It is well known that the substrate current of an MOS transistor can be used to monitor the hot carrier effects. Recent developments of substrate current modeling tend to use many fitting parameters to obtain an accurate result. However, the fitting parameters lack physical meaning and do not provide physical relationships with the device. The limited use of fitting parameters is a great achievement of this research. The derivation of the present substrate current model is based on device physics. The present substrate current model provides a better understanding of the parameters for substrate current modeling. Thus, a better correlation of substrate current and hot carrier effects can be achieved.; This dissertation provides a new maximum electric field model, a new length of velocity saturation model, a new threshold voltage and saturation voltage model for short channel devices. These new models constitute an accurate substrate current model. The accuracy of the substrate current model for MOSFET devices was verified by comparison with measurement. |