. ‘t L. a...» a 9...1.I!.:5.r..l;.. 2' ski-I53}! : s 2!! e . :- ‘ E, 1:: .» t. 5. plilif.‘ ! aid»... .. 2b. :31. nuwuuin .... V at... .5... R It! .n.‘v Lana-«y... 2 2.: .3 .Jiéas . e ..3§..£..%.§ .U‘l II.VI\‘Q&‘-‘fi a! ‘31. II .QL \Ixiso‘...itlI-.Ilfim \Ix ilt-lail: 1;. a . xti‘u‘..to£:'IOSI I“ {I Q1.|1-.£i\’!.p Infill... I. .15! .xv {Cl-.6! Iv .- Shclt 1100.0..‘2: 0 I!!! e. . 2..“ .l'l! :9 x g 2... ~ o I I '1. 3. .. Mal...» :. -T.SB.¥£§. . .2: uw is .4... ui~ch|I \IP. .. 1:013} 15!!!! :- .Gll.irl..‘ll!.0v1.i\uitvliltnoi nil! isiiilzvlts‘bofl ’II‘J.&.5:\1.XII:I.$..I) . ‘glnhtflisvrfllnll 111:. v. 1 i In! tittlriv! :‘O‘r‘fln’l IrIIIIQ. £01.: Q‘tivfifi {xiii}. .{ IVE‘CQ‘ Kill! I'vtl...n.i.ilno 2.. 1‘1"... 3...!!! In $515 LIBRARY Mic..ig:.n State 3‘ 3 University This is to certify that the thesis entitled COMPARISON OF EXTRACTION PROCEDURES FOR THE EXTRACTION OF ORGANIC IMPURITIES FROM SEIZED 3,4-METHYLENEDIOXYMETHAMPHETAM|NE (MDMA) TABLETS presented by SARAH CHRISTINE MEISINGER has been accepted towards fulfillment of the requirements for the MS. degree in Criminal Justice Mad Major ProfesTs‘o‘f's Signature 03/21 /a 1 Date MSU is an Affinnative Action/Equal Opportunity Employer PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 5108 K:lProj/Acc&Pres/ClRC/DateDue.indd ABSTRACT COMPARISON OF EXTRACTION PROCEDURES FOR THE EXTRACTION OF ORGANIC IMPURITIES FROM SEIZED 3,4- METl-IYLENEDIOXYMETHAMPHETAMINE (MDMA) TABLETS By Sarah Christine Meisinger Organic impurity profiling of seized 3,4-methylenedioxymethamphetamine (MDMA) tablets is used to link tablets by identification of common synthetic routes and common production batches. Currently, liquid-liquid extraction (LLE) is typically used to extract organic impurities from MDMA tablets, afier which the extract is analyzed by gas chromatography-mass spectrometry (GC-MS). This research investigated two alternative extraction procedures; headspace-solid phase microextraction (HS-SPME) and microwave-assisted extraction (MAE). HS-SPME offers more selective extractions than LLE, requires less sample preparation, and impurities are pro-concentrated on the SPME fiber. MAE offers highly efficient extractions, and requires fewer steps than LLE. MAE is often followed by HS-SPME for the selective extraction of the target analytes. In this research, a HS-SPME procedure was optimized using an empirical procedure. The optimized HS-SPME procedure was then used to extract impurities from several different exhibits of MDMA for impurity profile comparison. Additionally, a MAE procedure was developed and used in combination with HS-SPME to selectively extract impurities from the MAE extract. The HS-SPME and the MAE/HS-SPME procedures were compared with a LLE procedure in terms of the number of impurities extracted from a homogenized batch of seized MDMA. Both HS-SPME and MAE/HS- SPME resulted in more informative impurity profiles and show potential for impurity profiling applications. For my family, who always support me through every challenge I face, and for LTC Craig S. Harju, Sr., who always encouraged me to follow my dream of becoming a forensic scientist- Drive On, Sir. iii ACKNOWLEDGEMENTS I would like to thank my advisor Dr. Ruth Smith, for all her help and guidance throughout this project, and for all her input throughout the writing process. Finishing the writing process long-distance was not an easy task, and could not have been done without her constant communication and support. I would also like to thank my committee members Frank Schehr and (CJ Faculty) for graciously volunteering their time to read this thesis and offer their insight. Thank you to all the forensic chemistry students for attending my practices and offering additional input and support. I would like to acknowledge the Michigan State Police Forensic Science Division for supplying the program with MDMA tablets. Without their help, this project could not move forward. I would also like to acknowledge that a portion of this research was supported by Department of Justice Grant # 2005CKWX0466, which allowed for the purchase of supplies. This grant also allowed me to travel to the 2007 Midwest Association of Forensic Scientists meeting to present this research. Thank you to all the family and friends who encouraged and supported me throughout this process. 2: m TABLE OF CONTENTS List of Tables ........................................................................................................................ v iii List of Figures......................... ..................................... ix Key to Abbreviations ............................................................................................................. xi Chapter 1: Introduction .................................................................................................... 1 1.1 MDMA ....................................... 1 1.1.1 MDMA Abuse Statistics ........................................................................................ 2 1.12 MDMA Trafficking in the US. .................................... 3 1.2 Clandestine Production of MDMA _ _______________ 5 1.2.1 Common Synthetic Routes of MDMA," 5 1.2.2 Clandestine Production of MDMA Tablets __________ 6 1.3 Profiling of Illicit Drugs ................................................. 7 1-3-1 Physical Drug Profiling ......................................................................................... 7 1.3.2 Compositional Profiling" _________________ _9 1.3 3 Chemical Impurity Profiling ....................................... 10 1.4 Headspace- -Solid Phase Microextraction (l-IS- -SPME) 14 1.5 Microwave-Assisted Extraction (MAE) ____________________ 15 1.6 Research Objectives ....................................................................... 16 1.7 References .................................................................................... .-.18 Chapter 2: Theory ................................................................................ 21 2.1 Solid-Phase Microextraction (SPME) .......................................................................... 2 1 2 2 Deve10ping a SPME Procedure ..................................................... 23 2 2 1 Sampling Procedure .............................................. 23 2 2 2 Contributing Factors to Extraction Efficiency ................................................... 24 2 2 2 1 Fiber Coating ..................... 25 222.2 Fiber Thickness .................................................... 26 2.2.2.3 Sample Heating _ .............. 26 2224 Additional Factors ......................................... 26 2.3 SPME for Drug Impurity Profiling 28 2-4 Microwave-Assisted Extraction (MAE) ................................. 32 2.4.1 MAE Heating and Partitioning .......... 33 2 4 2 MAE Extraction Solvent ......................................... 34 2 4 2 1 Solvent Polarity ............................................................. 34 2 4 2 2 Dielectric Constant of Solvent ............................................. 35 2423 Additional Factors ................................ 35 2.4.3 MAE/HS-SPME ......................... 37 2.5 Applications of MAE and MAE/l-lS-SPME" 38 2.6 Gas Chromatography-Mass Spectrometry (GC-MS) ,,,,,,,,,,,,,,,, 39 2.7 Pearson Product Moment Correlation ,,,,,,,,,,,,,,,,,,,,,,,, “.42 2.8 References Chapter 3: Materials and Methods 3.1 Physical Description of MDMA Tablets 3.2 Optimization of Headspace-Solid Phase Microextraction (HS-SPME) Procedure" 3.2.1 Buffer Preparation" 3. 2. 2 General HS- SPME Procedure 3. 2. 3 Empirical Optimization of HS- SPME Extraction Time and Extraction Temperature 3.3 HS-SPME for Impurity Profiling from Multiple MDMA Exhibits 3.4.1 General MAE Procedure 3.4.2 Development of Microwave Extraction Time and Extraction Temperature 3.4.3 Development of HS- SPME Extraction Time and Extraction Temperature .. ____51 Following MAE 3.5 Liquid-Liquid Extraction (LLE) 52 3.6 GC- MS Parameters 53 3.6.1 GC- MS of HS SPME Extracts 3. 6. 2 GC- MS of LLE Extracts 3. 7 Pearson Product Moment Correlation (PPMC) Coefficients 3. 8 References," 53 54 54 56 Chapter 4: Results and Discussion: Optimization of Headspace-Solid Phase Microextraction (HS-SPME) _______ 57 4.1 Optimization of HS-SPME Procedure for MDMA Sample in Aqueous Buffer 4.1.1 Empirical Optimization of HS-SPME using Carbonate Buffer (0.05M, pH 10) 4.1.2 Empirical Optimization of HS- SPME using Phosphate Buffer (0.1M, pH 7) 4.1.3 Comparison of Carbonate and Phosphate Buffers for HS- SPME .. Impurity Extraction 4.1.4 Precision of Optimized HS-SPME Procedure 59 62 69 72 4.2 Comparison of HS-SPME and Liquid-Liquid Extract-ion (LLE) for the Extraction of Organic impurities from MDMA ,,,,,,,,,,,,,,,,,,,,,,, 76 4.3 Impurity Profiling of Multiple MDMA Exhibits 4.4 Summary 4.5 References“ Chapter 5: Results and Discussion: Development of Microwave-Assisted Extraction/Headspace-Solid Phase Microextraction (MAE/HS-SPME)_ 5.1.1 General Procedure 81 85 _88 89 89 89 5.1.2 Development of MAE using Phosphate Buffer (0.5M, pH 7) as Extraction Solvent vi ..90 5.1.3 Development of MAE using Carbonate Buffer (0.05M, pH 10) as Extraction Solvent ...................................... 94 5.1.4 Comparison of Phosphate Buffer and Carbonate Buffer as Extraction Solvent for MAE ..................................... 5.2 Development of HS- SPME Following MAE _97 99 5.3 Comparison of MAE/HS- SPME with Liquid- Liquid Extraction (LLE) and HS- SPME for the Extraction of Organic Impurities from MDMA _ 5.4 Summary" 5.5 References Chapter 6: Conclusions and Future Work 6.1 Conclusions 6.2 Future Work IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII Appendices... ............... APPENDIX A: Common Synthetic Routes for MDMA" APPENDIX B: Common Impurities from MDP2P Synthesis APPENDIX C: Comparison of Impurities Present in Five Exhibits vii __ _119 _ 125 105 112 114 -- 115 115 124 124 126 LIST OF TABLES Table 4.1. Comparison of optimized HS-SPME procedures using two different buffers by the number of impurities, additives and components extracted ,,,,,,,,,,,,,,,, 70 Table 4.2. Precision of HS-SPME extraction procedure over four days .................... 73 Table 4.3. PPMC coefficients for replicate HS-SPME extractions over a period of four days ..................................................................................................................... 74 Table 4.4. Comparison of HS—SPME and LLE by the number of impurities, additives and components extracted. Peaks in the chromatogram not listed in the table are due to fiber or column bleed ................................................................ 78 Table 4.5. Physical characteristics of various exhibits of MDMA tablets ,,,,,,,,,,,,,,,,, 82 Table 5.1. Summary of impurities, additives and components extracted from homogenized MDMA using phosphate and carbonate buffers as extraction solvents ............................................................................................................................ 97 Table 5.2. Precision of MAE/HS-SPME at different HS-SPME extraction temperatures based on RSDS of impurity peak areas ................................................. 103 Table 5.3. Comparison of RSD values for various impurities extracted by PIS-SPME alone versus MAE/HS-SPME ................................................................... 104 Table 5.4. Comparison of LLE, HS-SPME and MAE/HS-SPME by the number of impurities, additives and components extracted. Peaks in the chromatogram not listed in the table are due to fiber or column bleed .................................................... 