Density Gradient Method

Density Gradient Method is the most widely used quantum correction technique in commercial TCAD — it extends the drift-diffusion equations with a quantum pressure term derived from carrier density gradients, repelling charge from the interface and recovering quantum confinement behavior without solving the Schrodinger equation.

What Is the Density Gradient Method?

• Definition: A quantum correction approach that adds a gradient-of-density dependent term to the carrier quasi-Fermi potential, creating an effective repulsive force that pushes the inversion charge peak away from the semiconductor-dielectric interface.
• Physical Interpretation: The correction term represents a quantum pressure analogous to the Bohm quantum potential, arising from the kinetic energy cost of spatially confining a quantum particle.
• Tunable Parameter: A single fitting parameter (gamma) controls the strength of the correction and is calibrated to match Schrodinger-Poisson calculations for representative gate stack configurations.
• Tunneling Capability: Unlike some quantum correction methods, density-gradient can also model gate tunneling current within a fluid simulation framework, making it uniquely versatile.

Why the Density Gradient Method Matters

• Industry Standard: The density-gradient model is the default quantum correction in Synopsys Sentaurus and Silvaco Atlas, making it the most widely deployed quantum correction in commercial semiconductor design.
• C-V Accuracy: By pushing the inversion charge centroid away from the interface to its quantum-mechanically correct position, the method reproduces split-C-V measurements and inversion capacitance data with good accuracy.
• Threshold Voltage Correction: Energy quantization-induced threshold voltage shifts of 30-100mV at advanced nodes are captured by the density-gradient correction, closing the gap between uncorrected simulation and measurement.
• Gate Leakage Modeling: The density-gradient method is used to model direct tunneling and Fowler-Nordheim tunneling current through thin gate dielectrics as part of retention and reliability analyses.
• Nanowire and FinFET: Multi-gate geometries with strong quantum confinement in two lateral directions benefit especially from density-gradient correction, as the classical error is amplified by confinement from multiple interfaces.

How It Is Used in Practice

• Parameter Calibration: The gamma parameter is extracted by fitting the density-gradient inversion charge profile to a Schrodinger-Poisson solution for the target gate stack, then applied uniformly across the simulation domain.
• Coupled Iteration: The quantum pressure term is added to the drift-diffusion iteration loop, converging simultaneously with the standard carrier and Poisson equations without major solver changes.
• Verification: Corrected threshold voltage roll-off and subthreshold swing versus channel length are compared against split-lot measurements to validate the calibration.

Density Gradient Method is the practical standard for quantum correction in industrial TCAD — its combination of physical accuracy, computational efficiency, and commercial tool availability has made it the default quantum enhancement for advanced-node device simulation.