The Semiconductor Module provides dedicated tools for the analysis of semiconductor device operation at the fundamental physics level. The module is based on the drift-diffusion equations, using isothermal or nonisothermal transport models. It is useful for simulating a range of practical devices, including bipolar transistors, metal-semiconductor field-effect transistors (MESFETs), metal-oxide-semiconductor field-effect transistors (MOSFETs), insulated-gate bipolar transistors (IGBTs), Schottky diodes, and P-N junctions. A predefined Schrödinger Equation interface allows the modelling of quantum-confined systems such as quantum wells, wires, and dots.

Multiphysics effects often play important roles in semiconductor device performance. The Semiconductor Module enables you to easily create models involving multiple physical effects. For example, thermal effects within a power device can be simulated by adding a heat transfer physics interface (available in the COMSOL Multiphysics® software platform). Optical transitions can also be incorporated to simulate a range of devices such as solar cells, light-emitting diodes (LEDs), and photodiodes (some may require the Wave Optics Module).

You can choose to make use of the finite element or finite volume method when modelling the transport of holes and electrons in the Semiconductor Module. Each method has its set of advantages and disadvantages:

• Finite Volume Discretisation: Finite volume discretisation in the modelling of semiconductor devices inherently conserves current. As a result, it provides the most accurate result for the current density of the charge carriers. The Semiconductor Module uses a Scharfetter-Gummel upwinding scheme for the charge carrier equations. It produces a solution that is constant within each mesh element, so that fluxes can only be constructed on the mesh faces that are adjacent to two mesh elements. Yet, as products in the COMSOL Product Suite are based on the finite element method, this can make it a bit more challenging to set up multiphysics models.
  
  
• Finite Element Discretisation: The finite element method is an energy-conserving method. Consequently, current is not necessarily well conserved with this technique. To obtain accurate currents, it may be necessary to tighten the default solver tolerances or to refine your mesh. In order to help with numerical stability, a Galerkin least squares stabilisation method is included when solving the physics in semiconductor devices. One advantage of modelling semiconductor devices with the finite element method is that you can more easily couple your model to other physics, such as heat transfer or solid mechanics, in a single model.