平版印刷术
全息术
极紫外光刻
光学
光刻
抵抗
薄脆饼
X射线光刻
干涉光刻
投影(关系代数)
材料科学
干扰(通信)
计算光刻
相(物质)
计算机科学
光电子学
制作
物理
算法
纳米技术
电信
病理
频道(广播)
医学
量子力学
替代医学
图层(电子)
作者
Serhiy Danylyuk,Valerie Deuter,Maciej Grochowicz,Jan Biller,Sascha Brose,Thomas Taubner,Detlev Grützmacher,Larissa Juschkin
摘要
Nowadays, EUV projection lithography has been proven effective for high-volume manufacturing of microchips. In parallel, high-resolution nanopatterning has been demonstrated utilizing interference lithography [1]. However, the former suffers from the complexity of projection optics, and the latter is limited to periodic structures. The presented approach is free of imaging optics and moreover allows for printing arbitrary (non-periodic) structures. Taking advantage of iterative designing of synthetic holograms, the described idea enables creating dedicated optical structure that can be applied for proximity lithography with EUV radiation. The method does not require a sophisticated optical system but necessitates numerical computation of a holographic mask, which gives desired intensity distribution at wafer. It is an inverse problem: for known intensity distribution at the wafer a design of holographic mask has to be inferred. The light field distribution in the plane of the mask can be calculated using phase retrieval methods based on Gerchberg-Saxton algorithm. The process can be described as iterative propagation of light field between mask and wafer planes at which certain constrains are applied: limited number of phase levels, minimal element size on the mask due to the fabrication process, correlation between absorption and phase-shifts and also the resist response. Due to the appropriate optical properties, a photoresist has been chosen as phase shifting material allowing for patterning of arbitrary mask structures. For the realization of the holographic phase shifting mask we used two phase-shifting levels. The fabrication process of the designed mask and experimental results of its characterization are also presented and discussed.
科研通智能强力驱动
Strongly Powered by AbleSci AI