How CO2 Extraction Shifts an Herbal Profile

CO2 extraction — formally supercritical fluid extraction (SFE) — applies carbon dioxide under elevated pressure and temperature until it reaches a supercritical state, behaving simultaneously as a liquid and a gas. In that state it penetrates plant material the way a gas does while dissolving compounds the way a liquid solvent does. The result is a concentrated extract whose chemical profile differs meaningfully from what ethanol or water would pull from the same starting plant.

The technique is not new to industry — it has long been used in coffee decaffeination and hop extraction — but its application to botanical extraction for dietary supplements has grown as manufacturers seek to isolate specific compound classes without residual solvent concerns. Understanding what the process actually selects for, and what it leaves behind, is central to interpreting any label that references a CO2 extract.

How Supercritical CO2 Moves Through Plant Material

Carbon dioxide reaches its supercritical point at approximately 31 °C and 73.8 bar of pressure. Below those thresholds it is either a gas or a liquid; above them it occupies a phase where the distinction collapses. In that supercritical phase, CO2 carries a low viscosity (allowing it to diffuse into dense plant tissue quickly) alongside a liquid-like density that gives it genuine solvating power.

The extraction vessel holds milled or ground plant material. Supercritical CO2 flows through the vessel, dissolving target compounds as it goes. The enriched fluid then passes into a separator where pressure is reduced. Because CO2 reverts to a gas at ambient conditions, it simply evaporates, leaving behind the extracted compounds without requiring an evaporation step that applies sustained heat. That absence of prolonged heat exposure is one of the defining features distinguishing CO2 extracts from many steam-distilled or heat-assisted ethanol extracts.

Operators control selectivity by adjusting pressure and temperature. At lower supercritical pressures, CO2 behaves more like a nonpolar solvent, favoring lipophilic compounds: essential oils, terpenes, carotenoids, and fat-soluble pigments. Raising pressure increases CO2 density and expands what it can dissolve, pulling in moderately polar compounds. Adding a small percentage of a co-solvent — typically ethanol — extends polarity further still, reaching more hydrophilic constituents such as certain flavonoids and phenolic acids. The ability to tune selectivity by adjusting physical parameters, rather than switching entirely to a different chemical solvent, is the core mechanical distinction of the method.

Once the CO2 evaporates in the separator, the collected extract is a concentrated fraction of the plant's chemistry. Depending on target compounds and parameters chosen, it may appear as a viscous oil, a waxy solid, or a resin. That physical form itself signals which compound classes dominated the extraction.

Compounds, Organs, and Process Variables That Shape the Final Extract

Terpenes and terpenoids. These are among the most reliably captured compound classes in a standard (lower-pressure, nonpolar) CO2 run. Terpenes are volatile and heat-sensitive; conventional steam distillation can degrade or lose lighter fractions. CO2's low operating temperature preserves them more intact. This is why hops, cannabis-family botanicals, and aromatic herbs processed by CO2 often show a terpene profile closer to the fresh plant than a steam-distilled essential oil from the same source.

Carotenoids and fat-soluble pigments. Lycopene, beta-carotene, and related lipophilic pigments partition readily into supercritical CO2 at moderate pressures. Tomato and paprika oleoresins are commercially produced this way. The same principle applies to fat-soluble botanical pigments in supplement ingredients.

Essential fatty acids and waxes. At higher pressures, CO2 extracts lipid fractions including free fatty acids and surface waxes. This is relevant for seed-based botanicals such as sea buckthorn or saw palmetto, where the fatty acid composition of the extract is considered part of its active profile. Because no high heat is applied, oxidation of those fatty acids during processing is reduced compared to expeller pressing combined with heat.

Polar phenolics and glycosides. Standard supercritical CO2 without a co-solvent extracts these poorly. Flavonoid glycosides, hydroxycinnamic acids, and iridoids — the compound classes most often targeted in standardized herbal extracts such as those from valerian, milk thistle, or echinacea — require either a co-solvent or a separate polar extraction step. A CO2-only extract of these plants will therefore present a fundamentally different chemical fingerprint than an ethanol extract of the same plant.

