During our work in process optimization of cannabis production, we noticed a rate difference of decarboxylation for tetrahydrocannabinolic acid (THCA) and cannabidiolic acid (CBDA). The noticeable rate difference between the decarboxylation reactions of THCA & CBDA has been noted but is generally overlooked. [1a, 1b] We were curious to study and understand those rate differences. We took a two-pronged approach. First, we investigated the reaction mechanism in-silico. Secondly, we developed an in-process analytical tool to track decarboxylation in real time.

Fig. 2: Mid-IR Spectra: Example spectra of cannabis extract throughout the course of decarboxylation by the application of heat.

Fig. 3: Fit of IR spectra, based on PCR model, to HPLC reference data: Cannabinoid concentration plots over the course of decarboxylation for three separate decarboxylation reactions.

While the decarboxylation reaction has been studied [1a, 1b], [2a, 2b], [3a], and a mechanism has been proposed [3b], most of this work has been based on basic THCA analogs. Our calculations [4] for decarboxylation of THCA and CBDA show that: 

  • The rate determining step is the intramolecular protonation process from the carboxylic group to benzene ring.
  • The rate difference is due to steric effects caused by meta substitute rather than electronic effect from para substitute.
  • The steric effect was achieved by blocking the occurrence of an orthogonal carboxylic group conformation of the transition state (forthcoming research). 

For the in-process analytics we developed a Fourier transform infrared attenuated total reflectance (FT-IR-ATR) method in collaboration with PerkinElmer. We utilized cannabis extracts of different cannabinoid concentrations and decarboxylation ratios in the reaction setup. The extracts were heated in an oil bath equipped with an overhead stirrer. 

During an 80-minute reaction time we recorded extract temperature in regular intervals, measured mid-IR spectra every five minutes and collected aliquots for high-performance liquid chromatography (HPLC) analysis. For the IR spectra (Figure 2) principal component regression (PCR) was used to generate quantitative models for THCA and THC. 

Over the course of the decarboxylation reaction, the change in THCA and THC concentration can be observed (Figure 3). Cannabinoid concentrations predicted from IR spectra are overlaid with the HPLC reference values. 

The excellent fit between our IR model and the reference data demonstrates the utility and power of a low-cost, in-process analytical tool for real-time reaction analysis and control. 

In conclusion, we showed our multi-level approach to reaction analysis and process development to support the cannabis industry in their production challenges. 

If you’re interested in our work, reach out to us! We’re always happy to chat. You can also find us presenting continuations of this and other work at a handful of conferences in the coming weeks:
Aug 25 – 29: American Chemical Society National Meeting and Expo, San Diego
Sep 4 – 6: Cannabis Science Conference, Portland. Keynote by CEO Dr. Markus Roggen.
Sep 4 – 6MJBizCon Int’l, Toronto. Panel presentation by Scientific Adviser Dr. Glenn Sammis. 
Sep 23 – 24CannMed Conference, Pasadena. Presentation by CEO Dr. Markus Roggen.


[1a] Wang, M. et al. “Decarboxylation Study of Acidic Cannabinoids: A Novel Approach Using Ultra-High- Performance Supercritical Fluid Chromatography/ Photodiode Array-Mass Spectrometry”, Cannabis Cannabinoid Res. 2016, Volume 1: 262- 271. [journal impact factor = N/A; cited by 17]

[1b] Veress, T. et al. “Determination of cannabinoid acids by high-performance liquid chromatography of their neutral derivatives formed by thermal decarboxylation: I. Study of the decarboxylation process in open reactors”, J. of Chrom. 1990, Volume 520: 339-347. [journal impact factor = 4.169; cited by 43]

[2a] Kanter, S. et al. “Quantitative determination of Δ9- tetrahydrocannabinol and Δ9-tetrahydrocannabinolic acid in marihuana by high-pressure liquid chromatography”, J. of Chrom. 1979, Volume 171: 504. [journal impact factor = 4.169; cited by 25]

[2b] Taschwer, M. and Schmid, M. “Determination of the relative percentage distribution of THCA and Δ9-THC in herbal cannabis seized in Austria – Impact of different storage temperatures on stability”, Forensic Sci. Int., 2015, Volume 254: 167-171. [journal impact factor = 1.947; cited by 25]

[3a] Perrotin-Brunel, H. et al. “Decarboxylation of Δ9-tetrahydrocannabinol: Kinetics and molecular modeling”, J. of Mol. Struc. 2011, Volume 987: 67-73. [journal impact factor = 2.011; cited by 12]

[3b] Chuchev, K. and BelBruno, J. “Mechanisms of decarboxylation of ortho-substituted benzoic acids”, J. of Mol. Struc., (THEOCHEM) 2007, Volume 807: 1-9. [journal impact factor = 2.011; cited by 12]

[4] Unpublished work together with Weiying He and Paul Foth. 

Join our newsletter

Receive news about our latest research and more!