Störmer et al.
10.3389/fchem.2024.1397913
TABLE 1 Relevant European Standards for the extraction or migration testing of paper
Standard Short title
Description
Source
EN 645:1993
Cold water extract
Sample, cut or ripped in pieces, extracted with water 10 g/200 mL at 23 ° C ± 2 ° C for 24 h, shaking occasionally; the fi ltrate ( fi lled up to 250 mL) used for analysis
CEN (1993a)
Modi fi cation by BfR
Sample cut (not ripped), shaking not necessary, vacuum fi ltration with a glass fi ber fi lter (1.2 μm) instead of glass drip (10 – 16 μm), and pressing of the fi lter cake in case of high water absorption BfR (2022a)
Extracted with water at 80 ° C±2 ° C for 2 h. The procedure was similar to that of the cold water extract
CEN (1993b)
EN 647:1993
Hot water extract
Modi fi cation by BfR
Similar to that for the cold water extract but no comment on shaking BfR (2022b)
CEN (2007)
EN 15519:2007 Organic solvent extract
Sample cut, extraction with ethanol or isooctane 10 g/200 mL contact time and temperature depending on application 2 h or 24 h/20 ° C(short- or long-term contact) and 2 h/60 ° C (for hot contact), shaking occasionally; fi ltrate ( fi lled up to 250 mL) used for analysis Adsorbent MPPO (Tenax ® ) is spread on the surface of the sample (4 g/ dm 2 ), contact conditions according to application, and solvent extraction of MPPO Rapid extraction methods with isooctane and/or 95% ethanol 24 h at 40 ° C or 50 ° C depending on the polymer of the food contact layer; for polymer layer thicknesses up to 300 μm
CEN (2003)
EN 14338:2003 Migration using modi fi ed polyphenylene oxide (MPPO) as the simulant
CEN/TS 14234: 2002
Polymeric coatings on paper and board — guide to the selection of conditions and test methods for overall migration
CEN (2002b)
Regulation 10/2011 in its Article 18 states that “ the results of speci fi c migration testing obtained in food shall prevail over the results obtained in food simulant. ” This means simulation or prediction of migration should depict the situation with food as closely as possible, ideally matching with the upper bound margin of the determination in the food. The conventional assumption that the extraction values correlate to migration into food holds a non-negligible con fl ict potential as migration into food under realistic conditions may differ from the extract results. This becomes obvious for cut samples, which partially disintegrate during extraction. However, the discrepancy may occur in both directions: overestimation and underestimation. The example of per fl uoro compounds — which do not fi t into the simulant scheme as a worse case (Begley et al., 2008) — is given in BfR guideline (BfR, 2015). Merkel et al. (2018) compared the migration of primary aromatic amines (PAAs) from three paper napkins into four different food categories (wet, dry, acidic, and fatty) with the cold water extract results. In the food category pickled gherkins (aqueous – acidic), cold water seemed to be in seven of nine test cases suf fi ciently representative or even overestimating with a measured transfer into food (expressed as % of extract) ranging from 62% to 115%, depending mainly on the speci fi c amines, but was severely underestimative in two cases with 224% and 271% (PAA in both cases: aniline). Signi fi cantly less, or even no migration could be found into rice (dry), butter cookies (fatty), and cucumber (wet), respectively, with transfer ranging from 2% to 79% and many non-detects. Particularly striking is the difference between the results of cold water extract and butter cookies (fatty foods): only in three of nine test cases, PAAs were detectable in the food and, in which measurable transfer was ranging from 2% to 43% of cold water extract value, i.e., with extreme overestimation by the extract. Cold water extract is the sample preparation method proposed in the recently published standard EN 17163:2019 (CEN, 2019) for testing PAAs. For the organic solvent extract, the question arises as to whether the surface-related solvent extraction values can be directly
refers to Article 17 of Plastics Regulation 10/2011 and the exemptions for small ( < 500 g or mL) and very large packages ( > 10 kg or L) for which the conventional ratio of 6 dm 2 /kg is used. This concept can also be applied to paper (Beldi et al., 2023). Thus, cold and hot water extract results are recalculated to the surface area of the sample and the fi lling. The newer (or alteratively: more recent) organic solvent standard EN 15519:2007 already requires reporting the results related to the surface. Contrary to their name, the water and solvent extraction methods are not necessarily exhaustive (BfR, 2015). The extractive power will depend on the type and nature of the target substance (organic-polar, organic-unpolar, or inorganic), its physical – chemical properties, and its interactions with the paper sample. The name “ extraction, ” which is usually reserved for exhaustive methods, may cause confusion (Oldring et al., 2023). Taking these standards to depict “ worst case ” migration into food (BfR, 2015), it needs to be stated that these conventional methods might not match the real conditions of use. Confederation of European Paper Industries (CEPI) sees hot water extract testing as a close copy of the intended fi nal use and other extraction tests as mostly overestimating (CEPI, 2019/2021). Although CoE Technical Guide refers solely to these standard tests, German BfR (2015) recognizes this point and recommends a tiered approach using representative real foods in case of doubt or known overestimation or underestimation. Testing results in food have priority for the food regulatory assessment. For plastics, general testing requirements shall cover worst case of food contact applications and be even more severe, as speci fi ed in Annexes III and V of EU Regulation 10/2011. Similarly, the methods for paper are intended to simulate transfer into real foods conservatively and are slightly overestimating. However, for plastics, too severe tests can show conformity but not disapprove them. This is the case for alternative methods — the so-called screening methods — but might also be for (overestimating) conventional testing results. EU
Frontiers in Chemistry
04
frontiersin.org
Made with FlippingBook Digital Proposal Creator