STATISTICAL PROCESS ANALYSIS OF ION-IMPLANTATION

This paper will review the theory of control chart limits and report how well theory correlates with statistical analysis of process data taken from a high current implanter over a period of six weeks. In particular, the mean, moving range, and standard deviation will be computed from a population of 5-, 9-, and 45-site measurements of sheet resistance taken from implanted and annealed test monitors. Control chart limits will be calculated from both the moving range and population standard deviation. The efficiency of these statistics will be discussed relative to their ability to detect process anomalies. The intent of this paper is to enable present day ion implant engineers to understand and improve their existing data and trend analysis methodology.