Performance of the linear ion trap Orbitrap mass analyzer for qualitative and quantitative analysis of drugs of abuse and relevant metabolites in sewage water

Título artículo / Títol article: Performance of the LTQ-Orbitrap mass analyzer for qualitative and quantitative analysis of drugs of abuse and relevant metabolites in sewage water Autores / Autors Lubertus Bijlsmaa, Erik Emke, Félix Hernández, Pim de Voogt Revista: Analytica Chimica Acta, v. 768 (2013) Versión / Versió: Preprint de l’autor Cita bibliográfica / Cita bibliogràfica (ISO 690): BIJLSMA, L.; EMKE, E.; HERNÁNDEZ, F.; DE VOOGT, P. Performance of the LTQ-Orbitrap mass analyzer for qualitative and quantitative analysis of drugs of abuse and relevant metabolites in sewage water. Analytical Chimica Acta, v. 768 (2103), p. 102-110 url Repositori UJI: http://repositori.uji.es/xmlui/handle/10234/94472 1/22 Performance of the LTQ-Orbitrap mass analyzer for qualitative and quantitative analysis of drugs of abuse and relevant metabolites in sewage water Lubertus Bijlsmaa,§, Erik Emkeb, Félix Hernándeza, Pim de Voogtb,c,* 5 a Research Institute for Pesticides and Water, University Jaume I, Avda. Sos Baynat s/n, E12071 Castellón, Spain. b KWR Watercycle Research Institute, Chemical Water Quality and Health, P.O. Box 1072, 3430 BB Nieuwegein, the Netherlands. c Institute for Biodiversity and Ecosystem Dynamics, University of Amsterdam, P.O. Box 10 94248, 1090 GE Amsterdam, the Netherlands §Visiting scientist at KWR * Corresponding author: [email protected], Tel +31 20 5256565, Fax +31 20 5257431. 15 2/22 ABSTRACT This work illustrates the potential of liquid chromatography coupled to a hybrid linear ion trap Fourier Transform Orbitrap mass spectrometer for the simultaneous identification and quantification of 24 drugs of abuse and relevant metabolites in sewage water. The developed 20 methodology consisted of automatic solid-phase extraction using Oasis HLB cartridges, chromatographic separation of the targeted drugs, full-scan accurate mass data acquisition under positive electrospray ionization mode over an m/z range of 50 - 600 Da at a resolution of 30,000 FWHM and simultaneous MSn measurements to obtain information of fragment ions generated in the linear ion trap. Accurate mass of the protonated molecule, together with 25 at least one nominal mass product ion and retention time allowed the confident identification of the compounds detected in these complex matrices. In addition to the highly reliable qualitative analysis, Orbitrap analyzer also proved to have satisfactory potential for quantification at sub-ppb analyte levels, a possibility that has been very little explored in the literature until now. The limits of quantification ranged from 4 to 68 ng L-1 in influent sewage 30 water, and from 2 to 35 ng L-1 in effluent, with the exception of MDA, morphine and THC that presented higher values as a consequence of the high ionization suppression in this type of samples. Satisfactory recoveries (70 – 120%) and precision (< 30%) for the overall procedure were obtained for all compounds with the exception of metachlorophenylpiperazine, methylphenidate and ketamine. Isotope-labelled internal standards 35 were added to sewage samples as surrogates in order to correct for matrix effects and also for possible losses during sample treatment. The methodology developed was applied to sewage water samples from the Netherlands (influent and effluent), and the results compared with those obtained by LC-MS/MS with triple quadrupole. Several drugs of abuse could be identified and quantified, mainly MDMA, benzoylecgonine, codeine, oxazepam and 40 temazepam. Orbitrap also showed potential for retrospective investigation of ketamine metabolites in the samples without the need of additional analysis. Keywords 45 Drugs of abuse, accurate mass, Orbitrap analyzer, high resolution mass spectrometry, quantitative analysis, sewage water. 