Meaning
Determination of the relationship between gate voltage and the electrostatic potential at the semiconductor surface involves the numerical or analytical summation of the total charge density. Surface potential integration forms the basis for advanced compact models used in circuit simulators to predict the behavior of transistors. This technique moves beyond simple regional approximations to provide a continuous description of the device from accumulation to strong inversion.
Engineering teams at Chinese integrated circuit design houses rely on these models to ensure the accuracy of their timing and power simulations.
Calculation Procedure
Numerical integration of the Poisson equation across the semiconductor substrate yields the surface potential for any given set of terminal voltages. In the process of surface potential integration, the algorithm accounts for both mobile carriers and fixed ionized dopants. This calculation must be performed iteratively because the charge density itself depends on the potential.
Efficient numerical methods are used to solve these equations quickly within the simulation environment. This mathematical consistency ensures that the resulting drain current model remains valid across all operating temperatures and bias conditions without needing separate equations for different regimes.
Device Modeling
Linkage of the physical structure of the transistor to its terminal currents and capacitances is achieved through this physics-based approach. Models derived from surface potential integration are more robust than older threshold-based models when scaling to smaller gate lengths. They naturally handle the transition between different operating regimes without the need for empirical smoothing functions.
This accuracy is essential for the design of analog and mixed-signal circuits where precise voltage control is required.
Computational Accuracy
Verification of the model against experimental data involves comparing the predicted capacitance-voltage curves with measurements from real wafers. High-fidelity results from surface potential integration allow for the prediction of small-signal parameters such as transconductance and output resistance. These values are used to sign off on the performance of the design before it is sent to the foundry for fabrication.
Accurate modeling reduces the number of design iterations and shortens the time to market for new semiconductor products in China.