Abstract

EUV lithography is now considered to be the most likely technology to follow 193 nm deep UV lithography to satisfy the needs for computer chip manufacture with feature sizes of 70 nm and lower.¹ The roadmap for this technology, which is based on narrow-band (<3%), multi-layer coated reflective optics, calls for the development of stable, debris-free, light sources producing collecta- We in-band emission at 〜13.5 nm with power levels of 〜100 W.² This challenge is currently being met by two light source technologies, small dense plasma electric discharge lamps, and high repetition-rate laser plasma sources. The latter is now favored as the source of choice for the first engineering tests of an EUV lithography tool.³ For laser plasmas to be a viable choice for this application they must operate continuously at repetition rates of ~5 kHz with a pulse-to-pulse stability of better than 2% in a regime that protects the high NA (>0.25) collection optics from deleterious effects of target debris for periods of ~ 1 year. Moreover they must demonstrate a conversion efficient from laser light to in-band EUV plasma emission sufficiently high to (i) provide the projected required collectable power levels with viable commercial lasers and (ii) permit the overall cost of the source, including the laser, to remain within economic models of the overall EUV lithography stepper tool.

© 2002 Optical Society of America