[22a-P07-1~6] 7 Beam Technology and Nanofabrication(Poster)

[22a-P07-5] Study of secondary electron emission based on deep learning system

〇Masahiro Kusumi¹, Yoshihiko Hirai¹, Masaaki Yasuda¹ (1.Osaka Pref. Univ.)

Keywords:secondary electron emission, deep learning, work function

We constructed a simple secondary electron yield prediction system using deep learning. A database of experimental values was used as the learning data. The information to be input was acceleration voltage, atomic number, density, family, work function, etc. As for the prediction result, the increase / decrease in yield due to the change in atomic number could be reproduced relatively well. The prediction excluding the work function showed that the increase / decrease in yield was significantly different from the experimental value, suggesting a correlation between the secondary electron yield and the work function.