Project Details
Projekt
Identification of optimal corrosion inhibitors for bare and PEO-coated magnesium alloy by combining machine learning and robotic testing
Subject Area
Coating and Surface Technology
Term
since 2024
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 535656357
The major goal of the proposed project is a shift of paradigm for the development of active anti-corrosion coatings using the example of the specific magnesium alloy AZ31. We will demonstrate how machine learning (ML) combined with robotic testing can assist the identification of optimal corrosion inhibitors for bare and PEO-treated magnesium. Highly automated robotic testing will be used to create a large basis of experimental data characterizing the corrosion inhibition efficiency (IE) of a large number of different chemical compounds for AZ31. In parallel, we will develop a machine learning architecture that can benefit from the collected data to predict the corrosion IE of chemical compounds, paving the way towards computer-aided identification of promising corrosion inhibiting compounds. In doing to, our ML architecture will use also data from density functional theory (DFT) calculations, which help to reduce the amount of experimental data required for meaningful predictions. Particularly promising magnesium corrosion inhibitors identified in the course of this project will be examined in depth. Their corrosion inhibition mechanisms will be uncovered by employing a combination of electrochemical techniques accompanied by DFT calculations. The approach pursued in this project will pioneer a new paradigm for the automated and accelerated enhancement of corrosion protection, allowing the accurate and fast identification of the most suitable corrosion inhibitors for a given material from a large experimental database of potentially available corrosion inhibiting compounds.
DFG Programme
Research Grants
