[18a-Z16-1~8] 7.3 Micro/Nano patterning and fabrication

[18a-Z16-5] Study on process and martial design for nanoimprint lithography based on deep learning systems

Sou Tsukamoto¹, Kai Kameyama¹, Hideki Tanabe¹, Ryuhei Yamamura¹, Hiroaki Kawata¹, Masaaki Yasuda¹, 〇Yoshihiko Hirai¹ (1.Osaka Pref. Univ.)

Keywords:nanoimprint, design, deep learning

In order to design NIL processes for various materials, it is necessary to understand the physical and chemical properties of the materials, including their rheological properties, but sometimes it is difficult to measure them experimentally. Therefore, it has been necessary to design the process based on experience with limited knowredge. In this paper, we attempted to build a prototype of a neural network systems that can make sense of physical and chemical properties based on a knowledge database obtained from experiments and simulation analyses, and to construct a system that can estimate optimal material and process conditions.