No abstract

No abstract

Extreme ultraviolet lithography (EUVL) is a potential candidate for the next generation lithography techniques, which will use Xe or Sn as a main fuel to produce EUV light. However, the industry has favored to use Sn as main fuel in EUVL systems because of its high conversion efficiency over Xe. Sn has an advantage of producing more light, but on the other hand its condensable nature is a real threat to the reflective mirrors which are used to collect the EUV light at intermediate focus. Center for Plasma Material Interactions (CPMI) at the University of Illinois has studied plasma etching as a potential method of Sn removal from collector optics. RF-driven chlorine plasma is used to etch Sn from mirror samples. Previously we reported high selectivity of Sn over several EUV compatible mirror materials. The increased confidence in this technology had led us to perform cleaning experiments on real Sn contaminated samples exposed in an EUV source and the results obtained have been very encouraging. Small mock up shells (same as in the grazing incidence collector optics system) were constructed at CPMI and chlorine etching was performed at different samples placed at different locations on multi-shell collector mock up in ICP-RIE chamber. Post cleaning material characterization results of samples shows that chlorine can potentially clean Sn off of collector optics (Ru was used in this study as a mirror sample). Realizing this as a viable cleaning solution, we have stepped further and performed a full size cleaning test in the Xtreme's XTS 13-35 EUV source. Large mock up with appropriate dimension was placed in the EUV source chamber and the cleaning system was installed to etch Sn away from Ru surface. This study compares the cleaning results in a real system scale with the previous simulated system. The comparison shows how to improve the Sn cleaning system in the EUV source chamber. Results are encouraging and may enable source suppliers to integrate this technology in their respective sources. Cleaning rate was measured as >100nm/min using ion sputtered Sn samples.

No abstract

Tin ͑Sn͒ has the advantage of delivering higher conversion efficiency compared to other fuel materials ͑e.g., Xe or Li͒ in an extreme ultraviolet ͑EUV͒ source, a necessary component for the leading next generation lithography. However, the use of a condensable fuel in a lithography system leads to some additional challenges for maintaining a satisfactory lifetime of the collector optics. A critical issue leading to decreased mirror lifetime is the buildup of debris on the surface of the primary mirror that comes from the use of Sn in either gas discharge produced plasma ͑GDPP͒ or laser produced plasma ͑LPP͒. This leads to a decreased reflectivity from the added material thickness and increased surface roughness that contributes to scattering. Inductively coupled plasma reactive ion etching with halide ions is one potential solution to this problem. This article presents results for etch rate and selectivity of Sn over SiO 2 and Ru. The Sn etch rate in a chlorine plasma is found to be much higher ͑of the order of hundreds of nm/min͒ than the etch rate of other materials. A thermally evaporated Sn on Ru sample was prepared and cleaned using an inductively coupled plasma etching method. Cleaning was confirmed using several material characterization techniques. Furthermore, a collector mock-up shell was then constructed and etching was performed on Sn samples prepared in a Sn EUV source using an optimized etching recipe. The sample surface before and after cleaning was analyzed by atomic force microscopy, x-ray photoelectron spectroscopy, and Auger electron spectroscopy. The results show the dependence of etch rate on the location of Sn samples placed on the collector mock-up shell.