An infrared spectroscopic approach to characterise white powders, easily applicable in the context of drug checking, drug prevention and on‐site analysis. Issue 3 (24th November 2020)

• Record Type: Journal Article
• Authors: Deconinck, Eric; Aït‐Kaci, Camille; Raes, Andries; Canfyn, Michaël; Bothy, Jean‐Luc; Duchateau, Céline; Mees, Corenthin; De Braekeleer, Kris; Gremaux, Lies; Blanckaert, Peter
• Is Part Of: Drug testing and analysis. Volume 13:Issue 3(2021)
• Page Start: 679
• Page End: 693
• Publication Date: 2020-11-24
• Subjects: ATR‐(N)IR -- chemometrics -- illicit drugs -- mobile detection approaches; Drugs -- Analysis -- Periodicals; Drug testing -- Periodicals; Chemistry, Forensic -- Periodicals; 615.1901
• DOI: 10.1002/dta.2973
• Languages: English
• ISSNs: 1942-7603

Abstract

Abstract: More and more events, such as the summer music festivals, are considering the possibilities for implementing on‐site testing of psychoactive drugs in the context of prevention and harm reduction. Although the on‐site identification is already implemented by plenty of drug checking services, the required rapid quantitative dosing of the composition of illicit substances is still a missing aspect for a successful harm reduction strategy at events. In this paper, an approach is presented to identify white powders as amphetamine, cocaine, ketamine or others and to estimate the purity of the amphetamine, cocaine and ketamine samples using spectroscopic techniques hyphenated with partial least squares (PLS) modelling. For identification purposes, it was observed that mid‐infrared spectroscopy hyphenated with PLS‐discriminant analysis allowed the distinction between amphetamine, cocaine, ketamine and other samples and this with a correct classification rate of 93.1% for an external test set. For quantitative estimation, near‐infrared spectroscopy was more performant and allowed the estimation of the dosage/purity of the amphetamine, cocaine and ketamine samples with an error of more or less 10% w/w. An easily applicable, practical and cost‐effective approach for on‐site characterisation of the majority of the psychoactive samples encountered in Belgian nightlife settings based on IR spectroscopy was proposed. Abstract : ATR‐mid‐IR and NIR spectra were recorded for a series of white powders collected at summer festivals in the context of drug prevention. PLS‐models based on mid‐IR spectra could discriminate between amphetamine, ketamine, cocaine and other powders. PLS models based on NIR spectra can estimate the purity of amphetamine, ketamine and cocaine samples with an error within 10% m/m.