Innovative approach to identify location of AMC source in cleanroom by inverse Computational Fluid Dynamics modeling

Airborne Molecular Contamination (AMC) can cause serious impact on semiconductor manufacturing process. AMC source must be found and removed immediately once AMC level is detected high. Due to the complicated airflow circulation and contamination dispersion, the current methodology to detect the leak source for AMC removal is not effective. In the new approach, the inverse modeling methodology developed by adjoint equation is utilized through Computational Fluid Dynamics (CFD) simulation. It makes the quick identification of leak source possible, and thus quick responses can be made to remove the contamination. We verify the methodology successfully on a 2D cleanroom model.