A compact model has to fulfill several criteria to qualify as an industry standard model. In addition to accurate modeling of the real device physics, it is equally important for a model to behave reasonably well under different types of simulation environment. In this chapter, we rigorously test the BSIM-IMG model quality by performing several tests.

Testing BSIM-IMG Model Quality

In this study, the effect of using high-electron-mobility channel materials, such as GaAs, InAs, and InGaAs, on drive current in n-channel metal-oxide-semiconductor (nMOS) and p-channel metal-oxide-semiconductor devices is studied. Relative to silicon, these materials have lower electron effective mass, which leads to lower inversion capacitance (Cinv) due to their lower density of states. Despite their lower Cinv, high-electron mobility channel materials are shown to be promising alternatives to silicon in delivering higher drive current in nMOS devices.

Comparison of drive currents in metal-oxide-semiconductor field-effect transistors made of Si, Ge, GaAs, InGaAs, and InAs channels

In this letter, we present modeling results for germanium p-type FinFETs using the industry standard Berkeley Spice Common Multi-gate Field Effect Transistor (BSIM-CMG) model. The effect of perpendicular electrical field on hole mobility in germanium FinFETs is found to be different from silicon FinFETs. We present an updated Ge mobility equation to account for this difference. With this single update, BSIM-CMG agrees very well with the measured I-V data of Ge FinFETs with a gate-length from 130 to 20 nm. We conclude that a production quality standard model is available for simulation of circuits employing p-type Ge FinFET.

Testing BSIM-IMG Model Quality

References

Comparison of drive currents in metal-oxide-semiconductor field-effect transistors made of Si, Ge, GaAs, InGaAs, and InAs channels

Modeling 20-nm Germanium FinFET With the Industry Standard FinFET Model

Related Papers (2)

The case for thoroughly testing complex system dynamic models

Verifying and validating simulation models