STEAM DISTILLATION

What is steam distillation?

According to the review by Travis et al. (2026), steam distillation is a fundamental technique for isolating volatile natural products, especially essential oils, based on the principle that steam lowers the effective boiling points of immiscible volatile compounds through co-distillation. The review states that this allows extraction at temperatures below thermal degradation thresholds, the compounds vaporising together with water while heat-sensitive constituents are preserved.

What does a laboratory apparatus consist of?

The typical apparatus described by Travis et al. (Figure 14 of the review) comprises a round-bottom flask heated by a mantle, a distillation chamber packed with ground plant material, a condenser and a receiver. Steam generated from the boiling water passes through the plant matrix, ruptures the oil glands and volatilises the essential oil components. The mixed vapours condense into a biphasic distillate, a lighter essential oil layer above an aqueous hydrosol phase. According to the review, polar or water-soluble compounds “may” consequently be lost to the aqueous layer or to the surfaces of the apparatus.

How does it differ from hydrodistillation?

Wainer et al. (2022) write that both methods, hydrodistillation and steam distillation, use heat to vaporise the oil and let it leave the plant, the water and the oil separating through their difference in density. In hydrodistillation the lavender is placed in boiling water, whereas in steam distillation only the steam passes through it; the authors describe both methods as energy-intensive and time-consuming.

In the paragraph of their introduction devoted to hydrodistillation, Politeo et al. (2023) write that elevated temperatures and longer extraction time “may” lead to losses or chemical modifications of essential oil components.

How widely is it used?

According to Wainer et al. (2022), steam distillation is the extraction method currently used for most commercial essential oil production.

What does the European Pharmacopoeia list?

In the knowledge database of the European Pharmacopoeia (EDQM), monograph 02098 has the status “In use”; its titles are “Essential oils” in English, “Huiles essentielles” in French and “Aetherolea” in Latin. The same record carries the mention Pharmeuropa 32.2 and gives edition 10.7 as the last publication, in English as in French. The record also lists two on-going revisions, each at the state “1 - Draft”: one described as a “Revision to clarify requirements regarding the herbal starting material”, the other as a “Revision chromatographic profile”.

What did one industrial-scale study on Myrtus communis measure?

Kaya et al. (2020) used a factorial design to assess, at industrial scale, the effect of three variables on the steam distillation of essential oil from the herbs of Myrtus communis L. (name used by the source): boiler occupancy rate (50, 75 and 100 %), distillation duration (60, 75 and 90 min) and particle size (10, 20 and 30 mm, cut with a guillotine). The measured variable was the volume of oil obtained, in mL. The myrtle herbs were harvested during the full bloom period, then dried in the shade at room temperature.

What did the installation look like?

The distillation equipment comprised four units, two active and two passive: while the active units were distilling, the passive ones were emptied, cleaned and filled, then took over, which made continuous 24/7 distillation possible. Each unit was 3 m high with a water capacity of 500 L, and the water used was free of lime and heavy metals. An automation program recorded the pressure, temperature and time to which the plant material was exposed during distillation, and these “could” be traced back afterwards. According to the authors, a low temperature (98 °C) and a water vapour pressure of 1 bar were used to keep the quality and yield of the oil high, and steam at 101 °C circulated between the two walls to prevent the inner wall from cooling and from repeated cycles of evaporation and condensation.

Which variable changed the oil volume?

In this industrial-scale study, the oil volume rose by 171.10 % (from 275 mL to 745 mL) when the boiler occupancy rate went from its minimum to its maximum level at minimum distillation duration. It rose by only 4.11 % (from 730 mL to 760 mL) when the distillation duration went from its minimum to its maximum level at maximum occupancy rate. The Taguchi method, added to the experimental design, led the authors to select as optimum conditions an occupancy rate of 100 %, a distillation duration of 75 min and a particle size of 20 mm. Run 10, under these conditions, gave 750 mL of oil observed (793 mL predicted); its response was not the highest, but the most robust, stable and insensitive to noise factors.

