Machine Learning-Guided Etch Proximity Correction

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 296
  • Download : 0
Rule-and model-based methods of etch proximity correction (EPC) are widely used, but they are insufficiently accurate for technologies below 20 nm. Simple rules are no longer adequate for the complicated patterns in layouts; and models based on a few empirically determined parameters cannot reflect etching phenomena physically. We introduce machine learning to EPC: each segment of interest, together with its surroundings, is characterized by geometric and optical parameters, which are then submitted to an artificial neural network that predicts the etch bias. We have implemented this new approach to EPC using a commercial OPC tool, and applied it to a DRAM gate layer in 20-nm technology, achieving predictions that are 34% more accurate than model-based EPC.
Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Issue Date
2017-02
Language
English
Article Type
Article
Citation

IEEE TRANSACTIONS ON SEMICONDUCTOR MANUFACTURING, v.30, no.1, pp.1 - 7

ISSN
0894-6507
DOI
10.1109/TSM.2016.2626304
URI
http://hdl.handle.net/10203/222756
Appears in Collection
EE-Journal Papers(저널논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0