• DocumentCode
    41455
  • Title

    Litho-Friendly Decomposition Method for Self-Aligned Double Patterning

  • Author

    Mirsaeedi, Minoo ; Torres, Alvaro Javier ; Anis, Mohab H.

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Univ. of Waterloo, Waterloo, ON, Canada
  • Volume
    21
  • Issue
    8
  • fYear
    2013
  • fDate
    Aug. 2013
  • Firstpage
    1469
  • Lastpage
    1480
  • Abstract
    This paper establishes self-aligned double patterning (SADP) decomposition requirements and proposes a litho-friendly layout decomposition method. First, we explain the SADP-specific printability requirements that should be considered during decomposition. In silico experiments indicate that layout patterns that are printed by the trim mask may experience the highest levels of image distortion. Therefore, these patterns should be assisted by sidewalls of spacer patterns to improve printing robustness. Next, we present an integer linear programming-based decomposition method that avoids decomposition conflicts and sensitive trim edges simultaneously. To improve runtime and extend its application to partially decomposable layouts, too, a partitioning-based decomposition is also proposed. Our experiments reveal that the proposed methods decrease the total length of sensitive trim patterns and consequently reduce the total edge placement error significantly.
  • Keywords
    integer programming; linear programming; nanolithography; nanopatterning; SADP decomposition; image distortion; integer linear programming; layout patterns; litho-friendly decomposition; partitioning-based decomposition; self-aligned double patterning decomposition; silico experiments; Complexity theory; Dielectrics; Etching; Layout; Lithography; Robustness; Sensitivity; Double patterning; integer linear programming (ILP); layout decomposition; self-aligned double patterning (SADP);
  • fLanguage
    English
  • Journal_Title
    Very Large Scale Integration (VLSI) Systems, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    1063-8210
  • Type

    jour

  • DOI
    10.1109/TVLSI.2012.2212473
  • Filename
    6298981