Application of Raman spectroscopy in the detection of cocaine in food matrices

Abstract

While modern methodologies for the smuggling of illicit substances often tend towards more elaborate inventions, the simple practice of concealing cocaine hydrochloride within food matrices is becoming increasingly popular. This study was conducted to develop and optimize a Raman spectroscopic method capable of identifying and quantifying cocaine hydrochloride concealed within food matrices. Samples of cocaine hydrochloride were concealed within baking powder, cake mix and white rum, with identification of the drug achieved through a combination of manual and automated detection methods and comparison with a digital spectral library. Principal component analysis (PCA) and partial least squares regression (PLSR) were employed to qualitatively and quantitatively examine the collected spectra, allowing for traces of cocaine hydrochloride found within each matrix to be identified and quantified. For each of the solid matrices, Raman spectroscopy enabled rapid, non-invasive, and unambiguous identification of the concealed drug. For the white rum samples, due to strong matrix interference, full spectrum matching was not possible, but presumptive identification of cocaine in the samples was achieved at 100%. No false positives or miss-identified samples were recorded, Results of the PLSR analysis showed potential, however difficulties in obtaining accurate concentrations served to limit the method’s use in quantitative analysis.