A parametric investigation of a high current implanter for long term process performance improvements

The objective of this study is to understand and model the impact of implanter sub-system performance on the final process results. We have analyzed the time-dependent performance of a GSD implanter that has been in production for several years. Data analysis was done using multivariate statistical techniques by which process results are directly correlated to various hardware performance parameters, such as pressure or source operating and beam tuning parameters. This process data has been analyzed and we have found that a strong correlation exists in source operational conditions with various implanter specific metrics, as would be expected. The correlation with actual process results is a more complex problem, with many external factors affecting the final results, rather than a simple cause and effect relationship. Implementation of such a predictive process control model, could improve the performance and uptime while increasing device repeatability and subsequent yield