High performance liquid chromatography method development and chemometric analysis of ecstasy and cocaine

Consumption of illegal drugs of abuse remains a major social issue aligned with a \ud global law-enforcement priority. Forensic analysts are faced with the challenge of \ud continually developing sophisticated methods of analysis to combat the increasing \ud variability that occurs in illicit drug samples. Research work for this thesis has \ud focused on the development of High Performance Liquid Chromatography (HPLC) \ud methods for the analysis of major drug constituents associated with ecstasy and \ud cocaine illicit drug samples. Emphasis has been placed on method development with \ud strategies of univariate or multivariate experimental approaches used in the selection \ud and optimisation of procedures. Considerations with regard to the choice of \ud chromatographic factors, solutes under investigation and the provision of quality \ud assurance data throughout the research work have been the main criteria in methods \ud developed. Two HPLC methods were developed to qualitatively and quantitatively \ud assay for the major drug components and analogue derivatives found in ecstasy and \ud cocaine. Methods developed have undergone validation studies including intra- and \ud inter- reproducibility, accuracy, and linearity of calibration, limit of detection (LOD) \ud and limit of quantification (LOQ) and the use of internal standards. Applications of \ud methods to ecstasy and cocaine samples seized in Ireland ensured their suitability for \ud routine analysis of illicit drug samples.\ud As part of this study, chemical profiling of 183 ecstasy tablets seized in Ireland during \ud 2002-2004 were recorded as discrete data sets. Chemical data sets include both the \ud quantification and occurrence in individual tablets of the major amphetamine \ud components (i.e. MDA, MDMA, MDEA, MBDB methamphetamine and \ud amphetamine), adulterant components (i.e. caffeine, phenacetin, acetaminophen and \ud acetylsalicylic acid), excipients components (i.e. sucrose, glucose, lactose, fructose, \ud mannitol, sorbitol and inositol) and inorganic components (i.e. Al, Zn, Fe, Mg, Ca, Cr, \ud Pb, Na and K ). Chemometrics, including unsupervised methods of principal \ud component analysis (PCA), hierarchical cluster analysis (HCA) and Pearson’s \ud correlation coefficient, as well as supervised methods of linear discriminant analysis \ud (LDA) and artificial neural networks (ANN) was applied to the chemical data sets to \ud demonstrate the ability of the statistical approach to linking sample seizures. HCA \ud and ANN were the numerical methods that most efficiently distinguished between\ud 5HPLC method development & chemometric analysis of ecstasy & cocaine\ud linked and unlinked seizures. Eleven groups were identified from the chemical data \ud sets with group classification dependant on the main amphetamine, adulterant and \ud excipient components present. The benefits from this study can provide strategic \ud intelligence and an understanding of the operational level on the Irish ecstasy market \ud and help evaluate the changing profile or dynamics associated with this illegal market \ud supply