108 viii LIST OF FIGURES Figure 1.1. Chemical structure of MDMA ,,,,,,,,,,,,,,, 1 Figure 2.1. (a) Diagram of a SPME fiber and holder. (b) Expanded view of a SPME fiber exposed to a sample, the small circles representing analytes _________________ 22 Figure 3.1. MDMA tablet from the CJ-FS-05 exhibit. ,,,,,,, 46 Figure 4.1 Effect of extraction time on chromatographic resolution for (a) 30 minute extraction time at 60 °C and (b) 60 minute extraction time at 60 °C__60 Figure 4.2. Effect of extraction temperature on chromatographic resolution for (a) 30 minute extraction time at 50 °C and (b) 30 minute extraction time at 80 °C_,61 Figure 4.3. Effect of extraction temperature on chromatographic resolution for (a) unidentified impurity (RT 10.88 minutes) and column bleed peak (RT 10.83 minutes) and (b) ephedrine/pseudoephedrine (RT 13.05 minutes) and 3,4-MDDMB (RT 13.18 minutes) ,,,,,,,, __ ___________ 62 Figure 4.4. Effect of extraction time on chromatographic resolution for (a) 30 minute extraction time at 60 °C, (b) 50 minute extraction time at 60 °C, and (c) 60 minute extraction time at 60 °C_ _ 64 Figure 4.5. Effect of extraction temperature on chromatographic resolution for (a) 40 minute extraction at 40 °C, (b) 40 minute extraction at 70 °C, and (c) 40 minute extraction at 80 °C ,,,,,,,,,,,, 66 Figure 4.6. HS-SPME chromatograms using phosphate buffer: (a) 0.1M phosphate buffer (pH 7) at 70 °C for 40 minutes. (b) 0.5M phosphate buffer (pH 7) at 70 °C for 40 minutes ,,,,,,, 69 Figure 4.7. Comparison of chromatograms using two different extraction procedures. (a) LLE. (b) HS-SPME (carbonate buffer, 0.05M, pH 10; 30 minute extraction at 60 °C) _____ 77 Figure 4.8. HS-SPME Chromatograms of different exhibits of MDMA tablets using the optimized extraction with 0.05 M carbonate buffer (pH 10). (a) CJ-FS-Ol (b) CJ-FS-02 (c) CJ-FS-03 (d) CJ-FS-OS (e) MSU-900-01 ,,,,,,,,,,,,,,,,,, 83 Figure 5.1. Chromatograms comparing different MAE temperatures using phosphate buffer (0.5M, pH 7) with a 30 minute extraction time: (a) 90 °C, (b) 100 °C, and (c) 110 °C. *Peak at RT 12.15 minutes is due to fiber bleed 91 Figure 5.2. Chromatograms comparing different MAE times using phosphate buffer (0.5M, pH 7) at 100 °C: (a) 20 minutes, (b) 30 minutes, and (c) 40 minutes. *Peak at RT 12.15 minutes is due to fiber bleed 94 Figure 5.3. Chromatograms comparing different MAE extraction temperatures using carbonate buffer (0.05M, pH 10) with a 30 minute MAE extraction time: (a) 90 °C MAE extraction, (b) 100 °C MAE extraction, and (c) 110 °C extraction ,,,,,,,,,,,,,,,,,, 96 Figure 5.4. Comparison of HS-SPME extraction times for MAE/HSPME: (a) 5 minute HS-SPME extraction, (b) 10 minute HS-SPME extraction, and (c) 15 minute HS-SPME extraction _______ 101 Figure 5.5. Comparison of impurity profiles obtained for MDMA exhibit CJ—FS-OS using (a) LLE, (b) HS-SPME, and (c) MAE/HS-SPME for impurity extraction ,,,,,, 106 3,4-MDDMB amu CAR CCC CHAMP CHEDDAR CSA CW DEA DVB DVB/CAR/PDMS El GC GC-FID GC-MS HIDTA HS-SPME LLE MAE MAE/HS-SPME MDA KEY TO ABBREVIATIONS 3,4-methylenedioxy-N,N-dimethylbenzylamine atomic mass unit CarboxenTM central composite circumscribed Collaborative Harmonization of Methods for Profiling Amphetamine Type Stimulants Collaborative Harmonized European Database Determination of Amphetamine Relations Controlled Substances Act Carbowax® Drug Enforcement Agency divinylbenzene divinylbenzene/CarboxenTM/polydimethylsiloxane electron ionization gas chromatography gas chromatography-flame ionization detector gas chromatography-mass spectrometry high-intensity drug trafficking area headspace solid-phase microextraction liquid-liquid extraction microwave-assisted extraction microwave-assisted extraction/headspace solid-phase microextraction 3,4-methylenedioxyamphetamine xi MDEA MDMA MDP2P MDP2BP MDP2-propanol MS MSP NFLIS PA PCA PDMS PDMS/DVB PPMC PSE/EPH Rf RSD RT SPME UNID 3,4-methylenedioxy-N-ethylamphetamine 3,4-methylenedioxymethamphetamine methylenedioxyphenyl-Z-propanone 3,4-methylenedioxy-phenyl-2-bromopropane methylenedioxyphenyl-2-propanol mass spectrometry Michigan State Police National Forensic Laboratory Information System polyacrylate principal components analysis polydimethylsiloxane polydimethylsiloxane/divinylbenzene Pearson product moment correlation pseudoephedrine/ephedrine radio frequency relative standard deviation retention time solid phase microextraction United Nations unidentified xii CHAPTER 1 INTRODUCTION 1.1 MDMA 3,4-methyIenedioxymethamphetamine (MDMA), is more commonly known as the club drug ‘Ecstasy’. MDMA is a member of the phenethylamine drug class, which includes other stimulants such as amphetamine and methamphetamine. However, the methylenedioxy group located on the benzene ring of MDMA imparts hallucinogenic properties (Figure 1.1). MDMA was first synthesized by Merck in 1912, and was patented in 1914 as a precursor for a haemostatic substance, not as an appetite suppressant as has been previously reported [1]. Because it was at first only seen usefiil as a styptic compound, its pharmacological effects were not tested until 1927, although not in humans. Some of the first illicit MDMA tablets were seized in Chicago in 1970 [2], and since then, MDMA has had varying periods of popularity as an illicit drug. CH 1 3 N\ §(a) 1- Benzaldehyde l 2- Benzeneacetamide l 4 3- Safrole 3 I ‘i I i 1 l i ” i ii . l , l —i ii i ' - .l . l: l' l. i.‘ 5’. ii 1 1.1 ‘1 ll} .i 2‘ l j r ‘.l q l 8 10E+0771 I iil j hwil‘ Li H C - ,1‘ i T. I l r . ii g 03: 7 a -l 4 -1-s.L"‘AU3 -let...if“"A“'Uwvw’d't‘kflfl- Lg“, 3 1.0E+08 4 2 3 (b) l '1 l‘ 5 l. ! ’i H l" I‘ ’ ‘ : l” l ? l l l! :4 l ‘ i ll " l l l l J. 2 ‘ 1W1. . j H i‘ {l 1 7; ’iii ‘ l ‘1 1.0E+O712 l l: l‘ l ”ii I» !' . l “ ’3 v. ’j * P' ‘ ll - 11 if.“ ..iil't .3 Ail Ill! . Still? ..i In“‘Llieiiii’i‘iLU.iw‘wj'l‘d ,..\u. J .I , - ; 0 ...».,..-.—rm.r»»...vA.H—.vvn..’.-.+rna-~.-_.‘rffignaw—'rnpu-r‘r—rn...-LT..-.-Tw-.—..--—rrr.,—.~.-T-....I—vrfi*vm“-.fi—nfivn-. 0 25 Retention Time (min) Figure 4.6. HS-SPME chromatograms using phosphate buffer: (a) 0.1M phosphate buffer (pH 7) at 70 °C for 40 minutes. (b) 0.5M phosphate buffer (pH 7) at 70 °C for 40 minutes. 4.1.3 Comparison of Carbonate and Phosphate Buffers for HS-SPME Impurity Extraction Table 4.] summarizes the impurities, additives and components extracted using phosphate buffer (0.5 M, pH 7) and carbonate buffer (0.05 M, pH 7). In terms of the number of impurities extracted, the phosphate buffer extracted l7 impurities compared to 24 impurities extracted using the carbonate buffer. 69 Table 4.1. Comparison of optimized HS-SPME procedures using two different buffers by the number of impurities, additives and components extracted. Retention Prominent Phosphate Carbonate Tentative Peak Identity Time (min) m/z pH 7 pH 10 0.5M 0.05M *5.48/5.47 105, 91, 77, X X Benzaldehyde 65 8.27 135, 136, 77, X 3,4-methylenedioxytoluene 78 8.52 146, 105, 77, X 1,2-dimethyl-3-phenyl-trans- 91,117,131, aziridine 132 866/863 91, 92, 65, X X Benzeneacetamide 89, 134 9.43/9.50 77, 105, 51, X X Unidentified (UNID)1 106 *9.80/9.88 58, 91, 150, X X Methamphetamine 65 10.53 56, 91, 58, X UNID 2 65, 89, 146, 115, 130 11.58/11.54 162, 131, X X Safrole 103, 77, 104, 132, 78 10.86 72,91,131, X UNID3 70, 115 *12.36./12.34 149, 150, X X Piperonal 121, 65, 67, 91 *12.43/12.38 162, 149, X X lsosafrole 131, 103, 102, 77, 63, 91 *13.07/ 13.04 58, 77, 105, X X Ephedrine/Pseudoephedrine 146 /(Unidentified-phosphate)T *13.19/13.16 135, 136, 77, X X 3,4-MDDMB 178, 179, 58, 79, 105, 148 *l4.74/14.76 135, 178, 77, X X MDP2P 79, 136, 51 *14.95/ 15.03 135, 136, X X MDP2-propanol 180, 77, 78, 79, 106 70 Table 4.] (continued) 15.37 159, 131, X UNID-4 135, 91, 160, 128, 202 *15.69/15.79 58, 194, 77, X X MDMA 135 *15.88/ 16.82 72, 70, 77, X X MDEA 135 16.97 182, 105, 77, X Benzophenone 181, 51 17.66/ 17.65 195, 180, X UNID 5 165, 210, 72 17.92 190, 147, 3-methyl-6,7-methy1enedioxy- 148, 72, 13 5, 3,4-dihydroisoquinolin-1(2H)one I88, 208 18.31 58, 100, 72, X UNID 6 135 19.30/ 19.29 120, 121, X X Salicylic acid 138, 92 *19.96/19.84 194, 193, 55, X X Caffeine 109 20.90 162, 58, 135, X MDA Acetate 77, 135, 163 Total Impurities/Additives 17 24 * Indicates peaks that eluted at different retention times with the two different extraction methods. ’r Both ephedrine and pseudoephedrine are given as the provisional identity for the peak at 13.07/4 minutes, as these two compounds are diastereomers, and are difficult to differentiate by mass spectrometry alone. The phosphate extraction also had a third unidentified component that co-eluted with the pseudoephedrine/ephedrine. While the phosphate buffer and the carbonate buffer extracted many common impurities, two impurities were extracted only using the phosphate buffer: an unidentified impurity (RT 13.04 minutes) that co-eluted with pseudoephedrine/ephedrine and benzophenone (RT 16.95 min). These impurities likely have greater affinity for the headspace than the buffer at pH 7, indicating that these impurities likely have a pKa value over 7, thus making them more soluble in the buffer at higher pH values. Identification of the unidentified impurity is needed to determine its utility for impurity profiling. As mentioned previously, benzophenone could potentially be a contaminant from drying 71 THF, or as an additive as observed in tablets seen in New Zealand [5]. However, the phosphate buffer did not extract the impurities 3,4-methylenedioxytoluene, 1,2-dimethy1- 3-phenyl-trans-aziridine, and 3-methyl-6,7-methylenedioxy-3,4-dihydroisoquinolin-1(2H)one as well as five other unidentified impurities that were extracted with the carbonate buffer (RT 10.52, 10.86, 15.37, 18.32, 20.90 min). Again, the difference in pH likely affected the extraction of these compounds by increasing their affinity for the buffer solution over the headspace, effectively increasing the solubility of these impurities in the buffer. Despite improved chromatographic peak shape of the more abundant impurities (MDP2P, MDP2-propanol, MDMA, and MDEA) extracted with the phosphate buffer, fewer impurities were extracted than with the carbonate buffer. Thus, the optimized HS-SPME parameters using the carbonate buffer provided more informative profiles for this MDMA exhibit, and these parameters were used in all subsequent studies. 4.1.4 Precision of Optimized HS-SPME Procedure The precision of the optimized HS-SPME method of MDMA in carbonate buffer was assessed. A new DVB/CAR/PDMS fiber was used, and each 50 mg sample of MDMA homogenate was dissolved in 5 mL of 0.05 M carbonate buffer, pH 10. Samples were extracted by HS-SPME as previously described, using an extraction time of 30 minutes and extraction temperature of 60 °C. For the precision study, sixteen extractions were conducted over a four day period. Relative standard deviation (RSD) values based on peak areas of selected impurities are given in Table 4.2. Extraction 2 was eliminated from RSD calculations for day l, and extractions 10 and 12 were eliminated from RSD calculations for day 3, as some of the peak areas in these extractions were statistically determined to be outliers. 72 Table 4.2. Precision of HS-SPME extraction procedure over four days. Impurity (retention time in Relative standard deviation (%) of impurity minutes) peak area (Provisional identification) Day 1 Day 2 Day 3 Day 4* Benzeneacetamide (8.67) 6.7 12.7 7.1 5 .5 Methamphetamine (9.92) 17.0 7.0 7.5 7.3 Isosafrole (12.40) 9.1 11.1 7.5 2.5 Ephedrine/Pseudoephedrine (13.05)T 32.7 16.2 12.7 1 1.7 3,4-methylenedioxy- 12.3 5 .4 1.8 8.0 dimethylbenzylamine (13.18) MDP2P (14.80) 31.9 26.9 20.8 14.3 MDP2-propanol(15.02) 40.1 18.4 29.5 1 1.0 MDEA (16.79) 14.8 3.1 13.3 6.5 Salicylic acid (19.30) 28.7 3.6 16.6 35.7 *n=3ondays1,2and3,n=4onday4 1“ Both ephedrine and pseudoephedrine are given as the provisional identity for the peak at 13.05 minutes, as these two compounds are diastereomers, and are difficult to differentiate by mass spectrometry alone. The early eluting impurities (those with retention times before 14 minutes) and MDEA have RSD values mostly below 15% both within and between days, indicating acceptable precision of the HS-SPME procedure. These values are comparable with repeatability values obtained by Kongshaug, who reported RSD values ranging from 2.2 to 12.6% (n = 6) for HS-SPME of MDMA tablets dissolved in acetate buffer using a PDMS/DVB fiber [1]. However, in the study reported herein, the RSD values increase for the later eluting impurities (Table 4.2), which must be further investigated. One possible way for improving the precision of the later eluting impurities is to investigate the effect of the GC temperature program on the resolution and peak width in the later part of the chromatogram. Theoretically, by increasing the ramp rate of the temperature program for this part of the chromatogram, peak broadening will be reduced, with will ideally 73 improve resolution. However, a disadvantage of increasing the ramp rate is that it could cause co-elution of peaks, thus complicating their identification using GC-MS. The above precision data is based upon peak areas of individual impurities throughout the chromatogram. For impurity profiling, comparing the entire chromatogram rather than the peak areas of individual impurities is more useful. Pearson product moment correlation (PPMC) coefficients were calculated