Food packaging materials constitute an ever more threatening environmental pollutant. This study examined options to specifically assess the ecotoxicity of packaged wastes, such as cans, subjected to various experimental treatments (in terms of extraction media, time of exposure, and temperature) that imitate several basic conditions of the process of food production. The extracts were studied for their ecotoxicity with bioluminescent Vibrio fischeri bacteria. The first objective of this study was to find patterns of similarity between different experimental conditions; we used multivariate statistical methods, such as hierarchical cluster analysis, to interpret the impact of experimental conditions on the ecotoxicity signals of the package extracts. Our second objective was to apply best-fit function modelling for additional data interpretation, taking into account, that ecotoxicity for various temperature conditions is time- and temperature dependent. We mathematically confirmed that chemometric data treatment allows for better understanding how different experimental conditions imitating the real use of food packaging. We also demonstrate that the level of ecotoxicity depends on different extraction media, time of exposure, and temperature regime.
Keywords: assessment of extracts; best-fit function; canned food; chemometrics; ecotoxicity; food packaging.