ZHICHENG BIO
Epicatechin 90%: the catechin peak in your chromatogram may be a processing artefact
A flavan-3-ol certificate is a report on two chiral centres that the method used to produce it cannot resolve.
The short answer
(−)-Catechin is not a natural constituent of cacao — it is made during processing, by epimerisation of (−)-epicatechin at temperatures above about 70 °C. Beans from unripe and ripe pods contain only (−)-epicatechin and (+)-catechin, with (−)-epicatechin predominant. Fermentation causes decreases of more than 80% in both, plus the appearance of (−)-catechin; progressive roasting drives the same conversion further. On an achiral reversed-phase column, (−)-catechin and (+)-catechin co-elute as one peak, so a certificate reporting "catechin" and "epicatechin" is reporting on two constitutional isomers while staying silent about four stereoisomers.
- The molecules
- (−)-Epicatechin C15H14O6, Mr 290.27, CAS 490-46-0; (+)-catechin CAS 154-23-4; (−)-catechin CAS 18829-70-4
- What is naturally present
- In beans harvested from unripe and ripe cacao pods, "only (−)-epicatechin and (+)-catechin" were found, with (−)-epicatechin by far the predominant isomer (Payne MJ et al., J Agric Food Chem 2010;58(19):10518–10527, doi 10.1021/jf102391q)
- What fermentation does
- Substantial decreases — greater than 80% — in both catechin and epicatechin in fermented versus unfermented beans, together with the formation of (−)-catechin
- What roasting does
- Progressive loss of (−)-epicatechin and (+)-catechin with increasing roast, and increasing (−)-catechin; roasting in excess of 70 °C generates significant (−)-catechin, attributed to epimerisation of (−)-epicatechin
- Why the assay misses it
- Enantiomers co-elute on achiral stationary phases. Resolving them requires a chiral method — chiral capillary electrophoresis (Molecules 2007;12(7):1274–1288, doi 10.3390/12071274) or the AOAC First Action 2013.04 HPLC method for catechin and epicatechin enantiomers
- What this does not say
- Nothing here concerns the effects of any of these compounds. It concerns what a chromatogram can and cannot distinguish, and what that means for a purchase specification
Four stereoisomers, two peaks
Catechin and epicatechin are flavan-3-ols with two stereocentres, at C2 and C3. That gives four stereoisomers: (+)- and (−)-catechin, which are 2,3-trans, and (+)- and (−)-epicatechin, which are 2,3-cis. Catechin and epicatechin are diastereomers of each other and separate perfectly well on an ordinary C18 column, which is why every commercial certificate can report them as two numbers.
Within each pair, however, the two enantiomers are chromatographically identical on an achiral phase. (−)-Catechin and (+)-catechin come off the column at the same time, in the same peak, contributing to the same integrated area. The same is true of the epicatechin pair. So a method that reports "catechin 3.2%, epicatechin 90.1%" has measured two things and is silent about four.
For most raw materials this would be a technicality. For cacao-derived flavan-3-ols it is not, because the enantiomeric composition is a direct record of what was done to the material — and that record is being erased by the choice of column.
What the processing study found
Payne and colleagues, publishing in the Journal of Agricultural and Food Chemistry in 2010, tracked flavan-3-ol monomers through the actual processing chain — fermentation, drying, roasting and Dutch (alkali) processing — using methods that resolve the stereoisomers rather than lumping them.
In beans harvested from unripe and ripe pods, they found only (−)-epicatechin and (+)-catechin, with (−)-epicatechin by far the predominant isomer. Drying had minimal effect on levels. Fermentation produced substantial decreases — greater than 80% — in both catechin and epicatechin, and at the same time the formation of (−)-catechin, which had not been there before. Progressive roasting at low, medium and high roast conditions drove a further loss of (−)-epicatechin and (+)-catechin and a further increase in (−)-catechin, and the authors concluded that roasting in excess of 70 °C generates significant amounts of (−)-catechin, probably by epimerisation of (−)-epicatechin.
