The main task of device simulation is to compute the electrical characteristics of a semiconductor device for given specifications of its design. In this way, design variants can be studied on the computer without actually fabricating the device. Device simulation is a branch of Technology Computer Aided Design (TCAD) and represents an indispensable tool within the semiconductor industry. Classical device simulation is based on drift-diffusion transport theory and utilizes effective mass models for the electronic bandstructure. The basic equations consist of Poisson's equation and the carrier continuity equations. Robust and efficient numerical methods for the solution of these equations exist. Quantum correction models were introduced to account for the ever decreasing thickness of the conducting channel, the adoption of heterojunction devices has also called for the introduction of energy transport models. Applications of device simulation are manifold, and range from strain and bandgap engineering, as well as variability studies, to degradation modeling and life-time prediction.
The ongoing miniaturization of semiconductor devices has led to an exponential increase in production cost. Computer simulation of the electrical and thermal characteristics of devices provides a fast and inexpensive way to check designs and processes prior to fabrication. Simulation of semiconductor devices can be based on a semi-classical description such as the Boltzmann transport equation or, using the method of moments, other sophisticated transport models can be derived. For scaled devices as well as for the accurate inclusion of hot carrier effects, the Boltzmann transport equation has to be solved. This can be done either by using the Monte Carlo method or the Spherical Harmonics Expansion method. More on this topic can be found here.
For very small devices, however, quantum effects come into play and require the solution of the Schroedinger equation or the application of the non-equilibrium Green's function formalism. Alternatively, the Wigner equation can be used which provides a link between the classical and the quantum-mechanical descriptions of carrier transport. These methods, however, are computationally demanding. Check out our activities here.
Finally, non-ideality effects caused by defects have rapidly gained importance. Existing defects can become charged or discharged, defects can be created from precursors, and defects can be annealed. These challenges are tackled under our reliability activities and are supported by detailed experimental characterization efforts.
At the Institute for Microelectronics, research into the development of efficient classical device simulators, which need sophisticated models and three-dimensional simulation capabilities to describe the wide range of materials and devices, is ongoing. These simulators are bolstered by quantum-mechanical simulations, which are used to predict the behavior of advanced quantum devices and to find approximations which can then be adapted for efficient classical device simulations.