Epimerised catechins
What roasting actually does to a tea's catechins: not destroys them, flips them. The same molecules, mirrored at one carbon, sitting differently on the tongue.
What a drinker notices
- Astringency — a tactile sensation
Still drying, and drying differently. The sensation arrives a beat later and sits flatter and broader across the tongue than the sharper, more immediate grip of the epi- forms, which is a large part of why a well-roasted tea can be strong without ever feeling sharp. It is a texture rather than a taste throughout.
How much of it there is matters: What moves is a ratio rather than a total, so this is not a compound that gets stronger — it is one that takes over. A lightly fired tea reads as bright and grippy; the same leaf after a long roast reads as thick and mild at the same nominal strength, with no less of the group present.
What this gets credited for, wrongly
A GROUP, and the members differ. It also collects a misattribution from the other direction: the softness of an aged or heavily fired tea gets credited to this change alone when the sugars, the browning products and the loss of volatiles are all moving at the same time. This is one contributor among several, not the explanation.
What it is
The received account of roasted tea is that the fire burns the catechins away, which is why a roasted oolong or a hojicha feels softer than the leaf it was made from. That is not what happens. Under sustained heat a catechin can invert at a single carbon and become its non-epi partner — EGCG becomes gallocatechin gallate, EGC becomes gallocatechin — and the change is a rearrangement rather than a loss. The gallate group is still there, the molecule is still a catechin, and it still dries the mouth; it simply does so with a different edge and a different rate of arrival. This is a useful thing to know because it explains a pattern that the destruction story gets backwards: heavily fired teas are not low in catechins, and the softness a drinker notices in them is a change of character, not a subtraction.
Chemically: Flavan-3-ols that have flipped stereochemistry under heat — the non-epi partners of EGCG, EGC, ECG and EC.
Where it occurs in the plant
Present in small amounts in a fresh leaf and overwhelmingly a product of what happens afterwards. The fresh leaf's catechins are almost entirely the epi- forms; every subsequent hour of heat moves a fraction of them across, so the balance in a finished tea is a record of its firing history rather than of where it grew.
What processing does to it
In tea, processing is not a modifier applied to a fixed composition — it is what decides the composition. These are the steps that move this compound, and each links to the step itself.
- Hojicha roasting — Creates it
- Fast, hot drum roasting shifts a substantial part of the catechin pool across in minutes. It is the main reason hojicha reads as round rather than brisk while still being made from a green tea leaf.
- Charcoal roasting — Creates it
- Hours of low heat move the balance steadily. This is the slow half of what a long charcoal roast is doing to a yancha, working alongside the browning products rather than instead of them.
- Electric baking — Creates it
- The same shift under a controller, and the reason a repeatable oven programme gives a repeatable mouthfeel. Roasters who talk about baking a tea to a target texture are moving this ratio whether or not they describe it that way.
- Re-firing — Raises it
- Each re-fire nudges a little more of the pool across. A tea re-fired annually for a decade has a measurably different catechin balance from the day it was made, which is one of the ways stored oolong changes without anything being called ageing.
- Pulling the old fire (la lao huo) — Creates it
- Very high surface heat for very short intervals, repeated. It shifts the balance at the outside of the leaf hardest, which contributes to the layered impression of a tea finished this way — the first infusions read softer than the later ones.
- Two-stage drying (mao huo and zu huo) — Raises it
- The second, drier pass does most of the shifting, because the change runs faster in leaf that has already given up its moisture. It is one of the reasons the two-pass route gives a rounder tea than a single long dry at the same total heat.
- Final firing (hiire) — Raises it
- A heavy finishing fire on a Japanese green tea starts this shift, which is part of why a deeply fired sencha feels softer in the mouth than a lightly fired one from identical aracha.
- Steaming (Japanese fixation) — Leaves it broadly alone
- Steaming is hot and brief and almost entirely wet, and it leaves the epi- forms broadly where they were. That is one compositional reason a steamed green tea keeps the sharper, more immediate grip that a long-fired tea gives up.
How brewing brings it out
Extraction rate: Slow — needs heat and time
Like the catechins they came from, these need heat and time, and they are slower still. A roasted tea brewed cool gives up very little of them, which is the mechanism behind the standard advice to brew a heavy roast short and hot rather than long and warm.
Across infusions: Back-loaded and patient. A roasted oolong that tastes round on the second infusion can turn noticeably drier on the fifth, because this fraction is still arriving after the aroma has largely gone.
Water: Very hard water dulls these as it dulls the rest of the polyphenols, flattening both the colour and the grip. The effect is easier to notice on a roasted tea because there is less aroma sitting over the top of it.
Leaf, ratio and agitation: Broken leaf and agitation pull them forward as they do with any catechin. A roasted tea that has been shipped hard and arrived in pieces will brew drier than the same tea whole.
What you can change
- Brew short and hot rather than long and warm
- Pour off completely — the late infusions are where this arrives
- Rest a freshly roasted tea before judging its texture
- Avoid stirring or squeezing the leaf
Where this gets misdescribed
That roasting removes a tea's catechins and that this is why roasted tea is gentler. The pool is rearranged rather than removed, and the gentleness a drinker notices is a change in how the same family of molecules behaves in the mouth. The destruction story is tidy, widely repeated, and does not survive contact with what a heavily roasted tea actually feels like on the fifth infusion.
Teas this compound defines
Read next
- Catechins — The largest soluble fraction of a green tea leaf, the source of astringency, and the compounds oxidation turns into everything a black tea is.
- EGCG (epigallocatechin gallate) — The most abundant catechin in green tea, and the clearest single demonstration that astringency and bitterness are different things.
- EGC (epigallocatechin) — The catechin without a gallate group — bitter more than drying, and the partner EGCG needs to make a theaflavin.
- Melanoidins — The brown that heat makes. Why a roasted tea pours darker than the green tea it was, and why that darkness has nothing to do with oxidation.
- Soluble sugars — Present in small amounts, tasted only faintly, and indispensable as the fuel for every roasted aroma in tea.
What this page does not say
This record describes a structural change that heat makes to a family of compounds, and what that change does to the texture of a cup. Catechins reach most readers through a health heading; this is not that record, and it says nothing about what any form of them does in a body.
Sources
- TeaHQ editorial synthesis — TeaHQ
This page is TeaHQ’s own synthesis of well-established tea chemistry. No external technical document stands behind it, and that constraint is why there are almost no numbers here: directions, rankings and mechanisms are stated plainly, quantities are not, because a figure nobody can check would look more authoritative than it deserves.