Maillard reaction

The Maillard reaction is the chemistry that browns things when you cook them — bread crust, seared steak, toasted marshmallows, and roasted coffee. It is the single biggest reason a bag of coffee smells like anything at all. Green coffee seeds are pale, grassy and close to flavourless; almost everything a cupper later scores was built during roasting, and a large share of it was built by this one reaction.

It is a non-enzymatic reaction, which simply means no living enzyme is driving it — heat alone is enough.¹²³ That distinguishes it from the enzyme-driven browning you see when a cut apple goes brown on the counter.

What actually reacts

Two ingredients, both already present in the green seed:

• Reducing sugars — the sugars in the bean that have a chemically “open” end able to react. (Not all sugars qualify; sucrose, ordinary table sugar, has to break down first.)
• Amino acids — the building blocks of protein, and more broadly any compound carrying an amino group.

Under heat, the reactive end of the sugar joins onto the amino acid, and the result cascades: one join-up becomes a branching chain of hundreds of downstream reactions producing an enormous mixture of new molecules.¹²³ This is why it is described as a cascade rather than a single reaction — there is no one product, and the mixture is only partly characterised even now.¹

That dependency on sugars and amino acids is also why a seed that never ripened cannot brown properly. It simply never accumulated the raw material — which is the underlying cause of the pale quaker beans that show up in an otherwise even roast.

Where it sits in the roast

Maillard chemistry becomes fast at around 140 °C of bean temperature and runs quickly through roughly 140–165 °C.¹³ For scale, first crack — the audible pop that marks the start of the development stretch — sits near 196 °C. So the Maillard window opens well before first crack and is largely a pre-crack event.

Roasters borrow the name for that stretch of time. In the trade, the “Maillard phase” means the window between the end of the drying phase, when the beans turn from green to yellow, and the onset of first crack.³ That is a scheduling convention on a roast curve, not a chemical boundary — the reaction does not politely stop when first crack starts, and it is not the only chemistry running before it. The same stretch is called the browning phase in Coffee roasting.

To verify — where exactly the reaction "starts" is not a settled number

Coffee sources put meaningful Maillard activity at 140 °C, at 150 °C, and across a range of about 140–165 °C, and an American Chemical Society infographic behind the research for this note is reported as giving 150–200 °C as the main range.⁴ The spread is not really a contradiction — authors are choosing different thresholds for what counts as “significant” activity, and real onset shifts with bean density, moisture and how fast heat is being applied. Treat ~140 °C as the earliest credible figure and anything above it as one author’s threshold, not a measured constant.

What it produces

Colour. The brown pigments are melanoidins — large, dark, nitrogen-containing molecules formed at the end of the cascade.¹²³ The visible roast colour a colorimeter reads is largely melanoidin colour.

Aroma. The reaction produces the ring-shaped (“heterocyclic”) aroma molecules that dominate roasted smells. The families consistently named across sources are furans (sweet, caramel-like), pyrazines (nutty, earthy, roasted) and thiols (sulphurous, and extremely powerful in small amounts).²³ Many of the terms on the roasted, nutty and chocolate arms of the Coffee flavor wheel trace back here.

Body. Melanoidins are not only pigment — they also contribute to how thick and coating the brewed coffee feels.²

To verify — the specific compound figures rest on a single source

One coffee-research write-up consulted here supplies several precise claims that no second independent source in this research confirmed: that melanoidins make up 10–18% of roasted coffee’s dry weight; that pyrroles (sweet, slightly burnt) and pyridines (bitter, astringent) join furans and pyrazines as primary product families; and that a linked side-reaction called Strecker degradation yields 2-furfurylthiol, described there as the single most important odour compound in coffee.² The chemistry is plausible and widely repeated in roasting circles, but these are one author’s numbers. Treat them as claims pending a peer-reviewed review article.

To verify — the "how many aroma compounds" numbers openly disagree

Three different counts circulate, and they are not measuring the same thing. One coffee encyclopedia says about 800 volatile compounds are Maillard products in roasted coffee.³ A coffee-research write-up says roasted coffee contains over 1,000 identified volatile organic compounds of which the Maillard reaction accounts for 600+.² The existing Coffee roasting note here quotes over 800 volatiles developing across a full roast, from a third source. These cannot all be the same measurement — some count Maillard products only, some count the whole roast, and the identified total keeps rising as analytical methods improve. Do not quote any single figure as the number.