3/22 1. INTRODUCTION 50 The presence of drugs of abuse, unaltered or as metabolites, in the water cycle has spurred researchers on to investigate their occurrence in sewage water, surface water and drinking water [1-3]. Although concentrations found are generally low (sub µg L-1 level), data obtained from analysis of urban wastewater can be used to study consumption and usage trends in communities [4]. Furthermore, environmental loads can be calculated, as their 55 potential impact on aquatic organisms, human health and the environment may not be ruled out [1]. Most of the existing methods for determination of drugs of abuse in water are based on liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS), using triple quadrupole (QqQ) analyzers. Despite its excellent sensitivity and selectivity , this approach 60 also has some limitations [5, 6], the main being that other drugs, different from those included in the scope of the method, may be ignored in analyses, as analyte specific information is acquired and only the target analytes are normally detected and quantified. The increasing interest of using accurate mass High Resolution Mass Spectrometers (HRMS), e.g. Orbitrap and time-of-flight (TOF) instruments, in environmental sciences relies on its capability to 65 perform both targeted as well as non-targeted analysis, based on full-spectrum accurate-mass acquisition at good sensitivity [7]. Efficient screening strategies using HRMS have allowed the detection and identification of various drugs of abuse in environmental and wastewater samples, in some cases even without the need of reference standards, but with high confidence due to the high mass accuracy measurements [8, 9]. 70 Advantages of HRMS are widely recognized in qualitative analysis; however HRMSbased quantitative analyses have hardly been explored in the scientific literature until now. One of the classical criticisms concerns the relative low sensitivity and low linear dynamic range. This limitation was more evident in the first-generation instruments, e.g. first LC-TOF MS. However, the improved technology of the latest TOF instruments, i.e. higher sensitivity 75 and resolving power, and wider linear dynamic range, provides better quantitative capabilities. This has allowed quantification of pesticides, pharmaceuticals and illicit drugs in wastewater by using LC-TOF MS [10, 11]. As for Orbitrap instruments, good quantitative performances, i.e. high sensitivity and selectivity, have been demonstrated in some applied fields [5, 12-15]. 80 Nevertheless, to the best of our knowledge, the quantitative potential of Orbitrap has not been previously demonstrated for drugs of abuse in sewage water samples. The 4/22 determination of these compounds is complicated due to the complexity of the samples and low analyte concentrations. Sample pre-concentration is normally required, mostly based on solid phase extraction (SPE), but the key point is the quantification of analytes, which is 85 problematic in LC-MS based methods due to the strong matrix effects commonly observed for this type of sample matrices. The use of isotope-labeled internal standards (ILIS) is the approach most frequently applied to solve this problem, although its application is difficult in multi-residue multiclass methods where a large number of ILIS would be required. Typically the own analyte ILIS