Group Manager, Simulation
Fraunhofer Institute for Surface Engineering and Thin Films IST, DE



Andreas Pflug has studied Physics at University of Hannover and made his Diploma Thesis at the Institute for Solar Energy Research Hameln (ISFH). In the following, he worked at ISFH on modeling surface recombination dynamics in silicon solar cells. Since 2000 he is employed at Fraunhofer Institute for Surface Engineering and Thin Films (IST) where he made his PhD on „Simulation of reactive magnetron sputtering“ in 2006 and since 2008 he is head of the group „Simulation“ which is involved in modelling thin film deposition processes.​​

​Presentation ​

Multiscale Modeling of Sputter Deposition Onto 3D Substrates

The strive towards increased throughput, reliability and functional integration of coated products calls for improved deposition processes in terms of substrate size, precision, reproducibility and intrinsic performance of the coating materials. In parallel, economic and environmental targets ask for energy and resource efficiency of process, process chain and life cycle perspective. On the one hand, the application of detailed, model-driven simulation codes yields the required fundamental insights of the deposition process and thin film growth on an atomistic level. On the other hand, many industrial applications such as model-based in-situ process control or iterative optimization procedures require simulation codes, which are real-time capable. Such codes use simplified, data-driven models, calibrated to specific materials and process conditions. The internal model data either originate from experimental data logging or can be derived from more time-consuming model-driven simulation codes. The multi-scale simulation approach is demonstrated for a dual cylindrical magnetron sputtering process with rotating turntable. The task is to deposit an optical filter onto the convex side of a lens. The deposition setup includes specialized substrate holders and uniformity masks. A coupled simulation scheme yields the design of the optimized uniformity mask geometry: 3D Particle-in-Cell Monte Carlo (PIC-MC) simulations result in the relative erosion profile on the cylindrical sputter targets. Subsequently, the transport of sputtered material through the coater geometry is modelled via the Direct Simulation Monte Carlo (DSMC) method. Finally, a fast algorithm projects the deposition flux onto the moving and rotating substrate for arbitrary angles of the turntable rotation. The numerically optimized mask design enables the deposition of a band pass filter on a spherical lens including a tailored film thickness gradient for optical compensation at the non-centered positions. Further application of the coupled simulation framework for different coater and 3D substrate geometries is envisaged in the future.​​