their influence, i.e., DI-SPME 205 extraction, On-Fiber Derivatization, Fiber Desorption and Carry-Over avoidance (Table 206 S2). Afterwards, technical limitations to this innovative methodology were pointed out, 207 which narrowed down to 6 the number of parameters admitting further optimization. 208 Nonetheless, 4 of them, i.e., fiber coating, sample Ionic strength, sample pH and 209 derivatization temperature could easily be optimized by a one-factor-at-a-time 210 approach as they are discrete variables or otherwise consolidated references exist in 211 the scientific literature which drastically delimits the range of variation. Eventually, 212 only two parameters remained as significant, extraction and derivatization times, and 213 in need of further optimization. Thus, a response surface methodology (RSM), 214 consisting of a full factorial 22 with a central point repeated five times and extended 215 with 4 star points, was applied to them. Thus, a set of 13 experiments was randomly 216 performed. Afterwards, the software Statgraphics Centurion XVII was used to process 217 the acquired experimental data and mathematically fit it to a second order polynomial 218 model through the least squares method. 219 220 3 Results and discussion 221 8 3.1 Analytical method development and optimization for sewage and sludge 222 A selection of 22 PPCPs, in particular, 5 pharmaceuticals and 2 of their TPs as 223 well as 14 personal care products and 1 of their TPs, were initially chosen as target 224 analytes. 225 The protocol followed to develop and optimize the analytical method, including 226 an experimental design, is described in the supplementary data SD file. In brief, after 227 the GC-MS leg was developed, the sample pretreatment part of the methodology was 228 optimized. Hence all the parameters with a role during the Online DI-SPME – On-Fiber 229 Derivatization were identified and some technical limits were set. Afterwards, some 230 preliminary experiments were carried out in a one-factor-at-a-time approach to 231 optimize the Type of Fiber Coating, Sample Ionic Strength, Sample pH and 232 Derivatization temperature. Finally, as the extraction and derivatization time could 233 interfere with each other, a response surface method was designed based on a full 234 factorial 22 with a central point repeated five times and extended with 4 star points. 235 TS/N was selected as the response variable during the optimization, in order to get a 236 compromise among the performance of all the compounds. As a result, the optimum 237 value for the response variable obtained corresponded to an extraction time of 120 238 min and a derivatization time of 48 min. That is graphically shown in Figure 1. 239 After the optimization, ten of the initial target PPCPs, including the analgesics 240 acetaminophen and acetylsalicylic acid, the lipid regulator atorvastatin, and the 241 antibiotics amoxicillin, ciprofloxacin, levofloxacin, norfloxacin, sulfamethoxazole, 242 erythromycin and clarithromycin turned out to be unsuitable for their analysis by 243 Online DI-SPME – On-Fiber Derivatization – GC-MS, as they exhibited a very weak or 244 even no response whatsoever. Therefore, they were ruled out and not included in the 245 method. 246 The final methods, which allowed for the analysis of 12 PPCPs including 3 TPs, 247 are summarized in Sections 2.2 and 2.3. Representative SIM chromatograms, obtained 248 from MilliQ® water and sewage sludge samples spiked with the target PPCPs at 4 µg L-1 249 and 400 ng g-1, respectively, using the optimized method conditions, are illustrated in 250 Figure 2. 251 252 253 9 3.2 Method validation and application 254 3.2.1 Method validation 255 Several regulatory bodies (like the United States Food and Drug Administration 256 (FDA) [17] or Eurachem [18]), standardization agencies (like the International 257 Association of Official Analytical Chemists (AOAC International) [19]), and working 258 groups and committees (like the Food and Agricultural Organization/World Health 259 (FAO/WHO) [20]) have published guidelines and requirements for method validation. 260 In addition, the European Union adopted a decision [21] implementing a directive 261 concerning the performance of analytical methods and the interpretation of results. It 262 refers to animal products. However, it has been widely used as an illustrative reference 263 in the design of customized validation protocols for environmental analysis like in [22-264 25], as well as in the present study because of the lack of specific guidelines. 