to allow a pair-wise comparison between impurity profiles. Table 4.3 shows PPMC coefficients for all pair- wise comparisons by day, as well as the average relative standard deviation for each day and overall. Table 4.3. PPMC coefficients for replicate HS-SPME extractions over a period of four days. Day Extract Pair PPMC coefficient Average d: standard deviation 1 1 and 3 0.7531 0.5875 1: 0.1835 1 and 4 0.3902 3 and 4 0.6194 2 6 and 7 0.9565 0.9257 :t 0.0356 6 and 8 0.9339 7 and 8 0.8868 3 9 and 11 0.8925 0.8738 2h 0.0198 9 and 13 0.8760 11 and 13 0.8530 4 14 and 15 0.9505 0.9364 :t 0.0206 14 and 16 0.9358 14 and 17 0.9369 15 and 16 0.9666 15 and 17 0.9194 16 and 17 0.9094 When average PPMC coefficients were calculated for each day, day 1 was found to have the lowest average PPMC coefficient and the highest standard deviation. Based upon the aforementioned values, the average PPMC coefficient on day 1 is on the low end of the moderate correlation range. The individual PPMC coefficients for the 74 extractions on day 1 were also the lowest of the four days, with two of the coefficients corresponding to the moderate correlation range, while the pair-wise comparison for extractions l and 4 had a poor correlation. Examination of the chromatograms showed some inconsistencies among these four extractions. Extractions l and 4 had the impurities benzaldehyde and caffeine present at near baseline levels, while extraction 3 had these impurities present at significant levels. Additionally, retention time shift was observed in the MDP2P peak; in extractions l and 3, this peak was observed at RT 14.85- 1490 minutes, and in extraction 4 it was observed at 14.77 minutes. Another factor that could have contributed to these low correlation coefficients will be discussed later in this section. For days 2, 3, and 4, all of the PPMC coefficients for the pair-wise comparisons indicated a strong correlation among chromatograms on each of these days. Thus, for the replicates on these three days, the HS-SPME procedure can be considered relatively precise. While these PPMC coefficients do indicate a strong correlation, there are some factors to be considered that could fiirther improve the correlation among replicates from the same MDMA exhibit. One factor that may affect PPMC coefficients is retention time shifts in the impurity profiles. Even with the same number of data points in the chromatogram, the actual retention time at which data points are collected shift slightly from run to run. Thus, future research should employ a retention time alignment algorithm to correct for retention time shifi among extractions, which should help improve resulting PPMC coefficients. 75 Overall, the data is promising for the precision of HS-SPME of a MDMA sample dissolved in aqueous buffer. However, optimization of the GC temperature program, along with a retention time alignment algorithm, and a larger data set are needed to further assess the precision of this extraction procedure. 4.2 Comparison of HS-SPME and Liquid/Liquid Extraction (LLE) for Extraction of Organic Impurities from MDMA The optimized HS-SPME procedure was compared to a conventional LLE procedure reported in the literature, in terms of the number of impurities and the identity of the impurities extracted [6]. Liquid-liquid extraction was performed using six 200 mg samples of a homogenized batch of exhibit CJ-FS-05, with triplicate analysis of each of the six extracts. Visual comparison of the chromatograms indicates that while the LLE chromatogram has better overall chromatography (Figure 4.7a), fewer impurities are extracted when compared to the HS-SPME procedure (Figure 4.7b). A summary of the impurities, additives and components extracted using each procedure is given in Table 4.4. 76 100 7 1- Methamphetamine j (a) \ 2- Piperonal 6 3- 3,4-MDDMB = 4- MDP2P j 5- MDPZ-propanol 6- MDMA ‘ . 7- MDEA 1 ' 8- Caffeine g. . ~ . 9- Unidentified (19.48) g 'i 10- Unidentified (19.35) g ’ ‘ f3 11- Unidentified (22.48) a 50% l = l < Ti 1". l . ,, 0 .. .i i l ii 1.; .1 ll ’1 I ‘1‘ g M ii 1 l | i“ m :2 ‘ 5 ‘ ‘ I. ‘1 j “2‘ l l | l ‘1 . l w w i j ‘ .i‘ l I I [l i- i (1‘ l ( j l t" ‘ l i : H. ,:1 ‘i L3 ‘ill ‘i ‘ || 1“ 1O 0 Retention Time (min) 25 Figure 4.7. Comparison of chromatograms using two different extraction procedures. (a) LLE. (b) HS-SPME (carbonate buffer, 0.05M, pH 10; 30 minute extraction at 60 °C). 77 Figure 4.7 (continued) 100a (b) 8 1- Benzeneacetamide : \ 2- Methamphetamine q 3- Piperonal/Isosafrole a 2 (adjoining peaks) i . 4- 3,4-MDDMB 3 ll 5- MDP2P j , 7 ‘ 6- MDP2P-propanol m i 7- MDMA g a . l ‘( 8- MDEA g 1 R ‘i 9- Salicylic acid 3 ‘ ( l l 10- Caffeine < 50 1 l g ,; ii I it '5 J i‘ 1 ‘ l § ll ‘ H 5i “ 6 i li a l) .1 . 3 .i i ll .‘1 10 i; 1/4 i. . / ‘ l ’ m ‘l l i. ‘ ‘ ‘ . 1 . . . i1 . i l . l I 1 , . ., l_ ‘1 i‘ . i3 . ‘lih‘ ‘ll- [ladder .fl..lr‘t-di L 1 l , O—t. .7. ......».oV—.fi_a..;-~v_.~«.._...‘ ,,..,,..,:,.. ......u .e....r.r ”WW...” . w. "v...“ .._.. 0 Retention Time (min) 25 Table 4.4. Comparison of HS-SPME and LLE by the number of impurities, additives and components extracted. Peaks in the chromatograms not listed in the table are due to fiber or column bleed. Time SPME .47 1 9 77 8.27 l 77 X 8.52 146, 105, 77, 91, X 1 117 131 132 aziridine 1 89 91 1 77 1 51 l 56, 91, 58, 89, l 11 130 1 l , 1 , 1 1 78 l ,l 70, 115 11.39/12.34 149,150,121, 67 91 78 Table 4.4 (continued) *11.38/12.38 162, 149, 131, X X Isosafrole 103, 102, 77, 63, 91 *12.01/13.04 58, 77, 105, 146 X X Ephedrine/PseudoephedrineT *12.12/l3.16 135, 136, 77, 178, X X 3,4-MDDMB 179, 58, 79, 105, 148 *13.64/14.76 135, 178, 77, 79, X X MDP2P 136, 51 *l3.76/ 15.03 135, 136, 180, 77, X X MDP2-propanol 78, 79, 106 *l4.7l/15.79 58, 194, 77, 135 X X MDMA 15.37 159,131,135, 91, X 160, 128, 202 *15.71/16.82 72, 70, 77, 135 X X MDEA 17.65 195, 180, 165, X UNID 4 210, 72 17.92 190, 147, 148, 72, X 3-methyl-6,7-methylenedioxy-3,4- 13 5, 188, 208 dihydroisoquinolin-l(2H)one 18.31 58, 100, 72, 135 X UNID 5 *18.92/19.84 194, 193, 55, 109 X X Caffeine 19.29 120, 121, 138,92 X Salicylic acid 19.48 86, 58, 194, 72, X N-ethyl,N-methyl-(1,2- 91, 135, 234 methylenedioxy)-4-(2- aminopropyl)benzene 19.84 97, 70, 194, 72, X UNID 6 135 20.90 162, 58, 135, 77, X MDA Acetate 135, 163 22.48 87, 55, 129,185 X UNID 7 Total Impurities/Additives 15 24 * Indicates peaks that eluted at different retention times with the two different extraction methods. ”Piperonal was only observed in two sets of triplicates out of six separate liquid/liquid extractions. TBoth ephedrine and pseudoephedrine are given as the provisional identity for the peak at 12.01/13.04 minutes, as these two compounds are diastereomers, and are difficult to differentiate by mass spectrometry alone. The LLE procedure extracted only 15 impurities and additives, in comparison to 24 impurities and additives using the HS-SPME procedure. Some peaks existed in the chromatograms (Figure 4.9), but were not identified in Table 4.4 since these peaks were a result of fiber and column bleed. For the peaks that were present in both extracts, there were some slight differences in retention time. This is likely due to the fact during LLE extraction, the sample is injected directly onto the column, whereas the analytes must first desorb from the fiber and onto the column during HS-SPME, resulting in slightly longer retention times. Aside from different injection port temperatures (250 °C for LLE, 260 °C for HS-SPME), the two GC-MS temperature programs were identical. Overall, the LLE chromatogram is dominated by three relatively broad peaks: MDP2-propanol (RT 13.76 minutes), MDMA (RT 14.71 minutes), and MDEA (RT 15.71 minutes). Other impurities and additives present in the LLB extract at relatively significant levels include methamphetamine (RT 8.80 minutes), 3,4-MDDMB (RT 12.12 minutes) and caffeine (RT 18.92 minutes), which were also observed in the HS-SPME extract. All other impurities present in the LLB extract were at levels just above background, which makes assessment of precision difficult. In the LLE extract, MDP2P is observed as a very small shoulder on the leading edge of the MDPZ-propanol peak. MDP2P and MDP2-propanol co-elute in the LLB profile, likely due to the broadness of the MDPZ-propanol peak. Also, impurities such as benzeneacetamide and tablet additives such as salicylic acid, which were present in the HS-SPME profile, were not present in the LLB profile. In the LLE chromatogram, piperonal was only observed in two of the six LLEs performed for this study, and it was only observed at low levels just above the background in these two sets of triplicates. Piperonal was also only observed at low levels in the HS-SPME extracts, although it was extracted more reproducibly by HS-SPME than by LLE. MDP2P and piperonal are examples of the disadvantages that can be experienced with LLE; MDP2P is nearly 80 masked by the broad MDP2-propanol peak, and piperonal is only present at low levels and is not reproducible throughout several extractions. This could be due to the multiple steps and solvents involved in the LLB extraction, which could contribute to potential loss of the trace level impurities among extractions. As mentioned previously, MDP2P and piperonal are impurities that are indicative of the synthetic pathway used to synthesize MDMA, so it is important that these impurities are resolved and observed consistently. Conversely, three impurities (retention times 19.48, 19.85 and 22.48 min) were present in the LLE profiles but not in the HS-SPME profiles. However, these three impurities were small peaks that were just above the baseline. Although these three impurities were not extracted by HS-SPME, other impurities were extracted in the same retention time range. Hence, the lack of extraction of the three impurities is not simply due to low volatility preventing efficient partitioning into the headspace. It is possible that the pH of the buffer used in the LLE favored the extraction of these two impurities over the pH of the buffer used for HS-SPME, and at this pH of carbonate buffer, these impurities were not volatile enough to partition into the headspace. For the impurity at 22.48 min, it is likely that this is a high molecular weight compound with low volatility such as stearic or palmitic acid, which are long chain fatty acids commonly used as tablet binders [7]. It is possible that a combination of structure, polarity, and lack of volatility prevented these compounds from partitioning into the headspace or onto the fiber. Full identification of these impurities will help further determine the utility of these impurities to impurity profiling of MDMA. 4.3 Impurity Profiling of Multiple MDMA Exhibits 81 One of the goals of this research was to investigate alternative extraction procedures for organic impurity profiling of seized MDMA exhibits. The resulting organic impurity profiles can be used to link tablets to a particular synthetic production method, or associate tablets to a common batch. Impurities were extracted from five different MDMA exhibits using the optimized HS-SPME procedure: 50 mg MDMA sample in 5 mL carbonate buffer (0.05 M, pH 10), extracted for 30 minutes at an extraction temperature of 60 °C. Due to the limited supply of tablets, one tablet from each exhibit was homogenized, and three 50 mg aliquots were weighed into separate vials to allow triplicate extractions of each tablet. Extracts were analyzed by GC-MS to generate the resulting impurity profiles. Physical characteristics of the five exhibits are given in Table 4.5, and representative impurity profiles for each exhibit are given in Figure 4.8. Table 4.5. Physical characteristics of various exhibits of MDMA tablets. Exhibit Color Logo Width Height Edges Identifier /mm /mm CJ-FS-Ol Dark rose Baseball 8 4 Beveled (both sides) CJ-FS-02 Dark orange Omega 8 4 Beveled (logo side only) CJ-FS-03 Purple/pink Spade 8 4 Not beveled CJ-FS-05 Blue Omega 8 4 Beveled (both sides) MSU-900- Pale Lacoste 8 4 Beveled 01 pink/dark (crocodile) (logo side pink mottled only) 82 0 i (a) CJ-FS-Ol 0 10° (b) C] Fs 02 4. 1 0 100 2.2 ._ 2.2212 .2'- 2. (c) CI- FS 03 l T l * T -l d -1 Relative Abundance O. 2 .-_2_ 2. 10° 7 (d) CJ-FS-OS ° 1 (e) MSU-900-01 I l I l (I ll . i ‘I “l M III“ 211.1 51.2. .112» ill 111 i111 J * 3,4-methylenedioxy-N- I benzylmethylamine l w No Methamphetamine i ‘i' i I; l I t I 2 l 2 .2 2:2 i: all! 2 l2 .2 2 I * N-fonnyl-meth Lil-“J“...illi _ 1.22. J * N-formyl-meth No MDP2—pr0pan0' 1...