What did the oil contain?

In the oil obtained under the optimum conditions, GC-MS analysis showed two predominant components, α-pinene (33.14 %) and eucalyptol (55.09 %). The other compounds above 1 % were β-fenchyl alcohol (3.20 %), α-terpinenyl acetate (1.93 %), linalool (1.79 %) and the acetoacetic ester of 3(10)-caren-4-ol (1.08 %), the remaining identified components totalling 3.77 %.

What did a semi-industrial study observe, species by species?

In the study of Léva and Gál (2026), the plant material of the seven species was harvested between June and August 2025, each at a specific organ: fully expanded leaves for Mentha piperita (accepted in WCVP as Mentha × piperita L.) and Mentha spicata (accepted in WCVP as Mentha spicata L.); fresh flowering tops for Lavandula angustifolia (accepted in WCVP as Lavandula angustifolia Mill.), Salvia officinalis (accepted in WCVP as Salvia officinalis L.), Hyssopus officinalis (in WCVP a synonym of the accepted Dracocephalum officinale (L.) Y.P.Chen & B.T.Drew) and Achillea millefolium (accepted in WCVP as Achillea millefolium L.); and terminal twigs bearing fresh needles for Picea abies (accepted in WCVP as Picea abies (L.) H.Karst.).

What equipment was used, and when did a run stop?

The steam distillation was carried out with hand-operated, non-automatic, semi-industrial dry steam distillation equipment made of food-grade stainless steel. It had four main components: a 150 L wood-fuelled steam generator, a 600 L tank for the plant material, a 300 L condenser with a spiral condensation tube about seven metres long, and a Florentine-type separator for the essential oil and the floral water. Depending on the species, the fully filled 600 L tank held around 140–170 kg of plant material. As the equipment is fully manual, the steam flow rate could only be influenced by the quantity of wood burned, and it was about 9–10 L/h in every case. The equipment has no integrated digital sensors to log pressure and temperature directly; the operator adjusted the furnace draft to hold a condensation flow rate of 9.5 ± 0.5 L/h, which the authors use as a proxy for a uniform steam flow rate and thermal equilibrium across all extractions. Each run was stopped on an operational endpoint: the moment the volume of essential oil in the Florentine flask did not increase by more than 0.1 mL over a continuous 30-min observation window.

How long did the runs last, and what did they yield?

All laboratory hydrodistillations of the study lasted 90 min, whereas the semi-industrial steam distillations lasted from 2 to 7 h. The authors report that laboratory hydrodistillation of dried biomass gave significantly higher essential oil yields (up to 4.5 times greater in Mentha piperita) and shorter extraction times (1.5 h) than semi-industrial steam distillation of fresh material (2–7 h). They largely attribute this discrepancy to the removal of moisture during drying, which concentrates the secondary metabolites relative to the total mass. In Table 1 of the study, semi-industrial dry steam distillation of fresh material gave the following distillation times and essential oil yields, in mL per kg of fresh material: Picea abies, freshly ground needles and small branches, 4–5 h, 3.26; Achillea millefolium, flowers and stems of about 20 cm, 5–7 h, 0.92; Lavandula angustifolia, fresh aerial parts, 2 h, 10.31; Hyssopus officinalis, fresh aerial parts, 2 h, 1.27; Salvia officinalis, fresh aerial parts, 2 h, 2.93; Mentha piperita, fresh aerial parts, 2 h, 4.76; Mentha spicata, fresh aerial parts, 2 h, 3.32. In the same table, laboratory hydrodistillation of dried, ground material, with a distillation time of 1.5 h for all seven species, gave the following essential oil yields, in mL per kg of dried material: Picea abies, ground dried needles and twigs, 3.71; Achillea millefolium, dried ground flowers, 4.29; Lavandula angustifolia, 15.71; Hyssopus officinalis, 4.29; Salvia officinalis, 11.14; Mentha piperita, 21.43; Mentha spicata, 14.29 (these last five as dried ground aerial parts).

