Top antireflection coatings

Top antireflection coatings were first proposed by Tanaka and co-workers at Hitachi in 1990. Top antireflection coatings are typically made of perfluorinated compounds such as perfluoroalkanoic acids or PFAA, in particular, perfluoro-octanoic acid or PFOA (II), and perfluoroalkanesulfonic acids or PFAS, in particu- 

The first-generation 248-nm and 193-nm lithographic top antireflection coatings used perfluorooctyl sulfonate (PFOS) compounds, which have been determined to be toxic and persistent in the environment. Recently introduced top antireflection coatings are PFOS-free and are environmentally friendly. 

Top antireflection coatings serve multiple functions reflectivity control (Fig. 9.1), defectivity control (although they can also be a source of defects), across-wafer line width variation control, improvement of process latitude 

Jaenen, V. van Driessche, A.M. Goethals, N. Samarakone, L. Van den Hove, and B. Denturck, Evaluation of methods to reduce hnewidth variation due to topography for i hne and deep UV lithography, Proc. SPIE 1674, 681 (1992). 

The 1990s saw the introduction of commercial water-hased top antireflection coating materials that do not require a solvent treatment hut are removed during normal resist development in 0.26 N aqueous tetramethyl ammonium hydroxide developer solution or simply dissolved with a water rinse. Specifically, they were introduced at 248-nm wavelength (KrF) lithography, with the first chemically 

---

Poly(2-acrylamido-2-methyl-l-propanesulfonic acid and isopropylhexafluor-oalcohol), (IV), was prepared by Khojasteh et al. (4) and used as a top antireflective coating and barrier layer for immersion lithography. 

In spite of this quality, CFCs cannot be used as solvents for top antireflection coatings because they are now banned for their role in the depletion of the ozone layer. 

Antireflection coatings can be applied at the resist-substrate interface, in which case they are called bottom antireflection coatings (BARCs), or they can be applied at the surface of the resist, in which case they are called top antireflection coatings. The effect of top antireflection coatings on the swing amplitude S of resist materials is fairly well described by Brunner s formula ... 

Figure 9.1 PROLlTH/2 simulations showing the effects of top antireflection coating (AZ Aquatar ) and BARC (AZ BARLi ) on the swing curve of AZ 7700 resist on silicon wafer. (Courtesy of R. Dammel.)...

Figure 9.2 Process latitude (in terms of overlap depth of focus for 0.5- jLm lines patterned with Mine lithography) of UCB-JSR 1x500 resist with and without AZ Aquatar top antireflection coatings. Shown are the contour plots for the O.S-fjim line as a function of focus and exposure dose. ...

Consider a ray of light incident on a thin film of top antireflection coatings sandwiched between two semi-infinite media air (or water) and the photoresist (Fig. 9.4). Assuming normal incidence, at the boundary between two media, say, medium 1 (air or water) and medium 2 (top antireflection coatings), the... 

Figure 9.4 Schematic showing the transmission and reflectivity of light incident on a film stack of top antireflection coating (medium 2) coated on top of a photoresist film (medium 3) at normal incidence. The medium above the top antireflection coating is air (medium 1).

Equation (9.18) is the phase match condition for zero reflectivity at the resist-air interface, and its significance lies in the fact that it encapsulates within it the optimal thickness of the top antireflection coating that is associated with the least reflectivity at the resist-air interface. 

The condition in which R 0 occurs when the refractive index of the middle medium (top antireflection coating) is a geometric mean of the two outside media, provided ni[Pg.427]

Top antireflection coatings are typically coated very thin, in the range of a few tens of nanometers, over the resist. The optimum refractive index for top antireflection coating depends on the refractive index of the resist in the following manner ... 

Using a similar approach as in the case of the top antireflection coating above, and assuming normal incidence at the boundary between two media, for instance medium 1 (photoresist) and medium 2 (BARC), the reflection (rj2) and transmission (ti2) coefficients are given by the Fresnel equations above [Eqs. (9.5) and (9.6)]. Multiplication of the reflection amplitude with its complex conjugate yields... 

As seen from the above equation, the normalized standing wave intensity is not dependent on Wj and ki. This suggests that the change in the reflectivity at the top of the resist caused by a top antireflection coating can only lead to exposure dose changes (due to different amounts of light being reflected from the surface). 

---