Choosing a plant source...
Do some research. Go to the library. Better yet, go to a chem library at university. The data for many essential oils are in specialized chemistry books. You must do the research.

One idea is to try distilling the essence of cumin seeds, but do this only if you want the whole place to smell intensely like cumin for days . It's hard to overstate the intensity of the odor. Even if you love curry, you might lose all desire to eat curry for a while.

The aroma chemical is called cuminaldehyde. It's an oily liquid, p-isopropylbenzaldehyde. See Introduction to Organic Laboratory Techniques, 3rd Ed., by Pavia et al., 1988.

Extracting Essential Oils and Fragrances

Fig. 1. Syringa vulgaris, or common lilac, blooms for only a short time in late April or early May. It has a pleasant fragrance.

Different fragrances can have different degrees of stability to heat and oxygen. Some will distill effectively; some won't.

CAUTION: You alone are responsible for safety.

Information is provided for educational purposes only. You assume ALL responsibility for consequences that may arise.

If you choose to attempt anything described on this site, you do so entirely at your own risk. Please review the Terms of Use for this web site.

This article is subject to copyright. If you see it appear on another site, in whole or in part without due credit, please let us know .

Terminology / Definitions
Background Theory
Steam Distillation
Equipment Setup
Temperature Control
Some Fragrances Won't Distill
Other Considerations

Terminology / Definitions

Condenser - the long, often corkscrew-shaped passage that recondenses the steam and volatilized oils. It is one component of a distilling apparatus. Typically, a laboratory condenser has two side tubulations. Water hoses are connected to these; one goes to the faucet, the other to the drain. When the faucet is turned on, cool water runs through the outer jacket of the condenser.

Distillate - the liquid that ends up in the receiving flask when the steam (etc.) turns back to liquid in the condenser.

Distilling flask - the source vessel, containing the herb / spice and the water to be heated. Technically, in chemistry the term "distilling flask" refers to a round-bottomed flask with a long neck and a long side arm or tube.

Hydrosol - a colloidal dispersion of essential oil droplets in water: in other words, an emulsion. An oil-water dispersion may or may not separate by itself, depending on the stability of the emulsion. Thus, the hydrosol is not necessarily a by-product, but instead a crude product; here, it is simply the raw distillate.

In the case where a distinct oil layer separates after standing, the term refers to the watery liquid that remains behind. Some fragrance components are about as soluble in water as they are in oil, meaning some will remain in the aqueous layer after spontaneous separation of the oil layer. Again, the stability of the emulsion plays a role here. Because the distillate will almost certainly contain polar organic compounds, it's likely that some proportion of oil will remain permanently dispersed in the water layer. This would be a more or less permanent hydrosol.

Maceration - an extraction carried out by crushing the plant material and soaking in liquid such as water, glycerin, ethanol, or vegetable oil. This is normally carried out well below boiling temperature; however, hot liquids may be used to help break cell membranes and release the components. There is not a precise definition in this regard, since the word goes back at least to the 1400's.

Normally, maceration is for fragrances that don't withstand steam distillation. The more chemically "fragile" fragrances tend to be more stable in colder liquids, but solubility also decreases. The best extraction seeks a balance between solubility and heat-stability; however, one way to offset the heat problem is simply to soak materials for longer. Another is to use a homogenator (laboratory blender) or high-powered ultrasound device to rupture cells.

Solute - the solid, liquid, or gas that is dissolved in the solvent. Salt can be a solute in water; CO₂ can be a solute in water; fragrances can be solutes in alcohol; etc.

Solvent - the liquid that does the dissolving. In the broadest sense, a solvent needn't always be a liquid. Furthermore, in cases where multiple liquids are involved, the line between "solute" and "solvent" isn't always clear.

Source vessel - see "distilling flask". In traditional distilling terminology, this is often called the "distilling pot" or simply the "pot". However, since various bad actors have so thoroughly corrupted the King's English, the word "pot" has taken on a rather unfortunate connotation having nothing to do with a container. Corruption of language is one of the devil's many tools.

Background theory

Let us think back to inorganic chemistry. Supposing there were a solution containing dissolved salt in water, and supposing that solution were boiled, the dissolved salts would remain behind while the water would evaporate. In other words, the solvent would evaporate, while the solute would stay behind in the distilling flask. Thus, the distillate would be pure water.

Recall now fractional distillation from organic chemistry. When the solute is a liquid, such as alcohol, and the solvent is another liquid, such as water, the components can be separated by virtue of their different boiling points. As you may also recall, however, certain liquids will form azeotropes with water, or sometimes with each other. In practical terms, that means the same composition will carry over into the receiving flask. In other words, distilling such solutions will not separate them unless we take special steps to "break" the azeotrope. Obviously, we must accomplish this before or during distillation, not afterward.

