DTCO for DSA-MP Hybrid Lithography with Double-BCP Materials in Sub-7nm Node

In the sub-7nm technology nodes, as the mask cost for printing the dense via layers increases dramatically with conventional lithography technique, industries are actively looking for some alternative techniques, such as E-beam, EUV or nanoimprint. In recent years, directed self-assembly (DSA) has been demonstrated to be a promising candidate to reduce the number of masks due to its pitch multiply ability by grouping several vias in the same DSA guiding patterns. A novel idea has been proposed recently to further reduce the mask cost by applying double block copolymer (double-BCP) materials for DSA lithography. However, it is also discovered that new challenges are introduced for the DSA pattern assignment and decomposition problem with double-BCP DSA. This paper discusses the design technology co-optimization (DTCO) for BCP materials and DSA-MP hybrid lithography. We show that it is necessary to consider the pitch-range of BCP materials for the DSA pattern assignment and decomposition. We first propose a pitch-range optimization method for the BCP material to minimize the potential DSA decomposition conflicts. We then solve the double-BCP pattern assignment and decomposition by formulating it to a maximum weighted independent set problem, and obtain the optimal solution with integer linear programming (ILP). We also propose a bounded approximation algorithm to solve the ILP problem more efficiently. The experimental results demonstrate that our pitch-range optimization method can quickly find the pitch-range that can minimize the decomposition conflicts. With the optimized pitch-ranges, double-BCP outperforms single-BCP by reducing more than 500 conflicts for very dense design. In addition, our proposed approximation algorithm solves the problem 38× faster than the ILP formulation.