is used, as the use of analogue compounds as internal standards does not 90 always ensure an appropriate correction [10, 16]. In the present work, analytical methodology based on the use of SPE followed by LC coupled to a hybrid linear ion trap (LTQ) Fourier Transform (FT) Orbitrap MS, has been developed for the determination of 24 drugs of abuse and metabolites in urban wastewater. The acquisition of full-scan accurate-mass data by Orbitrap together with the simultaneous 95 MS/MS measurements permitted by LTQ is a powerful combination for confident identification and confirmation. As the excellent qualitative potential of Orbitrap analyzer is widely accepted in the recent literature, an additional and relevant objective was to demonstrate the quantitative capabilities of this HRMS, a feature that has been little explored until now. To the best of our knowledge, the quantitative potential of Orbitrap has not been 100 previously demonstrated for drugs of abuse in complex sewage water samples. 5/22 2. MATERIALS AND METHODS 2.1. Reagents 105 Drugs of abuse and metabolites reference standards: amphetamine, methamphetamine, 3,4-methylenedioxyamphetamine (MDA), 3,4-methylene-dioxymethamphetamine (MDMA, or ecstasy), 3,4-methylenedioxyethylamphetamine (MDEA), cocaine, benzoylecgonine, heroin, morphine, 6-monoacetyl morphine (6-MAM), methadone, codeine, -9tetrahydrocannabinol (THC), 11-nor-9-carboxy--9-tetrahydrocannabinol (THC-COOH), 11-110 hydroxy--9-tetrahydrocannabinol (OH-THC), ketamine, methylphenidate, oxazepam, diazepam, temazepam, nordazepam, desalkyl-flurazepam, meta-chlorophenylpiperazine (mCPP), and fentanyl were obtained from Lipomed AG (Arlesheim, Switzerland) as solutions in methanol (MeOH), ethanol (EtOH) or acetonitrile (ACN) at a concentration of 1 g L-1. Standard solutions of each compound were prepared at 36 mg L-1 in MeOH. A final mix 115 solution was made by diluting aliquots from every compound individually to a concentration of 3.6 mg L-1. Working mix solutions for calibration curves were made in MeOH. Before each analytical run, the calibration standards were diluted 10 times with ultrapure water resulting in a mix of water: MeOH (90:10 v/v) and were injected into the Orbitrap system. Final concentrations of standards ranged from: 0.7 to 288 µg L-1. 120 Deuterated compounds were purchased from Lipomed AG as solutions in MeOH, EtOH or ACN at a concentration of 1 g L-1 and were used as surrogate isotope labelled internal standards (ILIS) for quantification: amphetamine-d11, methamphetamine-d5, 3,4methylenedioxyamphetamine-d2 (MDA-d2), 3,4-methylenedioxymethamphetamine-d5 (MDMA-d5), 3,4-methylenedioxyethyl-amphetamine-d5 (MDEA-d5), cocaine-d3, 125 benzoylecgonine-d3, morphine-d3, 6-monoacetyl morphine-d3 (6-MAM-d3), 9tetrahydrocannabinol-d3 (THC-d3), 11-nor-9-carboxy-9-tetrahydrocannabinol-d3 (THCCOOH-d3), 11-hydroxy-9-tetrahydrocannabinol-d3 (OH-THC-d3), oxazepam-d5, diazepamd5, nordazepam-d5. A mixed ILIS working solution was prepared in MeOH and added to all calibration standards to get a final ILIS concentration of 72 µg L-1, as well as to the influent 130 and effluent sewage water samples prior to sample treatment (final ILIS concentration in sample of 360 ng L-1 and 180 ng L-1, respectively). All standard and working solutions were stored in amber glass bottles at -18 °C. 6/22 The ultrapure water was obtained by purifying demineralized water in a Milli-Q system from Millipore (Bedford, MA, USA). Formic acid (98 – 100%), HPLC-grade MeOH, 135 EtOH and ACN were acquired from Mallinckrodt Baker (Deventer, The Netherlands). Glass fibre filters (1 µm, type A/E) were purchased from Pall Corporation (Port Washington, NY, USA). Polyethersulfone filters (0.45 µm) with disposable setup were acquired from Nalgene (Rochester, NY, USA). SPE cartridges, built of a hydrophilic and a lipophilic monomer (Oasis-HLB; 6 mL, 140 150 mg) were purchased from Waters (Milford, MA, USA). Polytyrosine-1,3,6 standard used for mass axis calibration was purchased from Cs Bio Co. (Menlo Park, CA, USA). 