EUVL is the "preferred solution" for 45nm/32nm nodes
Research consortia presented their activities at the European EUV Workshop 2002 (Grenoble, 27/28 Febr. 2002) organised by MEDEA+
There were 130 participants attending this workshop, 65 from France, 28 from Germany, 17 from Japan and US. The others came from other European countries.
The goal of this workshop was to present to the European EUV community the EUV activities of Japan and the US, and to present to the representatives of these programmes the EUV activities in Europe. Hideo Setoya (CEO of ASET) gave an overview on the EUV activities in Japan, which are carried out by ASET, MIRAI and the companies Gigaphoton, Canon and Nikon. Chuck Gwyn (Director, EUV-LLC) presented the status and the results of the EUV-LLC consortium in the United States. Giang Dao (Director for Lithography at International Sematech - ISMT) reported on the activities on Next Generation Lithography and the EUV activities of ISMT. In his written concluding remarks he explicitly stated for the EUV Lithography Program at International Sematech that the
"NGL process resulted in EUVL as preferred solution for 45nm/32nm nodes"!
Three quarters of the workshop were dedicated to the presentations of European work and results. Overviews of the results achieved in the French project PREUVE and the IST project EUCLIDES were given. These projects paved the way for work in the MEDEA+ projects.
The presentations were structured along the main R&D activities of EUV Lithography: EUV-Source, Exposure Tool and Optics, and Mask Fabrication.
A source which emits a radiation at the right wavelength for EUV lithography is a very critical issue. Actually the August 2001 Sematech NGL Workshop identified it as the "highest risk item for EUVL". Optical elements for EUV radiation can only be made by mirrors, which consist of multi layers of thin films of alternating materials with different atomic numbers. The wavelength of the source, the film thickness and the atomic numbers of the elements of the thin films are dependent on each other. Previous studies in all the three economic regions revealed that the materials for the thin films should be silicon and molybdenum. This combination of materials is optimized for a wavelength of 13.4 nm.
There are various methods to generate radiation at 13.4 nm. In the MEDEA+ project on sources four alternatives are investigated. The goal of these alternatives is to drastically increase the intensity of the respective source. Up to now there has been no application which used such kind of a source, and the status of development therefore is still in its infancy. The intensity of the source has to be increased because a large amount of intensity is lost at the mirrors (only 70% of the incoming intensity is reflected, the de-magnifying optics consists of 6 mirrors). In addition, the sensitivity of the resist cannot be increased to very high values because this material must withstand the etching steps after lithography.
MEDEA+ speakers presented activities focussing on two source technologies in which the plasma, which emits the radiation, is exited by a laser source and two additional source types, where the plasma is exited by an electrical discharge. After one and a half years the status of the source development will be analysed and future work will concentrate on the most promising alternatives.
The next topic was the development of an exposure tool with the relevant optics. In the US programme EUV-LLC a lot of know-how has been accumulated in the whole field of EUV lithography. The exposure tools will, however, be fabricated by European and Japanese companies.
Europe's progress in this field with the two partners ASML and Zeiss is excellent. ASML is fully confident that the main specifications of the source for their a-exposure-tool will be met. And Zeiss impressed the audience with astonishing details of the technical challenge in manufacturing the EUVL mirrors for the demagnifying optics. Mapping the specifications for these mirrors (100 mm width) into the 'real world' of Germany's geography (1000 km north-south), the audience discovered that the maximally tolerable 'peak to valley' error of mirror surface roughness of 0.6nm corresponds to the shrinking of Germany's 2994m Zugspitze peak to just 6 mm high!
Another very difficult development is the mask technology for EUV lithography. An EUV mask can be several mm thick, which is an advantage compared to the stencil masks necessary for electron projection lithography. But an EUV mask is also a mirror and must start with an extremely flat mask substrate. An additional difficulty is that due to the very strong absorption of EUV radiation in any material no pellicle can be used to protect the mask against particles. That in turn creates the big problem of how to keep a mask clean during transport, usage, and storage. First results were presented on mask substrate fabrication and flatness measuring techniques, on mask blank preparation, on defect detection and defect repair and the technique to define the absorbing pattern on the mask. An intensive discussion followed as to whether some defects on the mask might only be detected with an EUV microscope which would be a very expensive instrument.
Generally speaking, for each of the topics it turned out that due to the smaller dimensions and the new techniques used for EUV lithography, a lot of new measuring techniques have to be developed. National calibration bodies like PTB in Berlin and NIST in the US are presently doing this work.
At the end of the workshop the representatives of the programmes in Japan and in the US were asked for their impressions about this European EUV Workshop 2002. They all mentioned that they were very much impressed by the activities in Europe, which up to now were not so evident to them. Especially the intensive R&D on sources for EUV was highly appreciated.
After the meeting it was recommended to organize such a workshop once a year, always alternating with the EUV symposium and moving from region to region.
As stated in one of the presentations during the European EUV workshop, "we need the worldwide information exchange to narrow options, arrive at reliable standards, convince tool suppliers and bring EUV lithography to life", for the economic benefits of Europe and of the entire electronics world.