How did the two kinds of oil differ overall?

The authors state that the chemical variations observed between their hydrodistilled and steam-distilled oils stem from a combination of overlapping processing variables: extraction scale, biomass moisture content (dried or fresh), degree of matrix fragmentation (ground or intact) and distillation kinetics over time. In their abstract, they report that GC-MS profiling showed hydrodistillation consistently yielding a more chemically diverse volatile profile than steam distillation. They write that their results “prove a universal physical trend”: hydrodistillation significantly enriched low-boiling oxygenated derivatives (oxygenated monoterpene alcohols and oxides), while steam distillation selectively preserved heavier, thermally sensitive hydrocarbon fractions, across all the taxonomic groups studied. According to the authors, the direct contact between water and plant material in hydrodistillation creates a harsher extraction environment that “can” promote the thermal degradation and hydrolysis of sensitive volatile compounds, which are better preserved under the milder conditions of steam distillation; for this reason, they write, steam distillation is the preferred procedure for obtaining essential oils industrially. In their results, the area percentages of oxygenated monoterpenes, sesquiterpenes and oxygenated sesquiterpenes were in general higher in the hydrodistilled oils, a substantial number of compounds being even absent from the steam-distilled oils; the opposite held for monoterpenes.

Picea abies

In the semi-industrial dry steam distillation of Léva and Gál (2026), the needles of Picea abies (name used by the source) are covered with epicuticular wax and the essential oil is stored in internal resin canals, so that the steam must first break down the waxy layer and penetrate the plant tissue. According to the authors, steam distillation is thus prolonged to about 4–5 h compared with hydrodistillation, after which no substantial amount of oil was collected. The main components of the steam-distilled oil were α-pinene, camphene, β-pinene, β-myrcene, β-phellandrene, D-limonene, borneol and bornyl acetate, with a combined chromatographic area of about 82 %; they were also present in the hydrodistilled oil, with up to five-fold lower area. Thunbergol was at 0.15 % in the steam-distilled oil and at 5.09 % in the hydrodistilled oil.

Achillea millefolium

For Achillea millefolium (name used by the source), cultivar ‘Proa’, described by the authors as a chamazulene-rich chemotype, chamazulene forms during heat treatment through the degradation of the naturally present matricin, which makes the distillation of yarrow the longest of the study, at least 5–7 h. The hydrodistilled oil contained 20.45 % chamazulene and the steam-distilled oil only 9.45 %, although the steam distillation had lasted at least 3.5 times longer. The authors judge three explanations plausible for this lower content: the installation did not create the optimal conditions for matricin to decompose; the steam temperature was not high enough to promote that decomposition; or the spiral tube of the heat exchanger, about 7–8 m long with a slope of about 10°, retained on its wall a large quantity of viscous chamazulene, which “could not” flow down fast enough by gravity and was missing from the collected oil.

Hyssopus officinalis

For Hyssopus officinalis (name used by the source), isopinocamphone (cis-pinocamphone) had a relative abundance of 39.57 ± 0.0495 % in the steam-distilled oil and 39.58 ± 0.0424 % in the hydrodistilled oil, and trans-pinocamphone of 8.11 ± 0.1202 % and 8.12 ± 0.1131 % respectively. The steam-distilled oil contained hedycaryol (2.17 %) and elemol (0.12 %), the hydrodistilled oil only elemol (10.4 %).

Lavandula angustifolia

For Lavandula angustifolia (name used by the source), chromatographic analysis revealed over ten additional compounds in the hydrodistilled oil compared with the steam-distilled oil. The steam-distilled oil had a full, pleasant, sweet lavender scent, while the hydrodistilled oil had a strong, fresh grassy smell alongside the lavender scent.

What changes with superheated steam?

