ABSTRACT

In order for any metrological method to be usable in the fab, there must be an automated and reliable method of extracting the data without any operator intervention or (worse) personal judgment. As with many methods of optical metrology, x-ray metrology (XRM) is not a ruler, nor is it a system with an output that is monotonically related to thickness or any other parameter. Interpretation from x-ray scattering theory is required. This chapter discusses the methods by which this may be achieved. Fortunately, as has been discussed in Part 2 of this book, x-ray theory is extremely sound and comprehensive. It is based upon few assumptions, all of which are easily validated in practice. In particular, the x-ray scattering factors are well known. As a consequence of the high energies of x-rays in comparison with energies of valence effects in solids, they are invariant with chemical or physical state to sufficient accuracy. We can therefore assert with confidence that if we have a wafer whose film structure is known, and a metrology tool whose characteristics, such as beam size and divergence, are known, we may calculate the expected diffraction or reflectivity pattern with high accuracy.