Presentation Information
[22p-52A-2]Exploration of Sublimation Materials for Pattern Collapse Mitigation using Machine Learning on Limited Experimental Data
〇Shogo Kunieda1, Yuta Sasaki1, Yosuke Hanawa1, Masayuki Otsuji1, Hitoshi Kamijima2, Toshiaki Shintani2, Shota Nakajima2, Ryo Yoshida3 (1.SCREEN Holdings Co., Ltd., 2.Research Institute of Systems Planning. Inc., 3.The Institute of Statistical Mathematics)
Keywords:
machine learning,materials informatics,semiconductor
The sublimation drying method is effective in suppressing the pattern collapse phenomenon after cleaning of micro semiconductor devices. Since the pattern collapse rate in the sublimation drying method is highly dependent on the sublimation material, it is important to find a sublimation material that can minimize the collapse rate. In this presentation, we report a series of results on the creation of a prediction model by machine learning on a small number of experimental collapse rate data, the selection of materials based on the prediction model, experiments on pattern collapse rates, and the effect of the selection method.