Top-view SEM image of [(a)–(c)] wet process and [(d) and (e)] dry process of template forming process. As the aspect ratio increases from 1.0 to 1.4 and 2.0 [(a)–(c)] respectively, the wet process may cause pattern shifting, irregular spacing between patterns, as shown by the red dashed circle in (b), and template collapse, as shown in the red dashed circle in (c). The dry process in (d) has no such problems under the aspect ratio of 2.0, but the linewidth in the dry process was thinner because of the lateral etch rate. And in (e), we demonstrated the possibility of achieving an aspect ratio of 3.5 by using our dry process.
Top-down and cross-sectional SEM images of templates formed by different etching recipes in our RIE tool. Overetching percentage (OE%) increases from top to bottom. Although the Ar etching has higher PMMA-to-PS selectivity than , it induces a wiggling line from the early stage of the etching, thus is not considered as a good candidate for the PMMA removal process. The recipes provided better smoothness and pattern fidelity than Ar. However, the higher power recipe resulted in a zigzag roughness on the top surface (as shown in the red dash circle) after long overetching.
Lateral and vertical polystyrene etch rate of RIE and RIE (a) Lateral PS etch line slope/2 (on each side) (convert from ) (b) . Vertical PS etch slope (convert to nm/min). Force for consistency. The anisotropy, vertical-to-lateral etch rate ratio, for RIE is 9.5. RIE has anisotropy 6.1. The average CD difference of the three recipes is about under 28% overetching.
(a) Line-edge roughness based on a top-down SEM image of PS template formed by plasma. Average LER is . The inset provides a magnified view of the smoothness and the image resolution. (b) Cross-sectional SEM image after Si etching using a fingerprint template produced by plasma. A vertical and smooth trench can be obtained. (c) Combined with chemically patterned brush technique, straight Si trenches or more complex patterns can also be obtained through the same process.
The etching recipes’ setting and vertical etch rates, based on thickness measurements on blanket coated test wafers before and after etching. All the RIE etch rate data were based on of etching except for the PMMA data, which were based on of etching.
Article metrics loading...
Full text loading...