Practical implementation of a Safe-by-Design framework: a proof of concept for early-phase hazard assessment of PFAS-free formulations

Source abstract

Abstract Per- and polyfluoroalkyl substances (PFAS) are widely used in various industrial sectors for their thermal stability, chemical resistance and surface-active properties. However, these same characteristics also contribute to their environmental persistence, potential for accumulation and adverse health effects, triggering increasing scrutiny across product life cycles. Within the EU-funded project PROPLANET, a tiered and iterative Safe-by-Design methodology was applied to PFAS-free coatings being developed for technical textiles, metal elements of food-packaging machinery and low-maintenance glass applications, at technology readiness levels (TRL) 3–5. The approach combined regulatory data mining, in silico screening and in vitro testing to identify potential hazard concerns and inform formulation-development decisions, while accounting for technical-performance requirements. Data mining from Safety Data Sheets (SDS) and the ECHA CHEM database were used to identify and exclude known harmful substances, including carcinogenic, mutagenic or reprotoxic (CMR) compounds and substances of very high concern (SVHCs), during the early stages of formulation development. Data-gap analysis indicated that the existing in silico models applied had insufficient coverage of the chemical space relevant to the substances under investigation, encouraging the development of custom-trained models. Two-tiered in vitro testing further characterized the toxicity profiles of several formulations throughout the project. One formulation intended for textile applications showed no hazard alerts under the tested conditions, whereas the formulations intended for glass and metal applications raised potential hazard concerns, requiring further investigation before progression to higher TRLs. This study presents the practical implementation of the Safe-by-Design methodology and the main results and inputs for decision making. At the same time, the practical challenges encountered during the implementation of the approach are discussed, together with the methodological gaps and grey areas. In conclusion, the structured, tiered and iterative approach implemented in this study supports the screening and prioritization of candidate chemicals and formulations in the early stage of their development, guiding the selection of PFAS-free chemicals and materials for further development in a SSbD context. In future work, analysing leachates, abraded particles, and chemicals released from coated products could provide an additional approach to assessing consumers and environmental safety during the use phase.