An alcohol-water mixture might begin to separate as desired, until we arrived at some azeotropic ratio of alcohol to water. From that point onward, the rest of the distillate would be identical in composition to the liquid in the source flask.

Steam Distillation

You may recall that many essential oils and fragrances are steam-volatile. In other words, the steam carries with it appreciable amounts of the desired compound.

One can pass steam through a container full of plant material, but the simplest way is to put the crushed plant material in water and boil it. The steam will carry away steam-volatile components.

Earlier we discussed azeotropes. We can think of such steam-volatile compounds as forming azeotropes as well.   There is no easy way to make the essential oil compounds come over separately from the water.

One might think it possible to attempt dry distillation of the plant matter, but there's a problem with that. Many of the fragrant compounds (e.g., aldehydes) are susceptible to oxidation; the temperatures involved in dry distillation would ruin these. In fact, the lack of temperature control can easily lead to charring of the plant material.   Dry distillation of plant matter can easily become pyrolysis, which entails permanent chemical changes that don't make very nice fragrances. It's probably a better idea all-around to use steam.

One can simply put some of the crushed herb in water and boil it. That is the most ancient method. When the steam re-condenses, the resulting liquid (i.e., the distillate) will also contain the essential oil.   This oil may remain inseparable by mechanical means, or it may partially separate on standing.   Don't forget that essential oil often represents only a tiny portion of the total weight of plant matter. It may require a kilogram or more of starting material just to end up with a couple milliliters of oil. The oil will not separate in a recoverable manner if there simply isn't enough of it.

Spontaneous separation of an oil layer is less probable when the oil contains polar organic compounds such as aldehydes, ketones, alcohols, and carboxylic acids. Chances are that it will! Some of these are the fragrant compounds themselves; others are nuisance components that do little else but hold the desired stuff in emulsion with the water. Either way, one can achieve a purified essential oil only by resorting to additional steps. In the organic chem lab (and also industrially), one would use a hydrophobic solvent. Small-scale work calls for a separatory funnel, shaking some distillate together with a solvent such as methylene chloride (dichloromethane, CH₂Cl₂). The methylene chloride layer, which separates readily from the water layer, would contain the essential oil. The organic layer could be evaporated on a slow steam bath in a fume cupboard. The less-volatile essential oil would remain behind, while the dichloromethane would evaporate.

Any organic solvent that won't dissolve appreciably in water could, in theory, be used for solvent extraction of the distillate. It is helpful to know the partition coefficient of the desired solute; that will be particular to the solvent system you're planning to use. Someone, somewhere in the vast chemical literature may have found this coefficient already.... or not. Also, the choice and grade of solvent depends on what you're going to do with the oil. It is up to you to know the appropriate solvent. We cannot recommend anything in particular. Please consult the literature. Keep in mind that industrial extractions often rely on solvents that wouldn't be appropriate elsewhere. Industry has developed reliable ways to remove solvents such as methanol (etc) that would otherwise present a serious toxicity problem. That's why, at least traditionally, many of the flavorings in use by the food industry have been extracted with solvents that don't belong anywhere near your food. See for example the Handbook of Solvent Extractions by Lo, Baird, and Hanson.

A few herbs and spices that lend themselves to steam distillation:

Caraway seeds contain d-carvone, a pungent ketone having a cyclohexene ring.
Cinnamon - if the small excess of cinnamon in last year's batch of pickled watermelon rinds is any indicator, cinnamon oil is hot stuff. Don't taste the distillate.
Cumin - source of cuminaldehyde... if you want the place to smell like "curry x 1,000".
Ginger - chop up some fresh ginger into tiny pieces and try distilling.
Lavender - a very popular choice for steam distillation
Lilac (Syringa vulgaris) - use the flowers. According to an article on Bojensen.netand the diagrams there, the primary fragrance components don't appear to have ester linkages. Something decent should survive, perhaps being amenable to separation with a column. Whether it will smell exactly like lilac is another matter.
Orange - the citrus oils are present mainly in the "zest" of a citrus fruit.
Peppermint - the smell is pungent ; according to legend, rats and mice will avoid the odor of peppermint.
Spearmint and Catnip are also in the mint family.

While we're on the subject: the concentrated oil of herbs and spices is not necessarily edible. Peppermint oil is a good example (don't eat it!). When in doubt, don't taste it. Do your research on this.

Equipment Setup

Take the time to set up your glassware properly. This may require some minor construction. Plan it out carefully; do not take shortcuts here. Stability of the setup is crucial. You are expected to know how to achieve this properly. Following are some reminders.

Because a Graham condenser has to be set up in the vertical position, there will be a significant height difference between one side and the other (Figure 1). The distilling half of the setup will be much higher off the bench than will the receiving half of the setup. A column (Vigreux, Hemple, etc.) would make up for some of the height differential, but this would change the composition of the distillate by fractionating one or more components away from others. That may or may not be desirable, depending on the fragrance(s) being sought.

