DC 10/99/Topic II.17/ Pg 10
GROWTH – DEDICATED CALL – 10/99
TOPIC II.17
Certified Reference Materials for specific migration testing of plastics for food packaging needed by industry and enforcement laboratories (Support to the development of)
1. Conformity with the Work Programme
This topic is directly related to the R&D activity Support to the development of Certified Reference Materials (CRMs) under the Competitive and Sustainable Growth Programme (1.1.3) for which expressions of interest have been called. It falls under the Generic Activity 1.1.3-6 Measurements and Testing.
2. Keywords
CRMs, specific migration, plastics for food packaging, additives, monomers, conformity testing with EU Directive 90/128/EEC
3. Summary of objectives and justification
The objective of the project is the development of the ability to produce and certify certified reference materials (CRM’s) for specific migration testing made of the most important polymers fortified with additives of variant physico-chemical properties. The materials shall be well characterised relating to the polymer, the concentration of the migrating additives in the material, their specific migration and their migration kinetics in different food simulants. These CRM’s are needed for testing and improving the accuracy of the specific migration tests and also for showing the comparability of different laboratories in industry and enforcement authorities. The CRM’s are also needed for the validation of substitute tests for olive oil for specific migration and of alternatives to migration testing, i. e. extraction tests and mathematical modelling to predict the migration. These alternatives help saving costs and time in compliance testing of plastic materials with the European food legislation.
4. Background
Food contact materials stand under the special care of the legislation to assure consumer protection. They must not transfer constituents in quantities which could endanger human health or bring about an unacceptable change in the composition of the foodstuffs[i]. Therefore the overall migration and the specific migration of certain monomers and in next future also additives from plastics intended for food contact is to be tested. The migration limits are laid down in the Plastics Directive 90/128/EEC[ii].
Migration testing is conducted in two distinct phases - the migration exposure itself (exposure of the plastic to the food simulant) followed by chemical analysis of the food simulant or gravimetric analysis in case of overall migration. It is the first phase that sets migration methodology apart and establishes it as a distinct discipline. The major source of error lies in the initial choice and execution of the migration exposure rather than in the subsequent chemical analysis.
There are numerous analytical methods for determining plastics monomers and additives in food simulants and it is possible to check the validity of these tests, in-house, by analysing spiked samples. In contrast, there are almost no reference materials available to check the migration exposure itself. The only certified reference materials in this field are CRMs 537, 538 and 539 which were certified in 1998 for use in testing for overall migration using the fat simulant olive oil[iii]. Although these materials are fit for the purpose as CRMs for overall migration tests, this test is intrinsically imprecise and an analytical tolerance of +/-30% is permitted in the regulations to allow for this[iv]. An uncertainty of +/- 30% is not acceptable in tests for migration of specific monomers and additives.
Further evidence for the unsatisfactory state of affairs can be obtained from the statistics of the FAPAS® scheme which conducts a regular check-sample exercises on overall and specific migration as well as on many other areas of food analysis including pesticides, mycotoxins, veterinary drugs, food additives, metals, etc. Of the 20 test series conducted, that for specific migration has the lowest percentage return rate, and the lowest rate of acceptable performance[v].
Migration testing in olive oil is not only rather imprecise, the test is also very time consuming and therefore expensive. Specific migrants are often difficult to measure in oil with the necessary sensitivity. Some migrants, for example hexamethylenediamine, react with constituents of the oil and are principally not measurable in oil. Therefore suitable substitute and alternative tests have to be found and validated. According to EC Directive 97/48/EC[vi] (Basic rules on migration testing) 95 % ethanol, isooctane and modified polyphenylene oxide (TenaxÒ) can be used as substitute tests if olive oil testing is technically not feasible. But, currently, these substitute tests are only validated on the base of overall migration data not of specific migration data[vii].
The normal contact time with the simulants is 10 days for materials intended for long time exposure to foods. Alternative extraction tests with contact times of 1 day using organic solvents have been developed[viii] to spare time and costs compared to olive oil testing. They also are to be validated for specific migration.
