Evaluation of a low-cost portable near-infrared spectrophotometer for in situ cocaine profiling

Talanta. 2019 Aug 1:200:553-561. doi: 10.1016/j.talanta.2019.03.091. Epub 2019 Mar 28.

Abstract

The potential of a low-cost (∼US$ 1000) portable near-infrared (NIR) spectrophotometer for in situ characterization of seized cocaine samples was evaluated. A set of 240 samples of cocaine seized in several regions and cities across Brazil by its federal police was employed in this study. These samples were previously analyzed by chromatography at the Forensic Chemistry Laboratory of the National Institute of Criminalistics in Brasília-DF for the contents of several constituents to chemically characterize the samples. A low-cost NanoNIR spectrophotometer (Texas instruments) was used to acquire the NIR spectra of the samples in the range 900-1700 nm. The spectra set was treated by the second derivative to construct and validate multivariate regression (Partial Least Square - PLS) and classification (software independent modeling of class analogy - SIMCA) models aiming to characterize the samples. Consequently, an informative toll for objective decision making could be used by the police agents to produce immediate answers to forensic questions raised at the point of seizing. Among those questions the most relevant are: does the seized sample contain cocaine? what is the cocaine form? what is its content? is the sample adulterated and/or diluted? what is the content of adulterant? is the sample significantly adulterated and/or diluted? what is the degree of oxidation of the cocaine? The results of this work allow to propose a NIR/chemometrics based analytical protocol providing fast answers to these questions with satisfactory confidence level for the purpose of reliably screen the seized samples.

Keywords: Cocaine hydrochloride; Free base cocaine; Green analytical method; In situ analysis; NanoNIR spectrophotometer; Seized samples.

Publication types

  • Evaluation Study

MeSH terms

  • Brazil
  • Cocaine / analysis*
  • Least-Squares Analysis
  • Software
  • Spectroscopy, Near-Infrared / economics*

Substances

  • Cocaine