In June 2026, the Antioxidants Research Group of the University of Lleida delivered a 19-page report to PONS with the complete phenolic profile of four extra virgin olive oil samples. Not just a total polyphenol figure: the individual identification and quantification of twenty-five compounds, one by one.

This page explains what was analyzed, using what method, what the results were, and what we do with that data. The results are published exactly as delivered by the laboratory, with their standard deviations and the reference of the analyzed batch.

Total polyphenol content is the figure reported throughout the industry, and it is a correct figure. But it describes a set without saying what is inside it.

Two oils can declare 650 mg/kg and have different compositions. The difference is not academic: the compounds that make up that figure determine the bitterness, the intensity of the peppery sensation in the throat, the oxidation stability, and the behavior of the oil in cooking.

Knowing which compounds are present and in what proportion requires a different analytical technique than the one used in routine quality control. That is what this work provides.

The study is based on the official method of the International Olive Council, the normative reference for determining phenolic compounds in olive oil. The preparation phase is the same: 2.5 grams of oil are weighed, 6 milliliters of n-hexane are added, and the phenolic fraction is separated from the fatty matrix by solid-phase extraction with diol cartridges. The retained compounds are eluted with methanol and injected into the chromatographic system.

The change lies in the detection. Routine control uses HPLC with a diode array detector, which identifies each compound by its retention time and its ultraviolet spectrum. The University of Lleida used ultra-high-performance liquid chromatography coupled to tandem mass spectrometry, with electrospray ionization in negative mode and selected reaction monitoring. Each compound is identified by its molecular mass and by the specific fragments it produces when it breaks down.

The practical consequence lies in the secoiridoid derivatives. They are the majority fraction of olive oil, and the official method quantifies them by equivalence, referring them to tyrosol or hydroxytyrosol because it does not have its own standards. In this study, they were measured with specific commercial standards for oleacein, oleocanthal, oleuropein aglycone, and ligstroside aglycone. Since these four compounds concentrate around 90% of the total phenolic content, measuring them instead of estimating them changes the reliability of the entire result.

The table shows the quantification of each compound in milligrams per kilo of oil, with its standard deviation. The values correspond to the analyzed batches, not to guaranteed minimums for the season.

In the three varieties, secoiridoid derivatives account for approximately 90% of the total phenolic content. When we talk about polyphenols in olive oil, we are, in practice, talking about secoiridoids.

These compounds are not present in the olive as they appear in the oil. They are formed during milling and malaxation, when the fruit's enzymes transform the original oleuropein and ligstroside. Oleacein and oleocanthal are products of this transformation, and they are also the compounds with the most associated scientific literature: oleocanthal is responsible for the burning sensation perceived in the throat.

Within this major fraction, the distribution changes completely depending on the variety.

Arbequina concentrates more than half of its secoiridoids in oleacein: 321 mg/kg, more than double any other variety analyzed. It also has the highest oleocanthal content of the group, 138 mg/kg. This is a polarized profile and explains its frank bitterness and defined pungency.

Koroneiki does the opposite: it distributes its secoiridoid fraction almost equally among the four main compounds. Oleuropein and ligstroside aglycones exceed 160 mg/kg each, the highest values in the study. It is also the variety with the most elenolic acid, 32.9 mg/kg compared to less than 10 in the other two.

Lecciana reproduces Koroneiki's balanced pattern with a lower total content: the two aglycones make up 62% of its secoiridoids. However, it provides the highest presence of lignans and flavonoids in the group.

Arbequina and Koroneiki declare almost identical totals, 652 and 627 mg/kg. Their profiles are nothing alike.

That is why a total figure does not describe an oil.

The report contrasts its results with those of the routine control laboratory, which applies the official HPLC-DAD method to the same samples. The differences range from 7% to 35% depending on the variety, and they do not all go in the same direction.

The largest discrepancy is in Arbequina: 652 mg/kg by UPLC-MS/MS versus 482 by the official method, a 35% difference. The origin is identifiable. The routine laboratory quantifies 65.6 mg/kg of oleocanthal and 200 of oleacein in that sample; the University of Lleida measures 138 and 321.

Quantification by equivalence underestimates these two compounds precisely in the variety that concentrates them the most. In Koroneiki and Lecciana, where the weight of oleacein is lower, the difference is reduced and changes sign.

The consequence goes beyond the number. By the official method, Arbequina is at the same level as Lecciana, 482 versus 483. By UPLC-MS/MS, Arbequina is the variety with the highest phenolic content of the three. The method not only changes the figure: it changes the classification.

This does not invalidate the official method. Each answers a different question. The IOC exists to verify quality and regulatory compliance with a standardized and reproducible procedure among laboratories, and it does so well. UPLC-MS/MS exists to characterize: to know what compounds are present, how many, and in what distribution. Both are necessary.

Regulation (EU) 432/2012 does not set its condition in milligrams per kilogram. It sets it per serving: 5 milligrams of hydroxytyrosol and its derivatives for every 20 grams of oil. This is the threshold that determines whether an oil can qualify for the authorized claim, and the report calculates it for each sample.

Characterization by variety is not a laboratory exercise. It determines specific agronomic and processing decisions.

It allows us to compare varieties based on what they contribute, not just how much they add, and to decide what to plant in new plots. It helps us adjust the harvesting time: the concentration of secoiridoids depends on the fruit's ripening stage, and advancing or delaying the harvest by a few days modifies the resulting profile. And it provides a validated reference against which to check the chromatograph readings that we have incorporated into our laboratory, in line with the milling process.

The agreement with the University of Lleida has a second open line of research: to determine which of these compounds are bioavailable in each variety.

This is ongoing research, with no definitive results yet. When they are available, they will be published here with the same traceability as those on this page.