The extraction vessel and separator hardware. Pressure ratings, flow rate, and separator temperature all influence yield and selectivity. Industrial-scale systems use multiple separators in series, each set to a different pressure, allowing operators to collect distinct fractions sequentially from a single run.

Where CO2 Extraction Produces Results People Do Not Expect

The most common misunderstanding is that CO2 extraction is categorically "better" or "more complete" than ethanol or water extraction. It is not more complete — it is more selective. For a plant where the bioactive compounds are predominantly nonpolar (terpenes, carotenoids, fatty acids), CO2 may capture a higher proportion of those targets with less co-extracted material. For a plant where the bioactive compounds are predominantly polar — glycosides, tannins, alkaloids with polar functional groups — CO2 without a co-solvent will underperform an ethanol extract on those specific markers.

This selectivity creates a labeling ambiguity. Two products from the same botanical can carry identical plant names while having almost no overlapping chemistry, simply because one was CO2 extracted and one was ethanol extracted. What determines an herbal supplement's potency is not the extraction method alone but which compounds are present and in what concentrations — and a CO2 extract optimized for terpenes from, say, a resinous root may contain negligible amounts of the water-soluble compounds a traditional preparation of the same root would deliver.

A second friction point involves co-solvent disclosure. When ethanol is added as a co-solvent to extend polarity, trace ethanol may remain in the final extract. Regulatory thresholds govern residual solvent levels, but the label "CO2 extracted" does not automatically mean solvent-free if a co-solvent was used. The term describes the primary extraction medium, not the complete process.

Yield is another practical limitation. Supercritical CO2 equipment is capital-intensive, and run times for some botanicals are long. For commodity ingredients where cost per kilogram matters significantly, CO2 extraction remains less common than ethanol or hydroethanolic methods. This means that for many widely used botanical ingredients, the CO2 extract is a specialty fraction rather than the standard commercial form, and reference data on its compound profile may be thinner than data on the ethanol extract.

Finally, the physical form of CO2 extracts — often oily or waxy — creates formulation challenges. Encapsulating a viscous oil requires different carrier strategies than tableting a dry ethanol extract powder, and that downstream processing can itself affect how compounds are released and absorbed.

What a Label Citing "CO2 Extract" Shows and What It Does Not

Under the U.S. Dietary Supplement Health and Education Act (DSHEA), manufacturers are responsible for the safety and accurate labeling of their products, but no dietary supplement undergoes FDA pre-market approval. A label stating "supercritical CO2 extract" or "SFE extract" describes the extraction process used; it does not constitute a government-reviewed or government-approved claim about the extract's composition or efficacy.

The label is required to list the botanical's common and Latin name, the plant part used, and the quantity per serving. It is not required to list the specific compound classes enriched or depleted by the CO2 process, nor is it required to disclose whether a co-solvent was used. A label therefore cannot be read as a complete chemical specification.

Third-party testing certificates of analysis (CoAs) go further than a label. A CoA from an accredited laboratory will typically identify and quantify specific marker compounds — terpene percentages, carotenoid levels, or fatty acid ratios, depending on the botanical — and confirm the absence of heavy metals, pesticide residues, and microbial contaminants above defined thresholds. A CoA does not, however, confirm that the extraction parameters used were optimal for any particular compound class; it reports what is present in the tested batch, not what the process was designed to maximize.

Third-party certification seals (from organizations that audit manufacturing practices and test finished products) indicate that what is on the label is in the bottle at the stated amount, and that certain contaminants fall below threshold. Holding such a seal is not the same as FDA approval, and it does not validate any structure-function claim on the label. The FDA's framework for dietary supplement labeling and claims is documented at its dietary supplements guidance pages, and the FTC oversees advertising claims separately.

CO2 extraction is best understood as a precision tool that enriches certain chemical fractions of a botanical while leaving others behind — a trade-off shaped entirely by the physical chemistry of the target compounds and the parameters the operator sets, not a universal upgrade over other methods.

Note: This explains how supplements work. This statement has not been evaluated by the Food and Drug Administration. This information is not intended to diagnose, treat, cure, or prevent any disease, and it is not a substitute for advice from a physician, pharmacist, or registered dietitian.