In the past several years, ArF immersion lithography has been developed rapidly for practical applications. ArF
immersion lithography is now researched actively and developed for the purpose of implementing the 45-nm technology
node. For the device designs involved immersion lithography, line width roughness (LWR) and film wettability are very
important criteria to control in the point of high resolution and defectivity.
Free radical polymerization in the presence of thiocarbonylthio compounds of general structure Z-C(=S)S-R provides
living polymers of predetermined molecular weight and narrow molecular weight distribution by a process of reversible
addition-fragmentation chain transfer (RAFT). A rationale for selecting the most appropriate thiocarbonylthio
compounds for a particular monomer type is presented with reference to the polymerization of methacrylates, styrenes,
acrylates, acrylamides, and vinyl acetate.
In this study, resist polymers with narrow polydispersity (PD) and controlled molecular structure were prepared using
controlled radical polymerization techniques, such as RAFT polymerization. PD index of polymers showed between
about 1.2 to 1.4 and in some instances, between about 1.1 to 1.2 or less. Additionally, each polymer chain has a RAFT
end group. That is the resulting polymer contains a chain transfer agent (CTA) moiety at each terminal end of polymer
backbone. It is possible that hydrophobic CTAs can be used to decrease the hydrophilicity of resist film.
ArF Immersion lithography is the most promising technology for 45nm node and possibly beyond. However, serious
issues in ArF immersion lithography for semiconductor mass production still exist. One of the issues is immersion
specific defects, which are caused by photoresist component leaching and residual water droplets. In order to minimize
immersion specific defects, preventing water penetration into the resist film is regarded as an important factor. Several
research groups have reported that higher receding contact angle reduced defectivity. High receding contact angle of
film surface prevent water penetration into the resist film due to the hydrophobic nature. Resist component leaching
phenomenon also can be caused by the water penetration into the film, so hydrophobic resist can reduce leaching
In this paper, to investigate chemical leaching from resist surface, we evaluated the leaching value of PAG anion and
contact angles of various polymers according to their hydrophobicity. Hydrophilicity of a polymer was changed by the
degree of hydrophobic group substitution to polymer chain. We measured receding contact angle with four different
resists composed of water-repellent functiona group. Receding contact angle of resist surface increased as the portion of
water-repellent functional group increased. Also, the leaching amount of PAG anion decreased as the receding contact
angle of film surface increased. We expect that higher receding contact angle prevents chemical leaching from resist film
by repelling water at the surface. We will report detailed results in this paper.
Immersion materials have to overcome immersion-issues for successful wet process introduction to semiconductor
mass production. Component-leaching issue is one of the most influential wet process huddles, which is related to
immersion-liquid and projection lens contamination as well as resist patterning performances. In this paper, we will
introduce our experimental results of leaching blocking effects resulted from the modification of polymer and additive
structures and from the application of top surface blocking layers. PAG-leaching level of resist film formed of low Tg
resin shows the highest meanwhile that of high Tg resin is the smallest leaching value. The interaction forces between
additives and resin platforms are the most important to prevent additives leaching to immersion liquid.
We have tested 3 different types of resin structures to modify the interaction forces between resin platform and resist
components especially PAG molecules and photo-generated acid molecules. We changed 2-hydroxyethyl
methacrylate(2-HEMA) contents to be 5, 10, 15% in our base resin, COMA-acrylate hybrid system to modify the
hydrophilicity of resist platforms. By mimicking immersion process to obtain wet-performance of their resists we have
obtained relative value of component-leaching. Interaction-force between resist platform and PAG was seemed to be
largest when resist component-leaching is least so that the pattern profiles become to be vertical. It was appeared that
the 5% 2-HEMA containing resin and TPS-Nonaflate PAG system showed the best performance because of its low
leaching resulted from their strong interaction forces.
Another polymer parameter to determine the component-diffusivity is glass transition temperature, Tg. Low Tg
means high mobility of resin by small thermal energy due to high free volume contents inside of the resist film which
can act as diffusion pathways of resist components. 10% MA resin system shows the lowest Tg, around 140 degrees C and the
most serious T-topping profiles.
Defectivity issue becomes more important in immersion process, which is most related to the partial leaching by
residual water and resulted in watermark defects. We evaluated the relative easiness of watermark formation on
different resist films by correlations between static and dynamic contact angles. We made 2 different resist films
composed of 2 different water-repellent functional group monomers. Static contact angles of developer as well as DI
water were measured. It was found indirectly that amphiphilicity of resist film surface makes positive effect on
watermarks formation, that is, more hydrophobic and more developer-philic resist film remains less residual water
Post exposure bake temperature sensitivity (PEB sensitivity) becomes important as the pattern pitch size shrinks gradually. There are several factors affecting the PEB sensitivity including acidity and diffusion of photogenerated acid, activation energy for deprotection reaction, free volume of base polymer, and so on. Our works were conducted as a part of the basic study for searching influential parameter of PEB sensitivity. We found that PEB sensitivity relies largely on not only acid diffusion parameter, but also the hydrophilicity of base polymer and protection group ratio. Also, we observed that bulkiness of deprotection group has great influence on PEB sensitivity. Detailed results will be reported in this paper.
Immersion lithography has drawn tons of interests as a potential solution for sub-65nm patterning. High refractive index liquid, which is filled in the gap between exposure lens and a photoresist, can improve a resolution through increased effective numerical aperture (NA) of the exposure system. Most attractive liquid for this purpose is water. Our works were conducted as a part of the basic study for immersion lithography and aimed for the verification of leached resist components by water. It was observed that leaching relies largely on the free volume of a polymer and anion size of photoacid generator (PAG). The larger free volume and the smaller anion, the larger T-top resist profile was generated. Additionally, effects of solvents, quenchers and polarity of the polymer were investigated. Detailed results will be reported in this paper.
Post exposure bake temperature sensitivity (PEB sensitivity) is getting important for below 100nm device. There are several factors affecting the PEB sensitivity including acidity and diffusion of photogenerated acid, stiffness and free volume of base polymer, and so on. Among them, the activation energy for deprotection reaction is regarded as the most critical factor. We have investigated the influence of protection group with various activation energies as well as Tg of polymer. Several different protection groups were incorporated into the polymer chain to modify activation energy of the resist. Also, we have investigated the influence of acid diffusion and quencher diffusion ability on PEB sensitivity. Three photoresists were formulated with different concentration of acid diffusion controller to asses the influence of acid diffusion on CD variation. And to evaluate the effect of quencher diffusivity on CD change, photoresist was formulated by adding amines having various different molecular size. Detailed results and new resist with reduced the PEB sensitivity will be reported in this paper.
We have been interested in the effect of the residual solvent on lithographic performance. The concentration distribution of solvent molecules along the film depth and the amount of residual solvents depend on their physical properties: evaporation rate, boiling point, viscosity, and so on. Since fast-evaporating solvent can make a dense skin-like layer at the top of the resist film, faster evaporation rate of solvent makes thicker film, while slow rate results in thinner film. And the amount of residual solvent is dependent of the evaporation rate of the casting solvent. The amount of residual solvent was verified by TGA method. It was found that the amount of residual solvent is a major parameter to determine film thickness, stiffness of resist pattern, acid diffusion length, and pattern profile shape.