Six years ago the 3-MCPD ester story started with the publication of Zelinkova et al. 1 showing for the first time that remarkably high amounts of 3-MCPD esters could be found in edible vegetable oils after refining. Later it was found that from the most commonly used edible oils especially palm oil had a high potential to form these esters. Already in this early state of this story the question was raised about the chlorine donor which is necessary for the formation of 3-MCPD esters. The knowledge about the source of the chlorine donor was considered to be a key point for the development of mitigation strategies to reduce the amounts of 3-MCPD esters in vegetable oils.
Water used for the generation of strip steam during deodorization was the first suspect to have an effect as a source of chloride, but Pudel et al. 2 showed that the type of water has no influence. Matthäus et al. 3 for the first time assumed that the chlorine donor has to be present in the oil in an oil-soluble form to enable the reaction with other precursors. They found amounts between 1 and 6 mg water-soluble chloride/kg in different palm oils, but it was unclear whether these water-soluble compounds react during heat treatment or whether the reactive chlorine donors have to be present in another form.
The breakthrough in the question regarding the source of chlorine came from a study of Nagy et al. 4. They used a holistic mass-defect filtering of isotope signatures to pinpoint completely unknown and unexpected chlorine-containing substances naturally present in various extracts of palm fruit and partially and fully refined palm oil. By this Nagy et al. found that palm oil contains a huge number of different chlorinated substances. In addition to many organic chlorine compounds they found several inorganic chlorides such as calcium chloride, magnesium chloride, iron [II] chloride and iron [III] chloride in palm oil. One important point in this approach is the so-called chlorination cascade. This means that during processing the initially relatively polar chlorinated palm oil constituents may get converted into more and more lipophilic forms (such as the MCPD esters) along the oil processing chain 4. Looking at the processing chain of palm oil, during the growing of the oil palm the first important input of inorganic chloride containing compounds takes place as fertilisers in form of potassium chloride or ammonium chloride to improve oil palm growth and bunches yield. During the growth these chloride compounds are taken up by the plant and accumulated in the fruit. On the other side a lot of drinking water is necessary for the further processing and here it is known that iron[III] chloride is used as a coagulant in water treatment in Malaysia 5. The organochlorines seem to be endogenously produced by the oil palm during maturation 4, 6, but Nagy et al. also showed that organochlorines initially present in oil palm fruits are transformed into more lipophilic forms during the further palm oil processing 4. Additionally Nagy et al. stated that organochlorines begin to decompose at T>120°C and at higher temperature (T>150°C) the formation of 3-MCPD esters starts.
Destaillats et al. 7 proposed as a formation pathway of 3-MCPD esters a thermal-catalysed decomposition of organochlorine compounds naturally occurring in palm oil into reactive chlorinated compounds such as hydrogen chloride during deodorisation. These compounds can react with acylglycerols resulting in the formation of MCPD diesters and the release of a free fatty acid from the intact triacylglycerol molecule.
The knowledge about the formation mechanism of 3-MCPD esters as a result of the decomposition of chlorine-containing compounds during deodorisation is of great importance for the development of mitigation strategies. It becomes clear that the removal of these compounds from the crude oil or even from the pulp during the early stage of processing can help reduce the formation of the esters drastically. In regard to the above the effect of washing crude palm oil with polar solvents like water or water/alcohol mixtures as suggested first by Matthäus et al. 3 can be explained. During the washing process chlorine-containing soluble compounds are removed from the oil, resulting in remarkably lower formation of 3-MCPD esters. Because of the conversion of more polar chlorinated compounds into more apolar species during the processing, Craft et al. 6 suggested that it would be better to remove the reactive chlorine species from the extracted palm fruit oil rather than from the crude palm oil. That means the earlier the removal of the chlorinated species takes place, the better for the mitigation of 3-MCPD esters.
This example shows that the knowledge about the compounds involved and the insight into the formation pathway can help explain the effect of the process steps such as washing palm oil. On the other side it helps to improve the development of mitigation strategies for the formation of 3-MCPD esters. Thus, the introduction of a washing step by Matthäus et al. 3 was the right approach, but it is only the work of Nagy et al. 4 and Destaillats et al. 7 that explains the background for the effect that the washing has.
Max Rubner-Institut, Federal Research Institute for Nutrition and Food, Working Group Lipid Research, Detmold, Germany