Illicit drugs analysis on chip...Illicit drugs analysis on chip The use of lab-on-a-chip technology for forensic applications Author: Brigitte Bruijns MSc Student Number: 5924553 Project:

Illicit drugs analysis on chip

The use of lab-on-a-chip technology for forensic applications

Author:Brigitte Bruijns MScStudent Number: 5924553Project: September 2010 - April 2011Colloquium: April 2011

Daily Supervisor:dr. ir. Wojciech Bula (UT/MCS/MESA+)

Supervisor:prof. dr. Han Gardeniers (UT/MCS/MESA+)

Second Reviewers:dr. Arian van Asten (NFI)

dr. Wim Kok (UvA/HIMS)

Brigitte Bruijns Illicit drugs analysis on chip Summary

Summary

One of the forensic traces that can be found at a crime scene are drugs of abuse. To get an indicationwhether there are illicit substances present in the unknown sample, presumptive colour tests can beperformed. Altough these tests are selective, they are not specific and give only an indication of thepresence or absence of drugs of abuse.

Micro-devices have become of interest to forensic scientists as these systems can speed up theanalysis, are compact, can easily be integrated, limit the risk of contamination and can be used bypeople who are not technically trained. Another advantage is the minimal amount of analyte materialneeded. The ultimate goal is to develop a so-called ”lab-on-a-chip” device that can be used for all thenecessary steps from sample preparation till detection.

The aim of this research project was to develop a micro-device for the analysis of illicit drugs1. Byintegration of UV-Vis spectroscopy more analytical information can be obtained from the analytes.

Another advantage is that lower detection limits can be obtained by UV-Vis spectroscopy incomparison with visual observation of the colour tests. The operational drug detection limit for visualobservation lies around 40 mg/mL, whereas for UV-Vis detection it is lower than 1 mg/mL.

This research shows promising results for the detection of illicit drugs on chip, although there arestill a couple of challenges.

1In this research drug analogues and legal highs were used to avoid procedures for obtaining the required licenses toperform research on real illicit drugs.

2

Brigitte Bruijns Illicit drugs analysis on chip Preface

Preface

This research project is part of a Masters in Analytical Sciences at the University of Amsterdam(UvA)/VU University Amsterdam (VU). This internship embraces a detailed study of the literatureas well as practical experience on a chosen subject. It aims to broaden and deepen the knowledge ofthe student.

The internship is a mandatory item of the curriculum of Analytical Sciences. The total timeduration of this project is 42 EC (1 European Credit corresponds to 28 hours). A colloquium todefend the thesis of the project must be held at the end. Grades will be given for the research itself,the thesis and the colloquium.

The project is carried out at the University of Twente (UT) in collaboration with the NetherlandsForensic Institute (NFI).

Hopefully the reader enjoys reading this thesis and appreciates the knowledge gained about allkinds of facets of lab-on-a-chip devices in forensics.

4

Brigitte Bruijns Illicit drugs analysis on chip Contents

Contents

Summary 2

Preface 4

List of abbrivations 9

Introduction 10

1 Drugs of abuse 141.1 Stimulants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141.2 Narcotics and depressants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141.3 Hallucinogens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

2 Drugs analysis 162.1 Colour tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2.1.1 Scott test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162.1.2 Marquis test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162.1.3 Simon’s test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

2.2 UV-Vis spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192.3 Lab-on-a-chip devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

3 Materials and methods 223.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

3.1.1 Chemicals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223.1.2 Stock solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223.1.3 UV-Vis spectrometer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223.1.4 Syringes and pumps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223.1.5 Microscopes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223.1.6 Photos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233.1.7 Chips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

3.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243.2.1 Colour tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243.2.2 Detection limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243.2.3 Calibration curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243.2.4 Chip A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243.2.5 Chip B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

4 Results and discussion 264.1 Colour tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

4.1.1 Scott test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264.1.2 Marquis test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264.1.3 Simon’s test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

4.2 Detection limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274.2.1 Scott test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274.2.2 Marquis test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274.2.3 Simon’s test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

4.3 Calibration curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304.3.1 Scott test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304.3.2 Marquis test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

6


4.3.3 Simon’s test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334.4 Remarks colour tests and UV-Vis spectroscopy . . . . . . . . . . . . . . . . . . . . . . 34

4.4.1 Colour tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344.4.2 Operational drug detection limit . . . . . . . . . . . . . . . . . . . . . . . . . . 344.4.3 Solvent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354.4.4 UV-Vis spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

4.5 Chip A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364.6 Chip B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374.7 Remarks detection on chip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

5 Legal highs 405.1 Mephedrone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405.2 MBDB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405.3 MDAI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415.4 Methylone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415.5 Ethylone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415.6 Butylone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415.7 4-Fluoroamphetamine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415.8 2,5-Dimethoxyamphetamines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415.9 2,5-Dimethoxyphenethylamines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

6 Materials and methods 436.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

6.1.1 Chemicals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436.1.2 Stock solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436.1.3 UV-Vis spectrometer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436.1.4 Pump . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436.1.5 Photos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436.1.6 Chips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

6.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446.2.1 Colour tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446.2.2 UV-Vis spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446.2.3 Calibration curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446.2.4 Chip C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

7 Results and discussion 467.1 Colour tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467.2 UV-Vis spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

7.2.1 Ephedrine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467.2.2 Diethylamine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467.2.3 Lidocaine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 497.2.4 MBDB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 497.2.5 4-Methylamphetamine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

7.3 Calibration curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 517.3.1 Ephedrine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 517.3.2 Diethylamine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 517.3.3 Lidocaine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537.3.4 MBDB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537.3.5 4-Methylamphetamine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 557.3.6 4-Fluoroamphetamine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

7


7.4 Remarks colour tests and UV-Vis spectroscopy . . . . . . . . . . . . . . . . . . . . . . 557.4.1 Colour tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 557.4.2 Operational drug detection limit . . . . . . . . . . . . . . . . . . . . . . . . . . 577.4.3 Solvent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 577.4.4 UV-Vis spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

7.5 Chip C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 607.5.1 Visibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 607.5.2 Mixing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 607.5.3 UV-Vis spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

7.6 Remarks detection on chip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 607.6.1 Colour tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 607.6.2 UV-Vis spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 637.6.3 Design of the chip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

Conclusion 64

Recommendations 65

Epilogue 67

A Project description 70

B Drugs of abuse 73

C Colour tests 75

D Detection limits by eye 76D.1 Scott test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76D.2 Marquis test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76D.3 Simon’s test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

E Detection limits by UV-Vis spectroscopy 78E.1 Scott test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78E.2 Marquis test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78E.3 Simon’s test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

F Designer drugs 79

G Chip designs 80

H Chip fabrication 82H.1 Process parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82H.2 Process summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

I PAC symposium 87I.1 Poster . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87I.2 Abstract poster . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

8

Brigitte Bruijns Illicit drugs analysis on chip Abbrivations

List of abbrivations

2C-B 4-bromo-2,5-dimethoxy-phenethylamine2C-I 4-iodo-2,5-dimethoxy-phenethylamineβk Beta-ketoε Molar extinction coefficientA AbsorbanceATR Attenuated total reflectanceBromodragonfly 1-(8-bromobenzo[1,2-b;4,5-b]difuran-4-yl)-2-aminopropaneButylone β-keto-N-methylbenzodioxolylpropylaminec ConcentrationCCD Charge coupled deviceCE Capillary ElectrophoresisCNS Central nervous systemCSI Crime Scene InvestigationDEP Direct electrospray probeDIOS Desorption/ionization on siliconDNA Deoxyribonucleic acidDOB 2,5-dimethoxy-4-bromoamphetamineDOM 2,5-dimethoxy-4-methylamphetamineEDMA 3,4-ethylenedioxy-N-methylamphetamineEthylone 3,4-methylenedioxy-N-ethylcathinoneFTIR Fourier Transform Infraredg GramsGC Gas ChromatographyGHB 4-Hydroxybutanoic acidHIMS Van ’t Hoff Institute for Molecular SciencesHPLC High Performance Liquid ChromatographyI Intensity of the lightIR Infraredl Path length of the cellL LitersLIF Laser induced fluorescenceLED Light-emitting diodeLOC Lab-on-a-chipLOD Limit of detectionLSD Lysergic acid diethylamidem MetersMethylone 3,4-methylenedioxymethcathinoneMephedrone 4-methylmethacathinoneMCS Mesoscale Chemical SystemsMBDB MethylbenzodioxolylbutanamineMDA 3,4-methylenedioxyamphetamineMDAI 5,6-methylenedioxy-2-aminoindaneMDEA 3,4-methylenedioxy-N-ethylamphetamineMDMA 3,4-methylenedioxymethamphetamineMS Mass SpectrometryNFI Netherlands Forensic InstituteODDL Operational drug detection limitPDMS Polydimethylsiloxanes SecondsT TransmittanceTLC Thin Layer ChromatographyUT University of TwenteUvA University of AmsterdamUV-Vis Ultraviolet-VisibleVU VU University Amsterdam

9

Brigitte Bruijns Illicit drugs analysis on chip Introduction

Introduction

The number and variety of forensic traces found at a crime scene is enormous. The term forensicscience is therefore very broad and can be divided in several expertise areas, such as DNA profiling,blood spatter analysis, explosives and illicit drugs. It is a clear desire of forensic investigators thatanalyses should be simple, fast, robust, cheap and have high sensitivity and selectivity. Devices withthese specifications that can be used directly at the crime scene are especially useful as they canprovide immediate information to the police investigators. Most ideal would be a mobile forensic labfor collecting, screening and analysis (identification/classification) of the evidence. So-called ”lab-on-a-chip” (LOC) systems can speed up the analysis, are compact, can easily be integrated, limit the riskof contamination and can be used by people who are not technically trained.

However, micro-devices for forensic investigation hardly exist2. Experts in LOC technology and/ornanotechnology do not have experience and knowledge about forensic science. On the other hand,forensic experts are in general not familiar with LOC devices. The two disciplines have not yet beencombined in order to obtain an LOC device for forensic research. Therefore the NFI and the MesoscaleChemical Systems (MCS) research group of the UT started a collaboration to explore the possibilitiesof the use of LOC technology in forensics. The original project description of this collaboration canbe found in Appendix A.

Possible application fields of LOC systems within forensic science are explosives, human biologicaltraces, toxicology, gun shot residues, arson and illicit drugs. Also screening of materials that can beused for (fake) powder letters, such as milk powder and powdered sugar, is a possible application ofmicro-devices.

The focus of this research project is on the analysis of illicit drugs. By adding a few drops of theScott3, Marquis and Simon’s reagent to the suspected analyte the presence of illicit drugs (such asamphetamines and opiates) can be indicated. These conventional tests are easy to perform, but theused chemicals are toxic, hazardous and/or corrosive and therefore a certain level of technical trainingis required. Appropriate training is also required to be able to interpret the observed colour changes.In this research these three widely used presumptive colour tests are integrated in an LOC device.

The report is divided in two parts: drugs of abuse and legal highs. Both parts contain the samestructure; first the theory is given and the used materials and methods are described whereafter theresults are discussed. The conclusion of all the results and findings for drugs of abuse as well as forlegal highs and further recommendations can be found at the end of the report in respectively thechapters Conclusion and Recommendations.

To get an insight in illicit drugs, in Chapter 1 drugs of abuse and the several types of illicitdrugs are discussed. The presumptive colour tests are explained in Chapter 2 as well as other drugsanalysis methods. In this chapter also the use of LOC devices within drugs analysis is described. Theexperimental part starts in Chapter 3, were an explanation is given about the used materials andmethods for drugs of abuse. The results and discussion points can be read in Chapter 4.

More information about legal highs can be found in Chapter 5. The theory of drugs analysis isthe same for drugs of abuse as for legal highs and therefore no separate chapter about drugs analysisis introduced. Also for the legal highs the used materials and methods are described and the resultsand discussion points are discussed in respectively Chapter 6 and Chapter 7.

2B. B. Bruijns. Forensic applications of nanotechnology, Literature survey on the use of nanotechnology and lab-on-a-chip devices for forensic investigations. October 2010.

3The official Scott test is a three step reaction; the one step test used in this research, which is also called the Scotttest, is better known under the name cobalt thiocyanate test [1].

10

Drugs of abuse

Brigitte Bruijns Illicit drugs analysis on chip Drugs of abuse

1 Drugs of abuse

There are four categories of legal offenses of acts with substances present in list I of the Dutch OpiumLaw, better known as hard-drugs. The first legal offense is the import and export of these substancesfrom the Netherlands. Preparing, editing, processing, selling, delivering, providing and transportingis illegal just like possessing the drugs of abuse. The manufacturing of substances in list I is thefourth legal offense [2]. The substances in list I are not only the pure compounds, but also the salts,enantiomers, ester and ethers of these substances [3].

For part b (hennep products) of list II (soft-drugs, sleeping pills and tranquillizers) there are twolegal offenses; trading of the drugs and the manufacturing, preparing, editing, processing, selling,delivering, providing and transporting of these substances [2]. It is only allowed to posses some ofthese substances in small amount for own use or in case of a prescription of a doctor. Other substanceson list II are mushrooms which contain psilocin or psilocybin [4].

Illicit drugs can be divided in a couple of different classes depending on the pharmalogical4 action,such as narcotics, stimulant, depressants and mind expanders, but also anabolic steroids (not in theOpium Law) [3, 5]. Examples of some drugs of abuse, their appearance and the molecular formulacan be found in Appendix B [3].

1.1 Stimulants

Amphetamines, such as amphetamine (speed in the Netherlands) and methamphetamine, are a class ofcentral nervous system (CNS) stimulants, which have a high psychological dependence and addictionpotential [6]. After intake the alertness and energy of the person are increased. These compoundsare therefore also known as ”uppers”. Fatigue and appetite are decreased by amphetamine-typestimulants [3, 6].

Negative side-effects are depression, also known as the burn-out, and inertia, caused by high and/orrepeated doses. Abusers can show amphetamine psychosis and they suffer from hallucinations andparanoia. Due to all these effects amphetamines were added to the Dutch Opium List in 1976 [3, 7].Other stimulants are cocaine or crack, the free-base form of cocaine and khat, but also tobacco, caffeineand the medicine Ritalin R© [5].

1.2 Narcotics and depressants

Narcotics and depressants are both sedative and slow down the CNS. Therefore depressants are alsoknown as ”downers”. The difference between these two classes of drugs is that narcotics also providepain relief [8]. Other effects of narcotics, such as the opiates morphine and heroin, are euphoria,sleepiness and slow breathing [3, 8].

Examples of other depressants and narcotics are 4-hydroxybutanoic acid (GHB), barbiturates, ben-zodiazepines, methadone, alcohol, tranquilizers and sleeping pills [3, 5]. Benzodiazepines are sedative-hypnotic and anxiolytic compounds and some of them are used as anti-epileptics and anesthetics [6].Examples of benzodiazepines are nitrazepam, bromazepam, flurazepam and diazepam [9]. Barbitu-rates are also used as sedative-hypnotic compounds, anesthesia and as medicine for epilepsy [6, 9].Examples of barbiturates are mephobarbital, meharbital, hexobarbital, thiopental and thiamylal [9].Also opium, that consists of a mixture of alkaloids such as codeine, morphine and papverine, is anexample of a narcotic [3, 5].

4The study of drugs, their sources, their nature, and their properties. Pharmacology is the study of the body’sreaction to drugs.

14

Brigitte Bruijns Illicit drugs analysis on chip Drugs of abuse

1.3 Hallucinogens

Hallucinogens or mind expanders are a class of drugs that have distortions of reality and hallu-cinations as effects [3, 5]. This class of compounds alter perception and mood of the user [6].Users describe their experience as an out of the body experience and therefore these drugs are alsocalled psychedelics [3, 5]. Most of the mind expanders are designer drugs [9]. Lysergic acid di-ethylamide (LSD), 3,4-methylenedioxymethamphetamine (MDMA, also known as ecstasy/XTC), 2,4-methylenedioxyamphetamine (MDA, also known as the love drug), ketamine, 4-bromo-2,5-dimethoxy-phenethylamine (2C-B) and magic mushrooms are examples of this class of drugs, although MDMAand MDA are only hallucinogens in a high dose [3, 5, 10].

15

Brigitte Bruijns Illicit drugs analysis on chip Drugs analysis

2 Drugs analysis

To obtain a high level of legally acceptable proof of identification of drugs of abuse, several entirelydifferent analytical techniques should be used. The techniques can be classified in three groups, whichare based on selectivity. The first group contains presumptive colour tests and the appearance ofthe substance. The next group consists of chromatographic techniques, such as Gas Chromatography(GC), Thin Layer Chromatography (TLC) or High Performance Liquid Chromatography (HPLC).Infrared (IR) Spectroscopy is an example of more selective spectroscopic techniques. With hyphenatedtechniques results from two (coupled) methods are obtained, which counts as two separate parameters.For example with Gas Chromatography Mass Spectrometry (GC-MS) the retention time as well asthe mass spectrum is used to identify the illicit substance [3, 11].

2.1 Colour tests

A colour test or spot test is a presumptive test that provides an indication of the presence or absenceof drugs of abuse. These tests are used in the field as a fast screening procedure, but are also used inlaboratories for this purpose [4, 12]. The colour tests are simple, quick and quite sensitive; the bestresults can be acquired with sample quantities of less than one mg or with the pure substance and theresult can be observed with the naked-eye [11, 12].

With the combination of the Marquis, Simon’s and Scott colour test about 90% of illicit drugsseized in the Netherlands can be indicated [4]. The chemicals needed and the methods for these colourtests are given in Table 1, although in literature also slightly other procedures are described [11, 12].The results of the three indicative tests with various substances can be found in Appendix C [11].

