Why are terpenes so hard to preserve during cannabis extraction?

Most terpenes, particularly monoterpenes like myrcene, limonene, and alpha-pinene, have low boiling points — many below 180 °C. Every extraction step involving heat, vacuum evaporation, or prolonged solvent removal creates conditions where these volatile compounds are lost before the final product is reached. Oxidation during handling adds a further degradation pathway on top of thermal loss.

Does supercritical CO₂ extraction preserve terpenes better than ethanol?

It depends on how the run is configured. Subcritical CO₂ conditions (lower temperature and pressure) selectively capture lighter terpenes, while supercritical conditions favour cannabinoids. A two-pass approach is often used to collect terpenes separately and recombine them. Cryogenic ethanol extraction at around -40 to -60 °C can also achieve good terpene retention, but the subsequent solvent-removal step requires careful vacuum control to avoid co-evaporating volatile terpenes along with the ethanol.

What happens to terpenes during decarboxylation?

Decarboxylation converts acidic cannabinoid forms (such as THCA and CBDA) to their neutral forms using heat, typically at 110–145 °C for varying durations. This temperature range overlaps directly with the boiling points of several monoterpenes, so significant terpene loss during decarboxylation is expected unless the step is conducted under vacuum to lower effective boiling points, or terpenes are collected and reintroduced afterward.

What is 'terpene reintroduction' and why does it happen?

Distillation and other high-purity processing steps separate terpenes from cannabinoids by design. To produce a full-spectrum or terpene-inclusive final product, manufacturers collect the terpene fraction during processing and add it back to the cannabinoid extract at a later stage. The resulting terpene profile may differ from the native plant profile, and independent analytical verification (typically via GC-MS) is the only reliable way to confirm what terpenes are actually present and at what concentrations.

How should terpene-containing extracts be stored to minimise further degradation?

The essential oil literature — which studies analogous volatile compound stability — consistently shows that amber glass with minimal headspace and inert gas flushing (nitrogen or argon) significantly outperforms clear plastic over six to twelve months. UV exposure accelerates photo-oxidation of monoterpenes, and temperature cycling can drive evaporative loss through imperfect seals. For regulated cannabis medicines, stability data to the claimed shelf life would be expected as part of any regulatory dossier.

Terpene Preservation in Extraction

By Tom Whitfield · 30 May 2026 · 7 min read

Volatility is the enemy, and most extraction processes don't take it seriously enough

Roughly 200 terpene compounds have been identified in Cannabis sativa, and the majority of them have boiling points below 200 °C. Some sit well under 160 °C. That matters because every mainstream extraction method , ethanol, hydrocarbon, supercritical CO₂, steam distillation , involves heat, pressure, or both, and each one creates conditions that favour evaporative loss before you've even started post-processing.

I spent a Friday afternoon last month at the State Library (I am constitutionally incapable of doing this kind of research anywhere noisier than a reference room) working through a cluster of analytical chemistry papers on terpene retention. What struck me is how rarely the pharmaceutical-grade extraction literature and the cannabis-specific literature talk to each other. They're solving the same physical chemistry problem; they just don't read the same journals.

So let's go back to basics: what are we actually trying to preserve, why, and what does the evidence say about how to do it?

What terpenes actually are , and why they degrade

Terpenes are a class of naturally occurring hydrocarbons built from repeating five-carbon isoprene units. Monoterpenes (two isoprene units, C₁₀) are the lightest and most volatile. Sesquiterpenes (three units, C₁₅) are heavier and somewhat more thermally stable. This structural difference is directly relevant to extraction design: a process that retains sesquiterpenes like beta-caryophyllene or humulene may strip monoterpenes like alpha-pinene, limonene, and myrcene almost completely.

Degradation pathways are worth naming properly. Thermal degradation is the most obvious: sustained heat breaks terpene structures, producing oxidised by-products. Oxidative degradation occurs when terpenes are exposed to oxygen during processing or storage; limonene, for instance, oxidises to carveol and carvone under ambient air exposure. Hydrolysis is less commonly discussed but relevant in aqueous extraction conditions. And then there's simply evaporation: because many monoterpenes have vapour pressures comparable to light solvents, they can be lost just from open-vessel handling at room temperature.

