Triggered by the regulatory need of the industry to demonstrate safe use of their alloy products from an environmental and health perspective, and by the significant lack of metal release data and its correlation to material and surface characteristics for iron- and chromium-based alloys, a highly interdisciplinary in-depth research effort was undertaken to assess the relation between material/surface characteristics and toxicity with main emphasis on stainless steel alloys. This thesis focuses predominantly on studies made on AISI 316L both as massive sheet and as powder particles, but includes also results for other stainless steel grades and reference metals and metal oxides.
The work comprises multi-analytical bulk and surface characterizations combined with particle characterizations and corrosion investigations, all correlated with in-depth kinetic metal release (bioaccessibility) studies as a function of route of manufacture, powder particle characteristics, surface finish, stainless steel grade, solution composition, pH, acidity and complexation capacity, as well as the presence of proteins. Speciation (chemical form) measurements were in addition conducted of released chromium, and of metal species in the surface oxide. Protein interactions were investigated in terms of adsorption, protein-metal complexation both at the surface and in solution, and the relative strength of protein-stainless steel surface interaction was addressed. In vitro and in vivo toxicological studies were conducted for the same inert-gas-atomized 316L powder sized < 4µm.
Bulk and surface oxide properties, such as phase, structure, morphology, chemical and electrochemical stability, protein-surface interactions, bioavailability of released metals, were all clearly evident to largely influence the metal release process and any induced toxicity. The route of manufacture was shown to strongly influence the bulk and surface oxide characteristics of stainless steel powders, hence also their electrochemical and catalytic properties, as well as the release/dissolution of metals from the powders (Papers VIII, XIII, XIV-XVII). The release of metals from both stainless steel sheets and powders was in general low compared to pure iron or nickel metal, and highly dependent on bulk and surface characteristics, the composition, complexation capacity and buffering capacity (and pH) of the solution, as well as on many experimental factors including time and sonication (Papers VI, VIII, XI, and XVII).
Surface-protein interactions strongly enhanced the release of alloy constituents (Papers IX, XI, and XVII). Iron was preferentially released (manganese in the case of inert-gas-atomized stainless steel powders) (Papers VIII, XI, and XVII). Protein-stainless steel surface interactions were most probably governed by chemisorption at given experimental conditions (Papers XI-XII). A strong protein-adsorption was evident for all stainless steel surfaces investigated, independent of protein charge, size or structure (Paper IX). Protein-metal complexes were formed both at the surface and in solution (Papers X-XII). Differences in protein charge and type resulted in varying degrees of interaction with differences in the extent of enhanced metal release as a consequence (Papers XI-XII). The inert-gas-atomized stainless steel powder sized <4 µm induced neither any significant increase of lysis of erythrocytes (rupture of red blood cells) nor any cytotoxicity, but resulted in a slight DNA damage in in vitro toxicity measurements (Paper VI). No adverse effects were however observed in an in vivo 28-day repeated-dose inhalation study on rats using the same powder (Paper VII).
The most important bulk, surface, particle, and experimental factors governing the bioaccessibility properties of stainless steel were identified and mechanistically elucidated. Detailed knowledge of all factors is essential for accurate hazard or risk assessment of metal alloys and enables read-across possibilities with materials of the same or similar characteristics. However, in cases where data is different from known systems for one factor or more, bioaccessibility data should be generated before any risk assessment is made.
Stockholm: KTH Royal Institute of Technology, 2012. , XIV, 38 p.
Omanovich, Sasha, Assoc.-Prof.