Extreme ultraviolet lithography at a wavelength of 13.5 nm has been prepared for next generation lithography for several
years. Of primary concern in EUV lithography is line edge roughness as well as high sensitivity. In recent years, various
types of resist, such as protected PHS resin resist and molecular resist, have been investigated. In order to reduce LER,
we have studied novel molecular resists which are promising alternative to polymeric photoresists for use as imaging
materials with improved resolution and line edge roughness. The work reported in this paper has focused on the
development of a new class of chemically amplified molecular resists that are composed of a single molecule which
contains all of the different functionalities desired in a chemically amplified resists. For the purpose of improvement of
the resist performance, we have designed the resist material of a protected polyphenol derivative (protected Compound
A). PAG moiety is bonded to Compound A to achieve uniform PAG density and to control the acid diffusion length in a
resist film. We analyzed uniformity of PAG density in a resist film by using gradient shaving preparation and TOF-SIMS
analysis. From the TOF-SIMS spectra, the ions intensities of the PAG moiety are almost constant from the surface to the
bottom of the film. Therefore, we can conclude that PAG is distributed homogeneously. Under e-beam exposure, a
100nm thick film of the PAG bonded molecular resist resolved lines down to 100nm. We also discussed the new design
for molecular resists, their synthesis and lithographic performance.
Our recent research effort has been focused on new top coating-free 193nm immersion resists with regard to leaching of
the resist components and lithographic performance. We have examined methacrylate-based resins that control the
surface properties of ArF resists thin films by surface segregation behavior. For a better understanding of the surface
properties of thin films, we prepared the six resins (Resin 1-6) that have three types fluorine containing monomers, a new
monomer (Monomer A), Monomer B and Monomer C, respectively. We blended the base polymer (Resin 0) with Resin
(1-6), respectively. We evaluated contact angles, surface properties and lithographic performances of the polymer blend
resists. The static and receding contact angles of the resist that contains Resin (1-6) are greater than that of the base
polymer (Resin 0) resist. The chemical composition of the surface of blend polymers was investigated with X-ray
photoelectron spectroscopy (XPS). It was shown that there was significant segregation of the fluorine containing resins
to the surface of the blend films. We analyzed Quantitative Structure-Property Relationships (QSPR) between the surface
properties and the chemical composition of the surface of polymer blend resists. The addition of 10 wt% of the polymer
(Resin 1-6) to the base polymer (Resin 0) did not influence the lithographic performance. Consequently, the surface
properties of resist thin films can be tailored by the appropriate choice of fluorine containing polymer blends.
With no apparent showstopper in sight, the adoption of ArF immersion technology into device mass production is not a matter of 'if' but a matter of 'when'. As the technology matures at an unprecedented speed, many of initial technical difficulties have been cleared away and the use of a protective layer known as top coat, initially regarded as a must, now becomes optional, for example. Our focus of interest has also sifted to more practical and production related issues such as defect reducing and performance enhancement. Two major types of immersion specific defects, bubbles and a large number of microbridges, were observed and reported elsewhere. The bubble defects seem to decrease by improvement of exposure tool. But the other type defect - probably from residual water spots - is still a problem. We suspect that the acid leaching from resist film causes microbridges. When small water spots were remained on resist surface after exposure, acid catalyst in resist film is leaching into the water spots even though at room temperature. After water from the spot is dried up, acid molecules are condensed at resist film surface. As a result, in the bulk of resist film, acid depletion region is generated underneath the water spot. Acid catalyzed deprotection reaction is not completed at this acid shortage region later in the PEB process resulting in microbridge type defect formation. Similar mechanism was suggested by Kanna et al, they suggested the water evaporation on PEB plate. This hypothesis led us to focus on reducing acid leaching to decrease residual water spot-related defect. This paper reports our leaching measurement results and low leaching photoresist materials satisfying the current leaching requirements outlined by tool makers without topcoat layer. On the other hand, Nakano et al reported that the higher receding contact angle reduced defectivity. The higher receding contact angle is also a key item to increase scan speed. The effort to increase the receding contact angle become very important issue for not only defectivity but also scanner throughput. Some of our experimental results along this line of study are also included in the report. The last topic covered is LWR (Line Width Roughness) as an essential leverage for performance improvement, especially for the smaller CD that immersion lithography is aiming to define. Our recent effort to find effect and working concept to reduce LWR with low leaching materials is also described.
Numerous resolution enhancement techniques have been introduced over the past few years as the design rule decreases rapidly. Among them are thermal reflow, SAFIER and RELACS just to name a few. Resist reflow is one of the simplest processes with a minimum process modification that only requires an additional baking step at or above its glass transition temperature after the contact holes have been developed. Since most of the methacrylic-based ArF resins have Tgs in vicinity of their thermal decomposition temperature, it is not desirable to expose the resins near Tg for a prolonged time. An approach to construct a resin that is physically or thermally viable, yet chemically stable is necessary and the easiest way of achieving this goal is to bring down Tg of the resin significantly so that there would be enough working space between thermal decomposition and glass transition. Out of several conceivable ways to lower the Tgs such as employing acrylic polymers, COMA type polymers etc., we have chosen to maintain the methacrylic platform because of its superior resolution capability. Our design strategy is to work on the pendent groups of methacrylic monomers to make polymer matrix more flexible. Thus, the incorporation of a more flexible unit, such as 2-methyl-2-adamantyloxycarbonylmethyl methacrylate, in our existing copolymer system reduced Tg almost by 30°C. In addition to its thermal property tuning ability, the resist sensitivity also has increased, presumably due to the out-stretched position of an acid labile protecting group for easy access of incoming acid molecules. Our newly developed resists based on the design concept showed a good C/H pattern profile and improved LER by reflow process at sub-70 nm node. We will discuss our newly designed materials in this paper in terms of material properties, resist characteristics and lithographic performances in relation to reflow processes.
We compared 2MAdMA(2-Metyl-2- Adamantylmethacrylate)/GBLMA((gamma) -butyrolactone methacrylate) resin system and 2EAdMA(2-Etyl-2- Adamantylmethacrylate)/GBLMA resin system. 2EAdMA/GBLMA resin system showed higher sensitivity, dissolution contrast and better adhesion to silicon substrate than 2MAdMA/GBLMA resin system. These results shows that 2EAdMA/GBLMA resin system is suitable for practical ArF positive resist.
Alicyclic groups are preferable resin components of ArF resists due to better dry-etching resistance and higher transparency at 193 nm. On the other hand, Alicyclic groups bring poor adhesion of ArF resists during wet development, because of their higher hydrophobic nature. To avoid the peeling problem diluted developer has been suggested to use. However, the compatibility with existing standard developer of i-line and KrF resists is necessary for the mass production. In this paper we compared two kinds of resists for the standard developer (TMAH 2.38%) application. The former has AdCEE unit and norbornene derivative/maleic anhydride alternating copolymer, together with relatively weak organic acid generating PAG. The latter having 2MAdMA/GBLMA copolymer and onium salt PAG shows better lithographic performance.