The main extraction methods and their terpene retention profiles

Supercritical CO₂ extraction is often described as "terpene-friendly" because it operates without residual solvent and allows fine-tuning of pressure and temperature parameters. That reputation is partly deserved, but with caveats. At subcritical conditions (lower temperature and pressure, around 35–40 °C and 80–100 bar), CO₂ selectively extracts the lighter, more volatile compounds including monoterpenes. At supercritical conditions (above 31.1 °C and 73.8 bar), selectivity shifts toward heavier cannabinoids. Most commercial runs targeting full-spectrum profiles use a two-pass approach: a subcritical first pass to capture terpenes separately, then a supercritical second pass for cannabinoids, with the fractions recombined downstream. Whether the recombined product accurately represents the native plant profile is genuinely debated in the literature, I'd be cautious about assuming it does without analytical verification.

Ethanol extraction is fast, scalable, and effective at pulling cannabinoids. It is also conducted cold (cryogenic ethanol at around -40 °C to -60 °C is now standard at pharmaceutical scale) precisely because lower temperatures reduce both terpene evaporation and co-extraction of unwanted waxes and chlorophylls. The subsequent ethanol removal step, rotary evaporation or falling-film evaporation; is where terpene loss becomes significant. Vacuum reduces the boiling point of ethanol, which helps, but any terpene with a boiling point close to ethanol's effective boiling point under that vacuum will be lost along with the solvent.

Hydrocarbon extraction (butane, propane, or blends) operates at very low temperatures and has historically produced extracts with strong terpene retention for this reason. Propane, with its lower boiling point, is particularly used when a more terpene-forward fraction is the target. The regulatory and safety picture around hydrocarbon solvents is more complicated in an Good Manufacturing Practice pharmaceutical context, where residual solvent limits are tightly specified.

Steam distillation, used for essential oil production generally, is poorly suited to cannabis terpene extraction. The prolonged heat exposure degrades thermolabile compounds and the aqueous environment promotes hydrolysis. It is worth knowing about historically, but it's largely been displaced for this application.

Post-extraction: where the losses compound

Even a well-designed extraction doesn't guarantee terpene preservation through the rest of the process. Winterisation (precipitation and filtration of waxes using cold ethanol) adds another solvent-removal step. Decarboxylation, converting THCA to THC or CBDA to CBD via heat, is essentially the point at which monoterpene loss is almost unavoidable if not carefully controlled. Published decarboxylation protocols range from 110–145 °C for varying durations; that range spans several terpene boiling points entirely.

Short-path distillation, used to produce high-purity cannabinoid fractions, by design separates terpenes from cannabinoids. The resulting distillate is a broad-spectrum or near-isolate product, depending on the cut points used, with terpenes largely absent unless reintroduced. This is why you see terpene "reintroduction" as a standard formulation step in many pharmaceutical-grade products; the original terpenes are either collected during distillation or, in some cases, sourced separately.

Reintroduced terpenes raise their own questions. A terpene fraction collected during CO₂ extraction and then recombined isn't necessarily identical in composition to what existed in the fresh plant. Some compounds that were present as precursors or as labile intermediates won't survive even the mildest collection process intact. For a product making a full-spectrum or entourage effect-adjacent claim, the analytical certificate matters enormously, but I'd argue that the industry's disclosure of that certificate to the end consumer remains inconsistent.

Storage: the degradation that happens after extraction

Terpene preservation doesn't end at the manufacturing step. Oxidation continues in the finished product, particularly if packaging allows oxygen ingress. UV light accelerates photo-oxidation of several monoterpenes. Temperature cycling during transport causes vapour pressure fluctuations that can drive terpene loss through imperfect seals.

The analytical literature on essential oil degradation during storage, which is a well-studied analogue; consistently shows that amber glass with minimal headspace and inert gas flushing (nitrogen or argon) outperforms clear plastic at six and twelve months. This isn't surprising given the underlying physical chemistry, but it's worth saying plainly because packaging decisions downstream of a careful extraction process can undo a lot of that effort.