265 Hence, five validation parameters, i.e., accuracy, ME, precision, sensitivity and 266 dynamic range were determined for all 12 target analytes included in the method 267 (clofibrate, 1,4-benzoquinone, methylparaben, ethylparaben, clofibric acid, ibuprofen, 268 propylparaben, salicylic acid, p-hydroxybenzoic acid, naproxen, triclosan, diclofenac) in 269 sewage and sludge. In addition, a carryover test was also performed to ensure the 270 absence of contamination between samples during the instrumental leg of the 271 analysis. Two meaningful levels of concentration per matrix −100 and 1000 ng L-1, and 272 100 and 400 ng g-1− typical for the target compounds in real sewage and sludge 273 samples, respectively, were tested for the four first parameters, as recommended by 274 [23, 24]. Each test was run in triplicate with the optimized method. The results, 275 average of both concentration levels, which are discussed below, are shown in Table 276 S4. 277 278 1) Accuracy: Absolute recoveries (%) were determined by comparing the 279 peak areas obtained from spiked samples analyzed using the optimized methods with 280 the areas obtained from direct injections (2 µL) of equivalent amounts of standards in 281 EA solutions. As both sewage and sludge can contain some of the target compounds, 282 non-spiked samples were also analyzed and the peak areas were afterwards 283 subtracted from the spiked samples in order to calculate the absolute recovery. Table 284 S4A shows very variable absolute recoveries for sewage. Hence, SPME supported good 285 16 [9] A. Ballesteros-Gómez, S. Rubio, Recent advances in environmental analysis, 474 Analytical Chemistry 83(12) (2011) 4579-4613. 475 [10] M.E.T. Padrón, C. Afonso-Olivares, Z. Sosa-Ferrera, J.J. Santana-Rodríguez, 476 Microextraction techniques coupled to liquid chromatography with mass spectrometry 477 for the determination of organic micropollutants in environmental water samples, 478 Molecules 19(7) (2014) 10320-10349. 479 [11] S. Risticevic, D. Vuckovic, H.L. Lord, J. Pawliszyn, Solid-phase microextraction, 480 Comprehensive Sampling and Sample Preparation2012, pp. 419-460. 481 [12] J. Li, Y.B. Wang, K.Y. Li, Y.Q. Cao, S. Wu, L. Wu, Advances in different 482 configurations of solid-phase microextraction and their applications in food and 483 environmental analysis, TrAC - Trends in Analytical Chemistry 72 (2015) 141-152. 484 [13] A. Peña-Álvarez, A. Castillo-Alanís, Identificación y cuantificación de 485 contaminantes emergentes en aguas residuales por microextracción en fase sólida-486 cromatografía de gases-espectrometría de masas (MEFS-CG-EM), TIP. Revista 487 Especializada en Ciencias Químico-Biológicas 18(01) (2015) 29-42. 488 [14] G. Basaglia, M.C. Pietrogrande, Optimization of a SPME/GC/MS method for the 489 simultaneous determination of pharmaceuticals and personal care products in waters, 490 Chromatographia 75(7-8) (2012) 361-370. 491 [15] E.M. Golet, A.C. Alder, A. Hartmann, T.A. Temes, W. Giger, Trace determination of 492 fluoroquinolone antibacterial agents in urban wastewater by solid-phase extraction 493 and liquid chromatography with fluorescence detection, Analytical Chemistry 73(15) 494 (2001) 3632-3638. 495 [16] S.R. Hughes, P. Kay, L.E. Brown, Global synthesis and critical evaluation of 496 pharmaceutical data sets collected from river systems, Environmental Science and 497 Technology 47(2) (2013) 661-677. 498 [17] USFDA, Guidance for Industry - Bioanalytical Method Validation, 2001. 499 [18] B. Magnusson, U. Örnemark, Eurachem Guide: The Fitness for Purpose of 500 Analytical Methods – A Laboratory Guide to Method Validation and Related Topics, 501 2nd ed., Eurachem, Leoben, 2014. 502 [19] AOAC, Guidelines for Single Laboratory Validation of Chemical Methods for 503 Dietary Supplements and Botanicals, in: O.P. Department (Ed.) 2002. 504 17 [20] FAO/WHO, Consideration of harmonized IUPAC Guidelines for the in-house 505 validation of method of analysis, in: C.A. Commission (Ed.) 2002. 506 [21] European Commission, Commission Decision 2002/657/EC implementing Council 507 Directive 96/23/EC concerning the performance of analytical methods and the 508 interpretation of results, in: T.C.o.t.E. Communities (Ed.) 2002. 