Water, ethanol, and supercritical carbon dioxide are the three workhorse solvents of botanical extraction, and the choice among them decides what an extract contains, what it costs, and what residues must be controlled. Water suits gentle traditional profiles, ethanol covers a broad middle spectrum, and supercritical CO2 delivers precise lipophilic fractions without solvent residue. Refine Biology, founded in 2007 as a natural ingredients manufacturer, applies separation and extraction expertise across botanicals including rhodiola, ginger, and ashwagandha at its three sites in China. This comparison lays out the real trade-offs so formulators can match the method to the target compounds.
Key takeawaysThe solvent defines the extract: water favors polar compounds, ethanol captures a broader middle range, and CO2 targets lipophilic fractions with tunable selectivity. · Operating economics differ sharply, with water cheapest in solvent but hungry in evaporation energy, and CO2 dearest in capital but cleanest in residues. [1] · Residue and safety profiles follow the method, so specifications must name the solvent and verify compliance lot by lot. · Matching the method to the target compounds beats defaulting to house equipment; ginger preparations show how much composition follows preparation. · Buyers should lock solvent, standardization basis, and residue limits in the specification before comparing prices across suppliers.
Every botanical extract begins with the same question: what liquid pulls the desired compounds out of the plant? The three classic answers are water, ethanol, and supercritical carbon dioxide, and each one produces a recognizably different ingredient from the same raw material. Choosing among them shapes potency, purity, cost, and even the regulatory story of the finished product.
The choice is rarely about finding the single best solvent. Water is gentle and traditional but unselective; ethanol is versatile but brings residue and flammability management; supercritical CO2 is precise and residue-free but capital-intensive. This article compares the three methods head to head so buyers can match the method to the molecule instead of inheriting whatever the supplier happens to run.
1. How each solvent works on plant material
Water extraction is the oldest method: hot or ambient water dissolves polar constituents such as polysaccharides, some polyphenols, and organic acids, leaving lipophilic compounds largely behind. It is familiar, food-compatible, and simple in solvent handling, which is why traditional decoctions and many tea-type ingredients still use it. Its weakness is selectivity: water pulls a broad sweep of soluble matter, including starches and sugars that later burden concentration and drying.
Ethanol extraction widens the capture range. Alcohol-water mixtures dissolve both polar and moderately non-polar compounds, making ethanol the versatile middle ground for polyphenols, flavonoids, saponins, and many alkaloid fractions. Operators tune the ethanol-water ratio to aim at chosen compound classes, and the solvent evaporates more readily than water. The costs are flammability management, explosion-safe equipment, and residual solvent limits that every lot must meet.
Supercritical CO2 works on a different principle. Above its critical point, carbon dioxide behaves like a tunable solvent whose dissolving power shifts with pressure and temperature, selectively carrying lipophilic fractions such as essential oils and certain oleoresins. When pressure releases, the CO2 simply vaporizes, leaving no solvent residue. The price of this elegance is high-pressure equipment, significant capital investment, and operating expertise that fewer facilities possess.
2. Selectivity, yield, and what ends up in the powder
Selectivity determines whether the finished extract concentrates the desired compounds or merely concentrates everything. Water gives generous yields of total solids but dilute marker levels, which suits traditional whole-profile products rather than potency promises. Ethanol typically delivers higher marker concentrations across a wider compound range, which is why standardized extracts so often specify an ethanol-water process.
CO2 brings the sharpest selectivity for suitable targets. By adjusting pressure, operators can fractionate aromatic oils from heavier resins, producing clean lipophilic ingredients with minimal co-extracted bulk. But CO2 cannot do everything: highly polar actives dissolve poorly in it, so a polyphenol-rich root that extracts beautifully in ethanol may yield almost nothing in pure CO2. Matching solvent polarity to target chemistry is the fundamental design decision.
Yield figures need careful reading across methods. A high total-solids yield with low marker content means large doses and heavy sensory load; a modest yield with high marker concentration means efficient dosing. Specifications should therefore report both the concentration ratio and the marker assay with its method, so buyers compare potency rather than powder weight. This is where Refine Biology becomes a practical tool: the method name alone predicts much of what the certificate should show.
