How do you detect sunflower, hazelnut or pomace oil in olive oil?

Seed oils show up in ΔECN42 and sterols, refined oils in stigmastadienes, pomace oil in waxes and erythrodiol plus uvaol. Hazelnut oil is the hardest to catch.

Sunflower and other seed oils in olive oil are detected through ΔECN42 and the sterol decision trees, refined oils through stigmastadienes and trans isomers, and pomace oil through waxes and the sum of erythrodiol and uvaol. Hazelnut oil is the hardest case, because its composition is close to that of olive oil and low levels can slip past routine parameters. The type of added oil is worked out by reading the parameters together, not from a single value.

Which parameter answers which suspicion?

The limits below are those for extra virgin oil in Annex 1 of the Turkish Food Codex Communiqué on Olive Oil and Olive Pomace Oil. The IOC standard and EU Regulation 2022/2104 use the same values for these lines.

SuspicionAsk for firstLimitDifficulty
Sunflower, cottonseed and other seed oilsΔECN42, Δ7-stigmastenol, campesterol, linoleic acidΔECN42 max 0.2; Δ7-stigmastenol max 0.5%; campesterol max 4.0%; linoleic 2.5-21%Sterol decision trees may miss low levels
Refined vegetable oilStigmastadienes, trans isomersStigmastadienes max 0.05 mg/kg; trans-oleic max 0.05%Refining strips the taste and aroma clues
Olive pomace oilWaxes, erythrodiol and uvaolWaxes max 150 mg/kg; erythrodiol and uvaol max 4.5%Low levels can stay under the limit
Hazelnut oilΔECN42, sterol ratios, additional triglyceride calculationsNo hazelnut-specific legal limitThe hardest group

The Communiqué labels the stigmastadiene line "detection of refined vegetable oils" and the ECN 42 line "detection of seed oils", so the choice of method is not arbitrary. Each line answers a specific suspicion.

How is sunflower oil in olive oil detected?

A 2023 study by Ege University researchers in the Turkish journal Gıda addresses this directly. They blended sunflower oil, palm olein and cottonseed oil into virgin olive oil. The regulatory decision trees on Δ7-stigmastenol and campesterol detected blends at 10% and above. ΔECN42 and FTIR separated even 1% blends, although FTIR lost sensitivity when the sample set was widened to other seasons and varieties.

Two practical points follow. Sterol values within limits do not mean "no blending", so ask for ΔECN42 as well. And percentages from one study are not legal thresholds; they can shift with other oils and methods.

Fatty acids alone can be weak evidence. The accepted range for linoleic acid is 2.5% to 21%. Adding a small share of linoleic-rich oil to an oil at the lower end of that range may not take the result outside it.

Why is hazelnut oil harder to catch?

The two oils have similar fatty acid profiles and similar minor components, such as tocopherols and sterols. The published work points the same way:

  • A 2000 paper in Grasas y Aceites reports that Δ7-stigmastenol and ΔECN42 detect hazelnut oil only at high levels of 20-25%, and that additional triglyceride calculations bring this down to 5%.
  • A 2007 validation in the same journal, involving 21 laboratories, suggests a cut-off of 8% for a triglyceride-based method and 10% for a sterol-based one.
  • A 2024 study notes that refined hazelnut oil remains hard to detect with conventional methods, particularly below 20%, and that filbertone, the characteristic hazelnut volatile, can be removed in refining.

DNA-based approaches such as qPCR are being researched, but they are not part of a routine analysis package. If you suspect hazelnut oil, say so on the request, and ask the laboratory which method it uses and what its lower limit is.

Which values reveal pomace oil?

The Communiqué's table separates the two clearly. Extra virgin allows up to 150 mg/kg of waxes and up to 4.5% erythrodiol and uvaol, while olive pomace oil is defined with waxes above 350 mg/kg and erythrodiol and uvaol above 4.5%.

The limit is subject to simple dilution: a small share of pomace oil can leave the total below it. In refined pomace oil, stigmastadienes and trans isomers help too, because refining affects those values. Under the Communiqué, pomace oil may never be called olive oil.

How should you write the request and read the result?

  1. State the suspicion. Name seed oil, refined oil, pomace or hazelnut instead of asking for a "purity analysis".
  2. Avoid single parameters. At least request ΔECN42, sterol composition, stigmastadienes, trans isomers and waxes together.
  3. Check the scope. See whether each parameter is within the laboratory's accreditation. We explain this in where olive oil is analysed.
  4. Remember the decision trees. Campesterol and Δ7-stigmastenol are assessed through the decision trees in Annex 3 of the Communiqué. The IOC standard also treats some oils slightly over these limits as authentic if the other conditions are met, so a small exceedance alone does not prove adulteration.
  5. Question the sample. No parameter rescues a sample that does not represent the batch; see how to sample olive oil.

We cover reading these values in how to read an analysis report.

Where does Olivex fit?

When another oil is blended into olive oil, an Olivex reading moves away from the reference, and you see this before taking delivery of the batch. The reading does not tell you which oil was added. The type and share of the added oil are investigated with the laboratory parameters above. The method is described on the technology page.

Sources

Meeting request

Let's try the analysis set in your operation and discuss where and how often you would be reading.

The first meeting reviews your product, sales channels and the risks you face, and recommends a suitable method.

Request a meeting