Chromium speciation analysis in foodstuffs: Ultra-trace speciation analysis of chromium in foodstuffs by high performance liquid chromatography coupled to inductively coupled plasma-mass spectrometry using species-specific isotope dilution.

It is nowadays largely accepted that the determination of the total content of potentially toxic elements (PTEs) in foodstuffs does not always provide sufficient information to support a correct risk assessment, especially when the PTE can be present in different chemical forms (i.e. chemical species). In such a case, information on each chemical species is necessary and hence speciation analysis is required. Chromium speciation analysis is a very relevant issue in food chemistry, on one hand due to the toxicity of Cr(VI) species and furthermore due to the analytical challenge to accurately determine Cr(III) and Cr(VI) in foodstuffs. The challenge is greater when the simultaneous determination of these two species is desired, as they are unstable with variations in temperature and pH. Actually, a polemic exists in the scientific community related to the possible presence of Cr(VI) (the most toxic Cr species) in foods. EFSA stated (2014) that in such sample Cr can be present only as Cr(III), because the foodstuffs matrices are reductive media and hence Cr(VI) incorporated into a diet is immediately reduced to C(III) (less toxic). The polemic in terms of Cr status in food was sustained by several authors who found Cr(VI) in several foodstuffs (bread, tea, etc.) but assessed also the formation of Cr(VI) by oxidation of Cr(III) during various culinary processing (e.g. bread toasting, etc.). Actually, it is reasonably assumed that the “discovery” of Cr(VI) in foodstuffs is attributed to analytical artefacts. Indeed, both Cr(III) and Cr(VI) are stable in different conditions, hence making their accurate determination very difficult. The clarification of this controversy using the species-specific isotope dilution (SS-ID), the only method that allows correcting the interconversions of species that may occur during the analytical procedure, is one of the main achievements of this thesis. The present PhD project aims at the development, validation and application of novel analytical approaches for the simultaneous speciation analysis of Cr(III) and Cr(VI) in foodstuffs such as dairy products, meat and meat products, bread, breakfast cereals and rice by high-performance liquid chromatography (HPLC) coupled to inductively coupled plasma-mass spectrometry (ICP-MS). Species specific-isotope dilution (SS-ID) was thoroughly employed for species quantification and for the correction of their potential mutual inter-conversion. The species extraction was carried out by complexation of Cr(III) with ethylenediaminetetraacetic acid (EDTA) and by derivatisation of Cr(VI) with 1,5-diphenylcarbazide (DPC) in the same analytical run. By this procedure Cr(VI) is not directly complexed with DPC, but is first reduced to Cr(III) and then subsequently Cr(III) is complexed by DPC. The HPLC separation of complexed species was carried out using a short microbore anion-exchange HPLC column, which allowed baseline separation of the two species in less than 3 minutes. The method was validated by means of the accuracy profile approach by carrying out six measurement series in duplicate on (six) different days over a timespan of two months. The quantification limit was 0.013 μg kg-1 for Cr(III) and 0.049 μg kg-1 for Cr(VI), respectively. he measurement bias corresponding to the validity domain ranged from 0.01 to 0.11%, whereas the coefficient of variation in terms of repeatability (CVr) varied from 2.9 to 11.6 % (depending on the analyte level) for Cr(III) and from 6.7 to 11.8% for Cr(VI). Similarly, the coefficient of variation in terms of intermediate reproducibility (CVR) ranged from 6.8 to 13.0% for Cr(III) and from 6.8 to 25.9% for Cr(VI), respectively. The method was successfully applied to the analysis of a selection of different types of food samples including dairy products, meat and cereals. Cr(VI) was not quantified in any of these samples while Cr(III) levels ranged between 0.22 µg kg-1 (infant formula milk) and 350 µg kg-1 (breakfast cereals with cacao). Cr(III) levels in all samples were comparable with the levels of total Cr determined in the same samples by ICP-MS. The results obtained in the various food samples analysed support the hypothesis that all Cr present in foods is in the form of Cr(III), clearly rejecting reports on the presence of Cr(VI) in foods. The results obtained with this work contributed positively to the current dispute over the speciation of Cr in foods.
The methodology was furthermore used to evaluate the potential impact of different culinary processes on the fate of Cr species in infant formula milk, semi-skimmed milk and bovine meat samples. The cooking procedures consisted of boiling the milk samples at 70ᵒC and 100ᵒC and frying the bovine meat without (95ᵒC) and with oil and (120ᵒC). The results showed no significant differences between the content of the Cr species in raw and cooked samples, hence indicating that no inter-species transformation happened during the processing. A risk analysis was performed taking into account the levels of Cr(III) found in cooked bovine meat and milk and showed an extremely low risk for consumers of these products. This work contributed to obtain more knowledge in terms of the Cr speciation in foodstuffs with the development of an important tool for assessing the risk of Cr in foods with total control over the interconversion of species that occur during the analytical procedure. It also contributed to resolve the current dispute over the presence of the toxic specie of Cr in food. For future work, the developed and validated method can be applied to other types of matrices, such as environmental or pharmaceutical.