3. Head-to-head comparison
| Factor | Water extraction | Ethanol extraction | Supercritical CO2 |
|---|---|---|---|
| Best target chemistry | Polar compounds, polysaccharides, traditional profiles | Broad middle range, polyphenols, saponins | Lipophilic fractions, aromatic oils, oleoresins |
| Selectivity | Low; co-extracts sugars and starches | Medium; tunable by alcohol-water ratio | High for suitable targets; tunable by pressure |
| Solvent residue story | None; water is the benign default | Must verify ethanol within limits every lot | None; CO2 vaporizes on depressurization |
| Capital intensity | Low; simple vessels and heaters | Medium; explosion-safe design required | High; high-pressure systems and expertise |
| Energy profile | Heavy evaporation load for dilute liquors | Moderate; ethanol evaporates readily | Compression energy; no evaporation burden |
| Typical positioning | Traditional teas and gentle botanicals | Standardized broad-spectrum extracts | Premium clean-label lipophilic fractions |
Use the row that describes the target compound: the method in the winning column is usually the right starting point for trials.
4. Residues, safety, and documentation
Every method carries its own residue and safety story, and specifications must address it explicitly. Water extraction has the simplest narrative with no organic solvent to clear, though microbial control of aqueous liquors demands attention during processing. Ethanol extraction requires residual solvent verification against applicable limits on every lot, plus documented control of the flammability risks in the facility. CO2 extraction sidesteps solvent residues entirely, shifting the assurance burden to pressure-system integrity and fraction consistency.
Regulatory evaluations increasingly look at manufacturing and purity as a pair. EFSA assessments of food additives weigh identity, manufacturing process, and purity together, which reflects a broader principle buyers can apply to botanicals: the method belongs in the specification, not in a footnote. [1] Certificates should therefore name the solvent, state the extraction and standardization basis, and report residue results with limits rather than bare pass or fail notes.
Change control deserves emphasis because method switches happen quietly. A supplier moving a product from ethanol to water extraction, or adjusting the ethanol ratio, creates a materially different ingredient even when the plant source is unchanged. Buyers should require advance notice of any process change and re-qualify the ingredient with fresh specification review and stability checks.
5. Choosing the method for a real formula
Start from the target compounds and the finished product, not from available equipment. A traditional calming tea concept points toward water extraction and its familiar profile; a standardized polyphenol capsule points toward ethanol-water tuning; a clean-label aromatic oil points toward CO2. Ginger shows why this matching matters: the NCCIH ginger resource describes how different preparations carry different constituent profiles, so two ginger extracts made by different methods can behave like different ingredients. [2]
Then test economics at the serving level. CO2 extracts look expensive per kilogram until their concentration and clean sensory profile cut the inclusion rate; water extracts look cheap until evaporation energy and dose bulk erase the advantage. Pilot samples in the real matrix, compared on dose cost and stability rather than material price, give the honest answer.
Finally, confirm the partner can run the chosen method at scale with consistent documentation. Refine Biology applies separation and extraction expertise across its botanical range at three coordinated sites in China, so buyers can source water, ethanol, and advanced separation products with one quality system behind them. The right question is never which solvent is best in general, but which solvent delivers the target compounds reliably at the serving economics the product needs.
Sources
Frequently asked questions
Can a supplier switch extraction solvents without telling customers?
The method changes which compounds end up in the extract and at what concentration, so switching solvents can shift marker levels, impurity profiles, and sensory character. Any change needs re-validation of the specification, stability checks, and updated documentation. Buyers should treat an extraction-method change as a new ingredient qualification rather than an administrative update.
Which extraction method costs the least to operate?
Water extraction generally carries the lowest solvent cost and the simplest residue story, but it demands large evaporation energy and can co-extract sugars and starches. Ethanol sits in the middle on equipment and energy, with residue limits to manage. Supercritical CO2 needs the highest capital investment but leaves no solvent residue and offers tunable selectivity.
How should formulators match a botanical to an extraction method?
Aqueous extracts often suit traditional teas and gentle botanical profiles, ethanol extracts suit broad-spectrum markers including many polyphenols, and CO2 extracts suit lipophilic fractions such as aromatic oils. Ginger preparations illustrate the principle: the NCCIH ginger resource shows how preparation changes composition, so the target compounds should dictate the method rather than habit.
What belongs in an extract specification about the process?
Confirm the solvent, the extraction conditions in outline, the concentration or standardization achieved, and the residue testing with limits for every lot. EFSA additive evaluations consider manufacturing and purity together, which is why process transparency belongs in the specification. Consistent certificates across lots matter more than any single impressive assay value.