-..11 l * PSEIEPH l I, I1 + 3,4-MDDMB ‘ (" l l ‘ if i‘ l l .‘ 2 _ _. i2|22 “1“" ‘22,ll 2'2_>2_2*2__v_2_2_2-22 I * Unidentified-14.29 5 i ..l . l . m— ' VV’7'VI . ~WJ+YIWWTW 25 Retention Time (min) Figure 4.8. HS-SPME Chromatograms of different exhibits of MDMA tablets using the optimized extraction with 0.05 M carbonate buffer (pH 10). (a) CJ-FS-Ol (b) CJ-FS-02 (c) CJ-FS-03 (d) CJ-FS-05 (e) MSU-900-01. It should be noted that while exhibit CJ-FS-OS was used in all previous optimization studies, slight differences are apparent in the impurity profiles previously generated compared to Figure 4.8d. This is likely due to the fact that in the optimization 83 studies, a homogenized batch of tablets from exhibit CJ-FS-OS was used compared to a single tablet from the exhibit as in Figure 4.8d. Following visual examination of the chromatograms, there were several similarities, however, there were sufficient differences to differentiate the exhibits. Appendix C compares the impurities extracted from each exhibit, which allows for a more detailed view of the similarities and differences among the exhibits. Impurities that were common to all exhibits were often observed at different retention times due to differences in peak width among the different exhibits (i.e. methamphetamine, observed at retention times of 9.76 min, 9.78 min, 9.87 min, and 9.86 min). All five exhibits contained MDP2P, which is a common precursor in the synthesis of MDMA, and is considered a List 1 chemical by the Drug Enforcement Administration. Piperonal and isosafrole were observed in all five exhibits, although piperonal was observed at low levels just above background. Both of these impurities are precursors for the synthesis of MDP2P. Since all five exhibits were found to contain MDP2P, it is not unusual that piperonal and isosafrole were also observed. Safi'ole, an isomer of isosafrole, was also observed in exhibits CJ -FS-01, CJ-FS-05, and MSU-900- 01. For the synthesis of MDP2P, safrole is extracted from sassafras oil and then isomerized to isosafrole. In the exhibits where both safrole and isosafrole is present, it is possible that the isomerization process did not go to completion. CJ-FS-03 was differentiated from the remaining four exhibits based on the absence of MDP2-propanol. This alcohol is formed during reductive amination synthesis of MDMA. This synthetic pathway typically uses MDP2P and methylamine as starting materials. During this synthesis, an imine is formed that is subsequently reduced to form 84 MDMA as the final product. If there is excess MDP2P in the reaction vessel, a side reaction can occur where the MDP2P is reduced to MDP2-propanol as the imine is reduced to MDMA. To reduce the occurrence of this side reaction and increase the yield of MDMA, NaBH4 is used as a reducing agent, which requires the reaction vessel to be cooled to very low temperatures; hence this synthesis is known as the “cold method.” At low temperatures, the reducing agent is more selective, which minimizes the side reaction producing MDP2-propanol. Hence, the absence of MDP2-propanol in exhibit CJ-FS-03 is an indication that the cold method may have been used to synthesize the MDMA in this exhibit. Methamphetamine and caffeine were observed in all exhibits with the exception of CJ-F S-01. Due to the relatively high levels of methamphetamine present, it is likely that the methamphetamine is added to the finished MDMA prior to tablet formation rather than occurring as a side product of MDMA synthesis. The exhibits that contained methamphetamine also contained various aziridine compounds, which are byproducts of methamphetamine synthesis, and not necessarily linked to the MDMA synthetic method [4, 8]. However, if a particular exhibit contains mixtures of controlled substances, the patterns of byproducts for each substance may differ depending on the synthetic method and purification of each drug before they are combined for tabletting, allowing for distinction between batches. Caffeine, which was also present in relatively high levels in all exhibits, is a common cutting agent used in tablets containing amphetamine-type drugs to enhance the stimulant effects. The presence or absence of additives and cutting agents also contribute to the overall pattern of the impurity profile, which can help distinguish or associate different seizures back to a particular clandestine laboratory. 85 Differences in the impurities present may indicate the use of different synthetic methods or starting materials. Overall, examining all five exhibits, the absence of methamphetamine and caffeine differentiated exhibit CJ -FS-01 from the other exhibits. Ephedrine/pseudoephedrine and 3,4-MDDMB are only observed in exhibit CJ-FS-05. MSU-900-01 and CJ -FS-01 did not contain MDEA, however, an unidentified impurity at RT 14.29 minutes was only observed in MSU—900-01. CJ-FS-02 and CJ-FS-03 had many similarities in their impurity profiles, with minor differences observed in impurities that were only present at low levels. While MDP2-propanol was not observed in CJ-FS- 03, MDP2-propanol was only observed in CJ-FS-02 at low levels. Although these minor differences do exist between CJ-FS-02 and CJ-FS-03, more than one tablet from each exhibit would have to be analyzed to definitively associate or differentiate these two exhibits. It is possible that if a batch of MDMA was not homogenized during tablet formation, there may be slight variations among tablets within a seizure. Hence, more than one tablet from each exhibit needs to be analyzed in order to increase confidence in potential links to common synthetic routes and/or common batches of MDMA tablets. 4.4 Summary A HS-SPME procedure was developed and optimized for the extraction of organic impurities from seized MDMA exhibits dissolved in aqueous buffer. For optimization, an empirical method of optimization provided a compromise between the number of impurities extracted and chromatographic peak shape. Carbonate buffer (pH 10, 0.05M) was found to extract more impurities than phosphate buffer (pH 7, 0.5M), although the phosphate buffer did offer slightly better chromatographic peak shape. Relative standard deviations of individual impurities for HS-SPME replicates of MDMA 86 in carbonate buffer were relatively similar to literature values reported for a similar method using a different buffer. Additionally, PPMC coefficients indicated that replicate extracts had strong correlation values within a day for three of four days of analysis. Thus, the developed HS-SPME procedure is fairly precise. When compared to a conventional LLE procedure, the developed HS-SPME procedure extracted more impurities, thus providing more information about the exhibit than the LLE profile. The developed HS—SPME extraction procedure was used to differentiate five seized MDMA exhibits based on differences in organic impurity profiles. Thus, HS-SPME was demonstrated to be a viable alternative to the conventional LLE procedure by providing a more informative impurity profile using approximately one-fourth the sample mass, and eliminating the need for use and disposal of organic solvents. 87 4.5 References [1] Kongshaug KE, Pederson-Bjergaard S, Rasumssen KE, Krough M. Solid-phase microextraction/capillary gas chromatography for the profiling of confiscated ecstasy and amphetamine. Chromatographia 1999;50:247. [2] Swist M, Wilamowski J, Zuba D, Kochana J, Parczewski A. Determination of synthesis route of 1-(3,4-methylenedioxyphenyl)-2-propanone (MDP-2-P) based on impurity profiles of MDMA. Forensic Sci Int 2005;149:181-192. [3] van Deursen MM, Lock ERA, Poortman-van der Meer AJ. Organic impurity profiling of 3,4-methylenedioxymethamphetamine (MDMA) tablets seized in the Netherlands. Science & Justice 2006;46(3):135-52. [4] Gimeno P, Besecier F, Chaudron-Thozet H, Girard J, Lamotte A. A contribution to the chemical profiling of 3,4-methylenedioxymehtamphetamine (MDMA) tablets. Forensic Sci Int 2002;127:1-44. [5] Microgram Bulletin. Recent unusual drug submission in New Zealand. Vol 40 No 12 Dec 2007. http://www.usdoi.gov/dea/programs/forensicsci/microgram/mg1207/mgl207.pdf (Accessed 27 May 2009) [6] Andersson K, Jalava K, Lock E, Huizer H, Kaa E, Lopes A, Poortman-van dar Meer A, Cole MD, Dahlén J, Sippola E, Development of a harmonized method for the profiling of amphetamines IV. Optimization of sample preparation. Forensic Sci Int 2007;169:64-76. [7] Baer I, Margot P. Analysis of fatty acids in ecstasy tablets. Forensic Sci Int 2009;doi:10.1016/j.forsciint.2009.03.015. [8] Qi Y, Evans 1, McCluskey A. New impurity profiles of recent Australian imported ‘ice’: Methamphetamine impurity profiling and the identification of (pseudo)ephedrine and Leuckart specific marker compounds. Forensic Sci Int 2007;169:173-180. 88 CHAPTER FIVE RESULTS AND DISCUSSION: DEVELOPMENT OF MICROWAVE-ASSISTED EXTRACTION/HEADSPACE SOLID-PHASE MICROEXTRACTION (MAE/HS—SPME) In this chapter, results fi'om several studies used to determine the viability of microwave-assisted extraction (MAE) followed by headspace solid-phase microextraction (HS—SPME) for the extraction of organic impurities from 3,4- methylenedioxymethamphetamine (MDMA) tablets are discussed. An abbreviated empirical procedure was used to determine parameters for MAE extraction time and extraction temperature for the extraction of organic impurities from MDMA. Both phosphate buffer (0.5 M, pH 7) and carbonate buffer (0.05 M, pH 10) as extraction solvents. Following development of the MAE procedure, extraction parameters for the subsequent HS-SPME step were investigated. The developed MAE/HS-SPME procedure was then compared with a conventional liquid-liquid extraction (LLE) procedure and the HS-SPME procedure developed in Chapter 4. 5.1 Development of Microwave-Assisted Extraction (MAE) Procedure 5.1.1 General Procedure The two extraction solvents investigated in the development of MAE parameters were phosphate buffer (0.5 M, pH 7) and carbonate buffer (0.05 M, pH 10). These buffers were chosen as starting point based upon their utility for the HS-SPME optimization previously described in Chapter 4. The two buffers were compared for MAE/HS-SPME based upon the number and potential identity of the impurities extracted from seized MDMA tablets. 89 As with previous extractions from Chapter 4, 50 mg of homogenate from the CJ- FS-OS batch was used for each MAE/HS-SPME extraction. Based on manufacturer recommendations for an aqueous extraction solvent, microwave extraction temperatures ranging from 90 to 110 °C in increments of 10 °C were investigated using MAE ramp and hold times of 15 minutes (30 minute total extraction). Total extraction time (ramp time + hold time) was then investigated in a similar empirical manner. MAE ramp and hold times of 10, 15, and 20 minutes were investigated giving total extraction times of 20, 30 and 40 minutes. After MAE, extracts were allowed to cool in the sealed vessel to 75 °C, which was 5 °C above the subsequent HS-SPME extraction temperature. After transferring 5 mL of MAE extract into an amber vial, the extract was preheated for 5 minutes in a 70 °C water bath with stirring, and the divinylbenzene/CarboxenTM/polydimethylsiloxane (DVB/CAR/PDMS) SPME fiber was exposed to the headspace for 15 minutes. The fiber was retracted and analyzed by GC- MS using the GC-MS parameters as detailed in Chapter 3. 5.1.2 Development of MAE using Phosphate Buffer (0.5 M, pH 7) as Extraction Solvent Overall, for the investigation of MAE extraction time and extraction temperature using phosphate buffer (0.5M, pH 7), there was no significant visual difference in overall chromatography among the different time and temperature extractions. This is expected, as the HS-SPME parameters were kept constant (15 minute extraction at 70 °C) throughout the development of the MAE extraction procedure (Figure 5.1). 90 1- Benzaldehyde 35 (a) L I L L 2- Piperonal ,. l l “ 3- 3,4-MDDMB % H 4- Unidentified 14.05 I 7 5-Caffeine . 1 I I I1 I I I IL IL L. l L} ‘L I L L 1L1 I L . :L LL 1 . ’ ‘ L. L , .L L 1L i .L ( .I._.I|1.L:‘:I..L .=.L.. . [51 1:. 022 2 22 7| 'I '~ 353i (b) L I IL . l 3 ; ‘1‘ 5 E H i E 2 1L 11L 1 L 3 1 L L I I L < l‘ L I l . I 0 . I I I l E 1‘ L II ' LILL l L L5 1 a Ll |i 4 I . L‘ 1 1 1. : ~ 1‘ .4 ‘ . L. L2 9.4le ."4.,L , 411...... LL LL . a 35 (C) L L L 2 1 L L I L I i I I Ii I I I . . . ‘1. . I .1 i '1 i If '1 ‘ . I 4 L l 3 l I L . i I l I II 12 LILI II I 1 L.‘ ( I. ELL: 1 l I . . 3 L . .L‘ L L 14.4.4 ;‘.L.. L5L it . . 