< Back to previous page


Impact of DSA process variability on circuit performance

Book Contribution - Book Chapter Conference Contribution

Directed Self-Assembly (DSA) of block copolymers (BCPs) has been attracting a great deal of academic and industrial interest [1]. Recent major advancements have shown the technique’s strong potential for low-cost contact hole layer manufacturing in new generations of IC’s [2-4]. Via layer patterning through templated confinement of cylindrical phase BCP materials emerges as one of the most promising approaches for DSA implementation.Recent studies have shown considerable progress on controlling DSA process-related defects, like template shape variation, surface energy, placement accuracy and cylinder profile [3-4]. However, it is still not clear what is the actual impact of each of these factors on circuit design and performance. The main focus of this work is to provide circuit-level analysis of the impact of DSA variability effects on IC design. This analysis, based on the current status of process defectivity in DSA, helps to study and define the process optimization margins.In order to achieve this, we designed a framework which can extract full circuit parasitics and perform SPICE-level simulations in an iterative loop, using statistical distributions of process-level variability as inputs. Assuming the SRAM memory as our target circuit, we design SRAM arrays for the 5nm technology node where the via layers are decomposed with the DSA method. Monte-Carlo sampling of process-related effects, like CD and placement accuracy variation, is performed by our framework, which provides statistical distributions of SRAM performance values as outputs. These results are compared with the performance of SRAMs designed with other patterning options, such as immersion 193 nm multiple patterning and single patterning EUVL.[1] J. Y. Cheng, A. M. Mayes, and C. A. Ross, Nat. Mater. 3, 823 (2004)[2] H. Yi et al., Proc. SPIE, Vol. 8323, (2012)[3] J. Doise et al., Journal of Vacuum Science & Technology B 33, 06F301 (2015)[4] J. Doise et al., Proc. SPIE 9779 (2016)
Book: Proc. DSA2016
Publication year:2016