A highly transparent (60% transmittance at 120-nm thickness: abs. = 1 .85/tim), fluorine-containing, silsesquioxane-type resist for 157-nm lithography has been developed. When the resist was exposed with a 0.85-numerical--aperture (0.85-NA) microstepper and a phase-shifting mask, the high transmittance resulted in a steep profile for a 55-nm 1 : 1 line and space (L/S) pattern, as well as a feasible depth of focus (DOE) of 0.2 im for a 100-nm contact hole (C/H) pattern. By using a 157-nm bi-layer resist process, which employed 120 nm of silsesquioxane-type resist as the top layer and a 200-nm-thick organic film as the underlayer, a sub-100-nm C/H pattern could be successfully fabricated and transferred to a 400-nm-thick Si02 film by reactive ion etching (RIE). Neither pattern deformation during RIE nor residue after resist ashing was observed. The successful fabrication of a sub-100-nm C/H pattern in 400-nm-thick Si02 clearly demonstrated the advantage of the 157-nm bi-layer resist process for fabricating sub-65-nm-node semiconductor devices, especially for C/H fabrication or damascene processes.
The bilayer process we developed for 157-nm lithography uses a fluorine-containing silsesquioxane-type resist (F-SSQ). Gate fabrication is done by using a F-SSQ(90 nm)/organic film(200 nm)/poly-Si(150 nm)/SiO 2 (10 nm)/Si structure. The organic film works well as an anti-reflecting layer. Using a microstepper with a numerical aperture of 0.90 and optimizing the resist thickness, we made a 50-nm 1:1 line-and-space (L/S) pattern by using an alternative phase-shifting mask and made a 45-nm SRAM by using a chromeless phase lithography mask. Neither resist pattern footing nor undercutting was observed on the organic film. The reactive ion etching (RIE) selectivity between the F-SSQ and the organic film was sufficient (about 7), the resist pattern was transferred to the underlayer, and both 50-nm 1:1 L/S and 45-nm SRAM gate patterns were made using the organic film as an etching mask. Contact hole (C/H) fabrication is done by using a F-SSQ(105 nm)/organic film(400 nm)/tetraethyl orthosilicate (TEOS)-SiO 2 (1200 nm)/Si structure, and we made a 75-nm 1:1 C/H pattern by using the microstepper with a binary mask. The RIE selectivity was sufficient (about 15) for making high-aspect-ratio contact holes, and we made a 75-nm 1:1 C/H pattern in 1200-nmthick TEOS. This bilayer process is thus promising for making 65-nm-node semiconductor devices.
This paper describes the investigation on the feasibility of current coater/developer processes to the 157-nm lithography from the viewpoint of critical dimension (CD) control. The effect of coating, bake, and development process on the CD of a 157-nm resist, where fluorine is introduced in the side chain, is studied. A KrF and ArF resist is also used for comparison. Firstly, as for the coating process, the coverage performance and the film thickness uniformity of the 157-nm resist shows that the current coating methods are feasible to 157-nm resist, even though the 157-nm resist is highly hydrophobic. Secondly, as for the bake process, the post exposure bake (PEB) temperature dependence of CD for the 157-nm resist is found to be lower than that for 248 and 193-nm resist. This means that our current PEB temperature control system, which is suitable for 248 or 193-nm resist, is also effective for the 157-nm resist. Thirdly, as for the development process, it is found that a static puddle formation process shows a smaller line edge roughness (LER) than a dynamic puddle formation process. Therefore, the static puddle formation process, with the use of linear drive (LD) developer nozzle for instance, is attractive for the 157-nm resist process. Lastly, from the viewpoint of contamination control, it is found that the amine level should be controlled to be less than 0.1ppb in order to prevent the CD change during post exposure delay (PED) for the 157-nm resist.
The ammonia durability of the 157-nm lithography resists is still unclear due to the smaller target dimensions, thinner resist films, and variations in base polymer compared to those of 193-nm and 248-nm resists. It has not been determined what ammonia concentrations must be achieved in order to successfully process 157-nm resists. Until now, the ammonia durability of initial 157-nm resists during post exposure delay (PED) and during post coating delay (PCD) was compared to those of 193-nm and 248-nm resists. It was confirmed that all initial 157-nm resists had low ammonia durability. In this paper, the ammonia durability of newly developed 157-nm resists, that have improved transmittance and resolution, was evaluated during PED and PCD. Then, we found that the ammonia durability of these resists were not enough and that the ammonia concentration from exposure to development should be kept under 0.1 ppb. Thermal desorption spectroscopy results showed that resists with lower ammonia durability tended to have more amount of adsorbed ammonia than other resists. Furthermore, the ammonia durability of 157-nm resist couldn't be improved to the level of that of 193-and 248-nm resist by the adjustment amount of resist additives. Due to the low ammonia durability, it will be necessary to control the ammonia concentration below 0.1 ppb in processing equipment used in 157-nm lithography.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.