The basic purpose of this thesis was to find models for the analysis of ferroelectric memory cells. Starting from a comparison of existing approaches, new models, which match the constraints of computer aided device design, had to be developed. Now for the first time real two-dimensional simulation of ferroelectric materials with arbitrary geometry and material composition has become possible.
Special care was taken for the implementation of the new models into the device simulator MINIMOS-NT. Following the generic approach of MINIMOS-NT, a modular concept was found which will also be easily extendable for future simulation tasks, which is necessary as the knowledge about ferroelectric materials and the possible applications is still increasing.
In addition to the development and implementation of the two-dimensional algorithms, the transient properties were investigated, and a new model based on differential equations was developed. This model offers an excellent fit to the measured data for a frequency range starting from zero up to 1MHz.
In order to demonstrate possible applications of the new tools a series of concrete examples is provided. One of these examples builds a bridge from two-dimensional device geometry to circuit simulation, another one allows deep insight into the field properties inside an anisotropic crystal. The last example serves as the first profound analysis of a new, very promising nonvolatile memory cell, the FEMFET, and several critical issues concerning device performance and reliability are pointed out.
As earlier in this thesis a lot of research on ferroelectrics is carried out today. This will hopefully increase the availability of material data, thus allowing the extraction of the material parameters introduced in this thesis for a large number of different materials.
As outlined above this thesis was focused on the modeling of ferroelectric materials and the implementation of the resulting algorithms into a device simulator. This tool is now capable to allow a rigorous analysis of ferroelectric devices and opens the door to several critical issues concerning the development of the related memory cells. These include optimization of the device structure, extraction of possible operation schemes, and the development of compact models.