020 ..22. ...........2 . ...2_2 . . .. ..-.. ... .. ..I. . .. . ..2. ... . ..2 ..2.. 2 .225 Retention Time (min) Figure 5.1. Chromatograms comparing different MAE temperatures using phosphate buffer (0.5M, pH 7) with a 30 minute extraction time: (a) 90 °C, (b) 100 °C, and (c) 110 °C. *Peak at RT 12.15 minutes is due to fiber bleed. However, there were some differences in the impurities extracted at the different temperatures. At 90 °C (Figure 5.1a), piperonal, 3,4-methy1enedioxy-N,N- dimethylbenzylamine (3,4-MDDMB), and an unidentified impurity at RT 14.05 minutes were not extracted. Piperonal is a common starting material for methylenedioxyphenyl- 2-propanone (MDP2P), and 3,4-MDDMB is a byproduct indicative of the reductive amination synthesis for MDMA. Additionally, at 90 °C, benzaldehyde (RT 5.47 91 minutes) and caffeine (RT 19.86 minutes) were extracted at low levels just above baseline in comparison to the higher extraction temperatures. Benzaldehyde is a common solvent, and caffeine is a common additive in illicitly made tablets, both of which help contribute to the overall profile of the tablet. Because HS-SPME parameters were kept constant for this set of experiments, this indicates that a MAE extraction temperature of 90 °C is not high enough to effectively extract these impurities into the phosphate buffer for subsequent extraction by HS-SPME. It is possible that a 15 minute HS-SPME time was not long enough to extract piperonal, 3,4-MDDMB, and the impurity at RT 14.05 minutes, as this is a shorter HS-SPME extraction time than used in Chapter 4. However, these three impurities were extracted with a 15 minute HS-SPME following MAE at higher temperatures, so it is unlikely that HS-SPME was the limiting factor for the extraction of these impurities. There were no differences in the number or identity of the impurities extracted at MAE temperatures of 100 °C and 110 °C. F ifieen impurities and additives were extracted, compared to 12 impurities and additives extracted at 90 °C. Thus, 100 °C was chosen as the MAE extraction temperature, as higher MAE temperatures could potentially cause thermal degradation of impurities. While this was not observed in this exhibit, the lower extraction temperature was chosen to err on the side of caution, as not every MDMA exhibit will contain the same impurities. In terms of MAE extraction time, total MAE extraction times of 20, 30 and 40 minutes were investigated (Figure 5.2). While the same impurities were extracted at each extraction time, some impurities, such as 3,4-MDDMB (RT 13.18 minutes) and the unidentified impurity at RT 14.07 minutes (identical to the previously mentioned 92 impurity at RT 14.05 minutes), as well as the additive caffeine (RT 19.88 minutes) were extracted at levels just above baseline in the 40 minute extraction (Figure 5.2c). Additionally, in the 40 minute extraction (Figure 5.2c), the peak areas of impurities such as methamphetamine (RT 9.72 minutes), MDP2P (RT 14.67 minutes), and 3,4- methylenedioxyethylamphetamine (MDEA) (RT 16.51 minutes) decreased by approximately half compared to the 20 minute extraction (Figure 5.2a). Methamphetamine is a second controlled substance likely added to the product during tabletting, whereas MDP2P is a precursor for MDMA, and MDEA is a reductive amination byproduct that can occur during MDMA synthesis. A decrease in peak area can be detrimental to an impurity profile, as some impurities are present at trace levels. While the chromatogram did not appear to show any evidence of thermal degradation, this concept could be further investigated by the addition of an internal standard, to determine if the decrease in peak area is due to the longer microwave extraction time, or is due to variation in the sample. Thus, a 20 minute extraction time consisting of a 10 minute ramp time to 100 °C followed by a 10 minute extraction at 100 °C was chosen as the MAE extraction time for phosphate buffer (0.5 M, pH 7). The 20 minute extraction is the most time efficient, as the same impurities were extracted at all three extraction times. A full listing of impurities extracted by the phosphate buffer at the optimal MAE extraction time and temperature can be found in section 5.1.4 where impurities extracted by both buffers are compared. 93 1- Methamphetamine 35 L (a) I . i I 5 2-3,4-MDDMB . I I ‘. . I 3- Unidentified 14.07 L . I I I 4-MDP2P I . i II iI 5-MDEA ILI . iII . I ‘. 6- Caffeine I L ii I I I I I IIIL 4"I | I ‘ I i I L I , II II Iii. II . ,Iii II I 0 I , j II IZIIQIII. ILA. .Iii, BI I 7 b 1 35 . ( ) I/ I II I § II I ‘L 5 I m I I 2 II I I I i 3 , .D i < L Li L i “>’ IN I I . ‘ I . g L L I I I II . L I L o L, IL A 7 I I IiI '_ I2I3I9I I i .L I.“ E ‘63 l 35 (C) I I I i. I . 1 I I II I ii I L IL I IL.L I L II 4III IL I I 0 . A L LL IL._ 2L3L-‘ ....L .LL.L... . ‘64 A - HL m“; o 25 Retention Time (min) Figure 5.2. Chromatograms comparing different MAE times using phosphate buffer (0.5M, pH 7) at 100 °C: (a) 20 minutes, (b) 30 minutes, and (c) 40 minutes. *Peak at RT 12.15 minutes is due to fiber bleed. 5.1.3 Development of MAE Using Carbonate Buffer (0.05 M, pH 10) as Extraction Solvent The same empirical process was used to determine MAE parameters using carbonate buffer (0.05 M, pH 10) as the extraction solvent investigating the same MAE extraction temperatures and times as previously. Similar to observations with the 94 phosphate buffer, there were differences in the impurities extracted at different extraction temperatures Piperonal and methylenedioxyphenyl-Z-propanol (MDPZ-propanol) were not extracted at 90 °C (Figure 5.3a). Additionally, at 90 °C, several impurities were extracted at levelsjust above baseline. These include 2-phenyl aziridine (RT 7.09 minutes), isosafrole (RT 12.37 minutes), and an unidentified impurity at RT (18.71 minutes). Aziridine compounds are byproducts of methamphetamine synthesis, and isosafrole is a starting material for MDP2P. This indicates that a MAE extraction temperature of 90 °C does not effectively extract these impurities into the buffer from the tablet homogenate for subsequent extraction by HS-SPME. Again, because the HS- SPME parameters were kept constant for this study, and these impurities were extracted at higher MAE extraction temperatures, it is likely that the MAE temperature is the limiting factor. At MAE extraction temperatures of 100 and 110 °C (Figure 5.3b & 0), there were no differences in the number or identity of impurities extracted. At 100 and 110 °C, 26 impurities were extracted, compared with 24 impurities at an extraction temperature of 90 °C. Peak areas of impurities were also similar between the 100 and 110 °C extractions. Some differences included a decrease in the peak areas of benzeneacetamide (RT 8.63 minutes) by approximately one-third, the unidentified impurity at RT 9.51 minute by approximately two-thirds, and MDP2P (RT 14.74 minutes) by approximately one-fourth between the 100 °C and 110 °C extractions. Because only a few impurities appeared to be affected, further research with an internal standard is needed to determine if thermal degradation is possible, or if the difference in peak area is solely due to variation in the 95 “’1 F—fifi. a... sample. Therefore, 100 °C was chosen as the MAE extraction temperature for carbonate buffer (0.05M, pH 10). 35 1 I I IL 1-2-Phenylaziridine .L 5: I 2- Benzeneacetamide , I I. I 3-Unidentified(9.51) (a) I II I I 4- Piperonal/Isosafrole . I .‘i I I I (adjoining peaks) I I I 5-MDP2P 2 I I JI II 6—MDP2-propanol _ I I i I i II ‘I 7 7-Unidentified(18.71) ? I I IIII‘IIi I 0 r. I 1 .l3 n .4 ‘L ,II‘ II»: ii .IiIi . 1 I 5:I I i 35 i \II' . 8 i b L I g ‘ ( ) II I I I‘ I 'U . I S *1 i I I ‘i i g LI LI I II‘6II II .3 2 I I2 I Ii 7 g 1 ‘i I IL I i LI I I I m Li. 1 L 3‘ 4 L. I i W Ii I I I I i 0 ‘ 7 7 I I IIii ii 1 3L; 1 i4, .4 l-‘AI Ii . i . | I 35- I . I i i (C) I i I I I 5 ii I \I I I I ‘ L ’ 1" I , , I i 2 . LI II L L ‘I I I 1 L 3 #4 LI lLLL I LIIII II I I I 0 fl 4. _ ... fl, I 1,}... .— ..L [L I' ”in O 25 Retention Time (min) Figure 5.3. Chromatograms comparing different MAE extraction temperatures using carbonate buffer (0.05M, pH 10) with a 30 minute MAE extraction time: (a) 90 °C MAE extraction, (b) 100 °C MAE extraction, and (c) 110 °C extraction. Total extraction times of 20, 30 and 40 minutes were subsequently investigated at 100 °C. At all three MAE extraction times, chromatography appeared visually similar, and the same impurities were extracted. Hence, a 20 minute extraction consisting of a 10 L96 minute ramp time and a 10 minute hold time was chosen as it is the most time efficient, with an extraction temperature of 100 °C. These parameters are the same as determined for the extraction with phosphate buffer. 5.1.4 Comparison of Phosphate Buffer and Carbonate Buffer as Extraction Solvent for MAE Table 5.1 summarizes the impurities, additives and components extracted from the homogenized batch of MDMA using the phosphate and carbonate buffers with an extraction temperature of 100 °C and an extraction time of 20 minutes, which were the optimal extraction parameters from both buffers. Table 5.1. Summary of impurities, additives and components extracted from homogenized MDMA using phos hate and carbonate buffers as extraction solvents. Retention Prominent Phosphate Carbonate Tentative Identity Time m/z Buffer Buffer (minutes) (0.5M, pH 7) (0.05 M, pH 10) *5.47/5.42 105, 77, 51, 106 X X Benzaldehyde 7.05 1 18, 77, 91, 119 X 2-Phenyl aziridine 135, 136, 77, X 3,4- 823 78, 51 methylenedioxytoluene 146, 105, 77, X 2-ethyl-2-phenyl 8.46 117, 131 aziridine *8.64/8.59 X X Benzeneacetamide *9.39/9.43 77, 105, 51 X X Unidentified (UNIQl *9.73/9.83 X X Methamphetamine 56, 91, 58, 65, X N-benzyl—Z-methyl 10.44 146 aziridine 10.80 72,91,70,115 X UNID2 162, 131, 103, X Safrole 11.50 77, 132, 104 149, 150, 121, X X Piperonal *12.36/12.28 65 162, 131, 103, X Isosafrole 12.34 77, 104, 78 58, 77, 71, 91, X Pseudoephedrine/ 12.98 117, 105 Ephedrine‘f 56,71,91, 117, X UNID3 13.03 77, 105 97 Table 5.1 (continued) 135, 136, 77, X X 3,4-methylenedioxy *13.17/l3.10 178, 179 dimethylbenzylamine 104, 58, 118, X UNID 4 14.06 117 135, 178, 77, X X MDP2P *14.66/14.7O 79, 51, 136 135, 136, 77, X X MDP2-propanol *14.79/14.93 180, 78, 79 78, 77, 135, X X MDMA *15.54/15.71 136, 79, 105 56, 135, 77, X UNID 5 16.30 191,194,190 *16.52/16.74 72, 70, 77, 135 X X MDEA 195, 180, 165, X X UNID 6 *17.64/17.60 210 190, 148, 147. X 3-methyl-6,7- 188, 72, 204 methylenedioxy-3,4- dihydroisoquinolin- 17.87 1(2H)one 58,72,91,115, X UNlD7 18.71 135, 147 105,132,91, X UNIDS 18.77 77, 133 120, 91 , 77, X X Salicylic acid *19.30/19.23 121,138 194, 193, 55, X X Caffeine *19.85/19.81 195, 109 162, 58, 135, X MDA Acetate 20.86 77, 163 Total Impurities/Additives 15 26 * Indicates peaks that eluted at different retention times with the two different buffers. I Both ephedrine and pseudoephedrine are given as the provisional identity for the peak at 12.98 minutes, as these two compounds are diastereomers, and are difficult to differentiate by mass spectrometry alone. Visually, the phosphate buffer extraction does have overall better chromatography in terms of peak shape because at pH 7, the extraction of MDMA and methamphetamine is more efficiently suppressed. Because MDMA and methamphetamine are present in large proportions in the tablets compared to the trace impurities, they tend to overload the fiber, resulting in poor chromatography. However, in terms of the number of impurities and additives extracted by MAE/HS-SPME, 15 impurities and additives were extracted using the phosphate buffer compared to the 26 impurities and additives extracted by the carbonate buffer. In choosing the optimum buffer for HS-SPME, a compromise between chromatography and the number of impurities extracted must be made because the extraction of additional impurities by the carbonate buffer allows for a more informative impurity profile. Because impurity profiles are typically qualitative instead of quantitative, poor peak shape for some impurities can be tolerated in exchange for the extraction of additional impurities, as observed with the carbonate buffer. The use of MS allows for the toleration of poor peak shape in some impurities, as it enables definitive identification of the impurity despite its poor chromatography. Only one impurity (RT 13.03 minutes) extracted by the phosphate buffer was not extracted by the carbonate buffer. The difference in pH between the two buffers is likely why this