Food Standards Agency
Saturday 4 July 2009
Safer food better business banner
AZ-Directory
What's NewRSS
What is RSS?Listen
Listen to this siteTuesday 22 August 2006
This research project will develop and validate a HPLC-MS-MS method for the analysis of individual primary aromatic amines (PAA). The validated method will be applied to samples where there is a high packaging:food mass ratio and to coloured food contact articles.
Study Duration : May 2006 to Autumn 2008
Contractor : Leatherhead Food International
Many primary aromatic amines (PAAs) are considered toxic and some are considered to be possible causes of cancer in humans. They may be found in multilayer films which are commonly used in food packaging, although proper preparation and curing of these films means that the amines should no longer be present. They may also be found in coloured food contact articles. Migration of PAAs is subject to legal limits, and items can be screened using a colorimetric test. However there are certain drawbacks to this test; the PAAs are not measured individually, but are compared with a single compound, aniline. This means that there may be some inaccuracies if different PAAs give a different response in the test to aniline. It is also preferable to be able to identify which particular PAAs are present, if at all, as some compounds are more harmful than others. One objective of this project, therefore, was to develop and validate a method to identify and quantify individual PAAs in a range of sample types.
A further objective was to test the migration of PAAs from a multilayer film when it was exposed to food simulants at different ratios. This is relevant because there are many uses for multilayer films, ranging from small sachets e.g. a single serve sachet for vinegar, which will have a high packaging : food ratio, to large pouches of soup or pasta sauce with a low packaging : food ratio.
The final objective was to test the developed method against migration of PAAs from coloured food contact materials and articles.
Some work had already been published on the analysis of PAAs by LC-MS-MS, and this was used as a starting point for this project. Conditions were optimised for the analysis of 30 PAAs. Extraction and concentration methods were then developed for the analysis of those PAAs in food simulants (water, 3% acetic acid, 10% ethanol and olive oil). The performance of the developed extraction and analysis was tested by analysing replicate samples on different days.
Since migration testing can involve heating and storage, it was important to test whether the PAAs were stable in food simulants under these conditions. PAAs were added to food simulants and heated to 121ºC for two hours followed by storage at 40ºC for 10 days. PAAs were extracted and analysed to determine how much remained at different points during the regime.
The next stage was to extend the extraction and analysis from food simulants to real foods. Foods which might be sold in laminated packs were selected: vinegar, tomato sauce, mayonnaise and smoked salmon were chosen to give a mix of different food types, with different levels of fat, protein and acidity. PAAs were added to these and methods of extraction were tested and developed, then the performance was tested as for the simulants.
Testing of a multilayer film at different packaging : simulant ratios was carried out by first obtaining a film with measurable levels of PAAs. Therefore deliberately undercured film was obtained, tested for PAAs under standard conditions and then exposed to simulants at different ratios. PAAs were then measured in the simulants by the developed LC-MS-MS method.
Finally coloured food contact materials and articles, such as food trays, bowls and utensils, were obtained from a manufacturer and from retail outlets. These were screened for PAAs using the current colorimetric method. Articles which tested positive were then re-tested in simulants using the newly developed LC-MS-MS methods.
A method has been developed and tested to measure 30 PAAs in food simulants. Most PAAs were recovered well, and results could be reproduced on different days with good agreement. Extraction and quantification of PAAs from different foods was also satisfactory.
PAAs heated and stored in simulants proved to be stable in water, 3% acetic acid and 10% ethanol but not in olive oil. In practice, this means that olive oil may not be a suitable simulant for testing migration from articles which would be heated in real use.
An undercured film was obtained from a manufacturer, and tested for migration at different packaging : simulant ratios. Three PAAs were found to be present in the simulants after exposure. Migration increased with increasing packaging : simulant ratio, but not in a linear way. This suggests that the current practice of extrapolation of the result from low packaging : simulant ratio to high packaging : simulant ratio may lead to over-estimation of PAA migration.
Finally, a range of coloured food contact materials and articles were obtained and screened for PAA migration using the colorimetric method. Articles which gave a positive or tentatively positive result were tested using the LC-MS-MS method. There was no detectable migration from some of the articles suggesting that the colorimetric method may have given ‘false positive’ results due to the transfer of dye colour from the article to the test solution. One article, however, did contain high levels of PAAs, and so was subjected to further migration testing and analysis by LC-MS-MS. The results were forwarded to the Agency, who contacted the company involved. The company took prompt action to withdraw the product from the market. This demonstrates the use to which the method can be put to ensure consumer safety.
Find out what our other sites have to offer