| x (2007) | |||||||||||||||||
Abstract | |||||||||||||||||
| Chemical-mechanical planarization (CMP) and other manufacturing steps in very deep submicron VLSI have varying effects on device and interconnect features, depending on the local layout density. To improve manufacturability and performance predictability, we seek to make a layout uniform with respect to prescribed density criteria, by inserting area fill geometries (dummy fill features) into the layout. We review previous research on single-layer fill for flat and hierarchical layout density control based on the Interlevel Dielectric CMP model. We also describe the recent combination of CMP physical modeling and linear programming for multiple-layer density control, as well as the Shallow Trench Isolation CMP model. Our work makes the following contributions for the Multiple-layer Interlevel Dielectric CMP model. First, we propose a new linear programming approach with a new objective for the multiple-layer fill problem. Second, we describe modified Monte-Carlo approaches for the multiple-layer fill problem. Comparisons with previous approaches show that the new linear programming method is more reasonable for manufacturability, and that the Monte-Carlo approach is efficient and yields more accurate results for large layouts. The CMP step in Shallow Trench Isolation (STI) is a dualmaterial polishing process, i.e., multiple materials are being polished simultaneously during the CMP process. Simple greedy methods were proposed for the non-linear problem with Min-Var and Min-Fill objectives, where | |||||||||||||||||
Publication details | |||||||||||||||||
| |||||||||||||||||