2.2. Water samples 145 24-hours flow dependent influent and effluent composite-samples from different sewage treatment plants (STPs) located in the Netherlands were taken on the same weekend day, without accounting for lag-time. Samples were collected in amber glass bottles, and stored in the dark at 4 °C. Upon reception in the laboratory, the samples were immediately analysed. 150 2.3. Extraction procedure Prior to SPE, samples were vacuum filtered through 1 µm type A/E glass fibre filters, followed by 0.45 µm polyethersulfone (PES) filters with disposable setup. Subsequently, 200 mL of effluent sewage water, or 100 mL of influent sewage water sample, were spiked with a 155 mixed internal standard solution to give a concentration for each compound in sample of 180 ng L-1 and 360 ng L-1, respectively. SPE was performed automatically using a GX-274 ASPEC (Gilson). Oasis HLB cartridges were conditioned by washing and rinsing with 8 mL of ACN, 8 mL of MeOH and 8 mL of Milli-Q water. Samples were loaded onto the cartridges at 5 mL min-1, and then cartridges were washed with 8 mL of Milli-Q water and dried with 160 nitrogen for 15 min at a pressure of 1 bar. Analytes were eluted using 8 mL of MeOH at a flow of 0.5 mL min-1. The SPE eluates (MeOH) were evaporated to 200 µL at 35°C under a gentle stream of nitrogen. Then, 250 µL of Milli-Q water was added and the remaining MeOH (200 µL) evaporated. Evaporation of the extracts was performed automatically using Barkey 165 optocontrol (Germany). The final extract was then made up, by weight, to exactly 250 µL with Milli-Q water. As a final step, the volume was adjusted to 500 µL, by weight, with 7/22 water:MeOH (80:20 v/v) to achieve a final percentage of 10% MeOH. An aliquot of the sample extract (20 µL) was injected directly into the LC-LTQ FT Orbitrap system. 170 2.4. Liquid Chromatography A hybrid linear ion trap Fourier Transform (LTQ FT) Orbitrap mass spectrometer was interfaced to a Surveyor HPLC system, consisting of a Surveyor auto sampler model Plus and a Surveyor quaternary gradient HPLC-pump (Thermo Fisher Scientific, Breda, The Netherlands). Chromatographic separation of the compounds was made using an XBridge C18 175 column (150 mm x 2.1 mm I.D., particle size 3.5 µm) (Waters). The pre-column used was a 4.0 x 2.0 mm I.D. Phenomenex Security Guard column (Bester, Amsterdam, the Netherlands). The analytical column and the guard column were maintained at a temperature of 21 °C in a column thermostat. An optimized gradient was used at a constant flow rate of 0.3 mL min-1 using Milli-Q water (Solvent A) and MeOH (Solvent B) both with 0.05% 180 formic acid. The percentage of organic modifier (B) was changed linearly as follows: 0 min, 5%; 20 min, 100%; 30 min, 100%; 32 min, 5%. Between consecutive runs, the analytical column was re-equilibrated for 10 min. 2.5. LTQ FT Orbitrap mass spectrometry 185 An LTQ FT Orbitrap mass spectrometer (Thermo Electron, Bremen, Germany) was used. The LTQ part of this system was equipped with an Ion Max Electrospray Ionization (ESI) probe and operated in the positive ion mode. The conditions in ESI positive mode were: source voltage 4.0 kV, heated capillary temperature 300 °C, capillary voltage 30 V and tube lens 45 V. In