impurity only partitions into the headspace during extraction with the phosphate buffer. The pKa of this impurity could be preventing it from partitioning from the buffer into the headspace for extraction at higher buffer pH. If the identity of this impurity were known, it would provide better insight as to why it is only extracted with a buffer pH of 7 and is not extracted with a buffer pH of 10. All subsequent MAE/HS-SPME extractions were conducted using carbonate buffer (0.05 M, pH 10) as the MAE solvent, with a 20 minute total extraction at 100 °C. 5.2 Development of HS—SPME following MAE One of the goals of using MAE prior to HS-SPME was to increase the number of impurities extracted, due to the theoretical high efficiency of MAE. Head-space solid phase microextraction offers selective extraction of impurities following MAE. While incorporating MAE before .HS-SPME does add additional time to the analysis, combining 99 the efficiency of MAE with the selectivity of HS-SPME has potential for improving the extraction compared to HS-SPME alone. Once initial MAE parameters were determined, the HS-SPME parameters for the DVB/CAR/PDMS fiber were optimized. Following MAE extractions using carbonate buffer (0.05M, pH 10) with a 20 minute extraction at 100 °C, the vessels were allowed to cool to 5 °C above the subsequent HS-SPME extraction temperature, and a 5 mL aliquot of the extract was transferred to an amber vial. The vial was preheated for 5 minutes at 70 °C, and HS- SPME extraction times of 5, 10 and 15 minutes were investigated, holding the extraction temperature at 70 °C. Then, HS-SPME extraction temperatures of 60 and 70 °C were investigated with an extraction time of 15 minutes. These HS-SPME extraction temperatures were chosen based upon the previous HS-SPME studies described in Chapter 4. Following visual assessment of the chromatograms, at HS-SPME extraction times of 5 and 10 minutes, peak shape and chromatographic resolution were improved compared to the 15 minute extraction (Figure 5.4) in the retention time region between 14 and 18 minutes. This is potentially due to overloading the SPME fiber and subsequently the GC column by the more abundant impurities in this retention time region at longer extraction times. 100 Relative Abund ance 5 6 'I I3 I__\I \III 7 1- Piperonal 2- MDP2-propanol 3- Unidentified (16.30) 4- 3-methyl-6.7- methylenedioxy-3,4- dihydroisoquinolin-1(2H)-one I ’ 5- Salicylic Acid 6- Caffeine 7- MDA Acetate I. I , I l. ,, ‘ 4 l 25 Retention Time (min) Figure 5.4. Comparison of HS-SPME extraction times for MAE/HSPME: (a) 5 minute HS-SPME extraction, (b) 10 minute HS-SPME extraction, and (c) 15 minute HS-SPME extraction. However, the shorter HS-SPME extraction times affected the number of impurities extracted. The 5 minute HS-SPME extraction did not extract piperonal, an unidentified impurity at RT 16.30 minutes, 3-methyl-6,7-methylenedioxy-3,4- dihydroisoquinolin-l (2H)-one, salicylic acid, cafieine, and MDA acetate. Piperonal is a starting material for MDP2P, while 3-methyl-6,7-methylenedioxy-3,4- 101 dihydroisoquinolin-l(2H)-one is an impurity observed both in MDMA and MDP2P [l]. The presence of these impurities indicates the synthetic route for MDP2P which can also be used to link tablets to a common production source. The 10 minute HS-SPME extraction did not extract MDA acetate, and the MDPZ-propanol peak co-eluted with an adjoining siloxane peak, which complicated the mass spectrum of MDPZ-propanol. Thus, the 15 minute extraction was needed to extract all the impurities observed in earlier MAE/HS-SPME studies. Extraction temperatures of 60 and 70 °C were investigated using the 15 minute HS-SPME extraction time. With a HS-SPME extraction temperature of 60 °C, 22 impurities and additives were extracted, while at 70 °C, 26 impurities and additives were extracted. All the impurities and additives listed for the carbonate buffer in Table 5.1, were extracted at 60 °C with the exception of piperonal, the unidentified impurity at RT 18.71 minutes, and MDA acetate. Additionally, MDP2-propanol co-eluted with an adjoining fiber bleed peak in two of the three 60 °C extractions, and the impurity at RT 9.43 minutes had high levels of background ions present in the mass spectrum, indicating that a very low level of this impurity was extracted at 60 °C. It is likely that the higher extraction temperature is needed for these impurities to partition into the headspace for subsequent extraction by HS-SPME. Relative standard deviations (RSDs) were calculated for several impurities throughout the retention time range of the chromatogram (Table 5.2) to determine precision of the extraction of at extraction temperatures of 60 °C and 70 °C. At 60 °C, RSD values ranged from 3.3 to 32.3 %, while RSDs ranged from 1.5 to 12.0 %, for extractions at 70 °C. With the exception of 2-phenyl aziridine and benzeneacetamide, 102 which are more volatile impurities, RSD values were improved at an extraction temperature of 70°C. Because the majority of the improved RSD values occur for later eluting impurities, it is likely that the higher extraction temperature allows these impurities to partition into the headspace more efficiently, thus allowing the impurities to reach equilibrium between the sample and the headspace during extraction. Table 5.2. Precision of MAE/HS-SPME at different HS-SPME extraction temperatures based on RSDs of impurfiy peak areas. Impurity (RT in minutes) 60 °C (11 = 3) 70° C (n = 3) 2-phenyl aziridine (7.06) 9.8% 1 l.l% Benzeneacetamide (8.61) 5.0% 10.8% Methamphetamine (9.83) 3.3% 2.8% Safrole (11.51) 5.1% 2.3% 3,4-methylenedioxy-N,N- 13.7% 2.3% dimethylbenzylamine (13.10) MDP2P (14.71) 20.7% 12.0% MDP2-propanol (14.95) N/A 2.7% MDEA (16.74) 15.3% 1.5% 3-methyl-6,7—methylenedioxy-3,4- 27.6% 10.2% dihydroisoquinolin-l-(2H)-one (17.84) Unidentified (18.78) 32.3% 5.2% In comparison with RSD values of various impurities from HS—SPME alone (Chapter 4), the RSD values from MAE/HS-SPME were overall improved, especially for the less volatile impurities (Table 5.3). This could be due to a higher HS-SPME extraction temperature, as the optimized extraction temperature for HS-SPME alone was 60 °C; it could also be due to MAE resulting in a more homogenous extract for selective extraction of impurities by HS-SPME. While these RSD values are promising as an advantage for the MAE/HS-SPME procedure over HS-SPME alone, a full reproducibility study over several days must be performed to more fully compare the two procedures. 103 Table 5.3. Comparison of RSD values for various impurities extracted by HS-SPME alone versus MAE/HS-SPME. Impurity HS-SPME (n=3L* MAE/HS-SPME (n=3) Benzeneacetamide 12.7% 10.8% Methamphetamine 7.0% 2.8% 3,4-methylenedioxy-N,N- 5.4% 2.3% dimethylbenzylamine MDP2P 26.9% 12.0% MDPZ-propanol 18.4% 2.7% MDEA 3. 1% l .5% *RSD values for HS-SPME are reported from day 2 of the four day study. Pearson product moment correlation (PPMC) coefficients were used to assess precision of the extraction using the full chromatogram for the two extraction temperatures. Because PPMC coefficients take into account the entire chromatogram instead of individual impurities, these coefficients are a better representation of the overall precision of the MAE/HS-SPME procedure rather than RSD values of individual impurities. For triplicate extractions at 60 °C, PPMC coefficients ranged from 0.8474 to 0.9345, with an average value of 0.8954 i 0.0442. At an extraction temperature of 70 °C, PPMC coefficients for triplicate extractions ranged from 0.9124 to 0.9853, with an average value of 0.9375 3: 0.0414. For both extraction temperatures, these correlations are considered strong, and the standard deviations are not much larger than those observed with HS-SPME alone. Overall, the improved RSD values and high PPMC coefficients indicated that the MAE/HS-SPME procedure is promising. However, further . studies are needed to assess the precision of the combined MAE/HS-SPME procedure over several days. 104 Based upon these preliminary findings, the extraction parameters for HS-SPME following MAE was a 15 minute extraction at 70 °C. While this is a shorter HS-SPME extraction time than the 30 minutes at 60 °C needed for HS-SPME alone, the MAE does add additional time to the overall extraction procedure. 5.3 Comparison of MAE/HS-SPME with Liquid-Liquid Extraction (LLE) and HS- SPME for the Extraction of Organic Impurities from MDMA The developed MAE/HS-SPME procedure was used to extract impurities from a homogenized batch of MDMA from exhibit CJ-FS-OS and the resulting impurity profile was compared with chromatograms generated following LLE alone and HS-SPME alone. The HS-SPME and LLE chromatograms for comparison were generated using optimum HS-SPME parameters and a liquid-liquid extraction discussed previously in Chapter 4. The three impurity profiles are shown in Figure 5.5 and a comparison of the impurities, additives and components extracted by each procedure is given in Table 5.4. 105 100 «1(3) LLE 7\ 1-Methamphetamine -; 2- Piperonal 6 3- 3,4-MDDMB :3 4- MDP2P I 5- MDP2-propanol j 6- MDMA AI 7- MDEA j 1 . 8- Caffeine g j I i 9- Unidentified (19.48) g I I' 10- Unidentified (19.85) g 'I I 3 11- Unidentified (22.43) D 50 5: I I I < . . a: 3 I I I II 3% “I I I I I II IT? , I II II [I I ’I 5I I' iI II II I“ I3 I I . I "II II Ii 10 0 Retention Time (min) 25 Figure 5.5. Comparison of impurity profiles obtained for MDMA exhibit CJ-FS-OS using (a) LLE, (b) HS-SPME, and (c) MAE/HS-SPME for impurity extraction. 106 Figure 5.5 (continued) 10 Relative Abundance II 1.LL-.;li1.L;IJJJ_LiLil_ILlAJI.34-1L221...;1.. A_I_. I. . LL; 0 “I (b) HS-SPME Retention Time (min) 107 1- Benzeneacetamide 2- Methamphetamine 3- Piperonal/Isosafrole (adjoining peaks) 4- 3.4-M DDM B 5- MDP2P 6- MDP2P-propanol 7- Unidentified (15.37) 8- MDMA 9- M DEA 9- Unidentified (18.31) 10- Salicylic acid 1 1- Caffeine 11 i . JILAJIIHIJ, ,I ...k I _.. I‘ - ‘ OTOFVVUW'W¢WY'VW_-‘ 25 Figure 5.5 (continued) 100 7 (c) MAE/HS-SPME 3‘ 10 1-2-Phenylaziridine j 2- Benzeneacetamide I I 3- Methamphetamine 4- Piperonal/Isosafrole . I (adjoining peaks) I ‘ 8 5- 3,4-MDDMB . I . I 6- MDP2P 7- MDP2-propanol 8- MDMA I I 9- Unidentified (16.30) I II , 1o- MDEA 50 ’3 J I‘ ‘I 11- Unidentified (18.77) I I I, . I I II} 2 II II It? I a . . I I .-.. ..., _. .' .1..- :L'I4‘I .vII II'II'II Relative Abundance Retention Time (min) Table 5.4. Comparison of LLE, HS-SPME and MAE/HS-SPME by the number of impurities, additives and components extracted. Peaks in the chromatograms not listed in the table are due to fiber or column bleed. 3 1 77 78 77,91,117, 65, 89, 146, 108 Table 5.4 (continued) 10.57/11.54/11.50 162, 131, X X Safrole 103, 77, 104, 132, 78 10.86/10.81 72,91,131, X UNIDZ 70, 115 *11.39/12.34/12.29 149, 150, X” X Piperonal” 121, 65, 67, 91 *11.38/12.38/12.34 162, 149, X X Isosafrole 131, 103, 102, 77, 63, 91 *12.01/13.04/ 12.99 58, 77, 105, X X Ephedrine/Pseudoephedrine 146 *12.12/13.16/13.10 135,136, X X 3,4-MDDMB 77, 178 179, 58, 79, 105, 148 *13.64/ 14.76/ 14.72 135, 178, X X 3,4-methylenedioxy-2- 77, 79, 136, propanone 51 *l3.76/15.03/ 14.93 135, 136, X X 3,4-methylenedioxy-2- 180, 77, 78, propanol 79, 106 *l4.71/15.79/15.71 58,194, 77, X X MDMA 135 15.37 159,131, X UNID3 135, 91, 160, 128, 202 *15.71/16.82/16.75 72, 70, 77, X X MDEA 135 16.30 56, 135, 77, UNID 4 191, 194, 190 17.65/17.60 195, 180, X UNID 5 165, 210, 72 17.92/17.87 190, 147, X 3-methyl-6,7- 148, 72, methylenedioxy-3,4- 135, 188, dihydroisoquinolin-1(2H)one 208 18.31 58, 100, 72, X UNID 6 135 18.70 58, 72, 91, UNID 7 115, 135, 147 109 Table 5.4 (continued) 18.77 105, 132, X UNID 8 91, 77, 133 *18.92/19.84/ 19.80 194, 193, X X X Caffeine 55, 109 19.29/ 19.23 120, 121, X X Salicylic acid 138, 92 19.48 86, 58, 194, X N-ethyl,N-methyl-(l,2- 72, 91, 135, methylenedioxy)-4-(2- 234 aminogropyl)benzene 19.84 97, 70, 194, X UNID 9 72, 135 20.90/20.86 162, 58, X X MDA Acetate 135, 77, 136, 163 22.48 87, 55, 129, X UNID 10 185 Total Irmmrities/Additives 15 24 26 * Indicates peaks that eluted at different retention times with the three different extraction methods. ”Piperonal was only observed in two of six separate liquid/liquid extractions. T Both ephedrine and pseudoephedrine are given as the provisional identity for the peak at 12.01/ 13.04/ 12.99 minutes, as these two compounds are diastereomers, and are difficult to differentiate by mass spectrometry alone. The MAE/HS-SPME procedure extracted 26 impurities and additives