Near-infrared spectroscopy combined with chemometrics predicted 10 of 12 cannabinoid concentrations with r-squared values of 0.80-0.95, offering a rapid and economical alternative to chromatographic analysis.
Read this if you work in cannabis testing, quality control, or agricultural research.
r² 0.80-0.95Prediction accuracy for 10 of 12 cannabinoids using rapid NIRS spectroscopy
What the researchers found
Using dried, homogenized cannabis flower from 84 diverse accessions, NIRS predictive models achieved r-squared values of 0.80-0.95 for 10 of 12 therapeutically relevant cannabinoids. Models for major cannabinoids performed best. NIRS could discriminate between acid and neutral forms of cannabinoids and between C3-alkyl and C5-alkyl cannabinoids. The technique requires only solvent extraction and provides results in seconds using a diffuse reflectance device.
Why it matters
Current chromatographic cannabinoid testing is expensive and slow, limiting production efficiency, research, and regulatory compliance. NIRS could provide the cannabis industry with a practical high-throughput testing solution.
The numbers in context
- 84 diverse cannabis accessions
- 12 cannabinoids modeled
- 10 of 12 with r² = 0.80-0.95
- Major cannabinoids showed best performance
- Discriminated acid vs neutral forms
- Discriminated C3-alkyl vs C5-alkyl cannabinoids
How the study worked
Diffuse reflectance NIRS on dried, homogenized cannabis inflorescences. Chromatographic reference data for 12 cannabinoids. Partial least squares regression for predictive chemometric models. Cross-validation and independent prediction sets.
Who was studied
84 diverse Cannabis sativa accessions for analytical method development
What this study cannot tell us
Tested only on dried, homogenized flower — may not work on whole buds or other product forms. Model built on 84 accessions; larger datasets may improve accuracy. Minor cannabinoids at very low levels may be less accurately quantified.
How to read the evidence
Analytical method development with diverse reference panel and validated chemometric models. Strong proof of concept.
When this study was published
Published in 2023. Part of growing interest in rapid spectroscopic methods for cannabis analysis.
The bigger picture
Rapid, affordable cannabinoid testing technology could transform the cannabis industry by enabling quality control at every point in the supply chain, from farm to retail, without the delays and costs of laboratory chromatography.
Questions still open
- Can NIRS models be transferred between different spectrometer devices?
- Would NIRS work on fresh, undried cannabis flower?
- Could handheld NIRS devices enable field testing by growers?
Read the original research
Using a global diversity panel of Cannabis sativa L. to develop a near InfraRed-based chemometric application for cannabinoid quantification.
Scientific reports, 13(1), 2253
Citation
Gloerfelt-Tarp, Francine; Hewavitharana, Amitha K; Mieog, Jos; Palmer, William M; Fraser, Felicity; Ansari, Omid; Kretzschmar, Tobias. (2023). Using a global diversity panel of Cannabis sativa L. to develop a near InfraRed-based chemometric application for cannabinoid quantification.. Scientific reports, 13(1), 2253. https://doi.org/10.1038/s41598-023-29148-0