Classification of a target analyte in solid mixtures using principal component analysis, support vector machines and Raman spectroscopy. Marie-Louise O’Connell, a Tom Howley, b Alan G. Ryder, a* * * Marc N. Leger, a Michael G. Madden. b a Department of Chemistry/National Centre for Biomedical Engineering Sciences, National University of Ireland, Galway. b Department of Information Technology, National University of Ireland, Galway. ABSTRACT The quantitative analysis of illicit materials using Raman spectroscopy is of widespread interest for law enforcement and healthcare applications. One of the difficulties faced when analysing illicit mixtures is the fact that the narcotic can be mixed with many different cutting agents. This obviously complicates the development of quantitative analytical methods. In this work we demonstrate some preliminary efforts to try and account for the wide variety of potential cutting agents, by discrimination between the target substance and a wide range of excipients. Near-infrared Raman spectroscopy (785 nm excitation) was employed to analyse 217 samples, a number of them consisting of a target analyte (acetaminophen) mixed with excipients of different concentrations by weight. The excipients used were sugars (maltose, glucose, lactose, sorbitol), inorganic materials (talcum powder, sodium bicarbonate, magnesium sulphate), and food products (caffeine, flour). The spectral data collected was subjected to a number of pre-treatment statistical methods including first derivative and normalisation transformations, to make the data more suitable for analysis. Various methods were then used to discriminate the target analytes, these included Principal Component Analysis (PCA), Principal Component Regression (PCR) and Support Vector Machines. Keywords: Raman, Spectroscopy, Forensic, Classification, Chemometrics, Support Vector Machines. 1. INTRODUCTION Many spectroscopic tools have been incorporated into the field of forensic analyses. 1,2 Since the early 1990’s, technological advances in electronics; as well as materials have catapulted analytical methods, which were previously laboratory based into the field of in-situ investigations, one such method is Raman spectroscopy. 3,4 Based on the Raman effect, this technique provides a wealth of information about narcotic samples. Inelastically (or Raman) scattered light from a sample is analysed using a spectrometer and displayed in the form of a spectrum from which valuable information on the molecular structure and functional groups present may be obtained. 5,6 A change in the polarisability of a molecule is the only condition to be met for a substance to be Raman active. This requirement is unique to Raman spectroscopy and complimentary to IR spectroscopy, which requires a change in dipole moment for the substance to be IR active. 7,8 Although previously seen as simply a complimentary tool to IR spectroscopy, which was difficult to implement experimentally, technological advances such as diode lasers, sensitive CCD detectors, and inexpensive computing, have helped Raman spectroscopy to evolve into a rapid, non-contact, easily implemented technique. However, even with all these advances, the process of illicit narcotic analysis using Raman spectroscopy still faces several difficulties. Some of the materials used to bulk and/or bind narcotic active ingredients can be fluorescent thus obscuring the weaker Raman signal. 9 Using longer wavelength excitation minimises this effect but it still remains a problem in positive narcotic identification and quantification. In addition illicit narcotics can be mixed with several different excipients, all of which will have different Raman scattering efficiencies which can cause some (or all) of the Raman bands of the illicit narcotic to be obscured, making the process of identification or quantification more difficult. * alan.ryder@nuigalway.ie ; phone 353-91-492943; fax 353-91-494596 Proceedings of SPIE , the International Society for Optical Engineering, Vol. 5826, 2005.