Efficient and accurate RIE modeling methodology for BEOL 2.5D parasitic extraction

An efficient and accurate sensitivity based methodology is introduced for modeling reactive ion etch (RIE) in BEOL 2.5D parasitic extraction. Proposed methodology involves calibration of analytical equations based on layout parameters that are fitted to capacitance and sensitivity data from 2D field solver. Formulas are derived along with new capacitance and sensitivity equations in a 2.5D parasitic extraction framework. Calibration runtime is reduced due to sensitivity modeling while the overall BEOL accuracy is comparable to the case with no RIE effect. Proposed method has been validated over a wide range of technologies from 65nm to 20nm.