Large-Scale Simulations of Directed Self-Assembly with Simplified Model

• Published in: Journal of Photopolymer Science and Technology Vol. 26; no. 6; pp. 809 - 816
• Main Authors: Yoshimoto, Kenji, Taniguchi, Takashi
• Format: Journal Article
• Language: English
• Published: Hiratsuka The Society of Photopolymer Science and Technology(SPST) 01.01.2013; Japan Science and Technology Agency
• Subjects: Block Copolymer; Defect; DSA; Simulation
• ISSN: 0914-9244; 1349-6336
• DOI: 10.2494/photopolymer.26.809

Directed self-assembly (DSA) is considered as a candidate for future patterning technology for semiconductor manufacturing. The DSA utilizes the phase separation of block copolymer and provides further resolution enhancement by the use of chemically and physically pre-patterned surface. In order for DSA to be a viable lithography solution, it is crucial to realize defect-free manufacturing process. In this study, we utilize the so-called Ohta-Kawasaki (OK) model to simulate the morphological defects of block copolymer formed on the chemically and physically pre-patterned surface. The OK model has advantages of the relatively low computational expense, scalability for large-scale simulations, and compatibility to the other simulation models such as self-consistent field theory (SCFT) through common physical properties of the materials. As test cases, we investigated the lamella defects of symmetric diblock copolymer formed on the chemically and physically pre-patterned surface. For the chemically pre-patterned surface, the two-dimensional (2D) dynamic simulations were performed including the thermal fluctuations, and the time evolution of the lamella defects was characterized as a function of the surface interactive parameter. In the three-dimensional (3D) dynamic simulations of the physically pre-patterned surface, effects of the trench width on the formation of the lamella defects was examined. Our preliminary results demonstrate that various types of the DSA lamella defects can be reasonably predicted by the OK model. It is expected that by calibrating the surface interactive parameters with experimental data, the OK model could be applied to various large-scale DSA simulations, e.g., hotspot analysis over a large area, and design/process optimizations with numerous parameters.