compared to 24 for HS-SPME alone and 15 for LLE. Visually, the chromatograms for MAE/HS- SPME (Figure 5.5c) and HS-SPME (Figure 5.5b) appear similar, and thus the chromatographic discussion comparing LLE (Figure 5.5a) and HS-SPME is also true for the comparison of LLE and MAE/HS-SPME. In summary, the LLB chromatogram is dominated by several broad chromatographic peaks that could potentially mask impurities present at trace levels. Also similar to HS-SPME, MAE/HS-SPME extracts more impurities than LLE, thus providing a more informative impurity profile. The three impurities previously extracted only by LLE (RT 19.48 minutes, 19.84 minutes, and 22.48 minutes) were also not extracted by MAE/HS-SPME. Similar to the comparison with HS-SPME, it is possible that the pH of the LLB favored extraction of these impurities over the pH of the buffer used for MAE/HS-SPME. As stated previously, it is 110 likely that the impurity at 22.48 minutes is a high molecular weight compound such as a long-chain fatty acid, which is not sufficiently volatile for headspace analysis. Not every impurity extracted by HS-SPME was extracted by MAE/HS-SPME. While the two procedures had many impurities in common (Table 5.4), two impurities were extracted only by HS-SPME (RT 15.37 minutes and 18.31 minutes), and four impurities were extracted only by MAE/HS-SPME (RT 7.05 minutes, 16.30 minutes, 18.70 minutes, and 18.77 minutes). Only one of these impurities (RT 7 .05 minutes) was provisionally identified (2-phenyl aziridine); without knowing the identity of the other impurities, it is difficult to determine why they were extracted by one procedure and not the other, especially since these impurities are within a similar retention time range. For the impurities extracted by HS-SPME only, it is possible that the shortened HS-SPME extraction time following MAE was not long enough to extract these two impurities. For the impurities extracted by MAE/HS-SPME only, it is possible that the MAE more efficiently extracted these impurities into the liquid extract for subsequent headspace extraction. Additionally, while a full precision study was not performed for the MAE/HS-SPME procedure, initial RSD values for selected impurities across the retention time range were comparable to the RSD values observed for the HS-SPME precision study reported in the previous chapter. For the later eluting impurities, the RSD values observed with the MAE/HS-SPME procedure were slightly improved over HS- SPME alone. The PPMC coefficients observed with MAE/HS-SPME were comparable to those observed with HS-SPME alone, with only slightly larger standard deviations. However, the MAE/HS-SPME procedure needs to be fully optimized, after which a 111 precision study should be conducted over several days in order to fully compare these two procedures. 5.4 Summary Based upon this work, MAE/HS-SPME is a promising procedure for extracting organic impurities from seized exhibits of MDMA using carbonate buffer (0.05M, pH 10) as an MAE extraction solvent. Following an empirical procedure, the developed MAE/HS-SPME parameters were a 20 minute MAE at 100 °C followed by a 15 minute HS-SPME at 70 °C. The MAE/HS-SPME procedure was comparable to the optimized HS-SPME procedure (Chapter 4) in number and identity of the impurities extracted. Initial RSD values for impurities in the MAE/HS-SPME extracts (n=3) were also comparable to those obtained with HS-SPME, indicating similarities in precision between the two extraction procedures. Additionally, PPMC coefficients for triplicates of MAE/HS-SPME indicated strong correlation among replicates, similar to PPMC coefficients observed with HS-SPME alone. Afier comparing LLE, HS-SPME and MAE/HS-SPME for extraction of organic impurities from MDMA tablets, HS-SPME and MAE/HS-SPME extracted 24 and 26 impurities respectively, compared with 15 impurities extracted by LLE. In terms of total analysis time, the optimized HS-SPME procedure requires a 10 minute sample preheat time and a 30 minute extraction time, for a total of 40 minutes. The MAE/HS-SPME procedure requires a 20 minute MAE time, followed by a 5 minute sample preheat and a 15 minute HS-SPME extraction, also totaling 40 minutes. Sample preparation is essentially equivalent between the two methods. LLE requires that the sample be sonicated twice for 30 minutes each time, as well as centrifugation for 3 minutes, for a 112 total of 63 minutes. The GC-MS methods for all three extractions are similar in total analysis time. Thus, HS-SPME and MAE/HS-SPME are more time efficient than LLE in addition to extracting more impurities. However, fithher work is needed to determine whether the MAE/HS-SPME procedure is worth the extra expense and extraction step over HS-SPME alone. 113 5.5 References [1] Swist M, Wilamowski J, Zuba D, Kochana J, Parczewski A. Determination of synthesis route of 1-(3,4-methylenedioxyphenyl)-2-propanone (M‘DP-Z-P) based on impurity profiles of MDMA. Forensic Sci Int 2005;149:181-192. 114 CHAPTER SIX CONCLUSIONS AND FUTURE WORK 6.1 Conclusions In this research, headspace solid-phase microextraction (HS-SPME) was optimized for the extraction of organic impurities from 3,4- methylenedioxyrnethamphetamine (MDMA) tablets. An empirical optimization procedure was used to optimize the HS—SPME extraction time and extraction temperature for a sample of MDMA dissolved in aqueous buffer. Using carbonate buffer (pH 10, 0.05 M) resulted in a more informative impurity profile than phosphate buffer (pH 7, 0.5 M) in regards to the number of impurities extracted. Following empirical optimization of extraction time and extraction temperature, the optimal HS-SPME extraction procedure for carbonate buffer was a 30 minute HS-SPME extraction at 60 °C. A precision study of the HS-SPME of MDMA in aqueous buffer was performed to assess the precision of the extraction over several days. Relative standard deviations (RSDs) were calculated for impurity peak areas, and many were similar to RSD values reported in literature for a similar study involving HS-SPME of MDMA tablets in aqueous buffer. However, many of the RSD values increased for the later eluting impurities. This effect should be further investigated, possibly by changing the gas chromatography (GC) parameters for this retention time region. Pearson product moment correlation (PPMC) coefficients were also used to assess precision of the entire chromatogram among replicate extractions on each day. With the exception of day one, all of the PPMC coefficients within each day were considered to be strong correlations. 115 This indicates the procedure is relatively precise, but additional work is needed to assess the efficiency of the extraction. Despite the strong correlations, the PPMC coefficients can be further improved by retention-time alignment and peak area normalization of the chromatograms. In comparing HS-SPME and LLE, HS-SPME of exhibit CJ-FS-OS in carbonate buffer resulted in a more informative impurity profile than a conventional liquid-liquid extraction (LLE) procedure published in the literature. HS-SPME allowed for greater sensitivity, and extracted a greater number of impurities than LLE due to the pre- concentration ability and selectivity of the fiber. A more informative impurity profile allows for greater discrimination among exhibits or a greater ability to link exhibits to a common source. Additionally, HS-SPME eliminates the need for organic solvents, as the waste from the HS-SPME procedure is aqueous and does not require separate chemical disposal. Analysis of five exhibits of MDMA that differed physically in color and logo (with the exception of exhibits CJ-FS-02 and CJ-FS-OS which shared the same logo) indicated that organic impurities varied among different exhibits. This may indicate that different synthetic pathways were used to synthesize the MDMA in these tablets and/or the MDMA was synthesized in different clandestine laboratories. However, some impurities were common in the exhibits despite differences in physical appearance, which may indicate that laboratories produce tablets that differ in physical appearance from the same batch of MDMA. Despite physical differences, the impurity profiles of exhibits C]- F S-02 and CJ-FS-03 appeared visually similar and these two exhibits shared many of the same impurities. However, no conclusions can be made because only one tablet from 116 each exhibit was available for analysis. More tablets from each exhibit, as well as more exhibits, must be analyzed to determine the commonality of the impurities observed, which can be used to determine impurities that are the most valuable for profiling purposes. Identification of commonly observed impurities can then be used to determine the route used to synthesize the MDMA in the tablets. Overall, HS-SPME of MDMA in aqueous buffer was more effective for the extraction of organic impurities than the conventional LLE method in terms of the number of impurities extracted. However, these conclusions are limited to the exhibits of MDMA available at the time the study was performed. Further work is needed to determine the precision of the HS-SPME procedure within a single exhibit and among a larger sample set of exhibits. Analysis of more tablets from the same exhibit will also allow further statistical assessment of the procedure, and assessment of tablet variation within the exhibit. The second goal of this research was to investigate the potential of microwave- assisted extraction (MAE) combined with HS-SPME for the extraction of organic impurities from MDMA tablets. An abbreviated empirical method was used to develop MAE extraction time and extraction temperature parameters, which was followed by HS- SPME to selectively extract impurities from the MAE extract. The extraction procedure developed fiom this study was compared with both a conventional LLE procedure, as well as HS-SPME alone, in terms of the number of impurities extracted and the overall chromatography of the resulting impurity profile. The initial study indicated that MAE/HS-SPME is an effective procedure for the extraction of organic impurities fi'om MDMA tablets. Following an empirical process, 117 the chosen parameters for MAE were a 20 minute extraction consisting of a 10 minute temperature ramp to 100 °C followed by a 10 minute hold at 100 °C. Following MAE, a 15 minute HS-SPME extraction at 70 °C was used to selectively extract the impurities from the MAE extract. Of the two buffers studied, carbonate buffer (pH 10, 0.05 M) was found to be a better extraction solvent for MAE/HS-SPME as it extracted more impurities than the phosphate buffer (pH 7, 0.5 M). While the phosphate buffer did result in slightly better overall chromatography due to the suppression of MDMA extraction at pH 7, only extracted 15 impurities and additives were extracted, versus the 26 impurities and additives extracted using the carbonate buffer. Further work should be done to investigate the pH range between 7 and 10 to determine if there is a buffer pH that offers improved chromatography without sacrificing the number of impurities extracted. Relative standard deviations of impurity peak areas for triplicate MAE/HS-SPME extractions were improved over the RSD values observed from HS-SPME alone. Pearson product moment correlation coefficients were also calculated which resulted in high correlations among the triplicate extractions. This indicates that the extraction procedure is relatively precise for these initial studies. However, a full optimization and precision study is necessary firlly assess the precision of the MAE/I-IS-SPME procedure. Similar to HS-SPME alone, MAE/HS-SPME extracted more impurities than the conventional LLE procedure. The LLE procedure only extracted 15 impurities/additives, while HS-SPME and MAE/HS-SPME extracted 24 impurities/additives and 26 impurities/additives respectively. This is ideal, as the extraction of additional impurities increase the discriminatory abilities of the resulting impurity profile. Chromatography of the resulting MAE/HS-SPME chromatogram was also slightly improved over HS-SPME 118 alone. However, the procedure needs to be fully optimized and the precision assessed over several days to determine if the cost of the MAE and the extra analysis step over HS-SPME alone is worth the two additional impurities extracted. 