You can't do today's job with yesterday's methods and be in business tomorrow. - Anonymous
It should be obvious at this point that scatter measurement is a source of metrology not only for the laboratory, but also for many production applications. The purpose of this chapter is to review several of these applications. Some applications require only that changes in a production process are detected, with no need to quantify the change in terms of surface statistics or defect density. The semiconductor and computer disk industries are high-tech examples of situations where the increased use of light-scatter metrology is being pushed by tighter product requirements. Many general manufacturing applications rely on the experienced human eye to qualify a product. Some of those experienced eyes have reached retirement age, and scatter metrology, which amounts to a quantifiable measure of product appearance, offers a way to standardize quality control without retraining. These applications rely on the combination of high-speed, noncontact quantification of product characteristics that scatter measurement offers.
A large portion of this book has been devoted to obtaining surface-roughness parameters from measured surface scatter. The various conditions necessary for these calculations to be made accurately have been discussed in some detail. In the optics and semiconductor industries the smooth, clean, front-surface requirements are met in many situations. However, in other industries this is often not the case. Fortunately, in many applications it is not necessary to actually compute roughness (or BSDF) parameters. It is enough to be able to quickly detect a difference in scatter level that, through experience, can be related to a change in product quality. Experience is gained for these relative-measurement situations by examining good and bad product samples in a controlled laboratory situation. In this manner, the polarization, wavelength, incident angle, and scatter angles that optimize detection of changes in quality can be determined. The laboratory situation is used to simulate the desired in-process instrumentation.