the LTQ component of the instrument, the temperature was set to 26 °C and 190 helium was used as damping gas. All measurements were done using the automatic gain control (AGC) of the LTQ to adjust the number of ions entering the trap. Products ions were generated in the LTQ trap at a normalized collision energy setting of 40% and using an isolation width of 2 Da. Full-scan accurate mass spectra (mass range from 50 to 600 Da) were obtained at a 195 mass resolution of 30,000 FWHM (m/z 400). The total cycle time depends upon the resolution; at the selected resolution the total cycle time is about 0.55 s. The mass spectrometer operated under data-dependent-acquisition (DDA) mode during the complete chromatographic run, in which both MS and MSn spectra were acquired simultaneously. The instrument was initially set to operate in full-scan (‘survey’) mode with accurate mass 200 measurements. When an ion exceeded a preset threshold and corresponded to the target mass 8/22 list specified by the user, the instrument switched to product-ion scan mode (MSn) in the iontrap part with nominal mass measurements. In this way, relevant information for identification and confirmation, e.g. retention time, molecular weight and fragmentation, was obtained in a single analysis. All data were acquired and processed using Xcalibur version 2.1 software. 205 Mass calibration was performed with every batch run just prior to starting the batch by using flow injection of a Polytyrosine-1,3,6 solution ([M+H]+ 182.01170 / 508.20783 and 997.39781) at a flow rate of 10 µL min-1. Identification and quantification of target compounds was performed using the accurate mass of the protonated molecule within a mass window of 5 ppm. For confirmation 210 of the identity of the compounds, in addition to the accurate mass of the precursor ion, at least one nominal mass product ion was used together with retention time, which was compared with that of the reference standards (within 2.5%) [17, 18]. 2.6. Method validation 215 The performance of the method was evaluated in terms of linearity, limits of quantification, trueness and precision. The overall recovery (including sample treatment and potential matrix effects) was studied and evaluated. Instrumental linearity was estimated by analyzing standard solutions in triplicate. Satisfactory linearity was assumed when the coefficient of determination (r2) was > 0.99, 220 based on analyte/internal standard peak areas, except for those compounds that were quantified without ILIS (absolute response). Limit of quantification (LOQ): To facilitate the fourier transformation of the acquired frequency data and conversion to m/z in the orbitrap, noise is filtered out. This is why the common approaches to evaluate the limits of quantification do not apply [19]. Therefore, a 225 different approach was applied as previously reported by de Voogt et al. [20]. It is based on the matrix suppression of the deuterated analogue and the identification criteria [18] to reach enough identification points. The matrix effect (expressed as defined by Matuszewski et al. [21]) is calculated by using the area of the accurate mass signal of the deuterated standard, spiked before extraction (in matrix), divided by the average area of the deuterated standard in 230 the calibration curve (in solvent). By using the lowest standard visible in the calibration curve which meets all the identification criteria (typically the absence/presence of the confirmation product ion is the critical parameter) and corrected for the matrix suppression and the concentration factor, the LOQ can be determined. 