Table 1:Chemicals needed and the procedure for the performance of several spot tests [11].

One or two drops of the reagent(s) are added to the presumptive drugs.The tests are validated with drug standards dissolved in chloroform or methanol

at a concentration of 2.0 or 4.0 mg/mL (free-base).Crystals can be tested without further processing and tablets are crushed into fine powder.

Scott Dissolve 2 g of cobalt(II)thiocyanate in 100 mL of distilled water.

Marquis Carefully add 100 mL of concentrated sulphuric acid to 5 mL of 40% formaldehyde (v:v).

Simon’sA: Dissolve 1 g of sodium nitroprusside in 50 mL of distilled water and add 2 mL of acetaldehydeto the solution with thorough mixing.B: 2% sodium carbonate in distilled water.Add one drop of A to the drug followed by two drops of B.

2.1.1 Scott test

The Scott test with cobalt(II)thiocyanate is used to detect cocaine by producing a blue colour, ascan be seen in Appendix C. A colour reaction, with a greenish or bluish product, is also the resultof addition of ephedrine, pseudoephedrine, procaine, methadone, doxepin and other compounds [11].However, the exact structure of the cocaine-cobalt-thiocyanate complex is not known [13].

2.1.2 Marquis test

The Marquis test is an indicative test for a wide variety of basic drugs, such as morphine, MDMA,MDA, amphetamine and methamphetamine [13]. The reagent contains formaldehyde and concentratedsulphuric acid and can give several different coloured reaction products, as can be seen in AppendixC [12].

16


The Marquis test is the most versatile and widely used presumptive test for drugs analysis, althoughthe exact chemistry and occurring reactions and reaction products are not completely understood [14].The reaction mechanisms shown are proposed mechanisms and are not confirmed [4, 14].

Amphetamine and methamphetamine give an orange to brown reaction product [11, 12, 14]. Thecolour is coming from the linked carbenium ions, as is depicted in Figure 1 [13, 14]. Also amphetamine-type stimulants, such as 4-methylamphetamine, can give a colour reaction and the colour can varybetween yellow, red and green [11, 13].

Figure 1:Suggested reaction mechanism of Marquis reagent with amphetamines [13, 14].

A purple to violet colour appears when there are opiates, such as opium and morphine, present [11,13]. Two molecules of morphine and two molecules of formaldehyde form a dimeric product in thepresence of concentrated sulphuric acid. This product can be protonated and forms a carboniumoxonium ion that is responsible for the colour [13, 14]. The suggested reaction mechanism can be seenin Figure 2 [13, 14].

Figure 2:Suggested reaction mechanism of Marquis reagent with morphine [13, 14].

17


With the Marquis reagent it is possible to detect aromatic substances, such as methadone. Thesuggested reaction mechanism is depicted in Figure 3 and shows in the first step the reaction of thearomatic compound with formaldehyde to a carbonium ion. A second molecule can react due to thepresence of sulphuric acid and the product can thereafter oxidize. The end product is the pink-colouredcarbenium ion [13].

Figure 3:Suggested reaction mechanism of Marquis reagent with aromatic compounds [13, 14].

2.1.3 Simon’s test

The Simon’s test is developed for the detection of secondary amines. Addition of the reagent tomethamphetamine, MDMA or MDEA will give a colour reaction, but also other secondary amines,such as diethylamine and piperidine, will give a positive result [12]. The test can be used in addition tothe Marquis test to differentiate between primary amines (for example amphetamine) and secondaryamines (such as methamphetamine and MDMA) [14]. The combination of sodium carbonate, sodiumnitroprusside and acetaldehyde solution gives a blue reaction product, as can be seen in Appendix C,in case of a (false) positive result [11, 12].

Figure 4:Reaction mechanism of Simon’s reagent with secondary amines [13, 14].

18


The reaction mechanism of the formation of the blue Simon-Awe complex is depicted in Fig-ure 4 on the previous page [13]. The secondary amine and acetaldehyde form an enamine by anaddition-condensation reaction. The enamine reacts with sodium nitroprusside to an immonium saltas intermediate and subsequently reacts with water to form the Simon-Awe complex [13, 14]. Becausethere is an hydroxyl-group present in the neighbourhood of the amine, ephedrine and pseudoephedrinedo not show a blue reaction product [11].

2.2 UV-Vis spectroscopy

Ultraviolet-Visible (UV-Vis) spectroscopy is based on electromagnetic radiation in the wavelengthregion of 190 to 800 nm as can be seen in Figure 5 [15]. This technique can give information concern-ing the identity of the drugs as each compound has its own characteristic spectrum. However, thistechnique is not much used for identification, because there are only a few broad peaks present in thespectrum and therefore there is a lack of specific features. Contamination with other substances thatabsorb light in the same region can lead to misidentification. Detection with UV-Vis spectroscopy isa screening type of test due to these drawbacks [9, 16].

Figure 5:The electromagnetic spectrum [15].

The absorption of electromagnetic radiation can be considered to be a process with two steps. Thefirst step is the electronic excitation of the molecule: M1 +hv →M2. The lifetime of M2 is very shortand in the order of 10−9 s whereafter a relaxation process takes place. There are several relaxationprocesses that can take place, but the conversion to heat is the most common: M2 →M + heat. Alsofluorescence and phosphorescence are possible relaxation processes [14, 15, 17].

The wavelength of the absorption bands in the spectrum can be coupled to types of bonds asthe absorption of ultraviolet and/or visible light results from the excitation of bonding electrons.Therefore the obtained spectrum gives information about the functional groups of the analyte [15, 17].The transmittance (T) or absorbance (A) of a solution in a transparent cell is measured with UV-Visspectroscopy. The concentration of the analyte is linearly related to the absorbance. This linearitycan also be seen in Beer’s law [14, 15]:

A = −logT = logI0

I= ε · l · c

19


The molar absorptivity or molar extinction coefficient ε (L/mol·cm) multiplied by the path lengthof the cell l (cm) and the concentration c (mol/L) gives the absorbance A (no units). To compensatefor reflection and scattering losses the intensity of the light I is compared with a reference beam I0 orwith a reference signal [15, 17].

The linearity is in most cases valid, although at high concentrations of analyte this might not betrue anymore. Moreover when the refractive index of a solution changes significantly by concentrationchanges Beer’s law is not valid anymore [15].

To perform molecular absorption spectroscopy a continuum light source is required. At low pressurea deuterium or hydrogen lamp emits a continuous spectrum, with an output range of 160-800 nm and acontinuous spectrum between 190 nm and 400 nm. For the wavelength region 350-2500 nm a tungstenfilament lamp is very suitable. In order to obtain a stable radiation source close voltage control isrequired. Other light sources are light-emitting diodes (LEDs) and xenon arc lamps. By combiningdifferent LEDs a spectrum from 375 to 1000 nm can be obtained. A continuum spectrum of 200-1000nm can be acquired with a xenon arc lamp [15].

2.3 Lab-on-a-chip devices

In clinical diagnostics and bio-analysis already many LOC devices have been developed. An exampleis the i-STAT R© Portable Clinical Analyzer. The i-STAT R© is a hand-held device with a disposablecartridge for point-of-care measurements of metabolic parameters, such as sodium, potassium, glucose,chloride and lactate [18].

LOC devices are also introduced into the forensic field for differential extraction of female and maleDNA, DNA-analysis, analysis of explosive compounds and detection of chemical warfare agents [19,20, 21, 22]. The use of chips has not yet been introduced in the field of illicit drugs analysis. However,there are a lot of advantages using LOC devices, such as speeding up the analysis time and less volumeor weight is needed of the samples as well as the reagents [23].

Bell and Hanes developed a simple microfluidic device, which can be seen in Figure 6, to performseveral presumptive colour tests and a crystal test for the detection of illicit drugs. They focusedon the detection of methamphetamine, amphetamine, cocaine and oxycodone (an opioid). As colourreagent they used the Marquis, Simon and cobalt thiocyanate (in this report called Scott) test [23].For the crystal test, where the formation of crystals indicate the presence of illicit drugs, platinicchloride was used [14, 23].

Figure 6:The chip with three channels (A-C) with different channel designs for the colour reactions and one

channel (D) for crystal reactions [23].

20


The developed device has three channels with a different design for the presumptive colour testsand one channel for the crystal test. The design of channel A turned out to give the best results withrespect to intensity, reproducibility and sensitivity. About 50-125 pg of the illicit drugs was needed inorder to obtain a positive colour test in comparison with 5-100 µg for conventional spot plate tests.By using a reduced area for the colour change it turned out to be easier to observe the colour change.Bell and Hanes concluded that is was possible to perform the complete analysis within 15 s by usingless than 1 mg of sample for all three colour tests and the crystal test. Amounts in the pg range canbe detected, although the crystal reagent has more sensitivity than the colour tests [23].

In comparison with the current spot plate tests less analyte and reagent is needed and less waste isgenerated. Another advantages is that more analytical information is obtained from one single sample.Even people with no experience at all in carrying out colour tests and crystal tests could operate thedevice [23].

In addition, Bell and Hanes5 used an UV-Vis micro-spectrometer microscope in order to observe thecolour change by recording visible reflectance spectra. The spectrum of cocaine and cobalt thiocyanatecan be seen in Figure 7. From these kind of spectra Bell and Hanes confirmed that the colour changesobserved in the chip were the same as in the spot plate method, although no characteristic featurescan be seen in the spectrum [23].

Figure 7:Reflection spectra of cocaine and cobalt thiocyanate in a spot plate well and in the microfluidic

device. The blank spectrum is of the pure reagent obtained in the microfluidic device. [23].

5Personal communication with Bell learns that this research has not been continued after this one publication.

21

Brigitte Bruijns Illicit drugs analysis on chip Materials and methods

3 Materials and methods

3.1 Materials

3.1.1 Chemicals

Cobalt(II)thiocyanate, sulphuric acid, formaldehyde, acetylsalicylic acid, sodium nitroprusside di-hydrate, acetaldehyde, sodium carbonate, diethylamine and chloroform were obtained from Sigma–Aldrich Chemie BV. The ephedrine hydrochloride powder was obtained from the NFI. The sugar wasobtained from a sugar packet from Autobar.

3.1.2 Stock solutions

Stock solutions of the drug analogue acetylsalicylic acid of different concentrations were prepared withchloroform as solvent. Sugar appeared not to dissolve well in chloroform and is therefore preparedwith purified and deionized water (MilliQ) as solvent. Stock solutions of the drug analogue ephedrineof different concentrations were prepared with MilliQ as solvent. Experiments with diethylamine wereperformed with chloroform as well as with MilliQ as solvent6.

For the Scott test 1 g of cobalt(II)thiocyanate was dissolved in 50 mL MilliQ. For the Marquistest 50 mL of sulphuric acid was added to 2.5 mL of formaldehyde. Reagent A for the Simon’s testconsists of 1 g sodium nitroprusside dissolved in 50 mL MilliQ to which 2 mL of acetaldehyde wasadded. The second solution, reagent B, contains 1 g of sodium carbonate dissolved in 50 mL MilliQ.

3.1.3 UV-Vis spectrometer

Spectra were recorded with an HR4000 spectrometer from Ocean Optics with QP400-2-UV-Vis fibersand a Toshiba TCD1304AP linear CCD array detector in combination with a deuterium and a halogenlamp. SpectraSuite software was used to record and analyze the spectra in the wavelength region of200 to 1000 nm. For the recording of all the spectra an integration time of 300 ms was used and thenumber of recorded spectra was set at 1 scan. A Quartz glass cuvet was used with an optical pathlength of 1 cm.

In order to obtain a calibration curve a Tecan Safire2 reader with Megellan software was used. ANunc 96 optical polystyrene well plate and a 96 Quartz glass well plate were used as micro plate. Thewhole spectrum in the wavelength region of 370 to 825 nm for the polystyrene plate and of 230 to1000 nm for the Quartz plate were recorded with for both an integration time of 50 ms, wavelengthstep size of 1 nm and at least 3 scans.

3.1.4 Syringes and pumps

The syringes used for the introduction of the liquids into the chip, connected by capillaries to the chip,were Gastight #1750 syringes from Hamilton with a total volume of 500 µL. The syringe pump usedto control the fluid flow was a Harvard PHD 2000 Programmable pump.

A vacuum pump was used for the introduction of the liquids into the chip without the use ofsyringes.

3.1.5 Microscopes

A Motic SMZ-140 stereo microscope was used to observe and analyse the coloured reaction product ofthe colour tests. The microscope has a magnification range and a zoom ratio of respectively 1.0x-4.0xand 4:1.

6The concentration of only the analyte solution is given in this report for all experiments, independent of the volumeof reagent added to the analyte solution.

22


A Leica DMI 5000M inverted microscope was used to view and photograph the used chips. Themicroscope has a magnification range of 1.25x-100x. The photos were taken with Leica ApplicationSuite software.

3.1.6 Photos

All non-microscopic photos were taken with a Canon IXUS 105 photo camera. The IXUS 105 has a12 mega pixel CCD-chip and a lens with a zoom range of 28-112 mm.

3.1.7 Chips

The first chip design (chip A), made by Roald Tiggelaar, was used as proof of principle for theminiaturization of the colour tests. The channel in the Borofloat chip has a total channel length of372.6 mm. The channels are respectively 55 µm and 120 µm in depth and width as the channels wereisotropically etched. The design, as is depicted in Figure 8, consists of two inlets (one for the reagentsolution and one for the analyte solution) and one outlet.

Figure 8:Left: The design of chip A, which has two inlets and one outlet. Right: Microscopic (Leica DMI

5000 M) picture of the channels on chip A.

The second chip design (chip B), made by Wojciech Bula, was used as proof of principle for UV-Vis detection on chip. The original purpose of the chip was to perform analysis of reaction kineticsin a quench-flow multiline microreactor [24]. The glass-silicon-glass chip, as is depicted in Figure 9,consists of four inlets and eight outlets [25]. The channels are respectively 100 µm and 100 µm indepth and width. The radius of the UV-Vis window in the chip is about 250 µm and has an opticalpath length of about 100 µm.

Figure 9:Left: A picture of chip B, which has eight reaction channels. Right: Detail of chip B [25].

23


3.2 Methods

3.2.1 Colour tests

To test the principle of the presumptive colour test a few crystals of the pure drug analogue (acetyl-salicylic acid, sugar and ephedrine) or a droplet of the pure analyte (diethylamine) were placed on apetri dish or in a small beaker. One to two droplets of the reagent (Scott, Marquis and Simon’s) wereadded to the analytes and the colour change was observed visually.

The colour reactions were also observed by addition of 100 µL analyte solution with a concentrationof 50 mg/mL to 100 µL of reagent solution.

3.2.2 Detection limits

The visual detection limit of each analyte was determined by analysis of serial dilutions of the stocksolutions. A volume of 500 µL analyte solution was added to 500 µL reagent solution in a small beaker.

The spectroscopic detection limit of each analyte was determined by analysis of serial dilutionsof the stock solutions by UV-Vis spectroscopy (HR4000). A volume of 400 µL analyte solution wasused to obtain a liquid level high enough to acquire a reference spectrum. After recording and savingthe reference spectrum and the dark spectrum 300 µL of reagent was added to the cuvet. The totalvolume was chosen in order to prevent overflow of the cuvet. All spectra were directly recorded afteraddition of the reagent solution and analyte solution to each other.

The limit of detection (LOD) is defined as the lowest concentration of the analyte detectable inthree replicates. As a presumptive test must be fast, the peak in the spectrum must be visible withinapproximately one minute.

3.2.3 Calibration curves

The UV-Vis (Tecan Safire2) calibration curves were made by analysis of serial dilutions of the stocksolution. The whole spectra of 200 µL of each reagent, each solvent and each analyte solution wererecorded to obtain the background signal. The whole spectra of 100 µL analyte solution and 100 µLreagent were recorded to acquire the peak maximum. For the calibration curves also a solution of100 µL reagent solution and 100 µL analyte solution was used. As reference signal the spectra of thereagent, the solvent and the pure analyte solution were used. All spectra were directly recorded afteraddition of the reagent solution and analyte solution to each other.

3.2.4 Chip A

Chip A was used to analyse the mixing of the reagents with the analyte on the chip. The sulphuric acidin the Marquis reagent is too harsh for the connectors of the syringes and ephedrine was temporarilynot available. Therefore the proof of principle with these chips was only carried out with the Simon’sreagent and diethylamine as analyte.

As there are only two inlets the Simon’s reagent, made of reagent A and reagent B, is prepared inadvance in a ratio of respectively 1:2. Solutions of 5 mg/mL dietheylamine dissolved in MilliQ and inchloroform were used as analyte solutions. The fluid flow for both inlets, the Simon’s reagent and thediethylamine solution, was set to 0.5 µL/min.

The colour change was observed by connecting a capillary to the outlet, which was placed into asmall beaker. The colour change was also observed by using the Motic SMZ-140 stereo microscope.

To use the chip in combination with the Marquis reagent a vacuum pump was used to introducethe liquids into the chip. This was tried, to show the proof of principle, with the Simon’s reagent and5 mg/mL diethylamine dissolved in MilliQ as these chemicals are less harmful. Droplets of about 5µL of the solutions were disposed on the inlet via-holes.

24


3.2.5 Chip B

Chip B was used to analyse the possibility of UV-Vis detection on chip. As the reaction time was notof importance for this experiment, one of the inlets and all the outlets were covered with sticky tape.About 5 µL of only Simon’s reagent was placed at two of the inlets, the other non-covered inlet wasused to apply the vacuum on.

