In Situ Gas Analysis with Contamination Modeling for Diffusion Furnaces

摘要

In situ gas analysis has been shown to be a valuable tool for improving process operations in semiconductor device manufacturing. Analysis of the gas ambient in process chambers can be used to monitor the integrity of device manufacturing steps, diagnose causes of contamination upsets, baseline process recipes and validate PM efficiencies. The present work describes in situ analysis used together with a predictive contamination modeling capability to give improved gas purity control in horizontal furnaces during typical inert purge cycles. The horizontal furnace is typically purged with high purity inert gas while the furnace is open to the ambient cleanroom. These purge procedures have been shown to be ineffective at preventing atmospheric contamination from entering the furnace when the end cap is open. Use of contamination modeling to better understand the parameters that affect contaminant migration into the open furnace can lead to improve the purging techniques. In this study both numerical and experimental results indicate that conventional purge procedures, even at very high flow-rates, do not substantially reduce the contamination levels in a hot horizontal furnace when the furnace door is open. However, the purging is by far more effective when the furnace is operating at room temperature. The measured and calculated contamination levels are found to be orders of magnitude above the levels of the incoming inert purge gas.