15/22 4. CONCLUSIONS Analytical methodology based on full-spectrum accurate-mass and MS/MS acquisition provided by LC-LTQ FT Orbitrap MS has been developed for the simultaneous quantification and confirmation of 24 target drugs of abuse at ng L-1 levels in sewage water. Although Orbitrap is recognized as an excellent analyzer for qualitative purposes, its suitability to 420 perform quantitative analysis has not been much explored. In this work, Orbitrap has been applied for the first time to the quantitative analysis of drugs of abuse in sewage water. Our data showed that this analyzer can be used for the reliable quantification with almost the same sensitivity than the most commonly used methodologies based on LC-MS/MS with triple quadrupole. The quantitative applicability has been demonstrated by method validation and 425 the analysis of quality control samples included in each sample sequence, and also via a comparison with data reported by triple quadrupole analysis. In addition, MS data provided by Orbitrap have allowed retrospective analysis leading to an indication of the presence of two ketamine metabolites. In conclusion, this unique feature of high-resolution accurate-mass spectrometry demonstrates that ketamine is likely to be present in several samples. 430 16/22 ACKNOWLEDGEMENTS The authors wish to thank Juan V. Sancho of Research Institute for Pesticides and Water (IUPA) for his useful comments. 435 L. Bijlsma is very grateful to the KWR Watercycle Research Institute for allowing him to perform an internship as visiting scientist. The financial support from the Joint Research Programme of the Dutch water companies (BTO) and from the Generalitat Valenciana, Project: Collaborative Research on Environment and Food Safety (ISIC/2012/016) is gratefully acknowledged. 440 17/22 REFERENCES [1] E. Zuccato, S. Castiglioni, R. Bagnati, C. Chiabrando, P. Grassi, R. Fanelli, Water Res. 445 42 (2008) 961-368. [2] D.R. Baker, B. Kasprzyk-Hordern, J. Chromatogr. A 1218 (2011) 8036-8059. [3] M.R. Boleda, M. Huerta-Fontela, F. Ventura, M.T. Galceran, Chemosphere 84 (2011) 1601–1607. [4] A.L.N. van Nuijs, S. Castiglioni, I. Tarcomnicu, C. Postigo, M. Lopez de Alda, H. 450 Neels, E. Zuccato, D. Barcelo, A. Covaci, Sci. Total Environ. 409 (2011) 3564-3577. [5] M. Kellmann, H. Muenster, P. Zomer, H. Mol, J. Am. Soc. Mass Spectrom. 20 (2009) 1464-1476. [6] M. Krauss, H. Singer, J. Hollender, Anal. Bionanal. Chem. 397 (2010) 943-951. [7] F. Hernández, J.V. Sancho, M. Ibáñez, E. Abad, T. Portolés, L. Mattiolo, Anal. Bioanal. 455 Chem. 403 (2012) 1251-1264. [8] A.C. Hogenboom, J.A. van Leerdam, P. de Voogt, J. Chromatogr. A 1216 (2009) 510519. [9] F. Hernández, L. Bijlsma, J.V. Sancho, R. Díaz, M. Ibáñez, Anal. Chim. Acta 684 (2011) 96-106. 460 [10] J. Nurmi, J. Pellinen, J. Chromatogr. A 1218 (2011) 6712-6719. [11] I. González-Mariño, J.B. Quintana, I. Rodríguez, M. González-Díez, R. Cela, Anal. Chem. 84 (2012) 1708-1717. [12] J.A. van Leerdam, A.C. Hogenboom, M.M.E. van der Kooi, P. de Voogt, Int. J. Mass Spectrom. 282 (2009) 99-107. 465 [13] A. Kaufmann, P. Butcher, K. Maden, S. Walker, M. Widmer, Rapid Commun. Mass Spectrom. 25 (2011) 979-992. [14] S. Vogliardi, D. Favretto, M. Tucci, G. Stocchero, S.D. Ferrara, Anal. Bioanal. Chem. 400 (2011) 51-67. [15] R. Pinhancos, S. Maass, D.M. Ramanathan, J. Mass Spectrom. 46 (2011) 1175-1181. 470 [16] J.M. Marín, E. Gracia-Lor, J.V. Sancho, F.J. López, F. Hernández, J. Chromatogr. A 1216 (2009) 596-612. [17] O.J. Pozo, J.V. Sancho, M. Ibáñez, F. Hernández, W.M.A. Niessen, Trends Anal. Chem. 25 (2006) 1030-1042. 