From Plant to Bottle: How Essential Oils Are Made and Why Extraction Matters

Put two bottles of lavender oil side by side. Same botanical name, same amber glass, same claim of purity. One costs nine dollars and one costs thirty-four. Open them and they smell like different plants, because in a meaningful sense they are.

The gap between them is not branding. It is a chain of decisions made before either oil reached a bottle: which plant, cut at which stage, held how long before processing, run at what temperature, for how long, in what equipment. Every one of those choices leaves a chemical signature. Learning to read them is the difference between buying on faith and buying on evidence.

What steam distillation actually does

Steam distillation produces most of the essential oil on the market, and the physics are more interesting than the usual description suggests.

Many aromatic compounds boil at temperatures that would wreck them. Linalyl acetate, the ester that gives lavender its sweetness, boils around 220 degrees Celsius on its own. Heat lavender to that point and you get something scorched and sharp. Steam distillation gets around the problem through co-distillation: when two liquids that do not mix are heated together, they vaporize at a combined pressure that reaches atmospheric well below either one boiling alone. Volatile plant compounds ride over with water vapor near 100 degrees, far below the temperature that would destroy them.

The rest is mechanical. Steam passes through loosely packed plant material, carries the oil-bearing vapor into a condenser, and the cooled liquid separates into two layers. The oil floats. The aromatic water beneath it is the hydrosol, a co-product rather than a waste stream, and increasingly a formulation ingredient in its own right.

Why run time changes the oil

Compounds do not all come over at once. Lighter molecules distill early, heavier ones late, so an oil pulled off after thirty minutes has a measurably different profile from the same batch run for three hours. Producers who cut runs short to save fuel get a thin, top-heavy oil.

Harvest timing matters as much. In lavender the oil sits mostly in the calyx rather than the showy petals. A 2021 study in Molecules measured 1.29 percent oil by weight in the calyx against 0.10 percent in the corolla, roughly thirteen times more. Cut at full bloom for the photographs and you lose oil. Cut at ten to twenty-five percent of florets open and you keep it.

The methods that are not distillation

Distillation cannot handle everything. Four other approaches fill the gaps, and each produces something chemically distinct:

• Cold pressing. Citrus peel oil is mechanically expressed because heat flattens the bright top notes that make citrus worth having. Pressed oil keeps more fresh-fruit character and has a shorter shelf life as a result.

• Solvent extraction. Jasmine and tuberose do not survive steam at all. They are washed with a solvent such as hexane to produce a concrete, then an absolute. An absolute is not technically an essential oil, and an honest supplier labels and prices it as a separate product.

• Supercritical CO2. Above 31.1 degrees Celsius and 73.8 bar, roughly 1,071 psi, carbon dioxide behaves as both gas and liquid. It penetrates plant tissue like a gas and dissolves compounds like a solvent, then evaporates away cleanly when pressure drops. Because it works at 35 to 50 degrees, it preserves compounds that heat destroys.

• Enfleurage. Fat-based cold extraction, largely historical, though a few artisan producers still use it for delicate florals.

The differences are not academic. Chamomile is the clearest case. Matricine in the fresh plant converts irreversibly to chamazulene under prolonged heat, which is why steam-distilled chamomile is deep blue and a CO2 extract of the same flowers is not. Neither is wrong. They are different materials, and a formulator should know which is in the bottle.

Why two bottles of lavender are not the same oil

Species and chemotype

ISO 3515, the international standard for true lavender oil, sets compositional ranges: linalool 25 to 38 percent, linalyl acetate 25 to 45 percent, camphor at or below roughly 1 percent. Lavandin, the hybrid grown for volume, has its own standard under ISO 8902, which allows camphor at 6 to 8 percent. That one number is why lavandin smells sharper, and why it cannot quietly stand in for lavender without showing up on analysis.

Chemotype cuts deeper than species. Rosemary comes in 1,8-cineole, camphor and verbenone forms, all sold as Rosmarinus officinalis and all behaving differently in a formulation. Thyme splits into thymol, carvacrol, linalool and thujanol types with sharply different irritation profiles. For anyone formulating rather than simply enjoying a scent, chemotype matters more than the Latin binomial.

What the label does not tell you

The phrase therapeutic grade has no regulatory standing. No government body defines it, no agency certifies it, and any brand may print it. The same is true of clinical grade and certified pure. These are marketing terms, not quality tiers.

