Last updated on September 30, 2026 · Originally published May 2, 2018
Cannabis laboratories use different instruments for different jobs. Consumers want to know what cannabinoids and terpenes a product contains, while regulators require screening for contaminants such as pesticides and residual solvents. Gas chromatography (GC) is a mainstay for terpene profiling and residual-solvent analysis and forms part of many pesticide-testing workflows. It’s one piece of the wider practice of cannabis testing. Potency is the harder case: GC can also measure cannabinoids, but the instrument’s central requirement – heating the sample – can alter the chemical profile it is trying to measure.
How It Actually Works
GC separates compounds according to their volatility and how strongly they interact with the material coating the inside of a column. A heated inlet vaporizes the sample, while an inert carrier gas moves it through the column. Different compounds travel at different rates and reach the detector at characteristic retention times. Retention time helps identify a compound, while the size of the calibrated detector response is used to quantify it; mass spectrometry can add further identifying information. Liquid chromatography (LC) performs the same basic separation in a liquid mobile phase instead of a gas, which makes it better suited to compounds that are nonvolatile or heat-sensitive – exactly the compounds GC struggles with, since GC requires a compound to vaporize without decomposing along the way.
Helium, hydrogen, and nitrogen can all serve as carrier gases, and the choice is a real tradeoff, not a formality. Nitrogen can provide high efficiency but only across a narrow, relatively slow operating range; hydrogen supports faster separations while maintaining good efficiency across a much wider range of flow rates, but its flammability and, in some applications, its chemical reactivity require extra precautions; helium has traditionally offered a practical middle ground, though cost and supply concerns have pushed more labs to consider the alternatives. The column itself sits inside a temperature-controlled oven, which is one of several factors – alongside injection technique, flow control, column condition, calibration, and sample preparation – behind how reproducible a well-run GC method can be.
Why Cannabinoid Potency Is Complicated
Here’s the catch, and it’s not a small one. Getting a sample into a gas chromatograph means vaporizing it, and vaporizing it means heating it – at the injection port, sometimes past 200°C. Fresh and unheated cannabis flower typically contains much more THCA than THC. THCA is the acidic, non-intoxicating precursor that heat converts into THC – the same reaction that occurs when someone lights a joint. A GC’s heated inlet does that conversion too, whether the lab wants it to or not.
That’s a problem specifically for ordinary, underivatized GC analysis. In a 2025 study of cannabis oils, the particular underivatized GC-MS method tested converted only about 50–60% of THCA and CBDA to their neutral forms, producing lower total-cannabinoid recoveries than LC-MS or derivatized GC-MS – a result specific to that method and matrix, not a universal conversion rate for every GC run. [1] A separate 2025 analysis in Forensic Chemistry proposed that part of the problem comes from THCA and THC interacting with silanol sites on the heated inlet’s surface itself, and showed that “analyte protectant” chemistry can correct for it – bringing GC-MS results back in line with LC-PDA reference values. [2]
GC doesn’t inherently lose the acid-to-neutral split – it can preserve it with the right sample preparation. Derivatization can protect acidic cannabinoids through the heated inlet well enough for accurate GC analysis, and newer analyte-protectant methods can instead promote full, quantitative conversion for an accurate total-THC number. What ordinary, underivatized GC cannot reliably preserve is the sample’s original acid-to-neutral profile – the THCA-to-THC ratio the plant actually had going in. That’s a real part of why two labs can report different THC numbers for the same sample, and it’s a major reason liquid chromatography, which analyzes cannabinoids in solution without vaporizing them first, has become the preferred technique for routine potency testing. [3]
Where GC Is the Better Fit
None of this makes GC obsolete – it makes it particularly well suited to certain jobs and a poor fit for others. GC excels at terpene profiling and residual-solvent testing, where volatility is an advantage rather than a source of chemical transformation, and a deeper dive into pairing GC with tandem mass spectrometry covers how far that terpene work can go.
Pesticide screening is more of a split decision than a GC win. GC-MS/MS covers many volatile and thermally stable pesticides well – including the kind of broad, high-volume screening that demonstrates how wide GC’s reach can go – but LC-MS/MS is generally the better tool for pesticides that are less volatile or thermally unstable. A 2025 validation study from Canada’s National Research Council needed both platforms to cover its 96-pesticide panel. The study also found that cultivar-dependent matrix effects could substantially distort some LC-MS/MS results, particularly for etofenprox, while its GC-MS/MS method was less affected by differences among cultivars overall. [4] Comprehensive pesticide panels, in other words, often divide the work between both platforms rather than relying on either one alone.
Necessity drives innovation, and the detector side of GC keeps moving. Many cannabis labs run flame ionization detectors (FID); mass spectrometry is another common choice. A less common detector, vacuum ultraviolet spectroscopy (VUV), works as something close to a universal detector, since most chemical compounds absorb light somewhere in the vacuum-UV region. A 2018 study demonstrated GC-VUV identification of cannabinoids and their metabolites after derivatization was used to protect the acidic cannabinoids during GC analysis – the same underlying problem this piece keeps coming back to. [5] VUV’s absorption-profile approach can help distinguish structurally similar compounds that might otherwise co-elute, and with spectral deconvolution, can quantify some compounds without full baseline separation. Whether VUV offers enough practical advantage over established FID and mass-spectrometric methods for routine cannabis testing remains unsettled.
Smaller producers may use compact GC systems for internal process control before submitting products for compliance testing. Those measurements don’t replace compliance testing by a licensed laboratory, though – greater access to instrumentation only improves process control when it’s paired with validated methods, suitable reference materials, and trained operators. That honest picture of what GC can and can’t measure cleanly, more than the instrument’s price tag, is what actually moves cannabis testing toward the standardization the industry still needs.
References
- Franzin M, Di Lenardo R, Ruoso R, Addobbati R. Incomplete Decarboxylation of Acidic Cannabinoids in GC-MS Leads to Underestimation of the Total Cannabinoid Content in Cannabis Oils Without Derivatization. Pharmaceutics. 2025;17(3):334. doi:10.3390/pharmaceutics17030334
- Mulloor J, Wilson W, Sander L. New perspectives on THCA decarboxylation and accurate GC-MS quantitation of Total THC in Cannabis using analyte protectants. Forensic Chemistry. 2025;44:100668. doi:10.1016/j.forc.2025.100668
- Catani M, Felletti S, Buratti A, De Luca C, Cavazzini A. Perspectives and Pitfalls in Potency Testing of Cannabinoids by High Performance Liquid Chromatography (HPLC). LCGC North America. 2022;40(2).
- MacKenzie DA, Anyanwu AM, McRae G, Melanson JE. Quantitative determination and validation of 96 pesticides in cannabis by LC-MS/MS and GC-MS/MS. Analytical and Bioanalytical Chemistry. 2025;417:3959–3973. doi:10.1007/s00216-025-05918-9
- Leghissa A, Smuts J, Qiu C, Hildenbrand ZL, Schug KA. Detection of cannabinoids and cannabinoid metabolites using gas chromatography with vacuum ultraviolet spectroscopy. Separation Science Plus. 2018;1(1):37–42. doi:10.1002/sscp.201700005
Last updated September 2026 · Originally published May 2, 2018.
Sourced in part from DeepWeed’s analytical-chemistry research corpus (Franzin et al. 2025) · Reviewed by Nani Frenkel, chief editor.
