What the family is
Polyphenol is a broad chemical class, and in tea the dominant members are the flavan-3-ols, usually called catechins. Several are present together — the ungalloylated forms and the galloylated forms, of which epigallocatechin gallate is the most abundant in most green teas — and they occur alongside flavonol glycosides, phenolic acids and other related compounds. They are highly water-soluble, they make up a large share of what dissolves out of an unoxidised leaf, and they are the substrate on which every subsequent transformation in tea processing acts. Two things about them are worth holding onto. They are associated with astringency and with bitterness, which are separate sensations produced by the same family. And their level in the finished leaf is what oxidation, roasting and ageing all change, which is why they are the thread running through every other page here. It is worth adding that polyphenol is a chemical classification rather than a description of behaviour, and the class is enormous — the same word covers compounds in wine, cocoa, olives and a great many plants, with quite different properties. Saying that tea is rich in polyphenols therefore conveys very little on its own, which is why this page names the flavan-3-ols specifically and describes what they do in the cup. A class name is a starting point for a discussion and not a finding.
What they are associated with in the cup
Astringency first — polyphenols associate with the proteins in saliva, disrupting the lubricating film that coats the mouth, which is registered as drying and tightening rather than as a taste. Bitterness second, with the galloylated catechins generally reported as more bitter and more astringent than the ungalloylated ones. Colour third, since these compounds and their oxidation products are what give tea liquor most of its colour beyond the pale yellows of an amino-acid-rich leaf. And structure: a tea with very little of this fraction reads as flat and hollow, which is why the goal in brewing is never their elimination but their balance against sweetness and body. The house rule that astringency and bitterness are separate sensory axes is a statement about perception, not about chemistry — one family of compounds produces both. They also carry a substantial share of what a drinker experiences as strength, which is a further reason the fraction resists simple description. A liquor low in it reads as weak even at a high dose, because the sensations that signal intensity in tea are largely this family’s. That is why an over-shaded or over-cooled brew can taste sweet and insubstantial at once, and why brewing advice aimed only at reducing astringency will eventually produce a cup nobody wants.
Why brewing temperature matters so much for this fraction
Solubility and diffusion rise with temperature for everything in the leaf, but not by the same factor for every compound. The polyphenol fraction gains proportionally more from heat than the amino acids do, so hot water does not simply extract the same cup faster — it extracts a different balance, weighted towards grip and bitterness. This single asymmetry is the mechanism behind almost every temperature recommendation in tea. Brewing a shaded Japanese green tea at a low temperature is not gentleness, it is selection: it collects the amino acid fraction while leaving much of the polyphenol fraction behind. The same asymmetry explains why cold brewing produces a sweet, low-astringency liquor from leaf that would be harsh brewed hot, and why the two are genuinely different drinks rather than one being a slow version of the other. Time interacts with temperature rather than substituting for it. The fraction continues to come out steadily after the fast material has largely finished, so a long steep shifts the balance towards it even at a moderate temperature — which is why an unattended cup goes grippy rather than merely strong. The two levers therefore do different things, and adjusting both at once is why so many people cannot work out which change fixed their tea.
Why the level falls as oxidation proceeds
Oxidation is, chemically, the conversion of these compounds into other, larger ones. When leaf cells are ruptured, the enzymes held inside meet the catechins and oxygen, and the catechins are converted into quinones that then combine into theaflavins and, further along, into the much larger and less well-characterised thearubigins. So a fully oxidised black tea contains far less of the original catechin fraction than the green tea made from the same bush, and correspondingly more of the oxidation products, which have their own quite different sensory associations. This is why black tea and green tea made from the same leaf taste so unlike each other, and why the descriptor scale in this catalogue runs from unoxidised to fully oxidised: the scale is tracking exactly this conversion. The same conversion explains a practical brewing difference. Because a fully oxidised leaf holds much less of the fraction that responds so sharply to heat, black tea tolerates boiling water where green tea does not — the compounds most sensitive to over-extraction have already been transformed into something else. This is the mechanism behind the general rule that more oxidation means more tolerance of temperature, and it is why the rule is reliable rather than a convention.
What else changes the level in the leaf
Several agronomic and processing variables, and the directions are well established even where the magnitudes are not. Light drives their formation in the growing leaf, so shading a bush before picking is associated with a lower catechin level relative to amino acids, which is the agronomic basis of gyokuro and tencha. Leaf age matters: young buds and first leaves differ from mature leaf, and mature leaf carries more of the structural material and a different balance. Season matters, with the first flush after winter dormancy generally richer in amino acids relative to catechins than later pickings taken in strong summer light. Cultivar matters substantially. And processing matters most of all, since firing, oxidation, roasting and ageing each transform the fraction in a different direction. Storage belongs on the list as well. Over months and years the fraction slowly converts into larger compounds by a non-enzymatic route, which is why an old green tea is softer and duller than a fresh one and why a long-aged sheng loses the severe grip it had when young. The same change is loss in one category and development in the other, which is the recurring pattern across everything in this catalogue about ageing.
Why a single-compound story is wrong here
TeaHQ's own reading of well-attested general knowledge, with no single source behind it.
Popular writing tends to nominate epigallocatechin gallate as the compound that explains green tea, and that is a simplification in three ways. Several catechins are present together and their proportions differ by cultivar, season and processing, so the fraction behaves as a mixture rather than as one substance. The sensations attributed to it — astringency and bitterness — are also produced by other things, caffeine prominently among them for bitterness. And what a drinker perceives is a ratio: the same polyphenol level reads as harsh against low amino acids and as structured against high ones, which is precisely why shaded and unshaded teas from the same cultivar taste so different. Flavour is combinatorial. An inventory of compounds is a starting point, and the interactions are where the cup actually comes from. There is a further reason to be wary of the single-compound framing, which is where it usually comes from. Nominating one molecule as the explanation of a tea is a habit imported from a literature interested in that molecule for other reasons entirely, and it has shaped a great deal of consumer writing about green tea. A compound that is prominent in one discussion is not thereby the explanation of a flavour, and this catalogue keeps the two conversations apart deliberately.
What this page does not say
No quantity is given for any compound, in any tea, at any level. Published figures for catechin content vary widely with cultivar, season, leaf position, processing and the analytical method used, and a single number would assert a precision that does not exist. Nothing here describes what any of these compounds does in or to a person; that is a different subject and it is not this catalogue’s. What is asserted is the direction of a set of relationships — that these compounds are associated with astringency and bitterness in the cup, that their level falls as oxidation proceeds, that they extract preferentially at higher temperatures, and that shading and season shift their balance against the amino acids. The same restraint applies to comparisons between teas. Statements that one category contains more of this fraction than another are true only as broad tendencies and fail routinely at the level of any particular pair, because cultivar, season, plucking and processing move the figure more than category does. Where this catalogue records anything of the kind it records it as a direction with its reasoning attached, and where the evidence does not support a comparison it declines to make one.