Green analytical strategy for distinguishing Cannabis sativa L. chemotypes via planar chromatography and partial least squares regression

Cannabis sativa L. is the most distributed drug around the world. Accordingly, the choice of methods for the analysis of cannabis, and its products still remains a challenge for forensic scientists. High-performance thin-layer chromatography (HPTLC) represents a valuable tool for forensic purposes in discriminating between the fiber-type and drug-type of C. sativa. This research strives to evaluate whether any new chemical or statistical data would arise from groups/classes, and at the same time develop a green analytical approach with a combination of HPTLC and multivariate linear classification models. This integrated method seeks to extract maximum information from HPTLC chromatograms while prioritizing environmental sustainability. The greenness of the method was assessed using Analytical Greenness (AGREE) software, which confirmed significant improvements compared with traditional column chromatography techniques. Forty-three seized C. sativa samples were analyzed by two mobile phases as separate methods, using HPTLC for the differentiation of three main cannabinoids. Differences in cannabinoid profiles were examined using a partial least squares discriminant analysis (PLS-DA) model. The excellent classification of drug and fiber types of C. sativa types was confirmed by both PLS-DA models. Based on the AGREE software, the optimized HPTLC methods were shown to be green in regard to the usage of toxic solvents and reagents, miniaturization, automatization, energy, and solvent consumption. Moreover, this proof-of-concept approach demonstrated its potential to be used in forensic laboratories considering aspects of greenness, efficiency, effectiveness, and application on bigger datasets. It represents a new overview comprising different aspects of the forensic analysis of seized C. sativa samples.