The development of robust, high resolution, positive-and negative-tone resists is necessary for the eventual implementation of DUV lithography in a manufacturing environment. Great progress has been made in the last two years in the area of chemically amplified resists. Environmentally stable, high resolution and high thermal resistance negative-tone resists have been commercially available for three years. Prototype DUV positive-tone resists are now achieving better environmental stability through the use of partially blocked poly(pvinyl)phenol polymer and photoacid generators (PAG) based on organic onium salts. In this paper, our studies of various PAG structures in positive-tone DUV resists are reported. The PAG structure, M+ X-was varied such that M+ was either triphenylsulfoniuni (TPS+) or diphenyliodonium (DPfl), and X-was trif luoromethanesulfonate (TFA),-toluenesulfonate (TSA),-eamphorsulfonate(CSA-), and hexadecylsulfonate (HDSA).-The relative photospeed of these resists corresponded to the pKa of the photogenerated acid generated from the anion, TFA>TSA>HDSA>CSA. The TPSTSA-based resist showed the lowest diffusion coefficient, D = 2.7 x 10-5 ,uni2/s , as estimated from linewidth vs, postexposurebake(PEB) time plots. This paper also shows a unique advantage of onium salts in that they can show a maximum dissolution rate ratio, Rp/Rrnin, at less blocked polymer levels, where Rp is the dissolution rate of the blocked polymer, and R,77,1 is the dissolution rate of the unexposed blocked polyrner/5% onium salt. The dissolution rate ratio was 250 using a 10% blocked polymer, which approaches the inhibition capability of PAC/Novolak systems. This work on PAG structural variation has led to the development of XP-9402 positive DUV resist, which is linear to 0.225 um on a 0.53 NA excimer laser stepper at a photospeed of 46 inJ/cm2. This resist shows good postexposure delay stability for up to two hours, without the use of a covercoat. The second part of this paper discusses our work on negative DUV resists, where low molecular weight poly(p-vinyl)phenol (PVP) provides highest resolution performance in negative DUV resists. The PVP-based resists suffer from microbridging in 0.26 N TMAH developer. Finally, it is shown that with altered PVP, high resolution with negative DUV resists without niicrobridging in 0.26 N TMAH developer can be achieved.

No abstract

No abstract

The effect ofresin molecular weight on the lithographic properties ofnegative DUV resist is discussed. A wide range ofmolecular weights (3500 to 240000) of poly(p-vinyl)phenol was studied. Polymer dissolution rate vs. molecular weight followed a simple kinetic equation, with the kinetic order m =2.0. The photospeed of the resist was not strongly affected by the starting resin molecular weight; however, resolution decreased rapidly with increasing Mw. Also, the higher the starting molecular weight, the greater the tendency for the resist to form microbrdges between features. The microbridges could be as long as 1.0/imfor the highest molecular weight resin, Mw 240000. The lowest molecular weight resins, Mw 5000, showed no evidence ofmicrobridges even at the resolution limit of the stepper, 0.32wn, with 0.14jj development. However, higher normality developer, such as 0.261j TMAH, showed microbridges across 0.42wnfeatures with Mw 3500 resin. For comparison, a m,p-cresol novolak, also 3500 molecular weight, showed no evidence of microbridging in the higher normality developer. Ther reason for this difference s that the novolak does not crosslink as effectively as PVP does. Based on extraction experiments, it has been shown that the molecular weight at a sizing dose is 1 64000 for PVP and 6500 for the novolak. Thus, the novolak must react with the melamine primarily through an intrachain reaction, whereas the PVP-melamine reaction is an interchain reaction. Finally, a mechanism for microbridge formation is discussed. 152 / SPIE Vol. 2195 O-8194-1490-5/94/$6.OO Downloaded From: http://proceedings.spiedigitallibrary.org/ on 07/03/2016 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx

No abstract