18/22 [18] Commission Decision 2002/657/EC of 12 August implementing Council Directive 475 96/23/EC concerning performance of analytical methods and the interpretation of results. Off. J. Eur. Commun. L221/8 (2002). [19] A. Kaufmann, M. Widmer, K. Maden, Rapid Commun. Mass Spectrom. 24 (2010) 2162-2170. [20] P. de Voogt, E. Emke, R. Helmus, P. Panteliadis, J.A. van Leerdam, in: S. Castiglioni, 480 E. Zuccato, R. Fanelli (Eds), Mass Spectrometric Analysis of Illicit Drugs in the Environment, Wiley, New York, 2011, pp. 85-114. [21] B.K. Matuszewski, M.L. Constanzer, C.M. Chavez-Eng, Anal. Chem. 75 (2003) 30193030. [22] S. Castiglioni, E. Zuccato, E. Crisci, C. Chiabrando, R. Fanelli, R. Bagnati, Anal. Chem. 485 78 (2006) 8421–8429. [23] M. Stüber, T. Reemtsma, Anal. Bioanal. Chem. 378 (2004) 910-916. [24] R.J.B. Peters, A.A.M. Stolker, J.G.J. Mol, A. Lommen, E. Lyris, Y. Angelis, A. Vonaparti, M. Stamou, C. Georgakopoulos, M.W.F. Nielen, Trends Anal. Chem. 29 (2010) 1250-1268. 490 [25] N.G.F.M. van der Aa, E. Dijkman, L. Bijlsma, E. Emke, B.M. van de Ven, A.L.N. van Nuijs, P. de Voogt, RIVM report 703719064, Bilthoven, (2010) pp. 1-90. Available from URL: www.rivm.nl (accessed September 2012) [26] N.G.F.M. van der Aa, L. Bijlsma, E. Emke, E. Dijkman, A.L.N. van Nuijs, B.M. van de Ven, F. Hernández, J.F.M. Versteegh, P. de Voogt, Water Res. under revision. 495 [27] F. Hernández, M. Ibáñez, E. Gracia-Lor, J.V. Sancho, J. Sep. Sci. 34 (2011) 3517-3526. [28] K-C. Wang, T-S. Shih, S-G. Cheng, Forensic. Sci. Int. 147 (2005) 81-88. [29] L. Bijlsma, J.V. Sancho, E. Pitarch, M. Ibáñez, F. Hernández, J. Chromatogr. A 1216 (2009) 3078-3089. 500 19/22 Table 1: Exact masses of the protonated target drugs of abuse, nominal masses and relative abundance of product ions, together with their retention times and isotope labelled internal standards used for quantification. Compound tR Precursor ion [M+H]+ Product ion 1 Product ion 2 Internal standard (min) m/z m/z m/z RA (%) Amphetamine 10.28 136.11208 119.1 91.1 0.5 Amphetamine-d11 Methamphetamine 10.64 150.12773 119.0 91.1 9.0 Methamphetamine-d5 MDA 10.75 180.10191 163.2 MDA-d2 MDMA 10.90 194.11755 163.1 58.0 1.0 MDMA-d5 MDEA 11.66 208.13321 163.1 72.0 2.7 MDEA-d5 Cocaine 13.42 304.15433 182.1 150.2 2.6 Cocaine-d3 Benzoylecgonine 12.51 290.13868 168.2 272.2 4.8 Benzoylecgonine-d3 Heroin 13.07 370.16490 328.2 268.2 99.1 n/a Morphine 4.71 286.14334 201.1 229.1 51.9 Morphine-d3 6-MAM 10.33 328.15433 211.2 268.2 73.7 6-MAM-d3 Methadone 18.80 310.21654 265.1 247.2 0.1 n/a Codeine 9.10 300.15942 215.2 243.1 47.7 6-MAM-d3 THC 26.26 315.23186 259.2 193.2 76.7 THC-d3 THC-COOH 24.84 345.20604 327.2 299.3 6.1 THC-COOH-d3 OH-THC 24.48 331.22677 313.3 OH-THC-d3 Ketamine 12.43 238.09932 220.1 207.1 23.9 n/a Methylphenidate 13.61 234.14886 84.0 174.2 0.3 n/a Oxazepam 19.53 287.05818 269.1 241.1 3.9 Oxazepam-d5 Diazepam 20.62 285.07892 257.1 222.2 30.4 Diazepam-d5 Temazepam 19.85 301.07383 283.0 255.2 9.2 Nordazepam-d5 Nordazepam 20.13 271.06327 243.1 208.1 37.7 Nordazepam-d5 Desalkyl-flurazepam 19.67 289.05385 261.1 140.0 44.5 Nordazepam-d5 mCCP 13.62 197.08400 154.0 119.1 6.9 n/a Fentanyl 15.66 337.22744 188.2 216.3 5.6 Nordazepam-d5 RA: relative abundance of product ions. n/a: adequate internal standard was not available. 