Industrial Monitoring of Residue Deposition in Semiconductor Process Exhaust Pipelines Using Electrical Capacitance Measurements

Abstract

In semiconductor manufacturing, the accumulation of byproducts in exhaust pipelines under inadequate temperature control poses significant safety and operational risks. This study introduces an innovative approach employing an electrical capacitance measurement sensor system combined with an artificial neural network (ANN) to monitor residue buildup. The proposed method estimates the free volume index within industrial process exhaust pipes, enabling precise evaluation of residue deposition and gas phase fractions. Numerical simulations and field studies in semiconductor environments validate the model’s effectiveness, demonstrating accurate residue quantification and enhancing safety and operational efficiency. In numerical simulations, the error between true values and estimated values was within 1%, while the values estimated from experimental data showed an error within 5%. These findings underscore the robustness of the model in both controlled and real-world settings. This advancement offers a practical and reliable solution to mitigate hazards and optimize maintenance processes in semiconductor manufacturing.

Keywords

• Electrical capacitance measurement
• semiconductor process
• pipe deposition monitoring
• free volume index
• artificial neural network (ANN)