Project runtime: 01.12.2023 – 30.11.2026
2024-02-29_BMBF_KickOff_DigiChrom
Electroplating consists of a series of elementary steps (mass transport, charge transfer, surface diffusion, nucleation, and growth, etc.) which interact with one another. This results in complex process-structure-property relationships for the electroplated layer. At the same time, this offers many options to adjust the material properties in a targeted manner by changing the process parameters (e.g. current density, temperature, hydrodynamics) and adapting them to the technical requirements. In technical application, trial and error approaches are usually used to optimize a galvanic layer system, which is very unsatisfactory from a scientific and socio-economic perspective. Digital approaches and an associated ontology enable a much more goal-oriented the research into electroplating systems. To do this, workflows must be defined to systematically collect experimental and simulated data, enrich them with metadata and make them available to the community. Machine learning (ML) methods will be used to predict structure-property relationships from the process parameters. Digital simulations make it possible to provide data that is difficult to determine experimentally. The ontology and the associated materials data space are based on a combination of experimental and generated (simulation, ML) data. With this approach, missing knowledge can be derived from what already exists. This makes it easier to systematically examine process-structure-property relationships, demonstrate them explicitly and to develop new ones from existing relationships.
