Benzophenone may be present in cartonboard food-packaging materials as a residue from UV-cured inks and lacquers used to print on the packaging. It may also be present if the cartonboard is made from recycled fibres recovered from printed materials. A method has been devised to test for benzophenone in cartonboard packaging materials and to test for migration levels in foodstuffs. Packaging is extracted with solvent containing d(10)-benzophenone as the internal standard. Foods are extracted with solvent containing d(10)-benzophenone and the extract defatted using hexane. The extracts are analysed by GC-MS. For analysis of food, the limit of detection was 0.01 mg kg(-1) and the limit of quantification was 0.05 mg kg(-1). The calibration was linear from 0.05 to 20 mg kg(-1). The method for food analysis was validated in-house and it also returned satisfactory results in a blind check-sample exercise organized by an independent laboratory. The methods were applied to the analysis of 350 retail samples that used printed cartonboard packaging. A total of 207 (59%) packaging samples had no significant benzophenone (<0.05 mg dm(-2)). Seven (2%) were in the range 0.05-0.2 mg dm(-2) , 60 (17%) were from 0.2 to 0.8 mg dm(-2) and 76 (22%) werefrom 0.8 to 3.3 mg dm(-2). A total of 71 samples were then selected at random from the 143 packaging samples that contained benzophenone, and the food itself was analysed. Benzophenone was detected in 51 (72%) of the foods. Two food samples (3%) were in the range 0.01-0.05 mg kg(-1). A total of 29 (41%) were from 0.05 to 0.5 mg kg(-1), 17 (24%) were from 0.5 to 5 mg kg(-1) and three (4%) food samples exceeded 5 mg kg(-1). The highest level of benzophenone in food was 7.3 mg kg(-1) for a high-fat chocolate confectionery product packaged in direct contact with cartonboard, with room temperature storage conditions and with a high contact area:food mass ratio. When the mass fraction of benzophenone migration was calculated for the different contact and storage regimes involved, the attenuation effects of indirect contact and of low temperature storage were cumulative. Thus, there was a sixfold reduction in migration for indirect contact compared with direct contact, a sixfold reduction for chilled/frozen storage compared with ambient storage, and 40-fold reduction,for the two contact conditions combined.