The chip was placed in between two optical fibers of the HR4000 spectrometer, which were taped tothe table. The spectrum was observed without any solution in the chip and with the reddish colouredSimon’s reagent.

In order to align the fibers correctly a chipholder was developed, as can be seen in Figure 10. Thechip is placed in between the two black plastic blocks and can then be aligned and clamped betweenthe blocks by tightening of the screw. By placing the optical fibers in the holes of the two blocks thefibers are aligned and a spectrum can be recorded. The spectrum was observed without any solutionin the chip and with the reddish coloured Simon’s reagent as well as with the Simon’s reagent addedto 5 mg/mL diethylamine dissolved in MilliQ.

Figure 10:Side and top view of the chipholder with chip B.

25

Brigitte Bruijns Illicit drugs analysis on chip Results and discussion

4 Results and discussion

4.1 Colour tests

4.1.1 Scott test

The addition of the Scott reagent to the ephedrine solution and ephedrine crystals has a reddishproduct as result, similar to the Scott reagent itself. The observed colour reaction, which can be seenin Figure 11, is not in agreement with the colour given in Appendix C [11].

ephedrine crystals 50 mg/mL ephedrine

Figure 11:Scott reagent added to ephedrine crystals and ephedrine solution dissolved in MilliQ.

4.1.2 Marquis test

The addition of the Marquis reagent to the acetylsalicylic acid solution and acetylsalicylic acid crys-tals has a reddish product as result. The addition of the same reagent to the sugar solution andsugar crystals had a yellowish/brownish product as result. Both observed colour reactions, which aredepicted in Figure 12, are in agreement with the colours given in Appendix C [11].

acetylsalicylic acid crystals 50 mg/mL acetylsalicylic acid

sugar crystals 50 mg/mL sugar

Figure 12:Marquis reagent added to acetylsalicylic acid crystals and acetylsalicylic acid dissolved in chloroform

(top) and sugar crystals and sugar dissolved in MilliQ (bottom).

26



The addition of the Simon’s reagent to the diethylamine solution and pure diethylamine has a darkbluish/purplish product as result for chloroform as well as for MilliQ as solvent. The observed colour,which can be seen in Figure 13, is in agreement with the colour given in Appendix C [11].

100 µL A 100 µL A 100 µL A 50 mg/mL+ + + diethylamine200 µL B 100 µL 200 µL B

diethylamine +100 µLdiethylamine

Figure 13:Left: Pure Simon’s reagent. Middle: Simon’s reagent added to pure diethylamine.

Right: Simon’s reagent added to 50 mg/mL diethylamine dissolved in MilliQ.

4.2 Detection limits

4.2.1 Scott test

The observed LOD by eye is not determined for the ephedrine solution, since no colour change takesplace.

The reaction of ephedrine with the Scott reagent does not give a visual colour change. The Scottreagent has a colour which is too intense to monitor any peaks with UV-Vis spectroscopy. In order toget an indication of the spectrum 400 µL of Scott reagent was taken as reference signal. Additionally10-50 µL of 5 mg/mL ephedrine dissolved in MilliQ was added to the cuvet. The obtained signal is abit noisy, but no peak can be observed in the spectrum, as can be seen in Figure 14 on the next page.

Since no colour change takes place and no peak can be observed in the spectrum, also no LODcould be determined.

4.2.2 Marquis test

The observed LOD by eye is about 4.0 mg/mL (2 mg) for both the acetylsalicylic acid solution andthe sugar solution, as is depicted in Appendix D.

The reaction of sugar with the Marquis reagent gives a reaction product which shows a peak at250-320 nm with a maximum around 285 nm. Figure 15 on the next page shows the spectrum of thereagent with a sugar concentration of 5 mg/mL.

With a reaction time of approximately one minute the LOD is about 0.5 mg/mL (200 µg) for sugar,as can be seen in Appendix E. Drawback of this method is the occurrence of a highly exothermicreaction of the sugar solution with the Marquis reagent, as the MilliQ used as solvent reacts with thesulphuric acid in the reagent.

27


Figure 14:Absorption spectrum of 50 µL 5 mg/mL ephedrine dissolved in MilliQ and Scott reagent, with the

Scott reagent as reference (HR4000 spectrometer, this spectrum was obtained withQP600-2-SR/BX fibers.)

Figure 15:Absorption spectrum of 5 mg/mL sugar dissolved in MilliQ and Marquis reagent, with the pure

5 mg/mL sugar solution as reference (HR4000 spectrometer).

28


The reaction of acetylsalicylic acid with the Marquis reagent gives a reaction product which showsa peak at 390-590 nm with a maximum around 490 nm. Figure 16 shows the spectrum of the reagentwith an acetylsalicylic acid concentration of 5 mg/mL.

At a concentration of about 0.3 mg/mL still a little peak can be observed, which corresponds toa total of 125 µg acetylsalicylic acid. However, this peak was only observable with a reaction time ofmore than half an hour.

With a reaction time of approximately one minute the LOD is about 1.5 mg/mL (600 µg), as canbe seen in Appendix E. However, the exact limit is hard to determine with this method as the Marquisand the acetylsalicylic acid solution form a biphasic system, which does not disappear after mixing.Using MilliQ as solvent is not a solution as acetylsalicsylic acid hardly dissolves in water (about 3mg/mL at room temperature) [26, 27].

Figure 16:Absorption spectrum of 5 mg/mL acetylsalicylic acid dissolved in chloroform and Marquis reagent,

with the pure 5 mg/mL acetylsalicylic acid solution as reference (HR4000 spectrometer).


The observed LOD by eye is about 3.0 mg/mL (1.5 mg) for diethylamine dissolved in chloroform aswell as in MilliQ, as is depicted in Appendix D.

The reaction of diethylamine with the Simon’s reagent gives a reaction product which is hardto monitor with UV-Vis spectroscopy. The dark purple/blue colour of the reaction product was toointense to acquire a valid spectrum. The solution of only both reagent solutions gives a reddish/orange-like colour. This solution is also too intense to monitor properly with UV-Vis spectroscopy; the reagentsolution could be used as reference, but due to the strong colouration of this solution it is not possibleto obtain a stable reference signal.

Due to the use of chloroform as solvent a two layer system is formed, which does not disappear aftermixing. To overcome the problem of the two layer system diethylamine can be dissolved in MilliQ.But still the reaction product obtained has a colour which is too intense to acquire a spectrum.

29


In order to get an indication of the spectrum 400 µL of Simon’s reagent, in a A:B ratio of 1:2, wastaken as reference signal. Additionally 10-50 µL of 5 mg/mL dietheylamine dissolved in MilliQ wasadded to the cuvet. The obtained signal is very noisy, especially below 550 nm, but a broad peak canbe observed at 470-670 nm with a maximum around 570 nm. Figure 17 shows the spectrum of thereagent with 30 µL of diethylamine added to the reagent.

Figure 17:Absorption spectrum of 30 µL 5 mg/mL diethylamine dissolved in MilliQ and Simon’s reagent, with

the Simon’s reagent as reference (HR4000 spectrometer).

4.3 Calibration curves

Chloroform dissolves the polystyrene from which the micro plates are made of [28]. Therefore, it wasnot possible to obtain a calibration curve of the analytes dissolved in chloroform. The polystyrenemicro plate is transparent in the region from 370 nm till 825 nm. In order to analyse the analytesdissolved in chloroform and to obtain a spectrum below 370 nm also a Quartz glass micro plate wasused.

The Tecan Safire2 was used to record the calibration spectra as this spectrometer is more sensitiveand accurate than the HR4000 spectrometer. A small movement and/or misallignment of the cuvetin the holder of the HR4000 has a big influence on the exact (height of the) spectrum.

4.3.1 Scott test

No spectra were recorded with the Tecan Safire2, since no colour change as well as no peak in theUV-Vis spectrum could be obtained for the reaction of the Scott reagent with ephedrine.

4.3.2 Marquis test

The spectra of the Marquis reagent, MilliQ, and the sugar solution were recorded to obtain thebackground signal of the reagent, the solvent, and the analyte. The Marquis reagent shows a littlepeak around 310 nm and the sugar solution shows a peak around 980 nm, which is at the limit oftransparency of the plate. Therefore the spectrum of the Marquis reagent must actually be subtractedfrom the spectra of the analyte solution with the Marquis reagent. However, the exact chemistry ofthe reaction is not known and therefore it is not clear how to subtract the spectrum of the Marquisreagent.

30


The absorption spectra of the sugar solutions in the concentration series of 0.1-10 mg/mL show aclear peak at 285 nm, but besides this peak also little peaks can be observed at 440 nm and 490 nm.Those little peaks cannot be observed anymore at lower concentrations, as can be seen in Figure 18.A minimum can be observed at 250 nm.

With the Tecan Safire2 the LOD for sugar dissolved in MilliQ, at 285 nm, lies below 0.1 mg/mL(10 µg).

Figure 18:Absorption spectrum of 0.1-10 mg/mL sugar dissolved in MilliQ and Marquis reagent

and the spectrum of the pure Marquis reagent(Tecan Safire2).

A calibration curve, for which the spectra were measured in triplicate, is made at 250 nm, 285 nmand 490 nm. The calibration curve for the dilution series of 0.1-10 mg/mL is depicted in Figure 19 onthe next page.

Linear calibration curves could be made at the measured wavelengths. At the lower wavelengthsthe absorption at higher concentrations is too high to measure.

The spectra of the Marquis reagent, chloroform and the acetylsalicylic acid solution were recordedin order to obtain the background signal of the reagent, the solvent and the analyte. The Marquisreagent shows a little peak around 310 nm, the acetylsalicylic acid solution does not show a distinctivepeak, but only an elevated signal below 350 nm, and chloroform shows only an elevated signal below240 nm.

The issue with the acetylsalicylic acid is that it is dissolved in chloroform, which forms a twolayer system together with the Marquis reagent. The two layers can be mixed, but after a little whilea two layer system is formed again. Therefore it is not clear what is recorded during the UV-Vismeasurement.

Nevertheless, the absorption spectra of the acetylsalicylic acid solutions in the concentration seriesof 0.1-10 mg/mL were recorded. A clear peak is visible around 490 nm for high concentrations andbetween 300 nm and 330 nm for lower concentrations, as can be seen in Figure 20 on the next page.The peaks at the lower wavelengths could not be seen with the HR4000 since the signal-to-noise ratiowas too low in that range. No spectrum is subtracted in Figure 20 since it is not defined which of thetwo layers or if both layers are measured.

31


Figure 19:Calibration curves of 0.1-10 mg/mL sugar dissolved in MilliQ and Marquis reagent at

250 nm, 285 nm and 490 nm.

Figure 20:Absorption spectrum of 0.1-10 mg/mL acetylsalicylic acid dissolved in MilliQ and Marquis reagent

and the spectrum of the pure Marquis reagent(Tecan Safire2).

32


No calibration curve could be made for the dilution series of 0.1-10 mg/mL since no constant peakis observable in the spectrum and it is not sure which layer is measured.

With the Tecan Safire2 the LOD for acetylsalicylic acid dissolved in chloroform, at 300-330 nm,lies below 0.1 mg/mL (10 µg).


The spectra of the Simon’s reagent, MilliQ and the diethylamine solution, with MilliQ as solvent, wererecorded to obtain the background signal of the reagent, the solvent and the analyte. The Simon’sreagent, which is ten times diluted in order to get a less intense colour, shows an elevated signalbelow 630 nm and the dietheylamine solution shows a peak around 980 nm, which is at the limit oftransparency of the plate. Therefore the spectrum of the diluted Simon’s reagent must actually besubtracted from the spectra of the analyte solution with the Simon’s reagent. However, the exactchemistry of the reaction is not known and therefore it is not clear how to subtract the spectrum ofthe Simon’s reagent.

The purple colour obtained by addition of Simon’s reagent to diethylamine dissolved in MilliQ istoo intense in order to acquire a proper spectrum. The end product was therefore diluted with MilliQafter which a spectrum was taken. Due to the dilution of the reagent the spectrum below 550 nm isnow not noisy anymore and the peaks can be observed clearly.

The absorption spectra of the diethylamine solutions in the concentration series of 0.1-10 mg/mLshow a peak around 570 nm and a little peak around 450 nm and a minimum around 400 nm, as canbe seen in Figure 21.

With the Tecan Safire2 the LOD for sugar dissolved in MilliQ lies below 0.1 mg/mL (10 µg).

Figure 21:Absorption spectrum of 0.1-10 mg/mL diethylamine dissolved in MilliQ and Simon’s reagent

and with the spectrum of the diluted Simon’s reagent (Tecan Safire2).

A calibration curve, for which the spectra were measured in triplicate, is made at 360 nm, 400 nmand 570 nm. The calibration curve for the dilution series of 0.1-10 mg/mL is depicted in Figure 22 onthe next page.

33


Figure 22:Calibration curves of 0.1-10 mg/mL diethylamine dissolved in MilliQ and Marquis reagent at

360 nm, 400 nm and 570 nm.

It was not possible to make linear calibration curves at the measured wavelengths. This is possiblydue to the strong dependence of the colouration in time caused by ongoing reaction between theanalyte and the reagent.

No spectra could be recorded for the diethylamine solutions dissolved in chloroform, as it was notpossible to mix the chloroform with the Simon’s reagent; a two layer system is immediately formedback again after mixing.

4.4 Remarks colour tests and UV-Vis spectroscopy

4.4.1 Colour tests

All the reaction products have the colour as expected from the table in Appendix C, except forephedrine, which does not give a colour change with the Scott reagent [11]. All the reaction products,including ephedrine, are in agreement with the results obtained at the NFI. The colour change isclearly observable with the pure analytes as well as with the 50 mg/mL analyte solutions.

4.4.2 Operational drug detection limit

It is common to use a safety factor for the detection limit [11]. This operational drug detection limit(ODDL) is defined as the limit of detection multiplied by 10. The ODDL determined visually as wellas by UV-Vis spectroscopy for the various colour tests and used solvents can be found in Table 2 onthe next page.

The analyses were carried out with only the pure compounds (ephedrine, acetylsalicylic acid, sugarand diethylamine). Real drugs samples will contain all kinds of contamination which can provide falsepositive as well as false negative readouts.

34


Table 2:Operation drug detection limit of the analytes dissolved in chloroform and/or MilliQ.

Test Analyte Method ODDL Solvent

Scott Ephedrine

EyeNo solution made CHCl3- (no colour change) MilliQ

UV-Vis (HR4000)No solution made CHCl3- (no colour change) MilliQ

UV-Vis (Tecan)No solution made CHCl3- (no colour change) MilliQ

Marquis

Acetylsalicylic acid

Eye40 mg/mL (20 mg) CHCl3– (does not dissolve) MilliQ

UV-Vis (HR4000)15 mg/mL (6 mg) CHCl3– (does not dissolve) MilliQ

UV-Vis (Tecan)< 1 mg/mL (100 µg) CHCl3– (does not dissolve) MilliQ

Sugar

Eye– (does not dissolve) CHCl340 mg/mL (20 mg) MilliQ

UV-Vis (HR4000)– (does not dissolve) CHCl35 mg/mL (2 mg) MilliQ

UV-Vis (Tecan)– (does not dissolve) CHCl3< 1 mg/mL (100 µg) MilliQ

Simon’s Diethylamine

Eye30 mg/mL (15 mg) CHCl330 mg/mL (15 mg) MilliQ

UV-Vis (HR4000)No detection possible CHCl3No detection possible MilliQ

UV-Vis (Tecan)No detection possible CHCl3< 1 mg/mL (100 µg) MilliQ

The visual observation and analysis of the coloured reaction products is very subjective. Forexample diethylamine shows already at a lower concentration a colour change after addition of theSimon’s reagent, as can be seen in Appendix D, but this does not give the distinctive purple reactionproduct. Therefore it is better to determine the LOD with a more objective method, such as UV-Visspectroscopy.

The observed LOD for the analytes with the used methods is somewhat higher than suspectedfrom the LODs for illicit drugs, which are given in Appendix C. However, the used method for thedetermination of those LODs as well as the used method at the NFI is different [4, 11]. At the NFI aspatula tip of the analyte powder is added to an excess of reagent. The LOD is in that case lower asthe local concentration of the analyte is much higher in comparison with a diluted solution [4].

When dilution is needed in order to obtain a reliable spectrum it is important to start with theundiluted reagent. In cases where the reagent is diluted prior to the addition to the analyte no directcolour change will occur, therefore the reagent and the analyte must first be added to each otherwhereafter dilution can take place.

4.4.3 Solvent

To prepare drug standards for the colour tests in most cases chloroform is used as solvent [11]. Thisrelative non-polar solvent is not ideal for UV-Vis spectroscopy as it forms a two layer system with thereagents.

35


In some cases MilliQ can be used as solvent, but not all the illicit drugs dissolve in water, on theother hand an extremely exothermic reaction takes place with the Marquis reagent. Also methanolcan be used as solvent to make drug standards [11, 4]. However, dimethylsulphate is the reactionproduct of methanol and sulphuric acid, which is an extremely hazardous and toxic compound [29].Analysis on chip is then advantageous, since only small volumes are involved.

Another disadvantage of polar solvents, such as water and alcohols, is the possibility of a re-duced spectral fine structure resulting from the vibrational effects. It therefore mandatory that foridentification purposes always the same solvent is used for the reference spectrum as for the analytespectrum [15].

4.4.4 UV-Vis spectroscopy

In cases where the analyte concentration is too high, which is visible as an intense colour, the absorptionbands in the spectrum will saturate and the spectrum will show a noisy signal [15]. Therefore it wasnot possible to obtain a spectrum and determine the LOD for the Simon’s reagent with diethylaminewithout dilution.

