The role of time and storage conditions on the composition of hashish and marijuana samples: A four-year study

The aim of this study was to investigate the role of time and different real-life storage conditions on the composition of different varieties of cannabis products (hashish and marijuana). Six high-potency cannabis products constituted by herbal and resin materials containing different initial concentrations of delta 9-Tetrahydrocannabinol (THC) were employed for this study. Four representative samples were collected from each study material and were maintained for a prolonged time (four years) under different controlled storage conditions: (A) light (24 h) and room temperature (22 degrees C); (B) darkness (24 h) and room temperature; (C) darkness and refrigeration (4 degrees C); (D) darkness and freezing (-20 degrees C). The concentration of the three main cannabinoids, i.e. THC, Cannabinol (CBN, produced from the degradation of THC), and Cannabidiol (CBD), were measured by GC-FID around every 100 days along the four-year study. Significant changes in the THC (degradation) and CBN (formation) content were detected under storage conditions A and B, and almost 100% of THC was degraded after four years. A mono-exponential function was able to well fit both THC degradation and CBN formation, suggesting that these processes occur with a first order kinetics. Data treatment indicated that the storage temperature and light exposure had two different effects on the conversion of THC to CBN: temperature changed only the speed, light changed both the speed and the stoichiometry of this conversion. Models were proposed which allow to predict the storage time, if unknown, and the initial content of THC (i.e. the concentration of THC at the starting storage time), from the measurement of THC and CBN content at any time under storage condition A. Values predicted are more uncertain at larger storage times and have an accuracy of around 5-10%. These models were also tested on data reported in the literature, and can represent a starting point for further improvements. Prediction models may be helpful for forensic purposes, if the initial concentration of THC or the approximate age of a degraded material need to be estimated, or to plan the storage of delicate samples which need to be re-examined over time. (C) 2019 Elsevier B.V. All rights reserved.