Figure 1. Temporary setup of the glassware, only to illustrate the height differential.

To set up the apparatus for actual use, build a sturdy riser or shelf to elevate the left-hand support stand.

A C140 or C150 support stand is too short by itself. Figure 1 shows an extra-tall stand (C160), but even this ought to have a 2- to 6-inch riser built beneath it.

The base of a C140 would have to be elevated 10 to 12 inches off the bench.

Anchor all supports firmly!

It's possible to use a support stand mounted to a permanent riser or shelf, such that ten inches (approx.) will be added to the effective height.

The riser could be a block, a shelf, a stack of 2x8 sections (secured with nails or bolts) or something of that nature. Don't use a cardboard box. Don't use something fragile. Don't use anything that could be knocked over easily.

The author in the past has constructed various supports out of 2x4's, 2x8's, and so forth. Even short legs should have diagonal supports to prevent racking. If you don't have good carpentry skills, find someone who does.

Alternatively, procure a 1/2" steel rod at least 26 to 30 inches tall. Build this upright rod permanently or semi-permanently into your lab bench. Here, once again, you may need to build some kind of a riser to support your heat source; perhaps the upright rod could pass through a hole in the riser.

The right-hand portion of the apparatus (according to Figure 1, above) can be held with a support stand and clamps; it should not require any special riser. This half of the setup includes the Graham condenser and the receiving flask. You may need two clamps to hold these securely. Remember that a Graham condenser must be vertical to work properly. Don't forget to anchor the base of your support stand to the lab bench.

The end of your lab bench should have a sink for the condenser cooling hoses. A laundry sink typically has a standard hose thread on the outside. Obtain a hose barb (hardware store) and run some Tygon tubing from the faucet to the upper tubulation of the Graham condenser.   Run another piece of Tygon tubing from the lower tubulation into the sink drain. Make sure the cooling hoses will not come loose and spray water on your hot glassware!

Using (e.g.) a Vigreux or Hemple column in the setup will take away much of the height differential, but the Vigreux especially will alter the ratio of products in the receiving flask. Such a column might be useful to try separating the products later, especially if there are several fragrance components having different boiling point.

Temperature Control

When distilling essential oils, do not let the source vessel evaporate to dryness. This can cause charring of the plant material, which would ruin the product.

We have recommended on our web site to use a sand bath for many heating purposes, but it's better to use a boiling water bath or steam bath for essential oil recovery. The flask containing the herbs and the water is clamped in place so it's about an inch from the bottom of a lightweight metal pot. That pot is then filled with water. As the water in the metal pot boils, the heat transfers to the distilling flask. This limits the temperature in the distilling flask, ensuring that your herbs or spices will not become charred.

The ancients would have been able to use some form of the water bath, and it's likely they did. While you might simply try to make sure the source vessel never runs dry, there's always the possibility of isolated pockets of herb / spice / plant matter that remain dry. If any of these be directly against the glass, they could overheat. This could contaminate your fragrance with burnt odors. A water bath eliminates this danger.

Some Fragrances Won't Distill

As mentioned earlier, it's important to remember that certain fragrances cannot withstand steam distillation. Jasmine, for example, contains benzyl acetate, which is the ester of benzyl alcohol and acetic acid. Depending on the size, shape, and type of molecule, the ester linkage can be prone to hydrolysis. Benzyl acetate tends to hydrolyze upon exposure to steam or boiling water. Acidic or alkaline water will increase this effect. The result of acidic hydrolysis will be benzyl alcohol and acetic acid instead of (or in solution with) the fragrant compound desired. Alkaline conditions will yield the corresponding alkali acetate, plus benzyl alcohol.

Usually, neither the alcohol nor the carboxylic acid smell anything like the original ester. Sometimes the components of ester hydrolysis can have a rancid odor. This also explains the unpleasant transformation that seems to happen to the cheaper perfumes and colognes. Many of these are ester-based, whereas the more expensive ones tend to use aldehydes and ketones. The human body generates 100 to 250 watts of heat during activity.    Carboxylic acids, released by bacterial action on human sweat, accelerate the hydrolysis of these cheap fragrances.

Note that certain plant esters will still yield a fragrant product after distillation.   Thanks to steric hindrance and perhaps other factors, some esters are so resistant to hydrolysis that they can be distilled without appreciable change. Alternatively, the chemical characteristics of the ester and its components may be such that usable ester is still able to come over into the distillate..

Some fragrance compounds are quite sensitive to oxidation. They remain intact in the living flower or plant, but they change to other compounds when exposed to steam, boiling water, or hot air. One may notice, for example, that the distinctive odor and flavor of cilantro does not withstand drying.

