The potential of near infrared spectroscopy to estimate the content of cannabinoids in Cannabis sativa L.: A comparative study.

Cannabis has been one of the oldest source of food, textile fiber and psychotropic substances. Cannabinoids are the main biologically active constituents of the Cannabis genus, with a demonstrated medicinal value. Its production is becoming legalized and regulated in many countries, thus increasing the need for a rapid analysis method to assess the content of cannabinoids. Gas chromatography (GC) is the preferred analytical method for the determination of these compounds, although is a slow and costly technique. Near infrared spectroscopy (NIR) has the potential for the quantitative prediction of quality parameters, and also of pharmacologically active compounds, but no references about cannabinoids prediction has been previously reported. The aim of the present research was to develop a fast, economical, robust and environmentally friendly method based on NIR technology that allow the quantification of the main cannabinoids present in Cannabis sativa L. SAMPLES: A total of 189 grinded and dried samples from different genotypes and registered varieties were used. The content of the cannabinoids CBDV, Δ9-THCV, CBD, CBC, Δ8-THC, Δ9-THC, CBG and CBN were determined by gas chromatography. Spectra were collected in a dispersive NIR Systems 6500 instrument, and in a Fourier transform near Infrared (FT-NIR) equipment. The sample group was divided into calibration and validation sets, to develop modified partial lest squares (PLS) regression models with WINISI IV software with the dispersive data, and PLS models using OPUS 7.2 with the FT-NIR ones. Excellent coefficient of determination of cross validation (R2CV from 0.91 to 0.99), were obtained for the prediction of CBD, CBC, Δ8-THC, Δ9-THC, CBG and CBN, with standard error of prediction (SEP) values among 1.5-3 times the standard error of laboratory (SEL); and good for CBDV and Δ9-THCV cannabinoids (R2 values of 0.89 and 0.83, respectively) with the dispersive instrument. Similar calibration and validation statistics have been obtained with the FT-NIR instrument with the same sample sets, using its specific OPUS software. In conclusion, a methodology of quantitative determination of cannabinoids in Cannabis raw materials has been developed for the first time using NIR and FT-NIR instruments, with similar good predictive results. This new analytical method would allow a simpler, more robust and precise estimation than the current standard GC.