In addition to such alternative methods prediction of migration using mathematical models is a further tool[ix] to simplify the expensive experimental tests. The Commission intends to insert a clause in one of the future Amendments of Directive 90/128/EEC that the verification of compliance with the specific migration limits (SML) provided in the Directives may be replaced by the verification of compliance with the maximum permitted quantity of substance in the finished product (QM), which corresponds to the SML. The correspondence between the specified SML and the QM shall be established by a calculation based on adequate experimentation of the finished product or by the application of generally recognised diffusion models.[x]
In general, the mass transfer from a polymer into foodstuffs obeys Fick´s laws of diffusion and is generally predictable and a generally accepted equation exists for calculating the migration amount with any desired precision if the partition coefficient KP,F of the migrant between plastic packaging P and the food F and the diffusion coefficient DP of the migrant in P are known[xi] . For migrants with a high solubility in the foodstuff or simulant the value KP,F = 1 for the partition coefficient can be used and a worst case estimation is obtained in this way[xii]. The diffusion coefficient is a function of polymer type, molecular mass and shape of the migrant and of the temperature. Currently, there exist only a limited number of reliable DP-values due to the enormous requirements needed for the experimental determination. Therefore estimation procedures for diffusion coefficients are necessary. In order to facilitate this as much as possible, an empirical equation was developed for prediction of a priori not known diffusion coefficients with the key parameter denoted as AP as a characteristic property of the polymer describing its difusivity[xiii]. The whole approach of the DP-estimation is statistically based and the values for the parameter AP can be improved in conformity with experimental results. The collections of data from different sources show large statistical distributions. They lead to diffusion coefficients ranges in orders of magnitudes[xiv]. The main problem in reviewing literature is that often not all influencing parameters i. e. exact migration parameters or characterisation of the polymer can be picked out of the literature.
For the validation of substitute migration tests, alternative tests and migration modelling well characterised CRM’s with validated quantities of specific migrants in the material and their specific migration values in different simulants are needed. In addition to the test conditions, the specific migration depends on the polymer material and on the physical-chemical properties of the migrants like molecular weight and polarity. The polarity strongly influences the solubility of the migrants in the simulants or the polymer, respectively. Therefore these validations must be performed with reference materials produced from the most important polymers and spiked with different migrants representing so far as possible semi-volatile to non-volatile and polar to non-polar substances.
5. Economic and social benefits
Migration tests are performed by the plastics and packaging industries to ensure that their materials comply with the relevant legislation. The tests are also performed by regulatory authorities to ensure that consumers are protected from unacceptably high migration levels. It has been estimated[xv] that between 200 to 300 national, regulatory and industrial laboratories in Europe, undertake migration testing. For having confidence in the test results and for showing and improving the comparability of different laboratories, proficiency tests performed with reference materials are an important tool either for the regulatory or for the industrial and service laboratories. Especially for materials with migration values in the near of regulatory limits, the precision and the accuracy of the test method in each laboratory including the migration itself should be known to avoid wrong positive or wrong negative decisions in the sense of consumer protection and of economical damage. These parameters can be obtained with CRM’s and are essential for a valid assessment of the test results.
In Europe, the annual costs of compliance testing with EU food packaging legislation is estimated to be several million Euro. It imposes an enormous cost burden on industry, yet is essential to maintain the high standards of food packaging safety in Europe. Regular retesting of food packaging is required, for example, as recommended in the UK in the British Plastics Federation BPF Guide[xvi] making it an ongoing burden. Moreover, the packaging converter will often produce finished articles, films and bottles of different shapes and sizes, all of which need to be tested to demonstrate compliance with EC directive 90/128/EEC. The 5th amendment to the directive will add more than 200 substances to those already requiring specific migration tests, increasing this cost burden still further. In order to save costs for industry as for enforcement authorities, to obtain results in shorter times and in this way to improve the competitiveness of packaging industry and allow more effective enforcement of the legislation, cost reducing and quick test methods have to be further developed and validated. The development of alternative tests would be much easier with reference plastics with certified migration values of specific substances. In order to support alternative extraction tests and mathematical modelling for compliance purposes, well characterised CRM’s are required containing known amounts of additives which have been validated at an EU level by inter-laboratory tests for both starting concentrations of substances (polymer additives), and certified migration values into foods/food simulants. These materials will allow ‘benchmark’ migration values to be obtained, ratified at an EU level which can be used to validate these methods. Using these alternatives a cost reduction for compliance testing of 30 to 40 % can be estimated.