In the steam distillation set-up of Raza et al. (2024), water in an electrically heated round-bottom flask produced the steam, and 100 g of seeds of Foeniculum vulgare Miller (name as written by the source; accepted in WCVP as Foeniculum vulgare Mill.), in a cotton bag, were placed in a biomass flask set vertically above a Clevenger-type apparatus. The steam passed through the glandular trichomes of the seeds and extracted the oil, which was separated from the hydrosol in a separating funnel. The comparison hydrodistillation used 300 g of seeds (mesh size 80) mixed with 3 L of distilled water in a Clevenger-type apparatus, heated by an electric mantle for 3 h. The superheated steam distillation equipment comprised a stainless-steel superheated steam generator, biomass extraction chamber, condensers and chiller, and a glass hydrosol collection vessel. In it, 300 g of biomass were exposed for 60 min to steam generated at 75 psi and 150 °C, its temperature monitored by thermocouples, and the extraction was repeated five times.

What did the three methods yield from the seeds?

According to Raza et al., superheated steam distillation gave the largest quantity of essential oil from the seeds (5.24 %), against 3.47 % by steam distillation and 2.47 % by hydrodistillation. The authors write that superheated steam distillation at 150 °C yielded 53 % and 45 % more oil than hydrodistillation and steam distillation respectively, over a 60-min period.

How did the composition differ?

The oils obtained by hydrodistillation, steam distillation and superheated steam distillation contained 23, 20 and 17 identified compounds respectively. In all three, the major compounds were trans-anethole (58.48–66.46 %), followed by fenchone (8.4–9.8 %), estragole (6.23–7.13 %) and limonene (6.5–7 %). In Table 1 of the study, trans-anethole was at 58.48 ± 0.03 % by hydrodistillation, 62.38 ± 0.02 % by steam distillation and 66.46 ± 0.01 % by superheated steam distillation. β-Ocimene, fenchyl acetate, β-caryophyllene oxide, β-farnesene, α-copaene and phytol were absent from the oil obtained by superheated steam distillation. According to the authors, “it is possible” that these compounds were denatured by the high temperature of the superheated steam, but the concentration of each compound also depends on its volatility, its solubility in steam and the temperature and pressure of distillation.

What did a laboratory study on lavender compare?

Wainer et al. (2022) give two names to the same plant material, Lavandula × intermedia ‘Margaret Roberts’ and Lavandula × heterophylla, without choosing between them (in WCVP, two distinct accepted names: Lavandula × intermedia Emeric ex Loisel., L. angustifolia × L. latifolia, and Lavandula × heterophylla Viv., L. dentata × L. latifolia). The lavender was briefly kept in vases of water after harvest, then, shortly before the experiments, air-dried in the shade for a few days, cut into pieces of about 1 cm and weighed. The spike (flowers) and about 30 cm of stem were used, and all experiments were carried out in triplicate.

What did the laboratory set-up look like?

The steam distillation set-up was a two-neck round-bottomed glass flask with one opening at the top and one at the bottom, the bottom opening covered with mesh; 80 g of cut lavender were loaded through the top. Distilled water was boiled and the steam directed into the bottom opening; the oil-rich steam leaving the top passed through a Liebig condenser, then into a separator funnel. Each extraction ran until 200 mL had been deposited in the separator, which then held the oil floating on the hydrosol (aromatic water).

How long did extraction take, and how much energy did it use?

The extraction time, measured from room temperature until the separator reached 200 mL and excluding any pre-treatment time, averaged 57 min for steam distillation, 51 min for hydrodistillation and 49 min for cellulase-assisted hydrodistillation. The average energy consumption, estimated from the heater (assumed to run at 70 % of its maximum power rating) and the mixer, was 15.0 kJ/g for steam distillation, 13.4 kJ/g for hydrodistillation and 30.8 kJ/g for cellulase-assisted hydrodistillation. The authors write that hydrodistillation required slightly less energy than steam distillation because of its shorter extraction time.

What did the oils contain, and how did they smell?

