AVS 51st International Symposium
    Manufacturing Science and Technology Thursday Sessions
       Session MS-ThM

Paper MS-ThM2
Fault Detection and Classification Using RGA in Semiconductor Manufacturing

Thursday, November 18, 2004, 8:40 am, Room 303B

Session: Advanced Process Control
Presenter: J. Blessing, MKS
Authors: Y. Xu, IBM
J. Byrne, IBM
H. Clark, IBM
J. Parker, IBM
J. Blessing, MKS
Correspondent: Click to Email
A new advanced residual gas analysis (RGA) system was jointly developed by IBM and MKS-Spectra and implemented in IBM's 300-mm wafer fabrication facility for the purpose of process monitoring and fault detection. The RGA system was integrated into the Computer Integrated Manufacturing (CIM) architecture using IBM's Advanced Process Control (APC) third party interface, an Extensible Markup Language (XML) based messaging system. The APC third party interface presents a brand new APC/sensor integration methodology allowing the RGA application (or any other sensor for this matter) to have access to both tool level event/trace/recipe data and fab level process/wafer information. A centralized database is used to manage the RGA system and to store summarized RGA data. Several examples will be offered to demonstrate how the role of the RGA system has expanded far beyond traditional photoresist detection and tool vacuum integrity. Its ability to quickly detect the undesirable process variations and incoming contaminations (photoresist or non-photoresist) proves to be critical to improving product yield, tool availability and ultimately the profitability of the fab. RGA fault classification is another important topic that will be covered. This classification is required to facilitate the proper handling of various RGA faults. One approach under consideration is to apply multi-variable analysis techniques to RGA data analysis. Some preliminary results will be given in the discussion.