Micro-electromechanical systems (MEMS) are tiny systems (from a few micrometers to millimetres in size) that combine mechanical and electrical components. In the most general form, MEMS consist of mechanical microstructures, like sensors, actuators, gears, motors etc., and microelectronics, all integrated onto the same silicon chip. Microsensors detect changes in the systems environment by measuring mechanical, thermal, magnetical, chemical or electromagnetical information. Microelectronic components process this information and signal the microactuators to react and create some form of change to the environment. The major steps in the MEMS fabrication technology include film growth, doping, lithography, etching, dicing, and packaging. Devices are usually fabricated on Si substrates, thin films are grown on these substrates and are used to build active and passive components and interconnections between circuits. These films are: (1) epitaxial Si, (2) SiO₂, (3) silicon nitride Si₃N₄, (4) polycristalline Si, and (5) metal films. To modify electrical or mechanical properties, films are doped with impurities by thermal diffusion or ion implantation.

A specific issue for MEMS applications are sacrificial layers used to form the mechanical components. After formation of the mechanical components the sacrificial layer is selectively etched away, leaving the desired pattern in the film. The chemical etching of the sacrificial layer is a crucial step in the fabrication of MEMS. For example, SiO₂ or phosphosilicate-glass (PSG) can be easily etched in a hydroflorid acid (HF) based solution. HF based etchants etch the material isotropic. In some cases anisotropic etching is required, which means that etchants like potassium hydroxide (KOH) etch faster in a preferred direction.
The focus of the work is the development of etching models for sacrificial etching and the simulation of the moving etch front, which is calculated by means of a diffusion equation (for the etchant transport) with appropriate initial and boundary conditions in combination with the level set algorithm. The solution of the diffusion equation delivers the etching speed neccesary for the level set method, which describes the moving boundary. The advantage of combining both methods is that a single grid can be used. The level set equation will be solved using the narrow band approach, which allows a fast calculation.