To obtain useful simulation results of HEMTs a simulation setup has to be found which is capable of modeling the most important properties of the device characteristics in a realistic manner. The first step is to establish a suitable device geometry which will be described in Section 5.1. The different regions of the geometry can be simulated by various models. In Section 5.2 the appropriate transport models for each semiconductor layer of the device will be investigated. The volume and interface models used in the device simulator have been described in Chapter 4.
In Section 5.3 the verification of the simulation results is described. The key problem is that the set of models and parameters to obtain a certain current is not unique. In principle all used geometric and model parameters are only known with a certain accuracy. Some of them have to be considered to be fitting parameters because they have a large influence on the simulation results within the range they are known. Therefore these parameters have to be identified. To obtain a consistent set of parameters for the simulation the results have to be compared to various data. These will be measured currents, capacitances extracted from measured Sparameters, and Monte Carlo simulation results. The comparison between measured and simulated drain current ID will be performed on the transfer characteristics with VDS = 2.0 V because it reveals the most important information on the DC characteristics in only one curve.
To minimize uncertainties the simulation setup will be developed for
a DHHEMT whose dimensions are known extremely well from process technology.
The same device will be used multiple times for investigations and comparisons
to other HEMTs. Therefore it will be referred to as HEMTref.
Next: 5.1 Composition of the Simulated
Device Geometry Up: Dissertation Helmut Brech
Previous: 4.3.2 Mobility
Helmut Brech 1998-03-11