Self-Optimizing Supercritical Fluid Extractor for the Recovery of Cannabinoids
Despite the promise of cannabis containing goods as medicines and consumer products, the lab-intensive and time-consuming extraction process impedes applications of these compounds. Consequently, there is an urgent need to develop more effective extraction methods to access these high-value materials. Carbon dioxide supercritical fluid extraction is the typical technique to recover cannabinoids from cannabis plants with high levels of enrichment. Unfortunately, anticipating the extraction inputs (CO2 flow rate, time, pressure, etc.) that will lead to the highest quality or recovery is difficult. This issue may be addressed through the application of machine learning algorithms that create relationships between the extractor inputs and outputs. The statistical models then offer a platform to suggest inputs that would lead to an improvement. Addressing this goal establishes the foundation for researchers to apply our methods to their extractor systems for the rapid prediction of the optimal extraction conditions.