In-situ computational–experimental study of velocity anisotropy and slurry hydrodynamics in Wafer-Pad CMP using machine learning prediction

Semiconductors are the indispensable core of modern intelligent systems, driving automation, artificial intelligence, and advanced computing. However, existing studies on Chemical Mechanical Planarization often overlook the combined influence of velocity anisotropy and slurry hydrodynamics on material removal uniformity. This research bridges that gap through computational modeling and machine learning, optimizing wafer–pad interactions and identifying the Pad–Wafer RPM Ratio as a key control parameter. A novel sweeping-rotation model is proposed to minimize pad wear and enhance surface planarity. The integration of AI-driven prediction establishes a foundation for smart, data-centric semiconductor manufacturing aligned with the principles of Industry 4.0.

Engr. Abdullah Hasni is a mechanical engineer and researcher specializing in propulsion systems, energy efficiency, and material science. He has published novel research with Springer Nature on the testing and optimization of twin turbojet engines and has presented at five international conferences, including IMEC. Abdullah works at National Refinery Limited, focusing on hydrogen production, Industry 5.0 applications, and CMP planarization processes. His research interests include aero-thermal systems, hydrogen energy, and advanced materials.