Immersion lithography has gone through its first phase of introduction and acceptance as the main solution for critical
layer lithography for 45nm node and beyond. In this phase, the industry has found that immersion technology has its
own unique challenges associated with introducing water as a medium between the projection lens and wafer. Resist
process qualification is once again under the spot light.
Due to the rapid introduction of immersion technology resist suppliers did not have sufficient time to
reformulate their standard ArF resist processes to be compatible with water while at the same time satisfying critical
imaging, etching and other requirements. For this reason a barrier (topcoat) had to be introduced in order to prevent
resist leaching as well as to produce a more desirable surface for water to glide over. Introducing a top-coat created
challenges for all parties involved: scanner manufacturers resist vendors and the end users. Since each manufacturer has
its own unique technology for introducing immersion water, top-coat/resist processes needed not only to meet the end
users' performance criteria but also meet each scanner manufacturer's requirements. Therefore material screening
process and process evaluation became an important factor in immersion technology processes. Defectivity became the
primary criterion for the resist process.
The responsibility of the scanner manufacturer is twofold: first, to produce a system compatible with many
different resist processes while not introducing additional defects, and second, to give resist manufacturers clear and
concise requirements for achieving performance. In this paper we show how we have met the industry's needs in this
area. First, we discuss the importance of material screening, including requirements for hydrophobicity, leaching, and
peeling. Second, we present defectivity and other experimental data from practical materials that fulfill all requirements.
Cases will be shown wherein an immersion process using commercially available resist processes introduces no
additional defects. Several of these now do not require a topcoat. We therefore show that the industry's needs have been
met with both topcoat and topcoat-less processes.