The enormous spikes in the spectra at lower wavelengths can be explained by the high absorptionof the analytes at these wavelengths. An UV-Vis spectrometer measures the transmission of a sampleand translates that to the absorption. This conversion from transmission to absorption is logarithmic,as can be read in Chapter 2. For a transmission of zero or nearly zero is the absorbance not welldefined, since the log of zero is not defined, with spikes in the spectrum as result [15, 17].

It is not possible to make a quantitative statement based on the calibration curves as the exactabsorption is very time dependent as well as on the ratio of the analyte solution and the reagentsolution. Only with a well defined calibration and when all the parameters are known and kept thesame it is possible to make a quantitative statement.

It is difficult to subtract the correct background signal, especially from the reagents, as the exactchemistry of the reactions is not known. Therefore it is not known if the reagent and analyte reactin a ratio of 1:1. It is also possible that the (coloured) reagent solution is not completely used inthe reaction. Additional difficulty it that the background spectra were taken from 200 µL reagentsolution. The absorption spectra were obtained with only 100 µL reagent solution and an additional100 µL analyte solution. Hence both spectra were not taken with an equal concentration of reagentsolution.

4.5 Chip A

The chip showed a good mixing performance; the solution in the beaker coming from the outlet hasa distinctive purple colour as is expected for the Simon’s reagent with diethylamine. Diethylaminedissolved in MilliQ as well as in chloroform show a good mixing performance; a two layer systemdoes not appear. Drawback of this chip is that due to the small channels the colour change cannotbe observed visually. Even with the Motic SMZ-140 stereo microscope no colours, the red Simon’sreagents and/or the purple reaction product, can be observed in the chip. Another disadvantage isthat the chip cannot be used in combination with the Marquis reagent, as the connections of thesyringes are not compatible with strong acids.

The method with the vacuum pump worked well; by placing a droplet of about 5 µL of both theanalyte and the reagent solution at the inlets, the solutions get sucked into the chip. Drawback ofthis method is that the colour at the outlet is not always visually observable as there is only a smallamount of solution present, because most of the solution gets sucked into the pump. Nevertheless,this experiment showed that the method with the vacuum pump works and that only a small volumeis enough to obtain an usable result.

36


4.6 Chip B

By placing the chip in between the two optical fibers it was possible to obtain a signal from the chip.However, the alignment of the two fibers was hard as no set up was available and the fibers weretherefore taped to the table. Hence it was not possible to obtain a sufficient reference signal with asresult that no spectrum could be taken.

By placing the chip in the chipholder the fibers are aligned and a spectrum of the light sourcecould be obtained. However, it was not possible to obtain a spectrum of the solutions. Even whenthe Simon’s reagent and the dietylamine solution were added to each other before introduction inthe chip no spectrum could be obtained. It might be that the radius of the UV-Vis window is toonarrow, and therefore the detection area might be too small, to obtain a sufficient spectrum. Anotherpossibility is that, due to the design of the chip, there is not enough solution or the solution is notequally distributed at the UV-Vis window. Also the material where the chip is made of can influencethe signal since not all materials are transparent in the UV region.

4.7 Remarks detection on chip

Both chips show very promising results, concerning mixing and UV-Vis detection, for the analysis ofillicit drugs on chip. However, a chip must be designed to fulfill all the requirements for the analysis.The material of the chip must be transparent in the whole wavelength range and the detection areamust be large enough in order to obtain a sufficient spectrum. Another requirement is that the threecolour tests can be carried out simultaneously.

37

Legal highs

Brigitte Bruijns Illicit drugs analysis on chip Legal highs

5 Legal highs

The term designer drugs (or research chemicals), such as 4-methylamphetamine and mephedrone,refers to a compound that has a basic structure similar to a scheduled drug, but has been modifiedchemically to create non-scheduled compounds to avoid the current drug laws. The effect of these legalhighs, of which some examples can be seen in Appendix F, are often similar to illicit drugs. Thereforeare these compounds also called legal highs as they are not yet, in the case of the Netherlands, in listI or II of the Dutch Opium Law [30, 31].

Examples of designer drugs are the beta-keto-amphetamines (βk-amphetamines) such as methy-lone, ethylone and butylone, which are structurally derived from amphetamines and other CNS stim-ulants; also the effect of these new drugs are comparable [32, 33]. Some amphetamine derivatives arehallucinogenic compounds, such as DOM and DOB, whereas others are non-hallucinogenic, such asMBDB [10, 34].

”Bromodragonfly” (1-(8-bromobenzo[1,2-b;4,5-b]difuran-4-yl)-2-aminopropane) is another new syn-thetic hallucinogenic drug, which has an LSD-like effect and known for its long duration of action [35].Legal highs can also be plant or animal based materials with hallucinogenic effects. Recently, a wholerange of so-called synthetic cannabinoids and cannabimimettics appeared as a legal alternative tocannabis. One example is ”Spice”; a smoking blend/mixture of plant material with different syntheticcompounds added to it [30]. Unfortunately already several deaths have been reported in relation tothese legal highs [30, 35, 36, 37].

5.1 Mephedrone

Mephedrone, also known as ”meow meow” or 4-methylmethacathinone, is a synthetic cathinone andpenethylamine derivative, which is in structure related to drugs as MDMA and amphetamine [30, 36,38]. It is a popular drug among experienced polydrug users associated with the dance music scene andthe most commonly used designer drug [31, 38]. It is available as a hydrochloride salt, which is white,yellowish, beige or brown coloured and is sold as plant food and as bath salt [31]. The compound isavailable as pure powder, but also as tablet/capsule and intake can therefore be by oral ingestionsas well as nasal insufflation [36]. It can also be dissolved in a liquid or the powder can be wrappedin cigarette paper and swallowed and, although less popular, are intramuscular and rectal routes alsopossible [31, 38].

However the effects are about equal as for MDMA or cocaine, mephedrone may be more addictive,although not much is known about the effects and pattern of use [37, 38]. The acute toxicity ofmephedrone is similar to what is reported for MDMA and cocaine. Undesired side effects of this drugare sweating, palpitations, headaches and nausea. Some people also experienced visual and auditoryhallucinations or an increased sex drive [36, 38]. In comparison with cocaine people report a longerlasting and better high [38].

At the moment mephedrone is controlled in a number of European countries [36, 38].

5.2 MBDB

Methylbenzodioxolylbutanamine (MBDB), also known as Eden, has a structure that is closely relatedto MDMA. The effects of this drug on emotion and empathy are also similar to MDMA, althoughMBDB is less potent than MDMA. It is suggested that compounds such as MBDB, due to the differentpharmacology, should be placed outside the class of hallucinogenic or psychedelic amphetamines.MDBD is therefore known as an entactogen and also MDMA and even MDA exhibit entactogenicactivity [10].

40


5.3 MDAI

In the scientific literature not a lot can be found about 5,6-methylenedioxy-2-aminoindane (MDAI).The compound, which is an entactogen, is structurally related to MDA and MDMA, but has differentpharmacological properties. The compound was developed in 1990, but became popular around 2009.MDAI is available as powder and can be taken via the oral, rectal or intranasal route [31]. Most ofthe information available about the effects of MDAI is from users who share their experiences aboutthis recently available of drug on the internet.

5.4 Methylone

Methylone (3,4-methylenedioxymethcathinone or βk-MDMA), also known as Explosion, is an ecstasy-like compound which appeared at the end of 2004 in the Netherlands. The drug was initially sold asa room odorizer and could be bought in plastic tubes with about 5 mL of liquid [39, 40]. It is aftermephedrone the most used legal high [31].

Despite the behavioural and pharmacological similarities between methylone and MDMA, the effectof these two drugs are different. The potency is about the same, but methylone shows antidepressant,pleasant and positive effects, but without the so-called unique magic of MDMA [39]. Although peopleshould bear in mind that these effects are very subjective, because most of the available informationare reports of users [36, 39]. There is not a lot known about the risks of this drug and due to thesimilarities with MDMA identical harmfull effects cannot be excluded [40].

5.5 Ethylone

The compound 3,4-methylenedioxy-N-ethylcathinone (ethylone or βk-MDEA) is one of the syntheticcathinone derivatives of which not much is known chemically and biologically. It is an entactogen, butis also classified as psychedelic and the compound is the beta-keto analogue of MDEA [30].

5.6 Butylone

Also the compound β-keto-N-methylbenzodioxolylpropylamine (butylone or βk-MBDB) is one of thesynthetic cathinone derivatives of which not much is known chemically and biologically. It is an entac-togen, but is also classified as psychedelic and the compound is the beta-keto analogue of MBDB [30].

5.7 4-Fluoroamphetamine

The mono-substituted fluoroamphetamines, where a hydrogen atom of the phenyl ring is substitutedby a fluorine atom, are designed to increase the lipophilicity and to enhance the passing of the bloodbrain barrier. The pharmalogical effect is comparable with amphetamine; these compounds havetherefore a potential for abuse [41].

5.8 2,5-Dimethoxyamphetamines

Both 4-bromo-2,5-dimethoxyamphetamine (DOB) and 2,5-dimethoxy-4-methyl-amphetamine (DOM)pertain in the class of the 2,5-dimethoxyamphetamines. These drugs are available as tablets, powders,liquids and even blotters [33]. DOB is mainly available in the form of tablets or blotters [42].

With the Marquis test DOB shows a yellowish end product, which turns into green [43].At the moment are the 2,5-dimethoxyamphetamines, due to the abuse potential, controlled in most

countries [33, 42].

41


5.9 2,5-Dimethoxyphenethylamines

Examples of the class of 2,5-dimethoxyphenethylamines are 4-bromo-2,5-dimethoxy-phenethylamine(2C-B) and 4-iodo-2,5-dimethoxy-phenethylamine (2C-I). 2C-B is available as powder and in tabletform [43].

The effect of 2C-B is similar to mescaline, psilocybin and LSD, although there is not a lot of infor-mation about the pharmacologic and toxicologic properties of the 2,5-dimethoxyphenethylamines [33,43].

With the Marquis test 2C-B shows a yellowish end product, which turns into green [43].At the moment 2C-B and other 2,5-dimethoxyphenethylamines are controlled in a lot of coun-

tries [43].

42


6 Materials and methods

6.1 Materials

6.1.1 Chemicals

Methylone hydrochloride, ethylone hydrochloride, butylone hydrochloride, 4-fluoroamphetamine and5,6-methylenedioxy-2-aminoindane (MDAI) were obtained from LGC Standards. Cobalt(II)thiocyanate,sulphuric acid, formaldehyde, acetylsalicylic acid, sodium nitroprusside dihydrate, acetaldehyde, sodiumcarbonate, diethylamine and chloroform were obtained from Sigma–Aldrich Chemie BV. Ephedrinehydrochloride7, lidocaine8 and 4-methylamphetamine sulphate and methylbenzodioxolylbutanamine(MBDB) were obtained from the NFI.

6.1.2 Stock solutions

Stock solutions of the legal highs of different concentrations were prepared with MilliQ as solvent9.For the Scott test 1 g of cobalt(II)thiocyanate was dissolved in 50 mL MilliQ. For the Marquis

test 50 mL of sulphuric acid was added to 2.5 mL of formaldehyde. Reagent A for the Simon’s testconsists of 1 g sodium nitroprusside dissolved in 50 mL MilliQ to which 2 mL of acetaldehyde wasadded. The second solution, reagent B, contains 1 g of sodium carbonate dissolved in 50 mL MilliQ.Reagent A and reagent B were premixed before each experiment in the ratio of 2:110.

6.1.3 UV-Vis spectrometer

The UV-Vis spectra were recorded with an HR4000 spectrometer from Ocean Optics with QP600-2-SR/BX fibers and a Toshiba TCD1304AP linear CCD array detector in combination with a deuteriumand a halogen lamp. SpectraSuite software was used to record and analyze the spectra in the wave-length region of 200 to 1000 nm. For the recording of all the spectra an integration time of 100 mswas used and the number of recorded spectra was set at 1 scan. A Quartz glass cuvet was used withan optical path length of 1 cm.

UV-Vis spectra were also recorded with a Tecan Safire2 reader with Megellan software. A 96Quartz glass well plate was used as micro plate. The whole spectrum in the wavelength region of 230to 1000 nm was recorded with an integration time of 50 ms, wavelength step size of 1 nm and at least3 scans.

6.1.4 Pump

A vacuum pump was used for the introduction of the liquids into the chip without the use of syringes.

6.1.5 Photos

All photos were taken with a Canon IXUS 105 photo camera. The IXUS 105 has a 12 mega pixelCCD-chip and a lens with a zoom range of 28-112 mm.

6.1.6 Chips

Several Borofloat chips were designed to carry out the three colour tests simultaneously. The eightdesigns differ in channel width, number of inlets, number of detection windows, radius of the detectionwindow, the number of meanders and some of the designs also include a Tesla mixer.

7Ephedrine is a precursor of methamphtamine.8Lidocaine is a common cutting agent of cocaine.9The concentration of only the analyte solution is given in this report for all experiments, independent of the volume

of reagent added to the analyte solution.10At the NFI a one step reaction is performed with the premixed reagent instead of a two step reaction.

43


All channels have a depth of 22 µm, and therefore an optical path length of about 22 µm, andcontain one outlet for the application of the vacuum. The design and features of all the different chipscan be found in Appendix G. The chips were made by Stefan Schlautmann and the fabrication processof these chips can be found in Appendix H.

6.2 Methods

6.2.1 Colour tests

One to two droplets of the reagent (Scott, Marquis and Simon’s) were placed on a petri dish orplastic plate. A few crystals of the pure legal high (methylone hydrochloride, ethylone hydrochlo-ride, butylone hydrochloride, 4-fluoroamphetamine, MDAI, ephedrine hydrochloride, lidocaine, 4-methylamphetamine sulphate and MBDB) were added to the analytes and the colour change wasobserved visually. The reagent is placed on the plate before the analytes to ensure that the plate isnot contaminated. The same procedure was performed on the drug analogues (acetylsalicylic acid,sugar, ephedrine and diethylamine) as a reference.


To get an indication of the spectra of the different compounds with the different reagents a fullspectrum is recorded with the HR4000 UV-Vis spectrometer. A volume of 400 µL reagent solutionwas used to obtain a liquid level high enough to acquire a reference spectrum. The reagents werechosen as a reference, since the Scott and the Simon’s reagents already have an intense colour. Afterrecording and saving the reference spectrum and the dark spectrum 300 µL of analyte solution, with aconcentration of 5 mg/mL, was added to the cuvet. The total volume was chosen in order to preventoverflow of the cuvet. All spectra were directly recorded after addition of the reagent solution andanalyte solution to each other.

In case of a colouration which is too intense to acquire a proper spectrum, 400 µL of the reagentsolution was taken as reference signal and additionally a small volume of 5 mg/mL analyte solutionwas added to the cuvet.

No spectra were recorded with the HR4000 of MDAI, methylone, ethylone, butylone and 4-fluoroamphetamine, since only a small amount (10 mg) of analyte was available of these legal highs.

6.2.3 Calibration curves

The UV-Vis (Tecan Safire2) calibration curves were made by analysis of serial dilutions of the stocksolution. The whole spectra of 200 µL of each reagent, each solvent and each analyte solution wererecorded to obtain the background signal. The whole spectra of 100 µL analyte solution and 100 µLreagent were recorded to acquire the peak maximum. For the calibration curves also a solution of100 µL reagent solution and 100 µL analyte solution was used. As reference signal the spectra ofthe reagent, the solvent and the pure analyte solution were used. The solutions with Scott reagentor Simons reagent were ten times diluted, since the Scott and the Simons reagents already have anintense colour. All spectra were directly recorded after addition of the reagent solution and analytesolution to each other.

6.2.4 Chip C

At first it was analysed if a colour can be visually observed in the chips, especially for the chips thathave a big detection window (radius of ≥ 750 µm). About 5 µL of several solutions were placed at allthe inlets of the chips.

44


The used solutions with different colours were premixed Simon’s reagent with 5 mg/mL diethy-lamine in a ratio of about 1:1 (dark purple), an ink solution (bright red), reagent A of the Simon’sreagent (dark red), premixed Scott reagent with 5 mg/mL diethylamine in a ratio of about 1:1 (lightblue) and premixed Marquis reagent with 5 mg/mL acetylsalicylic acid in a ratio of about 1:1 (red).The vacuum pump was used to introduce the liquids into the chip.

To check if the analyte solution and the reagents will mix in all the chips about 5 µL of Simon’sreagent was placed at all the reagent inlets and about 10 µL of 5 mg/mL diethylamine solution wasplaced at the analyte inlet of the chips. The vacuum pump was used to introduce the liquids into thechip. The Simon’s reagent and the diethylamine solution were used as a proof op principle, becauseprevious experiments showed that the purple end product can be visually observed in the chips.

To obtain the UV-Vis spectrum about 5 µL of the reagents and about 10 µL of 5 mg/mL analytesolution were placed at the inlets of the chips. The vacuum pump was used to introduce the liquidsinto the chip. The chips were placed in the chipholder to align the two optical fibers of the HR4000spectrometer for detection by UV-Vis spectroscopy.

45


7 Results and discussion

7.1 Colour tests

The results of the addition of the three colour reagents (Scott, Marquis and Simon’s) to the analytes(drug analogues and legal highs) are depicted in Table 3 on the next page. Some compounds showa coloured end product with more than one colour reagent. The observed colour reactions are inagreement with the results of the NFI.

7.2 UV-Vis spectroscopy

7.2.1 Ephedrine

When ephedrine is added to the Simon’s reagent a bluish end product is obtained. However, no changeof colour can be observed when 300 µL of 5 mg/mL ephedrine solution is added to 400 µL Simon’sreagent. Only at higher concentrations a bluish end product can be obtained.

