Continuity Equation is the particle conservation law for electrons and holes in a semiconductor — it states that the time rate of change of carrier density at any point equals the difference between the divergence of carrier current flow and the net recombination-generation rate, forming one of the three fundamental equations of semiconductor device simulation alongside Poisson and the current density equations.
What Is the Continuity Equation?
- Definition: For electrons: dn/dt = (1/q) nabla·J_n + G - R; for holes: dp/dt = -(1/q) nabla·J_p + G - R, where J_n and J_p are electron and hole current densities, G is the generation rate, and R is the recombination rate.
- Physical Meaning: Carrier density at a point increases if more carriers flow in than flow out (positive current divergence for electrons) or if generation exceeds recombination. Carrier density decreases if carriers flow out faster than in or if recombination dominates.
- Steady-State Form: Setting dn/dt = dp/dt = 0 gives the DC conditions: current divergence equals net recombination-generation rate everywhere. This allows TCAD to find the equilibrium or steady-state carrier distribution.
- Transient Form: The full time-dependent equation governs switching transient response — how fast carriers redistribute when gate voltage changes, how quickly stored charge is removed from a forward-biased diode, and how the photoconductance decays after a light pulse.
Why the Continuity Equation Matters
- Physical Completeness: Without carrier continuity, device simulation would violate charge conservation — carriers could appear or disappear without physical cause. The continuity equation ensures that every electron and hole is accounted for as it moves, recombines, or is generated throughout the device.
- Transient Simulation: Circuit switching speed is determined by how fast minority carriers respond to changing gate and bias voltages. Transient continuity equation solution provides rise times, fall times, and turn-off delay predictions essential for timing-critical circuit design.
- Leakage Current Prediction: In steady-state reverse-biased junctions, the continuity equation balances zero current divergence against net generation in the depletion region to predict the thermal generation leakage current — the primary source of off-state power and DRAM refresh requirements.
- Solar Cell Analysis: The continuity equation for minority carriers under illumination determines the spatial distribution of photogenerated carriers, which carriers reach the junction to contribute to current, and which recombine before collection — the foundation of solar cell efficiency modeling.
- Carrier Lifetime Extraction: Photoconductance decay experiments directly measure the transient solution of the continuity equation with zero current divergence (isolated sample) — the decay time constant equals the effective minority carrier lifetime.
How the Continuity Equation Is Solved in Practice
- Discretization: On a finite-element or finite-difference mesh, the divergence of current and the G-R terms are discretized at each mesh node, converting the PDE to a set of algebraic equations solved simultaneously with the Poisson equation.
- Scharfetter-Gummel Scheme: The standard discretization for the electron and hole current density in the continuity equation uses the Scharfetter-Gummel scheme, which correctly handles the transition between diffusion-dominated and drift-dominated transport and avoids artificial numerical diffusion at high fields.
- Newton Coupling: In fully coupled (Newton) device simulation, the Poisson equation and two continuity equations (six unknowns: phi, n, p and their updates) are solved as a block system at each Newton step, providing robust convergence for most device operating conditions.
Continuity Equation is the particle bookkeeping law that makes device simulation physically rigorous — by enforcing that carriers are neither created nor destroyed without explicit generation-recombination physics, it ensures that all simulated device behavior respects charge conservation and that switching transients, leakage currents, and photogenerated carrier distributions are all computed with the internal consistency required for reliable device design.