There is one other factor to hinder recovery of certain fragrances: some fragrant components may not be steam-volatile in the first place.   They would remain behind in the source vessel, bound to the original plant material.

For any of the above-mentioned reasons, there are some herbs and spices that are no good for steam distillation. They must be extracted directly with organic solvents or with supercritical carbon dioxide, the latter being generally out of reach of the small-scale operator. Once again, it is your responsibility to find out what solvents are / aren't appropriate for extracting oils. You must also be fully aware of any potential toxicity or allergy considerations.

Outside of major industry, the solvent extraction of essential oils for human use is normally done with vodka, neutral grain spirits (C₂H₅OH with about 5% water) and/or vegetable glycerin. Obviously, denatured alcohol and rubbing alcohol are out of the question.

Many materials will extract into vegetable oil, especially with extended soaking in a warm place.

Other Considerations

Methyl benzoate is a fragrant ester present in certain plants, including snapdragons. It also happens to be the odor that alerts a specially-trained canine to the presence of a certain very popular but non legal alkaloid that tends to come from South America [1]. According to Dudareva et al., methyl benzoate is "the most abundant scent compound detected in the majority of snapdragon varieties" [2].    Another paper notes that methyl benzoate is present in Stephanotis floribunda [3], also known as Waxflower, Madagascar Jasmine, and Hawaiian Wedding Flower. Although various submerged agendas tend to undermine Wikipedia's reliability in certain areas, it's probably safe to believe Wikipedia's claim that S. floribunda is popular in wedding bouquets [4].

We have not tested whether snapdragons (etc) will actually cause a false positive with a canine; however, the potential is obviously there. Indeed, canines have falsely alerted to perfumes and other innocuous objects (as in the case Horton v. Goose Creek Independent School District) [5]. Methyl benzoate is still methyl benzoate, no matter where it's found. If the bulk flowers yield no hits, it would be interesting to test various extracts of the flowers (e.g., EtOH tincture; glycerin extract; chilled aqueous homogenate; etc.).

Perhaps the cast of Mythbusters would be willing to treat themselves and their luggage with snapdragon extracts and submit to a canine test. Just to keep it objective and interesting, let's see if they can get through a Peruvian airport staffed by strangers who don't know it's a test.   No worries, right?

This writer is aware that competent scientists in LE are working on the problem of false positives. In the meantime, though, it is of some concern that glassware having the scent of (e.g.) snapdragons might be hard to explain to someone who lacks familiarity with natural fragrances and yet is in the position of possibly hauling someone away based on a canine sniff. While we certainly support the apprehension of the bad elements in our society, we are also for the protection of the innocent. Working with fragrances and essential oils is a legitimate, wholesome pursuit enjoyed by many Americans.

We need to start reversing the trend toward a society in which the only safe thing to do is shuffle papers, watch TV, and do what everyone else does. Otherwise, we're going to be in bad shape. Our greatest inventions were discovered under circumstances that probably wouldn't even be permitted today;   consider the stories of Thomas Edison, William Perkin, and Charles Hall, to name just a few.   The outcome of building a society stacked against innovation and independent inquiry is to become marginalized, at best. If America doesn't do something to right her course, the prospects are a disaster.

Finally, whatever you do, stay away from those plants that are proscribed by law. It is up to you to find out which ones are off limits, and to comply with regulations. Stick with herbs and spices you can buy at the grocery store or from some other reputable source, and there shouldn't be any concern.

Wear your safety goggles, stay safe, and have fun.

Copyright

Terms of Use

References:

[1]. Waggoner P, Johnston J, Williams M, Jackson J, Jones M, Boussom T, Petrousky J. "Canine Olfactory Sensitivity...". Proceedings of SPIE 2937:216 (1997).

[2]. Dudareva N, Murfitt L, Mann C, Gorenstein N, Kolosova N, Kish C, Bonham C, Wood K. "Developmental Regulation of Methyl Benzoate Biosynthesis and Emission in Snapdragon Flowers". Plant Cell 12 :949-961 ( 2000).

[3]. Chen F, Effmert U, Hippauf F, Kiefer I, Noel J, Pichersky E, Piechulla B, Pott M, Ross J, Saschenbrecker S, Slusarenko A. "Biochemical and Structural Characterization of Benzenoid Carboxyl Methyltransferases Involved in Floral Scent Production in Stephanotis floribunda and Nicotiana suaveolens". Plant Physiology 135:1946-1955 (2004).

[4]. http://en.wikipedia.org/wiki/Stephanotis_floribunda

[5]. Lunney, Leslie A. "Has the Fourth Amendment Gone to the Dogs? Unreasonable Expansion of Canine Sniff Doctrine to Include Sniffs of the Home". September 2009.

Back to article

Ringstands

Distilling apparatus

Distilling set 24/40