Coffee Aroma in Depth: What Makes Coffee Smell the Way It Does

Coffee Guide

By MELB_NOTE

The short answer

Coffee aroma comes from volatile compounds that form mainly when the beans are roasted. Studies find more than a hundred of them in a single coffee, but only a few dozen drive the typical smell. They are fragile, so much of the aroma is lost within weeks of opening the bag, and how the coffee is stored changes how much is left.

This guide continues our series on coffee acidity in depth, coffee sweetness in depth, coffee bitterness in depth and coffee body and mouthfeel in depth. Those cover what the tongue tastes and what the mouth feels. This one covers what the nose picks up.

30-second summary

• Green coffee does not smell like coffee. The aroma is generated during roasting, largely through the Maillard reaction between sugars and amino acids.
• One study found 138 volatiles in 10 single-origin Arabicas. Furans and pyrazines each made up roughly a quarter to two fifths of them.
• In a study of four Arabicas, 85 compounds were aroma-active, and six of them were significant in every coffee when left out of a recombined aroma.
• In ground coffee kept in repeatedly opened packages, 35 to 45% of the volatiles were gone after 30 days, and over 80% in a paper sachet.

---

Coffee aroma is a mixture of volatile compounds made by roasting

The green coffee bean does not smell like roasted coffee. Smrke et al. (2022) note that the aroma is generated upon roasting, when sugars react with amino acids in the Maillard reaction to form a large number of volatile organic compounds. Zakidou et al. (2021) also name the Maillard reaction as the main route.

Smrke et al. also report that roasted coffee holds about 1% of its weight as carbon dioxide, which escapes over roughly a month. That is why freshly roasted coffee is often sold in bags with a one-way valve.

---

Only a few dozen of those compounds carry the coffee smell

The literature lists more than 1,000 volatile compounds in coffee. Zakidou et al. found 138 of them in 10 single-origin Arabicas, and note that only about 5% of all the volatiles are thought to be perceived by our olfactory receptors.

A different method asks the nose directly. Chen et al. (2025) used gas chromatography-olfactometry, in which a person sniffs the separated compounds as they leave the instrument. In four medium-roast Arabicas, 85 compounds were aroma-active. Smrke et al. summarise earlier work by saying that about 30 compounds are collectively responsible for the characteristic coffee aroma.

These counts are not one list that keeps narrowing. They come from different coffees, methods and thresholds, so treat them as an order of magnitude, not a fixed number.

---

Furans and pyrazines make up most of the volatile mix

Zakidou et al. grouped the 138 volatiles by chemical family. Furan derivatives made up about 24 to 41% and pyrazine derivatives about 25 to 39%, depending on the sample. Pyrroles were about 5 to 12%, esters 5 to 8%, pyridines 3 to 6% and phenols 2 to 4%.

Share of the mix is not the same as impact on the smell. A compound can be abundant and still matter little, while a trace compound with a very low odour threshold can matter a lot. The next section shows how researchers test that.

---

Six odorants stood out in a recombination test

Chen et al. ranked the aroma-active compounds by flavour dilution factor, which is how far the extract can be diluted before the smell disappears. Furfural, guaiacol and furaneol reached the top value of 2187. They also calculated odour activity values, which compare each compound’s concentration with its odour threshold. 4-Vinyl-2-methoxyphenol scored 761 in one coffee, and furaneol scored 250 in another.

They then mixed 17 odorants into an odourless base and compared it with the original coffees. The mixtures resembled the coffees overall, although smoky, roasted and woody notes were weaker in three of the four. Finally, they left out one compound at a time. Six compounds made a significant difference in all four coffees: γ-butyrolactone, hydroxyacetone, isovaleric acid, 4-vinyl-2-methoxyphenol, furaneol and guaiacol. Leaving out furfuryl alcohol made no significant difference in any of them.

The study used four medium-roast Arabicas, so the list is a snapshot, not a rule for every coffee.

---

Aroma fades fast once coffee is opened or ground

Trenzová et al. (2024) stored ground coffee in several package types and opened and closed each one repeatedly, to imitate a person using it up gradually. After 30 days, 35 to 45% of the original volatiles were gone whatever the package. A paper sachet was the worst, with a loss of over 80%.

Smrke et al. compared four ways of keeping whole beans after the bag was opened: an airtight canister, the original bag closed with tape, closed with a clip, and a bag with a built-in screw cap. The screw-cap package kept the aroma best. The clip, the tape and the canister all lost freshness faster.

Our inference, not a tested recipe: grind shortly before brewing, keep the package closed, and see our guide on how to store coffee beans.

---

Questions people ask

Where does coffee aroma come from?

Mostly from roasting. Sugars and amino acids in the green bean react in the Maillard reaction and form volatile compounds. Green coffee does not smell like roasted coffee.

How many aroma compounds are in coffee?

The literature lists more than 1,000 volatiles. One study found 138 in 10 Arabicas and another found 85 aroma-active ones in four. About 30 are usually described as carrying the typical smell.

Which compounds smell most like coffee?

There is no single one. In one recombination study, six compounds were significant in every coffee, including furaneol and guaiacol. Others may matter more in other coffees.

Does ground coffee lose aroma faster than whole beans?

The studies above did not compare them directly, so we cannot give a number. Both lost aroma after opening, and the package made a clear difference.

Is a vacuum or airtight container the best choice?

Not in the one study that tested it. A canister did worse than a screw-cap package. This was a single study, so treat it as a hint.

---

Sources and scope

---

Related guides

• Coffee Acidity in Depth
• Coffee Sweetness in Depth
• Coffee Bitterness in Depth
• Coffee Body and Mouthfeel in Depth
• How to store coffee beans