The Use of Multivariate Quality Control Chart in Solving the Thin-Film Thickness Uniformity Quality Problem - a Plasma Sputtering Case in TFT-LCD Manufacturing

Sputtering process is very important in the TFT-LCD manufacturing processes. During manufacturing, a glass (raw material) is plated with a thin metallic coating, which serves as the color frame for the LCD panels. The sputtering coating thickness affects the final quality of products, so the thin-film thickness uniformity issue has to be considered in this process. In the measurements of thin-film thickness uniformity diagnosis, the basis of diagnosis is the mean thickness of a thin-film which is measured in fixed positions on a glass. The measurement of quality in the sputtering process is apt to generate erroneous judgment and not to conform to the quality requirement of thickness uniformity effectively in the sputtering process. The issue of quality measurement of the thin-film thickness uniformity for (plasma) sputtering in TFT-LCD manufacturing is discussed. An effective method is provided in order to diagnose the conforming and nonconforming items. The mean, standard deviation and range of the thickness are adopted as the quality characteristics, as opposed to the general approach that only considers thickness mean. In this study, the x-bar control chart and multivariate quality control chart are used to diagnose the thick quality of sputtering process separately. These application studies demonstrate that, in comparison to traditional control charts, the multivariate quality control chart is more accurate and efficient.