Development of Luminescence Tests to Identify Irradiated Foods. Project N1701

This is the final report of project N170l, commissioned by MAFF from October 1990 until March 1992. The principle aims of the project, at a time when statutory changes involving strict labelling rules were anticipated, were to consolidate the thermoluminescence technique already developed at SURRC and to investigate it's extension to fruits and vegetables. The TL procedures for mineral separation were further developed by the incorporation of pre-concentration steps. These steps were introduced in order to improve the the sensitivity of the TL signal, particularly, where the separation technique resulted in low mineral yield for commercially clean samples. This has resulted in an order of magnitude increase in the absolute, TL, signal levels and provides a means of obtaining larger quantities of minerals for any further quantification, thus reducing the ambiguity of interpretation of data. An EC interlaboratory trial was instigated, using the full mineral separation method with reirradiation on a set of calibrated reference materials and paired (irradiated and unirradiated) samples of 12 commercial grade herbs and spices. Despite the diversity of experience and equipment employed, results from all laboratories showed that it was possible to determine which samples were irradiated. This demonstrated the strength of the separation method that its implementation could be successfully achieved in other laboratories. The procedure was then formally recognised and published, by MAFF, for detection of irradiated food for enforcement of UK legislation. As the mineral debris responsible for TL in herbs and spices occurs ubiquitously on all foodstuffs, which have been exposed to wind and soil, investigating the application of TL to fruits and vegetables was a natural extension of the previous work. An extensive survey was conducted of TL signals from fruits and vegetables, including exotic varieties. Minerals were separated from duplicated pairs of irradiated and unirradiated samples of 22 fruits and 20 vegetables. TL results demonstrated unambiguous discrimination between irradiated and unirradiated vegetables provided that concordance diagrams were used. For soft fruits in particular water based separation produced more variable results. However, subsequent analyses using a full density separation with HCI wash were more successful. It was recognised that whereas herbs and spices are largely protected from exposure to light during production and distribution, this is unlikely to be the case for fruits and vegtables. Since light exposure is known to reduce TL signal intensity, a set of illumination experiments was conducted to investigate the implications of optical bleaching for identification. Light boxes were constructed and characterised to simulated optical bleaching under controlled conditions. Two series of experiments were conducted with irradiated and control mangos. In the first experiment the effects of exposure to two different light sources at a level of 1 J cm-2 were investigated using 40 mangos. In the second experiment the influence of duration of exposure from 1 - 128 J cm-2 was examined for artificial daylight illumination of a further 96 mangos. The results of these studies show that although the TL signal is reduced as a result of exposure to daylight, there exists a residual unbleachable component comprising upto 40-50% of the original signal. In most cases the resulting TL will be distinguishable from background levels. As a result of this work it is now possible to extend TL detection protocols to a wide range of fruits and vegetables. Providing that recontamination with unirradiated minerals has not occurred after irradiation, the majority of treated fruits and vegetables are expected to be detectable. Positive signals will imply an irradiation treatment. There remains some possibility of false negative results from a small proportion of irradiated products.