509 [22] M. Gros, M. Petrović, D. Barceló, Tracing pharmaceutical residues of different 510 therapeutic classes in environmental waters by using liquid 511 chromatography/quadrupole-linear ion trap mass spectrometry and automated library 512 searching, Analytical Chemistry 81(3) (2009) 898-912. 513 [23] R. López-Serna, S. Pérez, A. Ginebreda, M. Petrović, D. Barceló, Fully automated 514 determination of 74 pharmaceuticals in environmental and waste waters by online 515 solid phase extraction-liquid chromatographyelectrospray-tandem mass 516 spectrometry, Talanta 83(2) (2010) 410-424. 517 [24] R. López-Serna, M. Petrović, D. Barceló, Direct analysis of pharmaceuticals, their 518 metabolites and transformation products in environmental waters using on-line 519 TurboFlowTM chromatography-liquid chromatography-tandem mass spectrometry, 520 Journal of Chromatography A 1252 (2012) 115-129. 521 [25] R. López-Serna, F. Ernst, L. Wu, Analysis of cinnamaldehyde and diallyl disulfide as 522 eco-pesticides in soils of different textures—a laboratory-scale mobility study, Journal 523 of Soils and Sediments (2015) 1-15. 524 [26] R. López-Serna, M. Petrović, D. Barceló, Development of a fast instrumental 525 method for the analysis of pharmaceuticals in environmental and wastewaters based 526 on ultra high performance liquid chromatography (UHPLC)-tandem mass spectrometry 527 (MS/MS), Chemosphere 85(8) (2011) 1390-1399. 528 [27] A. Jelić, M. Petrović, D. Barceló, Multi-residue method for trace level 529 determination of pharmaceuticals in solid samples using pressurized liquid extraction 530 followed by liquid chromatography/quadrupole-linear ion trap mass spectrometry, 531 Talanta 80(1) (2009) 363-371. 532 [28] Y. Yu, L. Wu, Analysis of endocrine disrupting compounds, pharmaceuticals and 533 personal care products in sewage sludge by gas chromatography-mass spectrometry, 534 Talanta 89 (2012) 258-263. 535 18 [29] R. Lopez-Serna, A. Jurado, E. Vazquez-Sune, J. Carrera, M. Petrovic, D. Barcelo, 536 Occurrence of 95 pharmaceuticals and transformation products in urban groundwaters 537 underlying the metropolis of Barcelona, Spain, Environmental Pollution 174 (2013) 538 305-315. 539 [30] Environmental Protection Agency (EPA), Transport and transformation test 540 guidlines. OPPTS 835.3240 Simulation test – aerobic sewage treatment: a activated 541 sludge units, 2008. 542 [31] S. Ortiz de García, G. Pinto Pinto, P. García Encina, R. Irusta Mata, Consumption 543 and occurrence of pharmaceutical and personal care products in the aquatic 544 environment in Spain, Science of the Total Environment 444 (2013) 451-465. 545 [32] C. Reyes-Contreras, V. Matamoros, I. Ruiz, M. Soto, J.M. Bayona, Evaluation of 546 PPCPs removal in a combined anaerobic digester-constructed wetland pilot plant 547 treating urban wastewater, Chemosphere 84(9) (2011) 1200-1207. 548 [33] C. Reyes-Contreras, M. Hijosa-Valsero, R. Sidrach-Cardona, J.M. Bayona, E. 549 Bécares, Temporal evolution in PPCP removal from urban wastewater by constructed 550 wetlands of different configuration: A medium-term study, Chemosphere 88(2) (2012) 551 161-167. 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 19 Figure 1: Response surface after applying an experimental design 22 + star + 5 central 568 points 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 20 Figure 2: Chromatograms obtained from A) 4000 ng L-1 MilliQ water and B) 400 ng g-1 598 sludge samples after the optimized methods were applied 599 A) 600 601 602 B) 603 604 21 Table 1: MS parameters for the final target compounds and internal standards 1IS Analyte Chemical Name Acquisition window # 2tR (min) 3SIM ions, m/z 1 Clofibrate 1 13.17 128 130 169 2 1,4-Benzoquinone 13.70 281 338 282 3 Methylparaben 2 15.23 209 210 135 4 Acetylsalicylic acid 15.28 195 237 135 5 Ethylparaben 16.02 223 224 151 6 Clofibric acid 16.02 143 271 185 7 Ibuprofen 16.38 263 264 117 1 Ibuprofen-d3 16.34 266 267 164 8 Propylparaben 17.04 237 238 151 2 Propylparaben-d7 16.96 244 245 152 9 Salicylic acid 17.50 309 310 195 3 Salicylic acid-d4 17.45 313 314 312 10 Acetaminophen 3 18.24 208 265 166 11 P-hydroxybenzoic acid 18.91 309 265 310 12 Naproxen 4 21.07 287 185 288 4 Naproxen-d3 20.95 290 188 207 13 Triclosan 21.66 347 345 200 5 Triclosan-d3 21.54 350 348 200 14 Diclofenac 5 23.73 352 214 354 6 Diclofenac-d4 23.73 356 218 358 1IS: Internal Standard; 2tR: Retention Time; 3SIM: Selected Ion Monitoring 22 1