The chemistry is straightforward once stated: heat epimerises the C2 centre of (−)-epicatechin, converting a 2,3-cis compound into a 2,3-trans one that retains the original C3 configuration — which is to say, into (−)-catechin rather than the naturally occurring (+)-catechin. A companion open-access paper in Chemistry Central Journal (2011;5:53, doi 10.1186/1752-153X-5-53) covers the stereochemistry across the same process chain, and the analytical groundwork for separating these enantiomers was laid by chiral capillary electrophoresis work published in Molecules (2007;12(7):1274–1288, doi 10.3390/12071274).
What this means when you are buying an extract
A "90% epicatechin" material has a source and a thermal history, and both are invisible on the certificate. If it derives from cacao, the fermentation and roast conditions determined how much (−)-epicatechin survived and how much became (−)-catechin. If it derives from green tea, the starting isomer profile is different again. If it is synthetic, the stereochemical outcome depends entirely on the route, and a non-stereoselective synthesis produces racemates.
None of these possibilities changes the number on an achiral assay. All of them change what is in the drum. This is the same structural failure that appears in theanine — an achiral method producing a correct answer to the wrong question — and it is why a specification for any material with a stereocentre needs a chiral line as well as an assay line.
There is a second, quieter consequence for anyone doing their own quality work. If you are comparing your incoming material against a literature reference value, and the literature was generated on unprocessed or lightly processed source material while your supply is heavily roasted, you are not comparing like with like — and the achiral assay will show good agreement while the actual stereoisomer distribution differs substantially.
The specification, and the method to name
Name the enantiomer in the identity line: "(−)-Epicatechin, CAS 490-46-0". Then set the assay: "≥ 90.0% by HPLC, on the dried basis", which is the total. Then add the line that does the work: "Enantiomeric composition by chiral HPLC (AOAC First Action 2013.04) or chiral capillary electrophoresis; report (−)-epicatechin, (+)-epicatechin, (+)-catechin and (−)-catechin individually".
Note the wording: report, not merely limit. For a first qualification you want the full distribution, because it tells you about the source and the process in a way no single limit does. Once you know what a good batch looks like from that supplier, you can convert the reported distribution into limits for routine release.
Ask for the botanical source and the thermal history in writing — cacao or green tea or synthetic; if cacao, fermented and roasted to what conditions; if synthetic, whether the route is stereoselective. Treat a change in any of these as requiring requalification, on the same logic as a route change for ergothioneine.
And when you read a chromatogram, remember what it is capable of telling you. A clean two-peak trace with excellent resolution and a 90% integration is a competent piece of analytical work that has not addressed the question you care about most.
Frequently asked
Is (-)-catechin naturally present in cacao?
Not in the unprocessed bean. Payne and colleagues (J Agric Food Chem 2010;58(19):10518-10527) found only (-)-epicatechin and (+)-catechin in beans harvested from unripe and ripe pods, with (-)-epicatechin predominant. (-)-Catechin appears during fermentation and increases with roasting; the authors attribute its formation above about 70 degrees Celsius to epimerisation of (-)-epicatechin.
Can HPLC distinguish (-)-catechin from (+)-catechin?
Not on an achiral column. The two are enantiomers and co-elute in a single peak on a conventional reversed-phase C18 method, contributing to the same integrated area. Resolving them requires a chiral method - chiral capillary electrophoresis, or the AOAC First Action 2013.04 HPLC method for catechin and epicatechin enantiomers in cocoa-based ingredients.
How much epicatechin is lost during cocoa processing?
Substantial decreases of more than 80% in both catechin and epicatechin were observed in fermented versus unfermented beans, with drying having minimal effect. Roasting causes further progressive loss of (-)-epicatechin and (+)-catechin with a corresponding rise in (-)-catechin, increasing with roast level.
What should an epicatechin specification require?
Identity as (-)-epicatechin with CAS 490-46-0; assay by HPLC with a minimum on the dried basis; and a separate line requiring the enantiomeric composition to be reported - (-)-epicatechin, (+)-epicatechin, (+)-catechin and (-)-catechin individually - by chiral HPLC per AOAC First Action 2013.04 or by chiral capillary electrophoresis. Also require the botanical source and thermal history in writing.
This page describes raw-material sourcing and specification. It is not a health claim and not medical advice. Last updated 2026-08-07.