Two types of sidegating effects are experimentally identified. One type of sidegating is observed in measurements of a test structure including not only a device and a sidegate but also a metal contact on the substrate. The current in the device decreases with the increasing negative sidegate voltage, and leakage current in the substrate reaches high levels. The other type of sidegating effect is measured with no substrate metal contact involved. The current in the device decreases after the sidegate voltage reaches a threshold, but the substrate leakage current remains low. Simulations are done using PISCES which proves to be a very useful and effective tool for investigating and understanding the underlying mechanisms of the observed experimental results. The inclusion of deep-level traps in simulations is essential, and a deep-level model based on SRH theory has been added to PISCES. Other factors such as mobility and injections of holes and electrons are also important in understanding sidegating and leakage current characteristics. Simulations show that sidegating effect is a consequence of increased trapping of electrons by deep-level centers, which creates negative space-charge near the channel-substrate interface of a device. In the presence of a metal contact, the increased electron trapping is caused by hole injection from the metal contact and electron injection from the sidegate, and the double-injection also leads to the high leakage current in the substrate. When the metal contact is absent, the increased trapping of electrons is caused by electron accumulation because of the negative-differential-mobility characteristics in the velocity-field relation of electrons in GaAs.
Yi Liu (liuyi@gloworm.Stanford.EDU)