Impact of trench width roughness on the graphoepitaxial assembly of block copolymers
Schematic of topographic assembly of cylindrical forming block copolymer. Photolithography is used to pattern photoresist before a reactive ion etch. A neutral brush is deposited before the is spun cast and annealed. The PMMA is subsequently removed using a DUV exposure and development in acetic acid to improve image contrast for analysis.
SEM images showing the impact of depth of focus on lithographically defined features. In all images, the target mesa width (wall thickness) and the target trench width are kept constant at 50 and , respectively. With each step out of focus, the roughness is dramatically increased until gaps appear in the wall itself.
(a) Scatter plot of the variation in the domain size vs the trench width roughness. The squares show five images from the same trench to estimate error. With the estimation of the error, no clear relationship between the area variation and TWR can be seen. (b) Image classification shown as a function of trench width roughness and incommensurability. Roughness plays at most a secondary role in defect formation.
SEM images showing defect classification and arranged to show impact of TWR and incommensurability. As one moves farther from a commensurate trench width, the number of defects increases, but increasing roughness at the same level of incommensurability shows no effect on defectivity. This shows that the major cause of defect formation is the incommensurable width of the trench, not the roughness (scale ).
Article metrics loading...
Full text loading...