Quantitative Model Based on Raman Spectra to Evaluate Contaminants in Crack (Cocaine).

The article describes a quantitative method applied to the Raman spectra of cocaine for analyzing the purity and the possible contaminants of the drug. The method is based on the scores and vectors Principal Components Analysis (PCA) and the cocaine sample is diluted in binary mixtures of selected contaminants. It has been shown that by calculating the score of each cocaine sample, the amount of cocaine in samples mixed with contaminants could be calculated with low prediction error.