Not the same thing as caramelization

These two get used interchangeably in café conversation and they are genuinely different reactions.¹²

	Maillard reaction	Caramelization
What reacts	Reducing sugars with amino acids	Sugars alone, breaking down under heat
Needs protein?	Yes	No
Typical onset	Around 140 °C13	Higher — reported around 160–170 °C2
Gives	Brown melanoidins plus a very wide aroma range	Sweet, caramel, eventually burnt-sugar character

Both are running in a roaster at once, and both make things brown, which is why they get conflated. The practical difference for a roaster is that Maillard chemistry is where the savoury, nutty, roasted, complex notes come from, while caramelization is narrower and more purely sweet-to-bitter.

Who Maillard was

Louis Camille Maillard (1878–1936) was a French physician and chemist who described the reaction in 1912, while investigating how amino acids and sugars behave together — he was not studying food at all.¹⁵ The mechanism was not properly mapped until decades later; the multi-stage model chemists still use was published by the American chemist John E. Hodge in 1953.³

To verify — 1910 or 1912?

Most sources, including two encyclopedia entries confirmed directly for this note and a coffee-chemistry write-up, date the description to 1912.¹⁵²³ One widely read coffee-education source is reported in the research behind this note as giving 1910 instead.⁶ The 1912 date is far better supported and is stated as fact above, but the discrepancy is real and would only be settled by checking Maillard’s original publication, which was not reachable here.

Why it matters beyond chemistry

Because the reaction needs sugars and amino acids, anything upstream that changes how much of either the seed carries can change how it browns — ripeness at picking, and to some degree what happens after harvest. And because the reaction is so temperature- and time-sensitive, the same green coffee taken through the browning stretch quickly versus slowly does not taste the same. That is the entire subject matter of roast profiling, and the reason roasters argue about development time at all.

Related

• Coffee roasting — the process this reaction sits inside; its browning phase is the Maillard window
• Agtron scale — the instrument scale that reads the brown colour this reaction creates
• Quaker — the pale defect bean that failed to brown because it lacked the sugars and amino acids to react
• Development time ratio — the roasting ratio argued over immediately downstream of this chemistry
• Coffee flavor wheel · Cupping — where the aromas built here get named and scored
• Coffee freshness and degassing — what happens to those aroma compounds after the beans leave the roaster
• Arabica vs Robusta — species differ in sugar and protein content, which is where browning behaviour starts
• Natural process · Washed process · Honey process — post-harvest choices that shape the sugars going into the roaster
• Third wave coffee · Specialty coffee — the movement whose lighter roasting keeps the roast closer to this window

Official links

• Official site — Specialty Coffee Association, Coffee Standards

Footnotes

1. Maillard reaction — Wikipedia — confirmed via WebFetch; source of the non-enzymatic sugar-plus-amino-acid definition, the 140–165 °C rapid range, melanoidins, the 1912 date and the caramelization distinction ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹

2. Jayarr Coffee — Maillard, Caramelization, and Strecker: The Three Browning Reactions — confirmed via WebFetch; source of the staged temperature model, the caramelization onset range, the pyrazine/pyrrole/pyridine descriptors, the melanoidin dry-weight share, Strecker degradation and 2-furfurylthiol, and the 600+/1,000+ volatile counts ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰

3. Roastopedia — Maillard Reaction — confirmed via WebFetch; source of the ~140 °C bean-temperature onset, the “Maillard phase” roasting convention, the furan/pyrazine/thiol families, the 800-compound figure, the 1912 date and John E. Hodge’s 1953 mechanism ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰

4. American Chemical Society — Why Does Your Coffee Taste and Smell Delicious? — cited via the research report; the document returned an access error when fetched directly, so its reported 150–200 °C range is flagged rather than stated ↩

5. Louis Camille Maillard — Wikipedia — confirmed via WebFetch; source of his dates (1878–1936), his profession and the 1912 description ↩ ↩²

6. Barista Hustle — What is the Maillard Reaction and Why is it Important? — cited via the research report; the page was not reachable when fetched, and is the sole source of the conflicting 1910 date ↩