For any product registered on the ARTG or supplied through the Special Access Scheme, stability data demonstrating terpene retention to the claimed shelf life would be expected as part of a dossier. Whether that data is routinely generated at the terpene-profile level, rather than just at the total-terpene or cannabinoid level, is a different question, and honestly one the TGA's guidance on cannabis medicines doesn't yet answer with great specificity. See the TGA's published guidelines at tga.gov.au for the current regulatory framework.

What the chemistry tells us about specific terpenes at risk

Myrcene (boiling point ~167 °C) is among the most abundant monoterpenes in many cannabis chemotypes and among the most thermolabile. Limonene (~176 °C) oxidises readily in air. Alpha-pinene (~155 °C) and linalool (~198 °C) both fall within decarboxylation temperature ranges. Terpinolene (~186 °C) and ocimene (~66–70 °C for the lightest isomers) are particularly vulnerable to evaporative loss during open-vessel steps.

Sesquiterpenes fare better across the board. Beta-caryophyllene (boiling point ~129 °C at 2 mmHg, approximately 260 °C at atmospheric pressure) and humulene (~106 °C at 1 mmHg) are structurally heavier, less volatile under standard processing conditions, and more resistant to oxidation. A product that still shows a healthy sesquiterpene peak on GC-MS but a flat monoterpene region has likely seen significant heat or solvent-removal losses somewhere in the chain.

Nerolidol, bisabolol, and eucalyptol each have their own degradation profiles worth looking up if you're working with a specific extract, the NIST Chemistry WebBook is a reasonable first stop for vapour pressure and stability data for individual compounds. A link is in the sources below.

Sources

, Tom Whitfield, Health librarian, references & fact-checking

Common questions

Why are terpenes so hard to preserve during cannabis extraction?: Most terpenes, particularly monoterpenes like myrcene, limonene, and alpha-pinene, have low boiling points — many below 180 °C. Every extraction step involving heat, vacuum evaporation, or prolonged solvent removal creates conditions where these volatile compounds are lost before the final product is reached. Oxidation during handling adds a further degradation pathway on top of thermal loss.
Does supercritical CO₂ extraction preserve terpenes better than ethanol?: It depends on how the run is configured. Subcritical CO₂ conditions (lower temperature and pressure) selectively capture lighter terpenes, while supercritical conditions favour cannabinoids. A two-pass approach is often used to collect terpenes separately and recombine them. Cryogenic ethanol extraction at around -40 to -60 °C can also achieve good terpene retention, but the subsequent solvent-removal step requires careful vacuum control to avoid co-evaporating volatile terpenes along with the ethanol.
What happens to terpenes during decarboxylation?: Decarboxylation converts acidic cannabinoid forms (such as THCA and CBDA) to their neutral forms using heat, typically at 110–145 °C for varying durations. This temperature range overlaps directly with the boiling points of several monoterpenes, so significant terpene loss during decarboxylation is expected unless the step is conducted under vacuum to lower effective boiling points, or terpenes are collected and reintroduced afterward.
What is 'terpene reintroduction' and why does it happen?: Distillation and other high-purity processing steps separate terpenes from cannabinoids by design. To produce a full-spectrum or terpene-inclusive final product, manufacturers collect the terpene fraction during processing and add it back to the cannabinoid extract at a later stage. The resulting terpene profile may differ from the native plant profile, and independent analytical verification (typically via GC-MS) is the only reliable way to confirm what terpenes are actually present and at what concentrations.
How should terpene-containing extracts be stored to minimise further degradation?: The essential oil literature — which studies analogous volatile compound stability — consistently shows that amber glass with minimal headspace and inert gas flushing (nitrogen or argon) significantly outperforms clear plastic over six to twelve months. UV exposure accelerates photo-oxidation of monoterpenes, and temperature cycling can drive evaporative loss through imperfect seals. For regulated cannabis medicines, stability data to the claimed shelf life would be expected as part of any regulatory dossier.