Anomalously weak antenna ratio dependence of plasma process-induced damage

Conventional antenna charging theory predicts that the net current drawn from a plasma glow discharge is proportional to the charge collecting area of the antenna. However, by correlating straightforward oxide breakdown statistics with typical yield data of antenna testers, we reveal a much weaker than proportional Antenna Ratio (AR) dependence. We observe this anomalous dependence both in our own as well as in yield data published by others. The data are originating from various process modules and plasma sources, from technologies with gate oxide thickness ranging from 8 nm down to 2.8 nm. The less than proportional AR dependence implies that extrapolation of large AR test structure results to predict small AR circuit behavior is not as straight-forward as previously thought