Impact of Solvent Polarity on Volatile and Non-Volatile Cannabinoid Recovery: A Multivariate GC-MS/LC-MS Extraction Optimization Study

Abstract

This study investigates the impact of solvent polarity on the simultaneous recovery of volatile and non-volatile cannabinoids using a multivariate extraction optimization framework supported by GC-MS and LC-MS analysis. Five solvents with varying polarity indices hexane, ethyl acetate, ethanol, methanol, and acetonitrile were evaluated to determine their efficiency in extracting neutral cannabinoids (THC, CBD, CBN) and acidic cannabinoids (THCA, CBDA, CBGA). Results showed that extraction efficiency increases with solvent polarity, reaching peak performance with methanol and ethanol, which provided the best balance between solubilizing hydrophobic and polar cannabinoid classes. Non-polar solvents demonstrated limited recovery of acidic cannabinoids, while highly polar solvents introduced variability due to co-extraction of matrix components. Multivariate modeling using Response Surface Methodology (RSM) and desirability functions revealed that solvent polarity was the most influential factor in determining total extraction performance, with optimal results occurring at polarity indices around 5.1–5.2. The study also highlights the complementary strengths of GC-MS and LC-MS: GC-MS demonstrated superior detection of volatile neutral cannabinoids, while LC-MS provided robust sensitivity for thermolabile acidic cannabinoids. Together, these analytical findings underscore the need for solvent-method alignment in cannabinoid profiling. Overall, the study contributes a comprehensive polarity-driven optimization framework that supports accurate, scalable, and analytically reliable extraction of cannabinoids for research, regulatory, and industrial applications.