Oils from hydrodistillation and from cellulase-assisted hydrodistillation contained more 1,8-cineole and linalool than oils from steam distillation, while the steam-distilled oil contained more camphor. The retention time reported for linalool was between 17.543 and 17.554 min, and that reported for terpinolene between 17.531 and 17.541 min; the authors write that these very similar values “suggest” that the linalool peak on the chromatogram “could have been” erroneously interpreted as terpinolene. In Table 1 of the study, steam-distilled samples 1 and 2 contained 18.55 and 14.18 wt% 1,8-cineole and 28.34 and 30.68 wt% camphor; sample 3 has no result, the analytical equipment having given obviously incorrect results, and the analysis “could not” be repeated. Hydrodistilled samples 4 to 6 contained 13.96, 15.6 and 27.91 wt% 1,8-cineole and 24.82, 36.74 and 24.97 wt% camphor. Hydrodistillation and cellulase-assisted hydrodistillation showed larger standard deviations than steam distillation, which gave more consistent results and was less sensitive to small changes in experimental parameters. The steam-distilled oils had the highest camphor content and a more “plant-like” smell, and were rated on average 3 out of 5, as were the hydrodistilled oils.

References

• Travis C.R., McMaster J., Rivas F., « Advances in Natural Product Extraction: Established and Emerging Technologies », Molecules 2026;31(7):1136, doi:10.3390/molecules31071136 — https://www.ebi.ac.uk/europepmc/webservices/rest/PMC13074414/fullTextXML
• Léva N., Gál E., « Impact of Extraction Scale and Method on the Chemical Profile of Essential Oils: A Comparative Study Between Laboratory Hydrodistillation and Semi-Industrial Dry Steam Distillation », Molecules 2026;31(12):2105, doi:10.3390/molecules31122105 — https://www.ebi.ac.uk/europepmc/webservices/rest/PMC13305851/fullTextXML
• Kaya D.A., Ghica M.V., Dănilă E., Öztürk Ş., Türkmen M. et al., « Selection of Optimal Operating Conditions for Extraction of Myrtus communis L. Essential Oil by the Steam Distillation Method », Molecules 2020;25(10):2399, doi:10.3390/molecules25102399 — https://www.ebi.ac.uk/europepmc/webservices/rest/PMC7287582/fullTextXML
• Wainer J., Thomas A., Chimhau T., Harding K.G., « Extraction of Essential Oils from Lavandula × intermedia ‘Margaret Roberts’ Using Steam Distillation, Hydrodistillation, and Cellulase-Assisted Hydrodistillation: Experimentation and Cost Analysis », Plants 2022;11(24):3479, doi:10.3390/plants11243479 — https://www.ebi.ac.uk/europepmc/webservices/rest/PMC9785002/fullTextXML
• Raza M.H., Ayub M.A., Zubair M., Hussain A., Saleem S. et al., « Comparative Study of Essential Oils Extracted From Foeniculum vulgare Miller Seeds Using Hydrodistillation, Steam Distillation, and Superheated Steam Distillation », Food Science & Nutrition 2024;12(12):10535, doi:10.1002/fsn3.4593 — https://www.ebi.ac.uk/europepmc/webservices/rest/PMC11667025/fullTextXML
• EDQM, European Pharmacopoeia, monograph details 02098 « Essential oils » (Aetherolea), consulted 02/10/2026 — https://extranet.edqm.eu/4DLink1/4DCGI/Web_View/mono/2098
• Politeo O., Popović M., Veršić Bratinčević M., Koceić P., Ninčević Runjić T. et al., « Conventional vs. Microwave-Assisted Hydrodistillation: Influence on the Chemistry of Sea Fennel Essential Oil and Its By-Products », Plants 2023;12:1466, doi:10.3390/plants12071466 — https://www.ebi.ac.uk/europepmc/webservices/rest/PMC10097284/fullTextXML
• Govaerts R. (ed.), World Checklist of Vascular Plants (WCVP), Royal Botanic Gardens, Kew — https://sftp.kew.org/pub/data-repositories/WCVP/wcvp.zip