7.2.2 Diethylamine

When diethylamine is added to the Scott reagent a bluish/greenish end product is obtained. Thecolour is too intense to acquire a proper spectrum and therefore 10-50 µL of 5 mg/mL diethylaminesolution was added to 400 µL Scott reagent. The reaction product shows a peak at 550-650 nm with amaximum around 590 nm and around 640 nm. Also an elevated signal is visible around 350 nm. Figure23 shows the spectrum of the Marquis reagent with 50 µL diethylamine solution with a concentrationof 5 mg/mL added to 400 µL Scott reagent.

Figure 23:Absorption spectrum of 50 µL 5 mg/mL diethylamine dissolved in MilliQ and Scott reagent,

with the Scott reagent as reference (HR4000 spectrometer).

When diethylamine is added to the Simon’s reagent a purplish end product is obtained. Thespectrum of this reaction product is already extensively discussed in the first part of the report aboutdrugs of abuse.

46


Table 3:The observed product of the reagents added to the analytes.

Scott Marquis Simon’s

No analyte (blank)

Ephedrine No reaction No reaction Bluish (dark)

Sugar No reaction Yellowish No reaction

Acetylsalicylic acid No reaction Reddish No reaction

Diethylamine Bluish/Greenish No reaction Purplish

Lidocaine Bluish (light) No reaction No reaction

MBDB Bluish (light) Purplish/Black Bluish (dark)

47


Table 3:The observed product of the reagents added to the analytes – continued.

Scott Marquis Simon’s

No analyte (blank)

MDAI No reaction Yellowish No reaction

Methylone No reaction Yellowish No reaction

Ethylone No reaction Yellowish No reaction

Butylone No reaction Yellowish No reaction

4-Methylamphetamine No reaction Reddish No reaction

4-Fluoroamphetamine No reaction Reddish No reaction

48


7.2.3 Lidocaine

When lidocaine is added to the Scott reagent a bluish end product is obtained. However, no change ofcolour can be observed when 300 µL of 5 mg/mL lidocaine solution is added to 400 µL Scott reagent.Only at higher concentrations a bluish end product can be obtained.

7.2.4 MBDB

When MBDB is added to the Scott reagent a bluish end product is obtained. However, no change ofcolour can be observed when 300 µL of 5 mg/mL lidocaine solution is added to 400 µL Scott reagent.Only at higher concentrations a bluish end product can be obtained.

When MBDB is added to the Marquis reagent purplish/black end product is obtained. Thereaction product shows a peak at 350-450 nm with a maximum around 400 nm. Figure 24 shows thespectrum of the Marquis reagent with the MDBD solution with a concentration of 5 mg/mL.

Figure 24:Absorption spectrum of 5 mg/mL MBDB dissolved in MilliQ and Marquis reagent,

with the Marquis reagent as reference (HR4000 spectrometer).

When MBDB is added to the Simon’s reagent a bluish end product is obtained. The colour is toointense to acquire a proper spectrum and therefore 10-50 µL of 5 mg/mL diethylamine solution wasadded to 400 µL Simon’s reagent. The reaction product shows a peak at 500-700 nm with a maximumaround 600 nm. Figure 25 on the next page shows the spectrum of the Simon’s reagent with 10 µLdiethylamine solution with a concentration of 5 mg/mL added to 400 µL Simon’s reagent.

7.2.5 4-Methylamphetamine

When 4-methylamphetamine is added to the Marquis reagent a reddish end product is obtained. Thereaction product shows a peak at 450-550 nm with a maximum around 500 nm. Other peaks can beobserved around 270 nm, 330 nm and 370 nm. Figure 26 on the next page shows the spectrum of theMarquis reagent with the 4-methylamphetamine solution with a concentration of 5 mg/mL.

49


Figure 25:Absorption spectrum of 10 µL 5 mg/mL MBDB dissolved in MilliQ and Simon’s reagent,

with the Simon’s reagent as reference (HR4000 spectrometer).

Figure 26:Absorption spectrum of 5 mg/mL 4-methylamphetamine dissolved in MilliQ and Marquis reagent,

with the Marquis reagent as reference (HR4000 spectrometer).

50


7.3 Calibration curves

There was not enough analyte material available to make a wide range of serial dilutions, which can bemeasured in triplicate at the Tecan Safire2. However, of each reagent, each solvent (only MilliQ) andeach analyte solution the whole spectrum was acquired to obtain the background signal. Additionally,spectra were taken in a small range of serial dilutions of the analytes and reagents which give a colouredreaction product.

Since no information was available about the solubility of MDAI, methylone, ethylone and butyloneno spectra could be obtained for these analytes, since there was not enough analyte material availableto test the solubility.

7.3.1 Ephedrine

When ephedrine is added to the Simon’s reagent a bluish end product is obtained. However, no peakscan be observed in the spectrum other than coming from the Simon’s reagent itself, as can be seen inFigure 27.

Figure 27:Absorption spectrum of 0.1-1 mg/mL ephedrine dissolved in MilliQ and Simon’s reagent

and the spectra of the diluted Simon’s reagent and pure 1 mg/mL ephedrine dissolved in MilliQ(Tecan Safire2).

7.3.2 Diethylamine

The absorption spectra of the diethylamine solutions with the Scott reagent in the concentrationsseries of 0.1-1 mg/mL show peaks around 470 nm, 590 nm and 640 nm, as can be seen in Figure 28on the next page. At lower concentrations also a peak around 270 nm can be observed.

The absorption spectra of the diethylamine solutions with the Simons reagent in the concentrationsseries of 0.1-10 mg/mL were again obtained, as the absorbance of the 10 mg/mL and the 5 mg/mLsolution showed some irregularities (Figure 21). Again, as can be seen in Figure 29 on the next page,the absorbance of the 10 mg/mL and the 5 mg/mL solution is not as would be expected around 570nm. The absorbance of the 5 mg/mL solution is higher than the absorbance of the 10 mg/mL solution.

51


Figure 28:Absorption spectrum of 0.1-1 mg/mL diethylamine dissolved in MilliQ and Scott reagent

and the spectra of the diluted Scott reagent and pure 1 mg/mL diethylamine dissolved in MilliQ(Tecan Safire2).

Figure 29:Absorption spectrum of 0.1-10 mg/mL diethylamine dissolved in MilliQ and Simon’s reagent

and the spectra of the diluted Simon’s reagent and pure 1 mg/mL diethylamine dissolved in MilliQ(Tecan Safire2).

52


7.3.3 Lidocaine

When lidocaine is added to the Scott reagent a bluish end product is obtained. However, no peakscan be observed in the spectrum other than coming from the Simon’s reagent or lidocaine itself, ascan be seen in Figure 30. Notable is the high absorbance of pure lidocaine around 270 nm.

Figure 30:Absorption spectrum of 0.1-1 mg/mL lidocaine dissolved in MilliQ and Scott reagent

and the spectra of the diluted Scott reagent and pure 1.0 mg/mL lidocaine dissolved in MilliQ(Tecan Safire2).

7.3.4 MBDB

When ephedrine is added to the Scott reagent a bluish end product is obtained. However, no peakscan be observed in the spectrum other than coming from the Scott reagent or MBDB itself, as can beseen in Figure 31 on the next page. Notable is the high absorbance of pure MBDB around 285 nm.

The absorption spectra of the diethylamine solutions with the Marquis reagent in the concentra-tions series of 0.1-10 mg/mL show a peak around 400 nm, as can be seen in Figure 32 on the nextpage. The peak around 285 nm overlaps with the peak of the pure MBDB.

The absorption spectra of the diethylamine solutions with the Simon’s reagent in the concentrationsseries of 0.1-1 mg/mL show peaks around 450 nm and 580 nm, as can be seen in Figure 33 on the nextpage. The peak around 285 nm overlaps with the peak of the pure MBDB and the diluted Simon’sreagent.

53


Figure 31:Absorption spectrum of 0.1-1 mg/mL MBDB dissolved in MilliQ and Scott reagent

and the spectra of the diluted Scott reagent and pure 1 mg/mL MBDB dissolved in MilliQ(Tecan Safire2).

Figure 32:Absorption spectrum of 0.1-10 mg/mL MBDB dissolved in MilliQ and Marquis reagentand the spectra of the Marquis reagent and pure 1 mg/mL MBDB dissolved in MilliQ

(Tecan Safire2).

54


Figure 33:Absorption spectrum of 0.1-1 mg/mL MBDB dissolved in MilliQ and Simon’s reagent

and the spectra of the diluted Simon’s reagent and pure 1 mg/mL MBDB dissolved in MilliQ(Tecan Safire2).

7.3.5 4-Methylamphetamine

The absorption spectra of the 4-methylamphetamine solutions with the Marquis reagent in the con-centrations series of 0.1-10 mg/mL show several peaks in the wavelenght region of 350-700 nm, as canbe seen in Figure 34 on the next page. However, these peaks are only visible at a concentration of 10mg/mL. The peaks visible at 1 mg/mL and 0.1 mg/mL, around 270 nm overlaps with the peak of thepure 4-methylamphetamine. Notable is the high absorbance of pure 4-methylamphetamine around270 nm.

7.3.6 4-Fluoroamphetamine

When 4-fluoroamphetamine is added to the Marquis reagent a reddish end product is obtained. How-ever, no peaks can be observed in the spectrum other than coming from the Marquis reagent or4-fluoroamphetamine itself, as can be seen in Figure 35 on the next page. Notable is the high ab-sorbance of pure 4-fluoroamphetamine around 270 nm.

7.4 Remarks colour tests and UV-Vis spectroscopy

7.4.1 Colour tests

Most of the reaction products have the colour as expected from the results of the NFI. It was notexpected that ephedrine would react with the Simon’s reagent as the structure contains an hydroxyl-group near the amine [11, 12]. It was expected that methylone, ethylone and butylone would give areaction with the Simon’s reagent, because these compounds are secondary amines [12]. Probably theamount of added analyte was not enough to achieve a colour reaction. There was only a small amount(10 mg) available for all the experiments, so the reaction could not be tested with a larger amount ofanalyte.

55


Figure 34:Absorption spectrum of 0.1-10 mg/mL 4-methylamphetamine dissolved in MilliQ and

Marquis reagent and the spectra of the Marquis reagent and pure 1 mg/mL 4-methylamphetaminedissolved in MilliQ (Tecan Safire2).

Figure 35:Absorption spectrum of 0.1-2.5 mg/mL 4-fluoroamphetamine dissolved in MilliQ and

Marquis reagent and the spectra of the Marquis reagent and pure 1 mg/mL 4-fluoroamphetaminedissolved in MilliQ (Tecan Safire2).

56


7.4.2 Operational drug detection limit

No ODDL could be acquired as only a small range of serial dilutions was used. However, with theTecan Safire2 still a characteristic peak can be observed in the spectrum at a concentration of 0.1mg/mL for most analytes. Therefore the LOD and ODDL for most analytes lies respectively below0.1 mg/mL (10 µg) and 1.0 mg/mL (100 µg).

An overview of the peaks in the UV-Vis spectra of the different analytes can be seen in Table 4.

Table 4:Overview of the peaks in the spectra of the different reagents and analytes.

Analyte Reagent Colour Peaks

- Scott Reddish 270 nm

- Marquis Transparent 310 nm

- Simon’s Reddish Below 650 nm elevated signal

Ephedrine Simon’s Bluish (dark) No peaks

DiethylamineScott Bluish/Greenish 270 nm & 470 nm & 590 nm & 640 nmSimon’s Purplish 400 (minimum) & 450 nm & 570 nm

Lidocaine Scott Bluish (light) 270 nm (lidocaine itself)

MBDBScott Bluish (light) 285 nm (MBDB itself)Marquis Purplish/Black 285 nm (MBDB itself) & 400 nmSimon’s Bluish (dark) 285 nm (MBDB itself) & 450 nm & 580 nm

4-Methylamphetamine Marquis Reddish 270 nm (4-methylamphetamine itself) & 350-700 nm several peaks

4-Fluoroamphetamine Marquis Reddish 270 nm (4-fluoroamphetamine itself)

For the analytes ephedrine and lidocaine with respectively the Simon’s reagent and the Scottreagent no peaks could be observed in the spectrum. A coloured end product can only be obtainedwith concentrations above 10 mg/mL. This also holds for MBDB with the Scott reagent and 4-fluoroamphetamine with the Marquis reagent.

Notable is the high absorption of lidocaine, MBDB, 4-methylamphetamine and 4-fluoroamphetaminebelow 300 nm.

When dilution is needed in order to obtain a reliable spectrum it is important to start with theundiluted reagent. In cases where the reagent is diluted prior to the addition to the analyte no directcolour change will occur, therefore the reagent and the analyte must first be added to each other whereafter dilution can take place.

7.4.3 Solvent

To prepare drug standards MilliQ is the most favourable solvent as it is easy to obtain, non-hazardousand non-toxic. Drawback of this solvent is that not all the legal highs dissolve in water and anextremely exothermic reaction takes place with the Marquis reagent.

Since the legal highs are relative new compounds within the field of forensic and analytical analysisnot a lot is known about the solubility of these compounds in different solvents.

57


Fig

ure

36:

Ab

sorp

tion

spec

tra

of1

mg/m

Lle

gal

hig

hs

dis

solv

edin

Mil

liQ

wit

hth

ed

iffer

ent

reagen

ts(T

ecan

Safi

re2).

58



The enormous spikes in the spectra at lower wavelengths can be explained by the high absorption ofthe analytes at these wavelengths. An UV-Vis spectrometer measures the transmission of a sampleand translates that to the absorption. This conversion from transmission to absorption is logarithmic,as can be read in Chapter 2. For a transmission of zero or nearly zero is the absorbance not welldefined, since the log of zero is not defined, with spikes in the spectrum as result [15, 17].

It is difficult to subtract the correct background signal, especially from the reagents, as the exactchemistry of the reactions is not known. Therefore it is not known if the reagent and analyte reactin a ratio of 1:1. It is also possible that the (coloured) reagent solution is not completely used inthe reaction. Additional difficulty is that the background spectra were taken from 200 µL reagentsolution. The absorption spectra were obtained with only 100 µL reagent solution and an additional100 µL analyte solution. Hence both spectra were not taken with an equal concentration of reagentsolution.

An explanation for the unexpected absorbance of the 10 mg/mL and the 5 mg/mL diethylaminesolution around 570 nm might be that there is not enough reagent present to react with the diethy-lamine. Which results in the same height of both peak in the spectrum around 570 nm.

Not for every analyte peaks could be observed in the spectrum. However, if the spectra of all theanalytes with at a concentrations of 1 mg/mL are compared with each other most spectra can bedistinguished from each other, as can be seen in Figure 36 on the previous page.

It is hard to make a quantitative statement as the exact absorption is very time dependent. Forexample MBDB with the Simon’s reagent becomes more intense in time, as can be seen in Figure 37.The second spectrum was recorded about 40 minutes later than the first spectrum.

Figure 37:Absorption spectra of 0.1-1 mg/mL MBDB dissolved in MilliQ and Simon’s reagents in time

(Tecan Safire2).

59


7.5 Chip C

The visibility of the coloured solutions in the chip, the mixing performance of each chip and UV-Visspectroscopy were performed and analysed on each chip. An overview of all the results can be seen inTable 5 on the next page.

7.5.1 Visibility

From all the coloured solutions only the premixed Simon’s reagent with the diethylamine solutioncould be observed by eye in all the chips, which can also be seen in Figure 38. In some of the chipsalso the premixed Marquis reagent with the acetylsalicylic acid solution could be seen, but only veryvaguely.

Figure 38:Left: Chip (C8) without any solution. Right: Chip (C5) with the premixed Simon’s reagent

with the diethylamine solution (photos made by Melvin van Melzen).

7.5.2 Mixing

Most chips, especially the more symmetric ones (C3-C6), show good mixing performance. The chipsthat are less symmetrical (C1-C2 and C7-C8) do not have a good flow profile; the solutions are notequally distributed through all the channels.


By placing the chips in the chipholder a spectrum of the light source could be obtained. The trans-mission of the Borofloat glass shows a cutoff below approximately 300 nm, as is expected [44, 45].However, no spectra could be obtained from the coloured reaction products in the chip.

7.6 Remarks detection on chip

7.6.1 Colour tests

Most coloured solutions could not be observed by eye in the chip; only very dark and intense colourcould be (vaguely) seen. Possibly the detection area or the total volume in the detection window istoo small to observe the colours. Another possibility is that the Borofloat glass layer on top of thedetection window is too thick, although the material is almost transparent in the visible region [44, 45].

60


Tab

le5:

Per

form

ance

ofth

ech

ips

C1-

C8.