Such a validated mathematical migration model is also a very useful tool for the Scientific Committee for Food (SCF) of the European Commission when evaluating SCF dossiers on new food contact polymers or additives. It allows a rapid estimation of migration to check the petitioners’ experimental data. Thereby can be ensured that the appropriate safety limits are assigned and sufficient toxicological data are provided.
Another beneficial application of the CRM’s is the comparison of the migration into simulants to that into real foods. The list of simulants for single groups of foodstuff is laid down in the EU Directive 85/572/EEC[xvii]. But when this list was established often a scientific base for the selection of the simulants was not available, so that the migration in the real foodstuffs is underestimated by the prescribed simulant in some cases. An example is milk and the simulant water[xviii]. The CRM’s can be used for the reevaluation of that list for an future amendment of the Directive.
The requirements for food packaging plastic materials are harmonised and are going on to be further harmonised in Europe. The test methods come to be standardised on European level at CEN. The results benefiting from the CRMs for specific migration shall be introduced both in European method standardisation as well as in European legislation.
6. Scientific and technological objectives
The objective of the project is the development of the ability to produce and certify CRM’s for specific migration testing is intended, as specified above, for quality assurance of testing laboratories and for the validation of migration test methods including quick and cost effective alternative testing and modelling methods. For this purpose the material should meet the following properties.
Due to the very different structures and properties (polarities) of the plastics, reference samples from the most important types of plastics are needed. They should include polyethylene (high and low densities), polypropylene, polystyrene, polyethylene terephthalate, polyvinyl chloride and polyamide. The geometric structure of the materials can differ. Mostly films will be appropriate but also other structures for example plaques are possible. The thickness should be dependent on the diffusivity of the material. Highly diffusive materials should be thicker than materials with a low diffusion. The materials used should be well described. It has been shown, especially for polyolefines, that density and orientation (PP) have a clear influence on the migration. The density of the material should be determined.
The polymers should be spiked with model compounds as specific migrants. The choice of the appropriate substances will be one of the crucial points of the feasibility study. The substances should so far as possible cover the range of different polarities and of the main chemical structures used, for example phosphite and phenolic antioxidants, ester plasticisers, light stabilisers, amide slip additives, monomers etc.. So they will differ in their solubility in the simulants. They also should cover a range of different molecular weights and volatility from semi-volatile to non-volatile. The model compounds should be additives or monomers used in food contact applications, which are listed in the Plastics Directive 90/128/EEC[xix] with specific migration limits or will in future adopted to it. Possible examples for such additive/plastic combinations are Di(2-ethylhexyl) phthalate and dibutylphthalate (DEHA and DBP) in PVC, glyceryl monostearate (GMS) in polystyrene or caprolactam in nylon. The model compounds have to be stable in the polymer, homogeneously distributed and compatible with the polymer. Especially the substances should not bloom out. The concentrations of the substances in the polymer should comply with concentrations used in practice but they should also be as high that they are measurable also in complex matrices for example in oil.
The reference materials shall be certified for the concentration in the plastic material (cP,0), for the diffusivity and for the specific migration values in official simulants[xx]. The concentration in the plastic material is a method independent value and characterises the migration potential of the material. The diffusivity shall be determined as migration kinetics into food simulants at a fixed temperature. The diffusivity is dependent on the application, especially on the solubility of the migrant in the simulant. The specific migration value shall be determined for a fixed time-temperature condition. These conditions will correspond to the test conditions laid down in the EU-Directive 97/48/EC[xxi]. For example, for a material intended for long time storage of food at room temperature the migration will be performed at 40 °C for 10 days.