Migration of Bisphenol A and Benzophenones from Paper and Paperboard Products Used in Contact with Food

Cumulative Dietary Risk Assessment of Benzophenone-Type Photoinitiators from Packaged Foodstuffs

Photoinitiators used in ultraviolet-cured ink may migrate from food packaging materials into food products. Therefore, we conducted a dietary risk assessment of exposure to benzophenone (BP)-type photoinitiators by quantifying and reducing uncertainties associated with the risk characterization. A total of 362 food packaging samples including 180 cereals, 136 fruit and vegetable juices, and 46 milk samples were subjected to fast pesticides extraction to determine photoinitiator residues. The average daily dose (ADD) of BP was the highest in the age group of zero to three years, with a P97.5 ADD of 2.56 × 10-4 mg/kg bw/day. The ADD of 2-hydroxybenzophenone (2-OHBP) was the highest in the age group of three to six years, with a P97.5 UB ADD of 3.52 × 10-5 mg/kg bw/day. The estimated UB P97.5 ADD for each age group was below the toxicological concern threshold of 0.0015 mg/kg bw/day. The cumulative toxicity of all BPs, evaluated using the MOET value, was at an acceptable level. Although the MOET value of BPs was above the safety limit in the foodstuffs studied herein, this result may be different if Taiwan were to follow regulation guidelines for BP-type photoinitiators based on the specific migration limit for the unmeasured BP residues in other foodstuffs.

Determination of five bisphenols in commercial milk samples by liquid chromatography coupled to fluorescence detection

The presence of five bisphenols, i.e., bisphenol F, bisphenol A, bisphenol B, bisphenol F diglycidyl ether, and bisphenol A diglycidyl ether, was monitored in commercial milk packed in plastic bottles marketed in Italy. The new validated method includes a solid-phase extraction procedure followed by liquid chromatography with fluorescence detection. All positive results were confirmed by liquid chromatography-tandem mass spectrometry analysis. The limits of detection and quantification and the recovery percentages indicated that the method is suitable for detecting bisphenols in milk at concentrations far below the legal limits. Of 68 commercial milk samples analyzed, no bisphenol was found in 27 samples (39.7%), and 41 samples (60.3%) contained one or more bisphenols. The bisphenol most frequently found was bisphenol F (36 samples, 52.9%) followed by bisphenol A (20 samples, 29.4%) and bisphenol B (6 samples, 8.8%). Taking into consideration the limits of detection, no sample contained either bisphenol F diglycidyl ether or bisphenol A diglycidyl ether.

Does the composition of urine change when collected from disposable diapers and other absorbent materials?

The free and conjugated urinary species of non-persistent environmental chemicals or their breakdown products are valid human exposure biomarkers. For convenience, disposable diapers and other absorbent materials are widely used to collect urine specimens from infants and young toddlers. However, the extent to which the different urinary species of the target analytes and other components are recovered after the urine is extracted from these absorbent materials is unknown. In this proof-of-concept study, we investigated the extraction recovery from disposable diapers, cotton pads, and gauzes of the free versus glucuronidated urinary species of three example chemicals: bisphenol A, triclosan, and 4-methylumbelliferone. Although the glucuronides were almost fully recovered, the free species were not. Our results suggest that, in addition to other sampling considerations, the binding affinity and extraction recovery of the target biomarkers to the material used to collect the urine should be considered. Alternative collection approaches that do not require such an extraction (e.g., urine bags routinely used in hospitals) may be worth exploring. Despite its shortcomings, having urinary concentrations for biomonitoring considerably strengthens the exposure assessment, particularly for infants and young toddlers, and the benefits of including biomonitoring data outweigh their potential limitations.