Controlling the morphologies of femtosecond laser-induced periodic surface structure on silicon by combining deep learning with energy deposition model

• DOI: 10.1016/j.matdes.2024.113021
• Publication Date: 2024-05-14T16:16:18Z
• Authors: Manlou Ye; Jingya Sun; Zhicheng Chen; Wenpan Tao; Yiling Lian; Zhuangge Yang
• Keywords: Deposition

As an important patterning method, femtosecond laser-induced periodic surface structure (LIPSS) has attracted widespread attention in recent years. Due to the complex physical processes involved laser scanning process, it is difficult predict required LIPSS morphologies, which hinders rapid customization of large-area patterns. In this study, a structural optimization method combining data-driven deep learning with energy deposition model was proposed control morphologies silicon. After processing under dynamic irradiation condition, four morphological types were well defined based on experimental results. Deep constructed extract data features by labeling and encoding sample data. The accuracy self-correlation validation reached 98.0% cross-correlation 91.9%. Our results show that distribution exhibits dependence deposition, joint effect effective pulse number accumulative fluence. Moreover, period groove (super-wavelength LIPSS) increases increase while LSFL (low spatial frequency slightly decreases deposition. Through online monitoring patterns future industrial applications.