Real documentation looks different. A batch-specific certificate of analysis whose number matches the bottle in your hand, the full botanical name, plant part, extraction method, country of origin, harvest date, and where relevant the chemotype. A generic analysis posted once on a website covers a batch that may be years gone.

How oils get adulterated, and how it is caught

Adulteration runs from crude to sophisticated. At the simple end, oils are cut with odorless carriers or extended with cheaper relatives: lavandin into lavender, cornmint into peppermint, geranium into rose. At the other end, synthetic isolates are added to bring a weak natural oil up to spec.

This is where consumer understanding of testing breaks down. A standard GC/MS report shows which compounds are present and in what proportion, which catches crude substitution. It cannot distinguish natural linalool from synthetic linalool, because they are the same molecule.

Chiral GC does. Many plant compounds exist as mirror-image forms, and plants build them in characteristic ratios while laboratory synthesis produces roughly equal amounts of both. Council of Europe guidance holds that authentic lavender should not exceed 12 percent S-linalool or 1 percent S-linalyl acetate. Cold-pressed bergamot runs above 99.9 percent optical purity on its linalyl acetate. A synthetic addition shifts those ratios even when the total percentage sits comfortably inside the ISO range.

Price is a chemistry signal

Yield explains most of what a bottle costs, and the spread is enormous. Rose otto runs 0.02 to 0.03 percent, which works out to between 3,500 and 5,000 kilograms of petals for one kilogram of oil. Bergamot needs roughly 200 kilograms of fruit per kilogram. Lavender manages 0.5 to 1.5 percent. Sweet orange is nearly a by-product of the juice industry.

Those ratios are fixed by botany, so they show up in price whether a brand likes it or not. When every oil across a range carries the same price, the expensive ones have been diluted or substituted. A cheap rose otto is not a bargain. It is a different product wearing the name.

For consumers and small wellness brands, understanding the steam extraction of essential oils makes it easier to evaluate how botanical freshness, temperature control, and equipment quality shape the aroma and purity of the finished product.

It is also why a growing number of small brands now distill in house, at least for signature materials. A copper alembic in the five to twenty liter range lets a formulator control harvest window, run length and temperature directly, and produces hydrosol alongside the oil. Copper is traditional for good reasons. It conducts heat evenly enough to avoid scorching, and it binds sulfur compounds that would otherwise dull the finished aroma.

Freshness is a safety question, not only a quality one

This is the part most often left out of clean beauty conversations.

Linalool and limonene, two of the most common constituents in natural oils, are weak sensitizers in fresh material. Exposed to air they autoxidize into hydroperoxides, which are potent ones. Karlberg and colleagues showed in 1992 that air oxidation of d-limonene creates allergens the fresh compound does not contain. A patch test study across six countries found that among 2,900 consecutive dermatitis patients, 281 reacted to oxidized limonene or oxidized linalool. Clinical reviews put it at around 10 percent of patients investigated for suspected contact dermatitis.

So an oil can become a skin risk it was not when bottled, and a product formulated with old oil carries that risk forward. Handling follows from the chemistry:

• Store in amber or cobalt glass, never plastic, which terpenes slowly degrade.

• Keep headspace minimal. Decant into smaller bottles as stock runs down, since oxygen in the bottle drives the reaction.

• Keep oils cool and dark. Refrigeration extends life considerably for citrus.

• Treat citrus as the shortest-lived category, conifers and delicate florals next, woods and resins as the most stable. Some resinous oils improve with age.

Phototoxicity, and why cold-pressed citrus needs care

Cold-pressed citrus oils carry furocoumarins, principally bergapten, which react with UVA and can burn exposed skin badly. IFRA and EU rules cap bergapten at 15 parts per million in leave-on products for sun-exposed skin, which for bergamot works out to roughly 0.4 percent in the finished formula. Furocoumarin-free bergamot, labeled FCF, is made by vacuum distillation that strips the phototoxic fraction while keeping most of the aroma.

Steam-distilled citrus oils do not carry the same risk, because furocoumarins are too heavy to come over with the steam. It illustrates the wider point neatly. Extraction method determines the safety profile, not only the smell.

What to ask before you buy

Five questions separate a supplier who knows their material from one reselling drums:

• What is the full botanical name, and where relevant, the chemotype?

• Which extraction method, and which plant part?

• Country of origin and harvest year?

• Can you send the batch-specific certificate of analysis matching this bottle?

• Has the oil been tested chirally, or by GC/MS alone?

A supplier who answers all five without hesitation is worth keeping. One who responds with grade language and nothing else has told you what you need to know.