Last updated on September 15, 2026 · Originally published February 27, 2017
Liquid chromatography separates the compounds in cannabis and cannabis products so they can be identified and measured. In testing laboratories, it is the technique behind most cannabis potency labels.
Simple Liquid Chromatography
At its most basic, liquid chromatography separates a mixture by passing it through a column packed with a stationary phase — a solid or a liquid held on an inert support — while a solvent carries the sample along. Different compounds interact with that stationary phase differently (through adsorption, ion-exchange, partitioning, or size), so they travel through the column at different rates and emerge separated, collected as fractions over time. Most cannabis potency methods use a specific version of this idea called reversed-phase LC, commonly run on a C18 column, chosen for how well it separates cannabinoids by polarity. Simple, gravity-fed LC still turns up in teaching labs and low-throughput settings, but it’s rarely what runs a compliance test today. That’s HPLC’s job.
High-Performance Liquid Chromatography
High-performance liquid chromatography, historically also called high-pressure liquid chromatography, uses pumps to push pressurized solvent through the column rather than relying on gravity. That lets it run at much higher pressure with smaller sorbent particles, giving it far greater resolving power than simple LC — which is why it’s widely used across pharmaceutical manufacturing, medicine, and research, cannabis testing included.
Why HPLC, specifically, for cannabis potency?
Cannabis has one property that makes the choice of instrument matter more than it would for most plants: in fresh cannabis, most potential THC and CBD occur as their acidic precursors, THCA and CBDA, which convert to THC and CBD mainly through heat — smoking, vaping, or baking.
That conversion, decarboxylation, is also what tends to happen inside a gas chromatograph (GC), the other major technique used in cannabinoid testing. GC vaporizes the sample to separate its components, and that heat generally decarboxylates THCA and CBDA before they’re measured — unless the sample is chemically derivatized first, an extra preparation step that lets GC distinguish the acid forms too. Without it, a standard GC run reports only a combined total, and even that total isn’t automatically reliable: incomplete conversion or degradation during the heated run can throw the number off.
HPLC avoids the problem at the source. It doesn’t vaporize the sample to separate it, so the acidic cannabinoids stay largely intact through the column. That’s why HPLC has become the standard for cannabis potency testing: it measures THCA and THC — and CBDA and CBD — as the separate compounds they actually are, without needing an extra derivatization step. A widely cited HPLC method paper in this field describes exactly this contrast: GC’s heat-driven conversion versus HPLC’s ability to measure the original, acid-and-neutral composition directly. [1]
This is also where the now-familiar 0.877 conversion factor comes from — and it’s worth being precise about what it does and doesn’t tell you. Regulators often require the theoretical amount of THC that could be formed through complete decarboxylation, so labs calculate “total THC” from the separately measured HPLC values: total THC ≈ THC + (THCA × 0.877), the factor accounting for the mass lost when THCA sheds its carboxyl group. It is a standardized estimate of maximum THC-equivalent content, not a measurement of the THC a consumer will absorb.
How Does a Peak Become a Potency Number?
As each cannabinoid leaves the column, the detector produces a peak. The laboratory compares that peak’s area against a calibration curve built from a certified reference standard, then corrects for the sample’s mass and any dilution to arrive at a concentration. Software can perform much of this calculation automatically, but only after the laboratory has established the separation method, the reference standards, and the calibration. The result is only as good as those three things.
Detectors and sample prep
Most cannabis HPLC systems pair with a UV or diode-array detector (DAD), which reads how strongly each separated compound absorbs UV light. Cannabinoids absorb ultraviolet light, but each one still has to be quantified against its own calibration standard — the detector doesn’t tell them apart on its own; the column’s separation and the calibration do that work. For lower-level or more selective work — pesticide residues in particular — HPLC is paired with mass spectrometry (LC-MS/MS) instead, providing much greater sensitivity and selectivity, although at greater cost and complexity. Many modern laboratories also use ultra-high-performance liquid chromatography, or UHPLC, which applies the same principle at higher pressures with smaller particles, usually producing faster and more efficient separations.
Sample preparation looks simple — homogenize, extract, filter, dilute — but it’s a major source of variation, not a formality. The flower has to be sampled and homogenized so the tested portion actually represents the batch; the cannabinoids have to be extracted efficiently, typically into methanol or ethanol; and the extract has to be diluted into the range the instrument is calibrated for. An error at any of those stages changes the final potency number, which is one reason two labs can report different results for the same flower even when both instruments are working correctly.
References
- De Backer B, Debrus B, Lebrun P, et al. Innovative development and validation of an HPLC/DAD method for the qualitative and quantitative determination of major cannabinoids in cannabis plant material. Journal of Chromatography B. 2009;877(32):4115-4124. doi:10.1016/j.jchromb.2009.11.004
- Pourseyed Lazarjani M, Torres S, Hooker T, Fowlie C, Young O, Seyfoddin A. Methods for quantification of cannabinoids: a narrative review. Journal of Cannabis Research. 2020;2:35. doi:10.1186/s42238-020-00040-2
