抄録
High-n immersion lithography has been under development for the past three years, the target being to extend Optical Lithography to the 32nm node and beyond. Feasibility studies have been generating results on all aspects of the technology. This paper reports results obtained in key areas: High-n Immersion Fluid performance and lifetime; High-n Final Lens material development; Defect generation and control.
High-n Immersion Fluids have been developed by companies such as DuPont and JSR. The fluids typically have a refractive index of 1.65 and will support a system numerical aperture of >1.55. These new fluids have been engineered to a point where they are now more transparent than water at the exposing wavelength of 193nm, but challenges remain. Test results are reported from the UV The studies indicate that the UV irradiation of the fluid at high doses causes a build up of carbonaceous contamination on the final lens element surface. Cleaning procedures have been developed and calculations of the expected cleaning schedule made.,br>A key requirement for this technology is the availability of high refractive index final lens element material. Final lens element material must have a refractive index >1.70 to support optics design with >1.55NA; it must also have sufficient optical transparency. The availability of viable high-n immersion fluid means that the supply of optical grade high-n material for the final lens element fabrication is the potential technology show stopper. High-n materials (>1.9) have been identified. Initial technical progress with materials such as LuAG and Spinel has been encouraging; the focus has been on obtaining starting materials that are pure enough to ensure a final optical material with a transparency that is high enough at 193nm wavelength. There is, though, a risk that availability of optical grade material will miss the window for the technology.
Defect generation is another major concern with this Immersion Lithography. We have studied the effect on defect generation of using the high-n immersion fluid instead of water. There are defect generating mechanisms specific to the use of high-n immersion fluid. High-n immersion fluid is much more difficult to purify than water, and the purification process can generate particles which must be removed.
The new high-n immersion lithography can be assessed with regard to the Semiconductor Industry roadmap. The relatively slow development of high-n optical material implies that the potential for the technology now lies in the context of Double Pattern Processing for application to sub-32nm nodes.
