Oolong Polyphenols

"Tea polyphenol" is not one compound. It is a family: flavanols (the catechins, the dominant share), flavonols and their glycosides, phenolic acids, and, once oxidation has acted on them, theaflavins and thearubigins. Weigh these against every other guide on this site and they are the common measure underneath: how far the leaf was oxidized, the astringency in your cup, the cup's own color, and the figure every health trial actually tests. Know this family and the rest stop being separate subjects.

The catechin family, and what each member is

Catechins belong to the flavan-3-ol subclass and share a common two-ring, oxygen-bridged skeleton. Six of them account for nearly all of a fresh leaf's catechin content: catechin (C), epicatechin (EC), gallocatechin (GC), epigallocatechin (EGC), epicatechin gallate (ECG), and epigallocatechin gallate (EGCG). Pooled across cultivars, EGCG is usually the largest single share of a fresh, minimally oxidized leaf, and its galloyl group gives it aggressive salivary-protein binding, among the strongest of the six alongside ECG, which is why the gallate pair carries most of a leaf's bite and EGCG remains the most studied of the family. A 2025 review of Chinese oolong samples⁷ found EGCG still the predominant catechin in the finished leaf too, ahead of EGC then ECG, though the same review found oolong carries the highest EGC concentration of any tea category it compared, ahead of white, green, and black tea (see Oolong and Health for that figure set against the trial evidence it feeds). A category average and a cross-category ranking are two different measurements. Ask which comparison a claim actually rests on before you repeat it.

The six sort into two functional camps that matter in the next section: catechol-type catechins (EC, ECG) and pyrogallol-type catechins (EGC, EGCG), named for the ring structure that decides how each pairs up once oxidation begins.

What kill-green freezes, and what runs past it

Oxidation is enzymatic, catalyzed by polyphenol oxidase (PPO)² the moment bruising ruptures the leaf's cells and lets the enzyme reach the catechins. PPO first turns a catechin into its reactive quinone form. From there, a catechol-type quinone pairs with the electron-rich B-ring of a pyrogallol-type catechin, and the joined product cyclizes into a theaflavin: a molecule built on a fused benzotropolone skeleton¹, the seven-membered ring that gives theaflavins their bright orange-red color. The pairing is specific, and worth holding in your head as a small table rather than a sentence.

Push the reaction further and theaflavins polymerize again into thearubigins, the brown compounds that give a fully oxidized liquor its depth. For decades thearubigins were described as an unresolved blend of large polymers; more recent mass-spectrometry work complicates that picture; one analysis, resolving the class by LC/MS/MS⁸, found the thearubigins it could identify were built entirely from comparatively low-molecular-weight compounds, under 2,100 grams per mole, not the sprawling giants the older literature assumed. The class is still only partly mapped. Take any single figure for it as provisional, the way the Order takes any figure it cannot verify twice.

Run the reaction to its end and roughly three-quarters of a leaf's original catechin content converts into theaflavins, thearubigins, and related oxidation products, which is what a fully fermented black tea does before kill-green ever intervenes (black tea is fixed by a separate, later step, not this one). Oolong's kill-green cuts in mid-reaction, so a finished oolong always carries all three: unconverted catechins, first-generation theaflavins, and some thearubigins, in a ratio that shifts continuously with how far you let kill-green wait. A Bao zhong fixed near the 8 to 12 percent floor is still mostly catechins. A Wuyi yancha fixed near 60 to 80 percent has traded much of that catechin pool for theaflavins and thearubigins, which is also why its liquor runs amber to red rather than pale gold.

Where oolong sits, without pretending to more precision than the record has

Published catechin percentages for oolong disagree with each other about as much as the category disagrees with itself: methodology, cultivar, season, and roast all move the number, and no two labs report quite the same range. State a single tidy figure here and you would be stating false precision, so the Order will not. What holds across every study, regardless of the exact numbers each one reports, is the order: green tea highest, oolong in the middle, black tea lowest, tracking the oxidation percentage directly rather than any one cultivar's chemistry. Trust the order. Distrust any single decimal claiming to have measured it exactly.

The astringency you taste is a protein reaction, not a taste

Astringency is a tactile sensation, not one of the five basic tastes: catechins, the gallate-bearing ECG and EGCG most aggressively⁹, bind salivary proline-rich proteins and cross-link them into a precipitate⁴ large enough to strip your mouth's normal lubrication, producing the dry, puckering grip a heavily oxidized or over-steeped cup leaves behind. This is a physical reaction, not a chemical burn, and the same leaf can taste different from harder water too, though not by turning sharper: dissolved minerals accelerate catechin breakdown and pull less of it out of the leaf to begin with, so a mineral-heavy brew typically comes out flatter, not more astringent (see What Mineral-Heavy Water Does to Oolong). Because theaflavins and thearubigins bind less aggressively than the catechins they were built from, a more heavily oxidized oolong is typically rounder and less astringent than a lightly oxidized one at the same leaf-to-water ratio. How far the leaf was oxidized, not steep time alone, is the lever worth learning if grip is what you are managing.

The cup that clouds when it cools, and clears again when you reheat it

One consequence of this chemistry you can watch on your own counter, no instrument required. Cool a strong, heavily oxidized cup and it turns faintly cloudy. Reheat it, and the cloud vanishes. The haze is a reversible, non-covalent complex forming between theaflavins and caffeine as the liquid cools, described in the chemistry literature as theaflavin dimers stacking against caffeine molecules⁵ until the resulting particles are large enough to scatter light. The more theaflavin a liquor carries, the more readily this "cream down" appears, so a heavily oxidized yancha will haze on cooling far more visibly than a jade Tieguanyin barely past the green edge. Nothing is lost when it happens. Heat the cup back up and the complex dissociates, the liquor clears, and the tea is chemically unchanged. Run the experiment yourself before you take the Order's word for it: cool a cup until it clouds, then warm it back to clarity, and you have verified the chemistry by reversing it.

Why the health trials measure these molecules specifically

Nearly every published trial on oolong and blood sugar, weight, blood pressure, or inflammation is, at the molecular level, a trial of some combination of these same compounds: EGCG and the other catechins in a lightly oxidized batch, theaflavins and thearubigins in a more heavily oxidized one. That is why oxidation percentage is the first variable this site's health coverage controls for, and why a trial's tea and a supermarket bag rarely carry the same dose. Oolong and Health reads the cardiovascular, bone, and inflammation record against this chemistry in full. The Oolong Weight-Loss Trials, The Oolong Diabetes Trial Everyone Cites, and The 1987 Paper That Invented GABA Oolong each test one specific claim built on it. Read this guide first. It is the chemistry those four pieces assume you already have.