Des

ign

Vis

ibil

ity

Mix

ing

UV

-Vis

Rem

ark

s

C1

Sim

on’s

+d

ieth

yla

min

e:Y

esM

ixin

gis

suffi

cien

t,N

osp

ectr

aco

uld

Th

eou

tlet

toIn

k:

No

bu

tth

eso

luti

ons

are

be

ob

tain

edap

ply

the

vacu

um

on

Sim

on’s

A:

No

not

equ

ally

dis

trib

ute

dw

as

too

small

Sco

tt+

die

thyla

min

e:N

oth

rou

ghth

ech

ann

els

on

this

chip

Marq

uis

+ac

etyls

ali

cyli

cac

id:

No

C2

Sim

on’s

+d

ieth

yla

min

e:Y

esM

ixin

gis

suffi

cien

t,N

osp

ectr

aco

uld

Th

eou

tlet

toIn

k:

No

bu

tth

eso

luti

ons

are

be

ob

tain

edap

ply

the

vacu

um

on

Sim

on’s

A:

No

not

equ

ally

dis

trib

ute

dw

as

too

small

Sco

tt+

die

thyla

min

e:N

oth

rou

ghth

ech

ann

els

on

this

chip

Marq

uis

+ac

etyls

ali

cyli

cac

id:

No

C3

Sim

on’s

+d

ieth

yla

min

e:Y

esG

ood

mix

ing

No

spec

tra

cou

ldT

he

inle

tsat

Ink:

No

per

form

ance

be

ob

tain

edth

em

idd

leof

Sim

on’s

A:

No

the

chip

are

Sco

tt+

die

thyla

min

e:N

oto

ocl

ose

toM

arq

uis

+ac

etyls

ali

cyli

cac

id:

Vagu

ely

each

oth

er

C4

Sim

on’s

+d

ieth

yla

min

e:Y

esG

ood

mix

ing

No

spec

tra

cou

ldT

he

inle

tsat

Ink:

No

per

form

ance

be

ob

tain

edth

em

idd

leof

Sim

on’s

A:

No

the

chip

are

Sco

tt+

die

thyla

min

e:N

oto

ocl

ose

toM

arq

uis

+ac

etyls

ali

cyli

cac

id:

No

each

oth

er

61


Tab

le5:

Per

form

ance

ofth

ech

ips

C1-

C8

–co

nti

nu

ed.

Des

ign

Vis

ibil

ity

Mix

ing

UV

-Vis

Rem

ark

s

C5

Sim

on’s

+d

ieth

yla

min

e:Y

esG

ood

mix

ing

No

spec

tra

cou

ldT

he

inle

tsat

Ink:

No

per

form

ance

be

ob

tain

edth

em

idd

leof

Sim

on’s

A:

No

the

chip

are

Sco

tt+

die

thyla

min

e:N

oto

ocl

ose

toM

arq

uis

+ac

etyls

ali

cyli

cac

id:

No

each

oth

er

C6

Sim

on’s

+d

ieth

yla

min

e:Y

esG

ood

mix

ing

No

spec

tra

cou

ldT

he

inle

tsat

Ink:

No

per

form

ance

be

ob

tain

edth

em

idd

leof

Sim

on’s

A:

No

the

chip

are

Sco

tt+

die

thyla

min

e:N

oto

ocl

ose

toM

arq

uis

+ac

etyls

ali

cyli

cac

id:

Vagu

ely

each

oth

er

C7

Sim

on’s

+d

ieth

yla

min

e:Y

esM

ixin

gis

suffi

cien

t,N

osp

ectr

aco

uld

Ink:

No

bu

tth

eso

luti

ons

are

be

ob

tain

edS

imon

’sA

:N

on

oteq

ual

lyd

istr

ibu

ted

Sco

tt+

die

thyla

min

e:N

oth

rou

ghth

ech

ann

els

Marq

uis

+ac

etyls

ali

cyli

cac

id:

No

C8

Sim

on’s

+d

ieth

yla

min

e:Y

esM

ixin

gis

suffi

cien

t,N

osp

ectr

aco

uld

Th

eou

tlet

toIn

k:

No

bu

tth

eso

luti

ons

are

be

ob

tain

edap

ply

the

vacu

um

on

Sim

on’s

A:

No

not

equ

ally

dis

trib

ute

dw

as

too

big

Sco

tt+

die

thyla

min

e:N

oth

rou

ghth

ech

ann

els

on

this

chip

Marq

uis

+ac

etyls

ali

cyli

cac

id:

Vagu

ely

62



With the Scott reagent as well as with the Simon’s reagent the reaction products have very intensecolours. To acquire a proper spectrum with UV-Vis spectrscopy the end product must be diluted,which is not possible with the current designs.

Since it was not possible to acquire a spectrum of the solutions in the chip, there is no method ofchecking if the reaction went well. It needs to be checked if there is no exchange of solutions betweenthe channels, which might give false positive or false negative results. In principle it is possible toanalyse the end products at the outlets, but the chips have only one outlet (to apply the vacuum on).

7.6.3 Design of the chip

The outlets of some of the chips, where the vacuum can be applied on, was too big or too small. Aradius of about 1000 µm is most desirable in combination with the used vacuum pump.

It is favourable when the inlet for the analyte solution is somewhat bigger than the inlets for thereagents. The analyte solution has to go through four channels, so a bigger injection volume is needed.The inlets must also be far enough from each other to prevent contamination, which is not the casefor chips C3-C6.

One detection window in the chip is enough, since it is not possible to observe the colour visually.The window should not be too big as it would be too hard to fill the window properly with thesolution that must be analysed. On the other hand, the window must be broad enough for the fibers;the window must be at least as broad as the core radius of the used fibers. The core radius of theused fibers for these experiments was 300 µm. However, a detection with a radius of about 1000 µmwould be desirable, because the chip can then be easily aligned in the chipholder.

The chips that have a symmetrical design, C3-C6, show a better flow profile of the solutions.The solutions flow more equally though the channels and therefore less change of unwanted mixingof solutions between channels. With the non-symmetrical designs not all channels have the sameflow-resistance.

One meander is enough to obtain good mixing performance of the chip. The integration of a mixeris not really necessary, although it slows down the flow, which makes the introduction of the solutioninto the chip by the use of the vacuum pump easier and more controlled.

The final conclusion of these experiments is that it was not possible to obtain a spectrum of asolution in the chip, because the optical path length within the chip appeared to be too small.

63

Brigitte Bruijns Illicit drugs analysis on chip Conclusion

Conclusion

The aim of this research project was to develop a chip for the analysis of drugs of abuse. The first stepto realise this was the analysis of the conventional colour tests with drug analogues and legal highs.Because UV-Vis spectroscopy will be integrated in the on chip analysis, the second step was to obtainthe UV-Vis spectra of all the analytes and the reagents that give a colour reaction. The third and laststep was the design of a chip to carry out the three colour tests simultaneously with the integrationof UV-Vis spectroscopy.

All the reaction products of the drug analogues and the reagents were as expected. The ODDL ofthe drug analogues lies around 40 mg/mL, which corresponds to a total amount of 20 mg of analyte.Also all the reaction product of the legal highs and the reagents were as expected, although for someof the reaction the amount of added analyte was too low to observe a colour change. No ODDL of thelegal highs was determines as only a small amount of these analytes was available.

The drug analogues showed very different spectra from each other; something that was expectedbased on the visual observed colour. Also most legal highs can be distinguished from each other basedon their UV-Vis spectrum. The ODDL of all the analytes with UV-Vis spectrometry lies below 1mg/mL, which corresponds to a total amount of less than 100 µg of analyte.

With chips (chip A and chip B) that were available in the MCS group it was demonstrated thatonly small volumes are needed to obtain a colour change and that a small meandering channel showsa good mixing performance. However, it was not possible to obtain UV-Vis spectra from solutions onchip (chip C), due to the small optical path length.

This research showed that by combining the presumptive colour tests and UV-Vis spectra moreinformation about the identity can be obtained of the suspected analyte, especially when good referencespectra of the illicit drugs are available. Concerning analysis on chip there are still a few challengesto overcome. The first hurdle is the solubility of the drugs of abuse. Currently, the colour tests areperformed on the pure powders, but for microfluidic analysis on chip the compounds must be dissolved.MilliQ would be favourable, but not all compounds dissolve in water and for some compounds thesolubility characteristics are not even known. Another challenge is the optical path length in the chip.By miniturazation less chemicals (reagents, analytes, but also waste) are needed. However, in orderto obtain a proper spectrum a long enough path length is required. The high absorbance, due to highanalyte concentrations or to intense colouration, is another problem that must be solved, although itmight be favourable for on chip analysis as less optical path length is required.

64

Brigitte Bruijns Illicit drugs analysis on chip Recommendations

Recommendations

The chips made from Borofloat are not transparent over the whole UV region, but almost all illicitdrugs have an absorption at 200 nm [44, 45, 46]. It is possible to make the chip of silicon nitride,however the limit of the used spectrometers is already somewhat above 200 nm [47, 48].

The drawback of UV-Vis spectroscopy on chip is the limited sensitivity due to the short pathlength. Laser induced fluorescence (LIF) is a good alternative as the sensitivity is several orders ofmagnitude greater than UV-Vis detection. Most drugs are not fluorescent, which is required for LIF,but the amine moieties in most illicit substances can be derivatised with reactive fluorescent dyes [49].However, such a derivatization would not be selective for illicit drugs and will give false positives withother amines.

Another option is to increase the path length by the use of a Z-cell [50, 51]. These cells, as canbe seen in Figure 39, have been made in a chip for capillary electrophoresis (CE) from polydimethyl-siloxane (PDMS) at the UT by Advanced Technology students. With this design they could increasethe optical path length from 50 µm to 500 µm [52].

Figure 39:The design of the Z-cell with the yellow line representing the pathway of the UV light [52].

As is discussed in Chapter 2 identification of an unknown illicit drug is only possible when severaldifferent analytical techniques are used with confirmatory results. Colour tests are a presumptive testas they are only selective and not specific. Therefore they must be used in combination with at leastone spectroscopic or two chromatographic techniques to make a statement about the identity of thedrug [3].

Fourier Transform Infrared (FTIR) spectroscopy is a commonly used technique at the NFI for thestructural analysis of an unknown drug compound [4, 53]. FTIR gives besides a qualitative result alsoa quantitative result and is popular due to the speed of analysis. Direct investigation of the suspectedliquid or solid, without any sample preparation, is possible with attenuated total reflectance (ATR)IR [12]. Chan et al. managed to obtain a chemical image of compounds within a microfluidic channelin a PDMS chip [54]. Microfluidic channels with ATR-FTIR are also used for the monitoring ofconcentrations of solutes and to monitor the kinetics of a reaction [55, 56, 57]. The MCS group hasdeveloped an ATR-FTIR chip, which may be used for this purpose.

CE is an analytical technique that proved to be very useful for the analysis of samples on chip [18,21, 58]. This technique can also be used to analyse drugs of abuse on chip [46, 49]. McCord et al.developed a method to analyse drugs of abuse by performing fluorescent derivatization of a sample incombination with LIF detection. Eosin isothiocyanate was used as derivatising agent for amphetamine,methamphetamine, ephedrine and ketamine. By the use of a microfluidic CE device they could sepa-rate and detect levels of about 20 ng/mL of phenethyl amines (except ketamine) on chip in less than1.5 minutes. With this method it is possible to detect trace amounts of phenethyl amines in bodyfluids such as urine. McCord et al. also separated nitro benzodiazepines, but the detection limitsobtained with indirect LIF were not sufficient for toxicological examinations.

65

Brigitte Bruijns Illicit drugs analysis on chip Recommendations

Overall conclusion of their research is that by coupling microfluidic detection with fluorescent tech-niques drugs of abuse can be analysed rapidly and with high specificity, although for the use in courtmore definitive techniques are required [49].

Also MS, often used in combination with GC, is a commonly used technique in the forensic worldfor the identification of illicit drugs [3, 12]. By using direct electrospray probe (DEP)/MS it is possibleto detect benzodiazepines in drinks [59]. Deng et al. described a chip-based CE/MS system for thedetermination of drugs in human plasma [60]. It is possible to make micro mass spectrometers, butthis research did not pass the stage of simulations [61]. Already developed mini-spectrometers have atoo limited mass resolution [62, 63]. Another option is mass analysis from chip as is described by vonEggeling et al. for the ProteinChip Array [64]. It is also possible to integrate porous silicon spots in acontinuous flow microreactor in order to perform matrix-free mass spectrometry (on a lab scale MS),also known as desorption/ionization on silicon (DIOS) [25].

UV-Vis spectroscopy is often used for quantitative measurements. However, quantitative researchof drugs of abuse is seldom necessary. In the Dutch Opium Law no statements are made concerningthe amount of drugs; every amount is prosecutable. The trade volume of 1 kg pure amphetamine isthe same as for 1 kg with only 10% amphetamine [3, 4].

The determination of drugs of abuse in body fluids, such as urine, blood and saliva, is not discussed,since that is part of forensic toxicology and not of illicit drugs.

66

Brigitte Bruijns Illicit drugs analysis on chip Epilogue

Epilogue

By doing this research project I have definitely improved my knowledge about all kinds of facets ofdrugs analysis, microfluidics and the combination of both. The literature survey has given me a goodinsight into the current state of the use of microdevices in forensic investigations. The fact that thereare not a lot of microfluidic systems available for forensic research makes it very interesting, but alsohard to develop an own device.

A lot of communication, discussions and knowledge transfer had to be carried out before thepractical part of the research could be started. Drug analogues had to be used in order to avoidprocedures for obtaining the required licenses to perform research on real illicit drugs. And of coursethere were some setbacks: a long delivering time of the chemicals, no small gloves for my small hands,people that also want to use the spectrometer, and more. But that is also part of doing research.

I have learned a lot from organising the symposium CSI: The Future. It was great to have suchenthusiastic speakers and the positive reactions from the participants. It gave a good insight in whatis going on in the forensic field regarding to lab-on-a-chip devices.

I would like to thank my (daily) supervisors, Han Gardeniers and Wojciech Bula, for all theirhelp, support and good suggestions during my research project. I would also like to say thank you toall the (ex-)members of the Mesoscale Chemical Systems Group at the University of Twente for alltheir help and of course their social support, especially Stefan Schlautmann for the fabrication of mychips. I also want to thank the people from the Netherlands Forensic Institute, especially Arian vanAsten and Jorrit van den Berg, for all the information and their sharing of knowledge. And I mustnot forget to thank Dirk Roelof Dekker from the Biophysical Engineering group for all his help andadvice concerning UV-Vis spectroscopy.

I really enjoyed this research project and the great opportunity to combine the three discipline offorensic science, analytical chemistry and lab-on-a-chip technology in one project.

Overview of visited conferences and symposiaName Date Place Details

The Analtyical Chal-lenge (KNCV)

2 November 2010 De Werelt - Lunteren

MicroNanoConference’10 (MESA+)

17 November 2010 University Twente -Enschede

”Zaalwacht”

What the eyes don’t see(Forensic Science)

18 November 2010 University of Ams-terdam - Amsterdam

Poster presentation:Enantioseparation ofdrugs by capillaryelectrophoresis

ASML excursion 25 November 2010 ASML - Veldhoven

CSI: The Future 10 December 2010 University Twente -Enschede

Organiser

Analyse en interpretatievan materiaal bewijs(TMFI)

11 Febrary 2011 DSM Resolve -Geleen

Curious (PAC sympo-sium)

3 March 2011 University Utrecht -Utrecht

Poster presentation:Illicit drugs analysison chip (Appendix I)

67

Appendices

Brigitte Bruijns Illicit drugs analysis on chip Appendix A

A Project description

Forensic application of nanotechnology

J.G.E. Gardeniers – UTP.J. Schoenmakers – UvAA.C. van Asten, A. Kloosterman, O. Schuring – NFI

IntroductionThrough the MESA+ Institute for Nanotechnology of the Technical University of Twente, the

Netherlands has made significant contributions to the breakthrough developments in the last decadein nanotechnology, microfluidics and ”lab-on-a-chip” systems. Currently, the institute employs 500specialists who work in over 1250 m2 of clean room space with state of the art nanotechnology equip-ment. This has resulted in over 40 high-tech commercial start-ups. Interestingly, up till now the useof the ”lab-on-a-chip” technology for forensic applications has not been explored. Given the impor-tance of crime scene investigation, the need for fast results (preferably in real time) to aid the tacticalinvestigation, the existence of often crucial trace evidence and the very strict quality requirements,the application of nanotechnology in forensic science seems very promising. In this project for the firsttime a MESA+ – NFI cooperation will be established to explore the forensic potential of nanotech-nology developed at Twente.

Forensic potentialIn several forensic expertise areas there is a need and a potential to develop mobile crime scene

methods for sampling, selecting, screening and analysing evidence. In general it can be stated thatwith the increased capability and quality within forensic laboratories there is also a strong need tobring science, innovation and quality control to the crime scene. Additionally, indicative chemicaland biochemical tests are still used frequently in the forensic laboratory environment. It is in theseforensic areas that ”lab-on-a-chip” technology could successfully be employed to bring new featuresopportunities.

Microfluidic and ”lab-on-a-chip” technology could offer new benefits with regard to:1. Quality and consistency of sampling (controlled sampling with minimal handling, degradation andcontamination)2. Miniaturization of sampling, screening and analysis (e.g. microtraces)3. Minimal handling and modification of evidence (sampling, screening and analysis require minimalamounts of material)4. Real time analysis on the crime scene for evidence screening and selection and for generation ofreal-time information for the tactical investigation5. Multi compound/functionality screening in a single test/device6. Reduced costs for chemical testing and sampling through cheap single-use kits and minimal use ofmaterials7. Ease of use also for non-specialists (e.g. police officers)8. Safety (no need to manual handling of chemicals)9. Automated interfacing with advanced analytical equipment in the forensic laboratory (e.g. massspectrometric techniques)

70


In a quick scan the following NFI expertise areas are identified that could benefit from nanotech-nology:1. Human Biological Traces- MiDAS project : DNA profiling in real time on the crime scene- Indicative testing into the origin of a human biological trace (saliva, blood, sperm, urine, faeces)2. Non-Human Biological Traces- Indicative testing : human or non-human traces?- Indicative testing into the origin of a non-human biological trace3. Mobile Forensic Team- Sampling and screening of traces and microtraces- Detection of blood traces- Sampling of DNA from finger prints4. Pathology and Toxicology- Sampling in and on the human body- Indicative testing for chemical substances in blood, urine and body fluids and tissues- Sensor technology to estimate time of death5. Gun shot residues- Sampling for GSR and SEM analysis- Indicative testing for GSR6. Illicit Drugs- Sampling of chemicals- Indicative testing for drugs and precursors7. Explosions and Explosives- Sampling of chemicals- Indicative testing for explosives and precursors (e.g. DROPEX kit)- Indicative testing for composition of charges in fireworks- Non-invasive or minimal invasive methods for safe sampling and detection8. Forensic Engineering- Gas sensors (CO, NOx, CO2, O2, NH3)- Environmental sampling, screening and analysis- Physical and chemical properties of industrial waste- Tracer technology : tracking waste streams through tagging and fast and selective detection on site9. CBRNE- Detection and identification of C agents on a terrorist crime scene- Detection and identification of B agents on a terrorist crime scene

Available microfluidic technologyBefore designing and developing tailor made microfluidic systems for forensic use it is sensible to

demonstrate the proof of principle by developing forensic applications from existing systems. Withinthe MESA+ institute a lot of knowledge and expertise has been generated on the use of microfluidicsystems for sampling, pretreatment, clean up and separation and chemical analysis. ”Lab-on-a-chip”methods have for instance been used for clinical testing, sample collection and preparation, and sensorapplications. Quite likely this methodology can already be used forensically by translating traditionalsampling and testing conditions to the microfluidic environment.

