Physics-Based Compact Modeling of Successive Breakdown in Ultrathin Oxides

In this letter, we present a physics-based compact SPICE model to predict statistical time-dependent dielectric breakdown (TDDB) in nanoscale circuits. In our model, an increase in the gate leakage current (I G_B D ) induced by TDDB is estimated using a quantum point contact (QPC) model depending on temperature. In addition, I G_B D is based on the statistics of time to breakdown (BD) (t BD ) and location of percolation path (x BD ) in the channel considering third successive BDs. We show that the model can be easily implemented to circuit simulators to predict the degradation of circuit lifetime. With the proposed model, we validated post-BD I-V characteristics with experimental data in ultrathin oxide technology.