UC San Diego

Scaling of physical dimensions faster than the optical wavelengths or equipment tolerances used in the manufacturing line has led to increased process variability. This in turn has led to unpredictable design, unpredictable manufacturing, and low yields. The result of these physical variations is variation in circuit metrics such as performance and power. Variations can be either systematic (e.g., metal dishing, lithographic proximity effects, etc.) or random (e.g., material variations, dopant fluctuations, etc.). The former can be modeled and predicted while random variations are inherently unpredictable. There are several pattern-dependent process effects which are systematic in nature. These can be compensated during physical design to aid manufacturability and hence improve yield. This thesis focuses on ways to mitigate the impact of systematic variations on design and manufacturing by establishing a bidirectional link between the two. The motivations for doing so are improved yield and manufacturability as well as reduced design guardband and cost. To improve manufacturability, we propose a detailed placement perturbation technique for improved depth of focus and process window. The technique facilitates downstream insertion of scattering bars and etch dummy features in the resolution enhancement process, reducing inter-cell forbidden pitches almost completely. We propose a systematic variation aware timing analysis methodology to reduce timing pessimism. The Proposed self-compensated design techniques achieve circuit robustness to focus variation. We also propose the use of small gate-length biases to reduce leakage power and leakage power variability. To reduce design guardbanding, we have proposed a methodology for power/performance analyses of the design based on lithography-simulation output. We also give the first method for performance-impact limited metal fill insertion. Finally, to reduce mask cost, mask data volume, as well as mask data preparation time, we propose a novel design-aware optical proximity correction (OPC) methodology