The following recent ”Lab-on-a-chip” developments are of specific forensic interest:- Micro needle systems for transdermal sampling- Electrowetting : droplet mixing and handling- Use of electric fields for sample pretreatment, fractionation and collection- Sample splitting through the use of parallel micro channels- Micromachined reactors and measurement of reaction kinetics

71


- Micromachined columns and micro LC systems- Sample pretreatment for MALDI-MS and MALDI imaging- Preconcentration of biofluids and use of micro probes for NMR analysis- In line proces monitoring of fermentation processes with chip based sensors- Catalytic gas sensors with tailor made selectivity- Commercial clinical applications (e.g. ion analysis in blood)- Commercial sensor applications (e.g. detection of C agents)

Project plan and goals1. Literature survey on the use of nanotechnology for forensic investigations2. Broad scoping exercise on the use of mobile sampling and analytical methods and indicative chem-ical and biomedical testing within the various forensic expertise areas3. Mapping suitable and available ”lab-on-a-chip” technology that could be used to develop forensicsampling, analytical and indicative testing methods4. Proof of principle: development of 1-2 forensic ”lab-on-a-chip” methods from available nanotech-nology and existing forensic methodology

72

Brigitte Bruijns Illicit drugs analysis on chip Appendix B

B Drugs of abuse

Several illicit drugs, their appearance and the cutting agents [3, 6, 14].Drug Appearance Cutting agents Remarks

NarcoticsSedative and strongly euphoric (a high).

Physical and psychological dependence and dosage needs to be increased due to tolerance.

Methadone Synthetic opioid first re-ported as a maintenancetreatment for opiate de-pendence

Morphine Morphine can be adminis-tered by intravenous, sub-cutaneous, oral, and rectalroutes

Morphine base is the im-mediate precursor of heroin

Heroin Light beige to dark brownpowder, in high puritycream coloured

Caffeine and/or paraceta-mol

Mainly produced in South-west America (Mexico andColumbia) , SoutheastAsia (Vietnam, Laos andThailand) and SouthwestAsia (Pakistan, Iran andAfghanistan)

StimulantsIncrease the energy and alertness.

Psychological dependence and dosage needs to be increased due to tolerance.

Amphetamine White to light brown pow-der or tablets

Lactose and caffeine

Cocaine White to cream colouredpowder

Mannitol, lidocaine andphenacetin

Extracted from leaves ofthe coca plant

Methamphetamine Pill, powder or crystals(Ice)

73

Brigitte Bruijns Illicit drugs analysis on chip Appendix B

Several illicit drugs, their appearance and the cutting agents – continued [3, 6, 14].Drug Appearance Cutting agents Remarks

HallucinogensDistorted perception and sensory illusions (out of the body).

Hardly addictive.

LSD Blotter papers Briefly used in psychother-apy

MDA Tablets (broad variance ofappearance)

Relative pure or mixedwith other active sub-stances (caffeine, am-phetamine)

MDMA Tablets (broad variance ofappearance) and crystalsin capsules

Relative pure or mixedwith other active sub-stances (caffeine, am-phetamine)

74

Brigitte Bruijns Illicit drugs analysis on chip Appendix C

C Colour tests

Colours produced by the Marquis, Simon’s and Scott test after five minutes with variousillicit drugs and other substances (within brackets the Munsell colour) and the operational

drug detection limits with CHCl3 as solvent [11].Scott Marquis Simon’s

(d)-AmphetamineStrong reddish orange(10R 6/12) – Dark red-dish brown (7.5R 2/4) -10 µg

Acetylsalicylic acidDeep red (5R 3/10)

Cocaine HClStrong greenish blue (5B5/10) - 60 µg

DiethylamineDark blue (2.5PB 2/6)

Ephedrine HClStrong greenish blue (5B5/10)

LSDOlive black (10Y 2/2) - 5µg

MDA HCl Black (black)

MDMA HClDark blue (2.5PB 2/6)

Methadone HClBrilliant greenish blue(5B 6/10) - 250 µg

(d)-Methamphetamine HClDeep reddish orange(10R 4/12) – Darkreddish brown (7.5R2/4)

Dark blue (2.5PB 2/6) -10 µg

Morphine monohydrateVery deep reddish purple(10P 3/6) - 5 µg

OpiumDark greyish reddishbrown (10R 3/2)

Pseudoephedrine HClStrong greenish blue (5B5/10)

SugarDark brown (5YR 2/4)

75

Brigitte Bruijns Illicit drugs analysis on chip Appendix D

D Detection limits by eye

D.1 Scott test

The observed LOD by eye is not determined for the ephedrine solution, since no colour change takesplace.

D.2 Marquis test

Left to right: Colour observed by addition of 500 µL of Marquis reagent to 5.0 mg/mL, 4.0 mg/mL,3.0 mg/mL, 2.0 mg/mL and 1.0 mg/mL acetylsalicylic acid solution dissolved in chloroform.

Left to right: Colour observed by addition of 500 µL of Marquis reagent to 5.0 mg/mL, 4.0 mg/mL,3.0 mg/mL, 2.0 mg/mL and 1.0 mg/mL sugar solution in dissolved water.

D.3 Simon’s test

Left to right: Colour observed by addition of 500 µL of Simon’s reagent(A and B with a ratio of 1:2) to 5.0 mg/mL, 4.0 mg/mL, 3.0 mg/mL, 2.0 mg/mL

and 1.0 mg/mL diethylamine solution dissolved in chloroform*.

76

Brigitte Bruijns Illicit drugs analysis on chip Appendix D

Left to right: Colour observed by addition of 500 µL of Simon’s reagent(A and B with a ratio of 1:2) to 5.0 mg/mL, 4.0 mg/mL, 3.0 mg/mL, 2.0 mg/mL

and 1.0 mg/mL diethylamine solution dissolved in water*.

* Only when a dark purple reaction product appears this is encountered as a positive reaction,although at lower concentrations still a colour change can be observed

77

Brigitte Bruijns Illicit drugs analysis on chip Appendix E

E Detection limits by UV-Vis spectroscopy

E.1 Scott test

The observed LOD by UV-Vis spectroscopy is not determined for the ephedrine solution, since nocolour change takes place.

E.2 Marquis test

Absorption spectrum of 2.00 mg/mL (green), 1.75 mg/mL (blue), 1.50 mg/mL (purple),1.25 mg/mL (red) and 1.00 mg/mL (yellow) acetylsalicylic acid solution dissolved in

chloroform and Marquis reagent.

Absorption spectrum of 1.00 mg/mL (orange), 0.50 mg/mL (red), 0.25 mg/mL (blue) and0.13 mg/mL (green) sugar solution dissolved in MilliQ and Marquis reagent.

E.3 Simon’s test

The observed LOD by UV-Vis spectroscopy is not determined for the diethylamine solutions, sincethe colour of the end product is too intense.

78

Brigitte Bruijns Illicit drugs analysis on chip Appendix F

F Designer drugs

The molecular structure of several designer drugs.

Amphetamine Methamphetamine 4-Fluoroamphetamine

Mephedrone

MDA MDMA MDEA

MBDB EDMA MDAI

Methylone Ethylone Buthylone

DOB DOM 2C-B

Bromodragonfly

79

Brigitte Bruijns Illicit drugs analysis on chip Appendix G

G Chip designs

Designs and features of the different chips.C1

Channel width: 100 µmNumber of inlets: 3 (reagents) + 1 (analyte)Number of detection windows: 2Radius detection window: 500 µm & 1000 µmNumber of meanders: 1Mixer: No mixer

C2

Channel width: 100 µmNumber of inlets: 3 (reagents) + 1 (analyte)Number of detection windows: 1Radius detection window: 1500 µmNumber of meanders: 1Mixer: No mixer

C3

Channel width: 100 µmNumber of inlets: 4 (reagents) + 1 (analyte)Number of detection windows: 2Radius detection window: 600 µm & 1000 µmNumber of meanders: 1Mixer: No mixer

C4

Channel width: 100 µmNumber of inlets: 4 (reagents) + 1 (analyte)Number of detection windows: 2Radius detection window: 400 µm & 750 µmNumber of meanders: 1Mixer: Tesla mixer

80

Brigitte Bruijns Illicit drugs analysis on chip Appendix G

Designs and features of the different chips – continued.C5


C6


C7


C8


81

Brigitte Bruijns Illicit drugs analysis on chip Appendix H

H Chip fabrication

H.1 Process parameters

Process parameters for a glass wafer [65].Step Process Comment

1 Substrate selection - CR112B ”Top”Borofloat Glass Supplier: Schott Glas wafer:(#subs007) www.schott.com/borofloat 1.1 mm

Type: Borofloat 33 ”Bottom”C.T.E.: 3.25 X 10−6K−1 wafer:Tglass: 525 ◦C 0.5 mmTanneal: 560 ◦CTsoftening: 820 ◦CDiameter: 100.0 mm ± 0.5 mmThickness: 1.1 mm & 0.5 mmEtch rate HF 25%: 1 µm/minEtch rate BHF (1:7): 20-25 nm/minEtchrate HF 1%: 8.6 nm/minSmooth side: second flat on the left side

2 Cleaning ultrasonic - CR116B/Wet-Bench 63 Time: 15 minIPA IPA VLSI(#clean014) time > 15 min

Spin drying3 Cleaning glass CR112B/Wet-Bench 133

(#clean005) HNO3 (100%)Selectipur: MERCKOnly use the dedicated wafer carriers and rod!Beaker 1: HNO3 (100%) 5 minBeaker 2: HNO3 (100%) 5 minQuick Dump Rinse < 0.1 µSSpin drying

4 Sputtering of Cr Eq.Nr. 37/Sputterke 10 nm, 1 min(Sputterke) Cr Target (gun #: see mis logbook)(#depo017) Electrode temp.: water cooled electrode

Ar flow: app. 80 sccm, pressure depending!Base pressure: < 1.0 e-6mbarSputter pressure: 6.6 e-3mbarPower: 200WDepositionrate = 15 nm/min

5 Sputtering of Au Eq.Nr. 37/Sputterke 200 nm, 4 min(Sputterke) Au Target (gun #: see mis logbook)(#depo036) Electrode temp.: water cooled electrode

Ar flow: app. 80 sccm, pressure depending!Base pressure: < 1.0 e-6mbarSputter pressure: 6.6 e-3mbarPower: 200WDepositionrate = 45-50 nm/minMAX THICKNESS: 250 NM

82


Process parameters for a glass wafer – continued [65].Step Process Comment

6 Lithography - CR112B/Suss Micro Tech Spinner (Delta 20)Olin 907-17 Hotplate 120 ◦C:(#lith057) Dehydration bake (120 ◦C): 5min

HexaMethylDiSilazane (HMDS):Spin program: 4 (4000 rpm, 20 sec)Olin 907-17:Spin program: 4 (4000 rpm, 20 sec)Hotplate 95 ◦C:Prebake (95 ◦C): 90 s

CR117B/EVG 620Electronic Vision Group 620 Mask Aligner:Hg-lamp: 12 mW/cm2

Exposure Time: 4 sec

CR112B/Wet-Bench 11Hotplate 120 ◦C (CR112B or CR117B):After Exposure Bake (120 ◦C): 60 secDeveloper OPD4262:Time: 30 sec in Beaker 1Time: 15-30 sec in Beaker 2Quick Dump Rinse < 0.1 µSSpin drying

7 Lithography - CR112B / Hotplate 120 ◦C Time: 5 minPostbake standard Time: 30 min(#lith009)

8 Etching of Au Wet - CR116B/Wet-Bench 2 Time: 1-2 minlow underetch KI (pa): MERCK 105043(#etch040) I2: MERCK 144761

KI:I2:DI = (4:1:400)Glycerine:Add 43 g KI, 4.5 g I2 and 150 ml glycerine to 300 mlDI waterTemp.: 55 ◦CQuick Dump Rinse < 0.1 µSSpin dryingEtchrates = 180 nm/minGycerine is added to prevent excessive underetch

9 Etching of Cr Wet CR116B/Wet-Bench 2 Time: 10 s(#etch034) Chromium etch LSI Selectipur: MERCK 111547.2500

Quick Dump Rinse < 0.1 µSSpin dryingEtch rates = 100 nm/min

83



10 Etching HF of Glass CR116B/Wet-Bench 2 25%(#etch031) HF (50%) VLSI: MERCK 100373.2500 Time: 15 min

Add HF (50%) to DI for diluted etch solutions 22 µmTemp.: 20 ◦CUse stirrerQuick Dump Rinse < 0.1 µSSpin dryingEtch rates (function of load):HF 5% = 0.1 µm/minHF 10% = 0.14 µm/minHF 30% = 1.9 µm/minHF 50% = 7.3 µm/min

11 Lithography - CR 112B/GBC 3500 PRO Laminator Non-bond sideLamination of BF 410 Ordyl BF 410 dry resist foilfoil(#lith045) Laminate BF 410 foil on both sides

Protect hotplate with Aluminium foilPut wafer on hotplate, 100 ◦C, 180 secRemove thick PET layer from BF 410 foilApply BF 410 foil with rollerProtect carry-paper with plain A4 paperClose carrier and laminateTemp: 130 ◦C (carry preset)Speed: 2 (carry preset)Remove and cool down waferCut the wafer out of foil

12 Lithography - CR 117B/EVG 20Alignment and Electronic Vision Group 20 Mask AlignerExposure BF 410(#litho044) Hg-lamp: 12 W·cm2

Exposure time: 20 sec (BF 410)

Remark: DSP alignment with foil on both sidesRemove the foil with a ”knife” to achieve a clear viewof the aligning marksAfter development proctect the aligning mask withtape again!

13 Lithography - ELTN7143/4/HCM Spray DeveloperDevelopment BF410 Na2CO3: MERCK 1.06392.0500foil Na2CO3:H2O = 15 g : 7.5 liters (+ 1 cup Antifoam)(#lith036) Temp: 32 ◦C

Time: 3 minRinsingSpin dryingDue to non-uniform development turn sample by 180◦

after half the timeSmall features might need longer development time

84



14 Powderblasting of ELTN10156/Powderblaster Through waferglass - low resolution For feature size > 100 µm(#etch020) Particles: 30 µm Al2O3

Pressure: 4.6 barMassflow: 3-12 g/minEtchrate appr. 91 µm per g/cm2

15 Foil removal CR116B Manually re-move bothfoils

16 Cleaning Ultrasonic - CR116B/Wet-Bench 63 Both wafersDI Removal of scribing particles(#clean001) DI water > 10 min

Quick Dump Rinse < 0.1 µSSpin drying

17 Cleaning glass CR112B/Wet-Bench 133 Both wafers(#clean005) Time: 15min

HNO3 (100%) Selectipur: MERCKOnly use the dedicated wafer carriers and rod!Beaker 1: HNO3 (100%) 5 minBeaker 2: HNO3 (100%) 5 minQuick Dump Rinse < 0.1 µSSpin drying

18 Borofloat BF33 CR116B/Wet Bench 2 Both wafersKOH-dip Application: Pretreatment for borofloat-wafer bond-

ingTime: 1min

KOH: MERCK 105019.500KOH:DI = (1:3)25wt% KOH: 500 g KOH pellets in 1500 mL DI waterTemp.: 75 ◦CStirrerKOH-dip of 60 secQuick Dump Rinse < 0.1 µSSpin dryingContinue immediately with bonding process

19 Manually aligning & Contact wafers manuallyPrebonding Apply light pressure with tweezers(#bond001) If necessary use tweezers to press out air-bubbles

Check prebonding by using IR-setup20 Bonding of Borofloat HO10156/BIOS Oven Material

@ 600 ◦C Type: Schott Borofloat 33(#anne017) C.T.E.: 3.25 X 10−6 K−1

Tglass: 525 ◦CTanneal: 560 ◦CTsoftening: 820 ◦CAnneal Program:upramp: 25-600 ◦C in 6 hrsanneal: 600 ◦C for 1 hrdownramp: 600-25 ◦C in 6 hrs

85



21 Dicing of a glass CR128C/Loadpoint dicing saw Dice 1.1 mmwafer Parameters dicing: from one side

Wafer work size: Then turn wafer110 mm for a standard 100 mm silicon wafer and dice 0.5 mmFeed speed: 4 mm/sec (for 1000 µm cut depth) from other sideX, Y values: correspond respectively to Ch1 and Ch2and those values are determined by mask layoutSaw type TC300

H.2 Process summary

”Top” glass wafer processing [65].Cross-section Process summery

Borofloat BF33 (or D263) glass wafer

Cr/Au sputtering on the bond side

Lithography and Au/Cr etching

Isotropic etching of glass in HF to create the microfluidic structures

Lithography and micro-powderblasting to create access holes on thenon-bond side

Stripping of resist and Cr/Au

Thermal bonding of the processed wafer to the capping wafer

86

Brigitte Bruijns Illicit drugs analysis on chip Appendix I

I PAC symposium

I.1 Poster

Poster for the poster presentation at the PAC symposium 2011 (Utrecht).

