Metallic foils are used in many modern applications such as, but not limited to, (flexible) batteries, (flexible) Printed Circuit Boards (PCBs), (flexible) solar cells, packaging and inside the skin of air and spacecraft as Multifunctional Composite Materials (MFCs) to enable, e.g. the implementation of sensors. In such applications, the copper foils and films are subjected to various (thermo-) mechanical loads ranging from compressive-tensile cycles inside the hull of an aircraft and vibrations in automotive applications to thermal loading in PCBs. To ensure these applications¿ safety and increase the lifetime of microelectronic devices, the (cyclic) material properties have to be determined in a way that allows the implementation of the material behaviour into numerical simulations. This is a crucial point since more sustainable (micro-) electronics are needed to tackle the pressing challenges of the 21st century, such as the climate crisis or the vast consumption of resources. To accomplish numerical implementation effectively, commonly used materials testing methods must be further developed. Foil-specific problems must be solved to enable precise measurements, especially the challenge of having a macroscopic sample in two dimensions, but a microscopic one in the third dimension has to be tackled. During tensile testing, foils will show effects, such as curtaining, which makes data extraction more complex. Therefore, this work presents a method that allows the precise determination of material parameters via a cyclic approach that uses a simple tensile testing device. The approach has been validated using multiple methods, especially nanoindentation led to good agreement with the measured values. Nanoindentation tips are susceptible to wear. For this reason, the tip wear has to be accounted for. This is usually done with an inverse method. For the first time, the wear behaviour of such indenters has been obtained using Explainable Artificial Intelligence (XAI). This will lead to a more precise materials characterisation, e.g., extracting the elastic properties from complex laminated samples. Laminated samples consisting of many copper and fibre-enhanced polymer layers were investigated, and numerical material models were calibrated in a physically meaningful way on low-cost experimental data combined with finite element simulations. This ensures easy implementation in different material laboratories and industrial settings. However, further research is still needed as the plastic and fatigue behaviour of foils, films, wires, and even sheet metal depend on the number of grains along the smallest dimension (thickness or diameter). For this reason, the fatigue behaviour of the layered samples was investigated, showing that copper foils are the failure introducing part of the composite. Several methods were used to characterise the foils, including scanning electron microscopy, X-ray diffraction, computed tomography and synchrotron stress measurements. The synchrotron data showed that the low-cost implementation of the model calibration was feasible as the deviatoric stresses in the different layers were in an acceptable range for numerical model calibration. Combining the developed numerical models with the experimental fatigue data led to precise fatigue life predictions for copper foils used in many applications.
|Translated title of the contribution||Weiterentwicklung der Charakterisierung mechanischer Eigenschaften von Metallfolien durch die Kombination von Experimenten, Simulationen und maschinellen Lernen|
|Publication status||Published - 2023|
Bibliographical noteembargoed until 27-02-2026
- Metallische Folien
- multifunktionale Kompositmaterialien
- Maschinelles Lernen