6.2 Future Work In this research, a HS-SPME procedure was developed for the extraction of organic impurities from a sample of MDMA in aqueous buffer. Additionally, the viability of MAE/HS-SPME for the extraction of organic impurities from MDMA tablets was assessed. Optimizing these procedures to extract the maximum number of impurities allows for an informative impurity profile that can provide the greatest ability to discriminate among exhibits or link exhibits to a common source or production method. However, additional work is needed to improve these procedures so that they can be used for profiling purposes to aid law enforcement investigations. For HS-SPME of MDMA in aqueous buffer, further improving chromatography, improving repeatability and reproducibility of replicate extractions, assessing the extraction efficiency, and confirming the identity of the impurities in the exhibits must be addressed. For MAE/HS-SPME, a full optimization of the procedure is necessary to address precision of the procedure over several days, and to determine if the procedure offers any significant advantages over HS-SPME alone. Additionally, analysis of a larger number of MDMA exhibits, as well as a larger number of tablets within a single exhibit, with the optimized extraction procedure would benefit both extraction procedures. In terms of precision, the initial RSD values for individual peak areas were promising, as many were similar to RSD values for a similar study reported in literature. However, the literature study did not optimize the HS—SPME method before employing 119 it, and only reported a range of RSD values without reporting the RSD values for each impurity. In this research, several peaks at later retention times had increased RSD values. One possible option for improving these RSD values is increasing the ramp rate of the GC temperature program in order to improve the chromatography by narrowing the peaks in this region of the chromatogram. However, increasing the ramp rate can also have a negative effect on chromatography, as it can cause co-elution of impurities that have similar retention times. Several different exhibits would need to be examined with the increased ramp rate to determine if co-elution would occur. Another possible method for firlly optimizing the GC program is to create a standard mixture of impurities or similar compounds than can be used to assess improvements in peak resolution and minimize peak broadening. The greater the retention time range spanned by the components of the standard mixture, the more accurate the GC program will be for the wide range of potential impurities in illicit tablets. Once the GC program is fully optimized, the standard mixture can then be used to assess extraction efficiency of the standard mixture. Further assessment of the precision of the HS-SPME procedure will also add weight to the association of exhibits that have similar impurity profiles. This study also offered a brief investigation into the potential of MAE/HS-SPME for organic impurity profiling; however, a firll optimization needs to be performed to fully assess its potential utility over HS-SPME alone. Initially, because only two buffer pH values were investigated in this study, additional buffer types and pH values could be investigated to determine the effect of buffer on the number of impurities extracted. 120 Because there are many parameters for the MAE and HS-SPME procedures (e. g. extraction time and extraction temperature for both procedures), an experimental design could be employed to identify extraction parameters that are statistically significant. Once the significant parameters are determined, a second experimental design could be conducted to optimize the extraction parameters concurrently. Similar to the optimization of HS-SPME alone, a standard mixture could then be used to assess the efficiency of the extraction. Once the MAE/HS-SPME procedure is fully optimized, a reproducibility study should be performed using both a standard mixture as well as a homogenized batch of seized MDMA tablets in order to statistically assess the precision of the procedure over several days. Aside from the high efficiency of MAE, the microwave reactor also has the capability of extracting several samples at one time, which would decrease overall analyst involvement if several samples could be extracted by MAE at once. This capability should be investigated to determine if several extracts extracted at one time remain stable over the course of a day. If the optimized MAE/HS-SPME extraction procedure is determined to be more precise than HS-SPME alone, and multiple extractions could be extracted in a single microwave run, the additional analysis time of the MAE step could be justified. Pearson product moment correlation (PPMC) coefficients were also used to assess the precision of both the HS-SPME and MAE/HS-SPME extractions. Because PPMC coefficients take into account the entire chromatogram instead of individual impurities, it can offer a more complete assessment of the extraction procedure. While the initial PPMC coefficients indicated high correlation among most of the HS-SPME replicate 121 extractions and all of the MAE/HS-SPME replicate extractions, additional steps can be taken to further improve the PPMC coefficients and their standard deviations, providing a more accurate evaluation of the extraction procedure. The first steps for improving the PPMC coefficients include applying a retention-time alignment algorithm to account for instrumental drifi between runs. Additionally, peak area normalization can be used to account for slight variations in impurity peak area among replicate extractions. Another potential step for improving PPMC coefficients involves eliminating the fiber bleed peaks from the chromatogram, which were another source of variation among replicate chromatograms. These fiber bleed peaks would often vary in size from extraction to extraction, which can decrease PPMC coefficients. If these peaks could be subtracted from the impurity chromatograms, a potential source of variation among extractions could be reduced or eliminated, which should result in increased PPMC coefficients. Standards of impurities commonly found in illicit MDMA tablets need to be analyzed under the optimized extraction parameters in order to provide a definitive identification of impurities. However, it is often difficult to obtain some of the impurities, especially those that are a result of reaction byproducts and do not have common uses. Tandem mass spectrometry (MS) can be used to elucidate the structure of impurities not previously identified by conventional mass spectral libraries. The GC-MS used in this research has tandem MS capabilities which should be further investigated. Additionally, a larger number of tablets from each exhibit, and more exhibits need to be analyzed using the optimized extraction procedure. Analyzing a larger number of tablets from an exhibit can provide information on the variation within a particular 122 production batch, which can help with the association of similar impurity profiles. Assessment of variation within a batch is important for cases in which tablets that appear physically similar are seized at different times. MDMA tablets are typically made in large batches, and if the batch is not completely homogenous before tabletting, tablets that appear physically similar could result in differing impurity profiles. If the variation within a particular batch is known, then tablets seized at different times may be more confidently linked based on impurity profiles. Analyzing a larger number of exhibits can be used to build a library of impurities and determine the commonality of certain impurities, which can aid in determining the potential synthetic route. Once a larger number of exhibits and tablets have been analyzed, statistical conclusions can be made on the potential synthetic route, which is necessary for linking common batches. Analysis of a greater number MDMA exhibits can eventually result in a searchable database of impurity profiles upon which exhibits could be linked based upon their synthetic route and similarities in the impurities present in the profile. This database could then be used to determine linkages among exhibits and provide information regarding a common production source for law enforcement agencies. 123 <35: Emma—92V 0:888:55-m-_»:o_._m-»xo_vo=o_buns—La.m 28.?m 0 Emoficxm 22.388on <55: 0 32922582 Ea: o of o /21 z\ o 50:28: 0 I 3.283: I o 0 I000: o ”maoficxm tee—0:3 <2Q2 ocowconA—anofig_EHuE-Z-NYv-§xo_BERfioEYN._ man—E /2_._ one EEEEE 038:3”— EGE 3.. 8:51 otofiim 58:80 "< 52.3.7... 124 Appendix B: Common Impurities from MDP2P Synthesis MDP2P O Piperonal and O I Saf role/Isosaf role MDP2P Synthesis 0 O 1 -( 3 ,4-methylenedioxyphenyl)-1 ,2-propanedione O \ /OH Piperonal (N itropropene) N MDP2P Synthesis 0 piperonal oxime O Saf role/Isosaf role MDP2P Synthesis 0 3,4-(methylenedioxy)-phenylpropane 125 APPENDIX C: Comparison of Impurities Present in the Five Exhibits Peaks present in the chromatograms not listed below are caused by fiber or column bleed. The number below the X in each column represents the retention time of that compound. Prominent CJ- CJ- CJ- CJ- MSU- Tentative Peak Identity m/z FS-Ol FS-02 F S-03 FS-OS 900-01 86, 58 X X N-formyl 4.22 4.29 methamphetamine 105, 77, 51, X X X X Benzaldehyde 106 5.40 5.48 5.44 5.47 135, 136, 77, X 3,4-methylenedioxy 51, 78 8.28 toluene 146, 105, 77, X X X 1,2-dimethy1-3-phenyl 117, 131,103 8.47 8.49 8.52 aziridine 91, 92, 65, 63, X X X X Benzeneacetamide 89, 119, 134 8.60 8.63 8.64 8.63 77, 105, 51 X Unidentified (UNID) 1 9.49 58, 91, 65, X X X X Methamphetamine 115, 119, 148 9.76 9.78 9.89 9.88 146, 105, 91, X X 1,2-ethyl phenyl aziridine 117, 77 10.03 10.05 103, 131,77, X UN1D2 58, 146, 102 10.37 56, 91, 58, 65, X N-benzyl-Z-methyl 115, 146, 147 10.53 aziridine 72,91,131, X UNID3 73, 70, 115 10.87 162, 131, 103, X X X Safrole 77, 104, 78 11.54 11.55 11.54 147,91, 119, X X UNID4 135 11.87 11.88 135, 135, 166, X UNID 5 165, 77 12.25 149, 150, 121, X X X X Piperonal 65, 63, 91 12.31 12.33 12.33 12.35 162, 103, 131, X X X X X Isosafrole 104, 78, 77 12.37 12.37 12.38 12.39 12.39 162, 154, 131, X X X X UNID6 103,104, 77 13.01 13.01 13.03 13.01 58, 77, 91, X Ephedrine/ 117, 104 13.04 PseudoephedrineT 135, 136, 178, X 3,4-methylenedioxy-N,N- 179, 77, 51 13.17 dimethyl benzylamine 135, 136, 134, X 3,4-methylenedioxy-N- 126 164, 165, 77 13.44 methylbenzylamine 164, 165,91, X UNID7 135, 149 13.75 176, 175, 89, X UNID 8 91, 63 14.29 135, 178, 77, X X X X X MDP2P 51, 79, 136 15.03 14.72 14.72 14.79 14.70 135, 136, 180, X X X X MDP2-propanol 77, 78, 79, 51 15.41 14.92 15.05 14.88 189,190,131, X UNID 9 147, 91,117 15.38 58, 77, 136, X X X X X MDMA 135, 195 16.05 15.75 15.70 15.80 15.70 194, 58, 135, X UNID 10 147, 89, 190 16.30 86, 205, 91, X UNID 11 170, 55, 105 16.39 72, 70, 107, X X X MDEA 135, 77 16.50 16.49 16.83 194,192,191, X UN1D12 58, 105, 161 16.49 194, 135, 58, X UNID 13 91, 95, 79 16.75 178, 194, 149, X UNID 14 163, 58, 222 16.83 194,161, 58, X UN1D15 77,91,204 17.17 202, 194, 188, X UNID 16 58, 91, 105 17.24 195,180, 210, X X X UNID 17 165, 245 17.61 17.62 17.66 195,167, 210, X X UNID18 165, 181, 236 17.70 17.75 190, 147, 148, X X 3-methyl-6,7- 135, 188, 58, 17.92 17.94 methylenedioxy-3,4- 194, 204 dihydroisoquinolin-1(2H)- one 135,107,181, X X UNID 19 210 18.56 18.57 189, 188, 135, X UNID 20 58, 77,136 19.11 175, 176, 160, X UNID 21 120, 147 19.28 120, 138, 121, X X X X Salicylic acid 92, 55 19.30 19.30 19.30 19.29 194, 193, 55, X X X X Caffeine 127 109,67, 195 19.82 19.83 19.85 19.82 162, 58, 135, X MDA Acetate 163, 77 20.89 182, 183, 91, X UNID 22 98, 58, 256 21.89 159,241,117, X X UNID23 185, 143, 242 21.99 22.01 232, 217, 215, X X UNID 24 202, 233 22.34 22.35 217, 232, 215, X X UNID 25 202, 198, 233 22.68 22.68 196, 197, 91, X X UNID 26 98 23.07 23.05 238, 239, 91, X X UNID 27 120, 119,148 23.51 23.51 168, 169, 135, X UNID 28 58 23.87 182, 162, 112, X X UNID 29 58, 98 24.98 24.99 194,178,161, X X UNID 30 135 25.23 25.24 182, 183, 135, X X UNID 31 98 25.74 25.74 196. 197, 98, X X UNID 32 135 26.77 26.77 163, 204, 135, X UNID 33 105, 77, 174 28.20 230, 147, 187, X X UNID 34 131, 105, 163 28.27 28.25 260, 395, 380, X X UNID 35 261, 204 29.22 29.21 '1' Both ephedrine and pseudoephedrine are given as the provisional identity for the peak at 13.04 minutes, as these two compounds are diastereomers, and are difficult to differentiate by mass spectrometry alone. 128 IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIII 3 1293 0306