505 20/22 Table 2: Method validation in influent (n = 4) and effluent (n = 4) sewage water Compound Influent Effluent Linearity Ra (%) CVb (%) LOQ (ng L-1) Ra (%) CVb (%) LOQ (ng L-1) r2 Instrumental LOQ (pg) Amphetamine 104 8 40 105 6 4 0.9960 58 Methamphetamine 98 4 15 92 7 5 0.9994 28 MDA 113 13 360c 92 4 158 0.9996 720 MDMA 102 4 12 97 7 4 0.9999 14 MDEA 101 4 17 98 5 4 0.9999 14 Cocaine 70 6 40 93 6 6 0.9999 14 Benzoylecgonine 111 8 10 93 7 2 0.9999 14 Heroin 70 20 19 72 21 7 0.9943 28 Morphine 102 5 360c 98 12 125 0.9996 144 6-MAM 117 11 19 119 15 7 0.9995 14 Methadone 73 19 45 76 5 6 0.9917 14 Codeine 90 15 19 120 26 7 0.9988 28 THC 109 11 360c 94 12 180c 0.9995 720 THC-COOH 102 7 33 90 5 7 0.9995 28 OH-THC 82 4 68 85 10 35 0.9988 58 Ketamine 48 21 10 81 7 2 0.9956 14 Methylphenidate 45 20 20 65 27 2 0.9969 14 Oxazepam 97 6 14 91 6 4 0.9994 28 Diazepam 100 4 18 94 7 6 0.9998 20 Temazepam 105 6 4 90 7 2 0.9980 28 Nordazepam 96 5 4 91 8 2 0.9995 28 Desalkyl-flurazepam 88 25 4 109 15 2 0.9994 28 mCCP 52 28 20 62 18 6 0.9911 14 Fentanyl 73 16 4 74 14 2 0.9964 14 a Trueness, estimated by means of recovery experiments. 510 b Precision, expressed as repeatability in terms of coefficients of variation. c These values were derived from validation experiments (for detailed explanation see text). 515 21/22 Table 3: Comparison of concentrations of drugs of abuse detected in influent and effluent sewage waters analyzed by LC-MS/MS with triple quadrupole (two different laboratories) and by the Orbitrap method presented in this work. Each number is the result of a single measurement of the pertaining sample using a specific MS detection technique Compound Influent sewage water (ng L-1) Effluent sewage water (ng L-1) Sample 1 Sample 2 Sample 3 Sample 4 QqQ 1a QqQ 2b Orbitrap QqQ 1 QqQ 2 Orbitrap QqQ 1 QqQ 2 Orbitrap QqQ 1 QqQ 2 Orbitrap Amphetamine 95 117 123 282 249 245 - - - - - - MDMA 21 96 144 - 56 86 84 88 137 50 54 76 Cocaine 439 296 dc 179 114 dc 2 - - 14 - 8 Benzoylecgonine 1178 1136 1637 528 645 615 26 19 45 85 77 99 THC-COOH 378 n/ad dc - n/a - - n/a - - n/a - Ketamine n/a -e - n/a - - n/a - 2 n/a 16 6 a Pre-treatment: centrifugation; Pre-concentration by SPE (Oasis MCX, 150 mg); pre-concentration factor: influent 10x, effluent, 50x; [29] b Pre-treatment: none; Pre-concentration by SPE (Oasis HLB, 200 mg); pre-concentration factor: influent 50x, effluent 250x; [25] 520 c d: detected d n/a: no data available e - : not detected 22/22 FIGURE CAPTIONS 525 Figure 1: The influence of intake volume (100, 300, 600 and 900 mL) on matrix effect in a typical influent Figure 2: Average matrix effects [21] observed in both influents (100 mL) and effluents (200 530 mL) for 5 different STPs. Figure 3: The mass deviation (mass drift) of the protonated ions of selected deuterated analytes in relation to the theoretical mass over a period of 55 hours. 535 Figure 4: LC-MS (ESI + mode) extracted-ion chromatograms (left) and MS/MS spectra (right) of drugs of abuse detected in influent (A) and effluent (B) sewage water from the sewage treatment plant of Amsterdam. Concentrations found in these samples were the following (influent and effluent, respectively); MDMA: 136 and 190 ng L-1; benzoylecgonine: 3701 and 155 ng L-1; THC-COOH: 431 and 22 ng L540 1; Oxazepam: 430 and 422 ng L-1. Arrows indicate chromatographic peak of MDMA. Figure 5: LC-MS (ESI+ mode) extracted-ion chromatograms of ketamine, norketamine and dehydronorketamine in an influent sewage water sample from Eindhoven 545 (retrospective search). 0.0 20.0 40.0 60.0 80.0 100.0 120.0 140.0 160.0 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 Matrix effect Morphine-d3 6-MAM-d3 MDMA-d5 MDA-d2 Amphetamine-d11 Methamphetamine-d5 MDEA-d5 Cocaine-d3 Benzoylecgonine-d3 Oxazepam-d5 Diazepam-d5 OH-THC-d3 THC-COOH-d3 THC-d3 Figure 1 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% THC-d3 Morphine-d3 Cocaine-d3 MDEA-d5 OH-THC-d3 THC-COOH-d3 Metamphetamine-d5 MDMA-d5 Amphetamine-d11 Benzoylecgonine-d3 6-MAM-d3 Diazepam-d5 MDA-d2 Nordazepam-d5 Oxazepam-d5 matrix effect Influent Effluent Figure 2