Advanced in-production hotspot prediction and monitoring with micro-topography

At 28nm technology and below, hot spot prediction and process window control on production wafers have become increasingly critical to prevent sensitive pattern geometries from becoming yield limiting defects as a result of process variation. We previously established a systematic approach to identify focus-sensitive hotspots, characterize their process window margins, use a focus variation map to predict patterning defect locations, and verify predictions by guided e-beam inspection [1]. The current paper establishes the impact of intra-die micro-topography and its correlation with best focus variations of hotspots in a production chip layout. For this purpose, we obtain high-resolution topography measurements from an offline tool, and determine pattern-dependent best focus shifts from litho simulations to compare against the measured best focus distribution. We exercise the entire prediction and guided verification flow for after-etch application on a production use case with full-stack topography wafers.