Cannabis GC–MS Bias: More Signal Does Not Necessarily Mean More THC

forensic toxicology working group members October 28, 2026 Forensic Toxicology

A larger THC signal in a plant extract can reflect analytical matrix effects rather than more THC. A GC–MS study found that biases previously attributed to incomplete THCA conversion could instead arise from differences between plant extracts and neat calibration solutions.

The experiment concerned plant analysis

Researchers investigated total-THC quantitation in cannabis extracts, including chromatographic separation, injection conditions, isotope-labeled standards and analyte protectants. The specimens were plant extracts and laboratory solutions, not human blood or urine.

Here THCA means tetrahydrocannabinolic acid, the plant precursor that can decarboxylate to THC. That name must not be confused with papers using “THCA” for the carboxy metabolite of THC.

Conversion and surface activity were separate mechanisms

Under the tested extract conditions, THCA converted efficiently to THC, including at relatively low injection temperatures. The authors concluded that decarboxylation was not the source of the observed bias in those experiments.

Instead, plant matrix increased analyte response compared with neat solutions. Surface adsorption, including effects associated with active silanol sites, contributed to response differences. External calibration was particularly vulnerable to such mismatch.

Why internal standards and protectants helped

Isotope-labeled compounds allowed the researchers to distinguish relevant species and study the analytical process. Analyte protectants reduced problematic adsorption and improved consistency between preparations.

This does not mean that every gas-chromatography method automatically achieves complete conversion or that every matrix effect disappears after adding one standard. The result belongs to the tested instrument conditions, materials and workflow.

Implications for a potency dispute

Review whether standards and samples behave comparably, whether the method separates relevant cannabinoids, and how conversion and adsorption were independently assessed. A strong calibration line in neat solvent cannot alone establish unbiased results in plant material.

The study offers a concrete explanation for some potency-measurement discrepancies. It does not establish the absorbed dose from consuming a product or permit conversion of plant total THC into a person’s blood concentration or impairment level.

For context, see our discussion of THC concentrations and the separate question of impairment.