87


I.2 Abstract poster

Illicit drugs analysis on chipThe use of lab-on-a-chip technology for a forensic application

Brigitte Bruijnsa, Han Gardeniersb, Wojciech Bulab, Arian van Astenc, Jorrit van den Bergc, WimKokd

aUniversity of Amsterdam/VU University Amsterdam∗bMesoscale Chemical Systems, University of TwentecNetherlands Forensic InstitutedVan ’t Hoff Institute for Molecular Sciences, University of Amsterdam∗ The research is carried out at the University of Twente in cooperation with the Netherlands ForensicInstitute

The number and variety of forensic traces found at a crime scene is enormous. The term forensicscience is therefore very broad and can be divided in several expertise areas, such as DNA profiling,blood spatter analysis, explosives and illicit drugs. It is a clear desire of forensic investigators thatanalyses should be simple, fast, robust, cheap and have high sensitivity and selectivity. Devices withthese specifications that can be used directly at the crime scene are especially useful as they canprovide immediate information to the police investigators. Most ideal would be a mobile forensic labfor collecting, screening and analysis of the evidence. So-called ”lab-on-a-chip” (LOC) systems canspeed up the analysis, are compact, can easily be integrated, limit the risk of contamination and canbe used by people who are not technically trained.

However, micro-devices for forensic investigation hardly exist[1]. Experts in LOC technologyand/or nanotechnology do not have experience and knowledge about forensic science. On the otherhand, forensic experts are in general not familiar with LOC devices. The two disciplines have not yetbeen combined in order to obtain an LOC device for forensic research.

In this research three widely used presumptive colour tests, which can indicate the presence ofdrugs of abuse, are integrated in an LOC device. By adding a few drops of the Scott, Marquis andSimon’s reagent to the suspected analyte the presence of illicit drugs (such as amphetamines andopiates) can be indicated. These conventional tests are easy to perform, but the used chemicalsare toxic, hazardous and/or corrosive and therefore a certain level of technical training is required.Appropriate training is also required to be able to interpret the observed colour changes.

Bell and Hanes[2] developed a simple microfluidic device to perform several presumptive colour testsand a crystal test for illicit drugs. They focused on the detection of methamphetamine, amphetamine,cocaine and oxycodone (an opioid). Bell and Hanes concluded that it was possible to perform thecomplete analysis within 15 sec by using less than 1 mg of sample for all three colour tests and thecrystal test. Amounts in the pg range can be detected, although the crystal reagent test turned out tohave more sensitivity than the colour tests. These preliminary results form the basis for this research.

In this research project, at first the detection limits of these colour tests are determined with druganalogues to avoid procedures for obtaining the required licenses to perform research on real illicitdrugs. A chip, designed for the mixing of two solutions, is used to analyse the mixing of the colour testchemicals with the drug analogues on the chip. The chip shows a good mixing performance, but dueto the small channels the colour change cannot be observed visually. Furthermore, the observationand analysis of the coloured reaction product by eye is very subjective. Therefore the detection limitis also determined by UV-Vis spectroscopy. The results will be presented at the symposium.

Next step in this research is the design and fabrication of an LOC device for the simultaneousanalysis of the three colour tests. The colour change must be clearly detectable by eye (or microscope)and the detection limits for the drugs analogues will be determined and compared with the conventionalmethod.

88


This research is part of the masters program of Analytical Sciences (University of Amsterdam/VUUniversity Amsterdam) and is carried out from September 2010 till April 2011.

References

B. B. Bruijns. Forensic applications of nanotechnology, Literature survey on the use of nanotechnologyand lab-on-a-chip devices for forensic investigations. October 2010.

S.C. Bell and R.D. Hanes. A Microfluidic Device for Presumptive Testing of Controlled Substances.Journal of forensic sciences, 52(4):884-888, 2007.

89

Brigitte Bruijns Illicit drugs analysis on chip References

References

[1] Y. Tsumura, T. Mitome, and S. Kimoto. False positives and false negatives with a cocaine-specificfield test and modification of test protocol to reduce false decision. Forensic Science International,155(2-3):158–164, 2005.

[2] Openbaar Ministerie. Aanwijzing opiumwet. http://www.om.nl/organisatie/beleidsregels/overzicht/drugs/@151408/aanwijzing_opiumwet, November 2000.

[3] J. van den Berg. Illicit Drugs & Forensic Science. Criminalistics, course of the master ForensicScience, University of Amsterdam, 11 March 2008.

[4] J. van den Berg. Drugs analysis. Visit Netherlands Forensic Institute, 4 November 2010.

[5] Sunshine Coast Health Center; Your Partner in Recovery. Drug chart. www.

sunshinecoasthealthcentre.ca/drug-information.html, June 2009.

[6] S.B. Karch. Addiction and the Medical Complications of Drug Abuse. CRC Press, Taylor &Francis Group, 2008.

[7] R. de la Torre, M. Farre, M. Navarro, R. Pacifici, P. Zuccaro, and S. Pichini. Clinical pharmacoki-netics of amphetamine and related substances: monitoring in conventional and non-conventionalmatrices. Clinical pharmacokinetics, 43(3):157–185, 2004.

[8] G. Glass. Narcotics: dangerous painkillers. The Rosen Publishing Group, 2001.

[9] O. Suzuki and K. Watanabe. Drugs and Poisons in Humans; A Handbook of Practical Analysis.Springer, 2005.

[10] D.E. Nichols. Differences between the mechanism of action of MDMA, MBDB, and the classichallucinogens. Identification of a new therapeutic class: entactogens. Journal of PsychoactiveDrugs, 18(4):305, 1986.

[11] C.L. O’Neal, D.J. Crouch, and A.A. Fatah. Validation of twelve chemical spot tests for thedetection of drugs of abuse. Forensic Science International, 109(3):189–201, 2000.

[12] United Nations: Office on Drugs and Crime. Recommended methods for the identification andanalysis of amphetamine, methamphetamine and their ring-substituted analogues in seized ma-terials. United Nations publication, pages 1–77, 2006.

[13] K.-A. Kover and M. Laudszun. Chemistry and Reaction Mechanisms of Rapid Tests for Drugsof Abuse and Precursors Chemicals. United Nations, SCITEC, 6.

[14] S. Bell. Forensic Chemistry. Pearson Prentice Hall, 2006.

[15] D.A. Skoog, F.J. Holler, and S.T. Crouch. Principles of Instrumental Analysis. ThomsonBrooks/Cole, 2007.

[16] S.H. James and J. J. Nordby. Forensic Science: An Introduction to Scientific and InvestigativeTechniques. CRC Press, Taylor & Francis Group, 2005.

[17] F. Ariese and C. Gooijer. Absorption. (Bio-)Molecular Spectroscopy, course of the master Ana-lytical Sciences, University of Amsterdam, February–March 2009.

[18] K.A. Erickson and P. Wilding. Evaluation of a novel point-of-care system, the i-STAT portableclinical analyzer. Clinical chemistry, 39(2):283–287, 1993.

90


[19] E.T. Lagally, C.A. Emrich, and R.A. Mathies. Fully integrated PCR-capillary electrophoresismicrosystem for DNA analysis. Lab on a Chip, 1(2):102–107, 2001.

[20] K.M. Horsman, J.M. Bienvenue, K.R. Blasier, and J.P. Landers. Forensic DNA analysis onmicrofluidic devices: a review. Journal of forensic sciences, 52(4):784–799, 2007.

[21] J. Wang, M. Pumera, M.P. Chatrathi, A. Escarpa, M. Musameh, G. Collins, A. Mulchandani,Y. Lin, and K. Olsen. Single-channel microchip for fast screening and detailed identification ofnitroaromatic explosives or organophosphate nerve agents. Anal. Chem, 74(5):1187–1191, 2002.

[22] J. Wang. Microchip devices for detecting terrorist weapons. Analytica Chimica Acta, 507(1):3–10,2004.

[23] S.C. Bell and R.D. Hanes. A Microfluidic Device for Presumptive Testing of Controlled Sub-stances. Journal of forensic sciences, 52(4):884–888, 2007.

[24] W.P. Bula, W. Verboom, D.N. Reinhoudt, and H.J.G.E. Gardeniers. Multichannel quench-flowmicroreactor chip for parallel reaction monitoring. Lab on a Chip, 7(12):1717–1722, 2007.

[25] W.P. Bula. PhD thesis: Microfluidic devices for kinetic studies of chemical reactions. Universityof Twente, 2009.

[26] Sigma-Aldrich R©. Acetylsalicylic acid. http://www.sigmaaldrich.com, 2010.

[27] Acros Organics. Acetylsalicylic acid. http://www.acros.be, 2010.

[28] Techno Plastic Products AG. Resistance to chemicals: Polyethylene, polypropylene, polystyrene.www.tpp.ch/technical_information/raw_materials/Resistenzliste.pdf, 2003.

[29] DuPont. Safety & emergeny information, dimethylsulphate. www2.dupont.com/Dimethyl_

Sulfate/en_US/index.html, 2010.

[30] S. Gibbons and M. Zloh. An analysis of the ’legal high’ mephedrone. Bioorganic & medicinalchemistry letters, 20:4135–4139, 2010.

[31] L. Karila and M. Reynaud. GHB and synthetic cathinones: clinical effects and potential conse-quences. Drug Testing and Analysis, pages 1–8, 2010.

[32] R.F. Staack and H.H. Maurer. Metabolism of designer drugs of abuse. Current drug metabolism,6(3):259–274, 2005.

[33] H.H. Maurer. Chemistry, Pharmacology, and Metabolism of Emerging Drugs of Abuse. Thera-peutic Drug Monitoring, 32(5):544–549, 2010.

[34] W. Dimpfel, M. Spuler, and DE Nichols. Hallucinogenic and stimulatory amphetamine deriva-tives: fingerprinting DOM, DOI, DOB, MDMA, and MBDB by spectral analysis of brain fieldpotentials in the freely moving rat (Tele-Stereo-EEG). Psychopharmacology, 98(3):297–303, 1989.

[35] M.F. Andreasen, R. Telving, R.I.D. Birkler, B. Schumacher, and M. Johannsen. A fatal poisoninginvolving Bromo-Dragonfly. Forensic Science International, 183:91–96, 2009.

[36] D.M. Wood, S. Davies, S.L. Greene, J. Button, D.W. Holt, J. Ramsey, and P.I. Dargan. Caseseries of individuals with analytically confirmed acute mephedrone toxicity. Clinical Toxicology,48(9):924–927, 2010.

91


[37] K.J. Lusthof, R. Oosting, A. Maes, M. Verschraagen, A. Dijkhuizen, and A.G.A. Sprong. A caseof extreme agitation and death after the use of mephedrone in The Netherlands. Forensic ScienceInternational (Article in press), xxx:xxx–xxx, 2011.

[38] AR Winstock, LR Mitcheson, P. Deluca, Z. Davey, O. Corazza, and F. Schifano. Mephedrone,new kid for the chop? Addiction, 106:154–161, 2010.

[39] M.R. Meyer, J. Wilhelm, F.T. Peters, and H.H. Maurer. Beta-keto amphetamines: studies onthe metabolism of the designer drug mephedrone and toxicological detection of mephedrone,butylone, and methylone in urine using gas chromatography–mass spectrometry. Analytical andbioanalytical chemistry, 397(3):1225–1233, 2010.

[40] R. Kikura-Hanajiri, M. Kawamura, K. Saisho, Y. Kodama, and Y. Goda. The disposition intohair of new designer drugs; methylone, MBDB and methcathinone. Journal of ChromatographyB, 855(2):121–126, 2007.

[41] P. Rosner, B. Quednow, U. Girreser, and T. Junge. Isomeric fluoro-methoxy-phenylalkylamines:a new series of controlled-substance analogues (designer drugs). Forensic Science International,148(2-3):143–156, 2005.

[42] A.H. Ewald, G. Fritschi, W.R. Bork, and H.H. Maurer. Designer drugs 2, 5-dimethoxy-4-bromo-amphetamine (DOB) and 2, 5-dimethoxy-4-bromo-methamphetamine (MDOB): studies on theirmetabolism and toxicological detection in rat urine using gas chromatographic/mass spectromet-ric techniques. Journal of Mass Spectrometry, 41(4):487–498, 2006.

[43] M.D. Cole, C. Lee, and N. Oxley. 4-Bromo-2,5-dimethoxyphenethylamine (2C-B): a review of thepublic domain literature. Science and Justice, 42(4):223–224, 2002.

[44] Prazisions Glas & Optik. Borofloat transmission. http://www.pgo-online.com/intl/katalog/curves/boro_kurve.html, 2010.

[45] Schott. Optical properties of borofloat – transmission. http://www.schott.com/borofloat/

english/download/optical_transmit.pdf, 2011.

[46] W. Qiang, C. Zhai, J. Lei, C. Song, D. Zhang, J. Sheng, and H. Ju. Disposable microfluidic devicewith ultraviolet detection for highly resolved screening of illicit drugs. Analyst, 134(9):1834–1839,2009.

[47] Ocean Optics. Hr4000 high-resolution spectrometer. http://www.oceanoptics.com/products/hr4000.asp, 2010.

[48] Tecan. Tecan Safire2 Brochure. pages 1–4, 2004.

[49] B. McCord, C. Turner, M. Blas, S. Bishop, M. Lerch, S. Dehere, and R. Pannepucci. Developmentof Microfluidic Devices for the Rapid Isolation and Detection of Drugs of Abuse.

[50] S.E. Moring, R.T. Reel, and R.E.J. van Soest. Optical improvements of a Z-shaped cell forhigh-sensitivity UV absorbance detection in capillary electrophoresis. Analytical Chemistry,65(23):3454–3459, 1993.

[51] JP Chervet, REJ Van Soest, and M. Ursem. Z-shaped flow cell for UV detection in capillaryelectrophoresis. Journal of Chromatography A, 543:439–449, 1991.

[52] J. Jansen, A. Louwes, P. Rossen, and M. van Schagen. Project report: Lab-on-a-Chip: The powerof separation. pages 1–6, 2006.

92


[53] Netherlands Forensic Institute. Vakbijlage: Identificatie van verdovende middelen. VakbijlagenNFI, pages 1–8, 2009.

[54] K.L.A. Chan, S. Gulati, J.B. Edel, A.J. de Mello, and S.G. Kazarian. Chemical imaging ofmicrofluidic flows using ATR-FTIR spectroscopy. Lab Chip, 9(20):2909–2913, 2009.

[55] J. Greener, B. Abbasi, and E. Kumacheva. Attenuated total reflection Fourier transform infraredspectroscopy for on-chip monitoring of solute concentrations. Lab Chip, 10(12):1561–1566, 2010.

[56] R. Herzig-Marx, KT Queeney, R.J. Jackman, M.A. Schmidt, and K.F. Jensen. Infrared spec-troscopy for chemically specific sensing in silicon-based microreactors. Analytical chemistry,76(21):6476–6483, 2004.

[57] P.O. Andersson, M. Lundquist, L. Tegler, S. Borjegren, L. Baltzer, and L. Osterlund. A NovelATR-FTIR Approach for Characterisation and Identification of Ex Situ Immobilised Species.ChemPhysChem, 8(5):712–722, 2007.

[58] A. Floris, S. Staal, S. Lenk, E. Staijen, D. Kohlheyer, J. Eijkel, and A. van den Berg. A prefilled,ready-to-use electrophoresis based lab-on-a-chip device for monitoring lithium in blood. Lab Chip,10(14):1799–1806, 2010.

[59] Y.C. Chen and A. Hu. Simultaneous determination of trace benzodiazepines from drinks byusing direct electrospray probe/mass spectrometry (DEP/MS). Forensic Science International,103(2):79–88, 1999.

[60] Y. Deng, H. Zhang, and J. Henion. Chip-based quantitative capillary electrophoresis/mass spec-trometry determination of drugs in human plasma. Analytical chemistry, 73(7):1432–1439, 2001.

[61] M.G. Blain, J.G. Fleming, and D.Y. Cruz. Micro mass spectrometer on a chip. United States.Dept. of Energy, 2005.

[62] S. Taylor and N. France. Miniature and micro mass spectrometry for nanoscale sensing appli-cations. In Journal of Physics: Conference Series, volume 178, page 012003. IOP Publishing,2009.

[63] W. Xu, N.E. Manicke, G.R. Cooks, and Z. Ouyang. Miniaturization of Mass SpectrometryAnalysis Systems. Journal of the Association for Laboratory Automation, 2010.

[64] F. von Eggeling, K. Junker, W. Fiedler, V. Wollscheid, M. Durst, U. Claussen, and G. Ernst. Massspectrometry meets chip technology: A new proteomic tool in cancer research? Electrophoresis,22(14):2898–2902, 2001.

[65] S. Schlautmann. Process docuement: Microfluidic Glass Chip Fabrication. 2009.

93

Illicit drugs analysis on chip...Illicit drugs analysis on chip The use of lab-on-a-chip technology for forensic applications Author: Brigitte Bruijns MSc Student Number: 5924553 Project:

Documents