Ueda S., Kawada Y., Horio K.
Shibaura Institute of Technology, JP
Keywords: breakdown voltage, buffer layer, deep acceptor, GaN HEMT, high-k passivation layer
AlGaN/GaN HEMTs are now receiving great interest for application to high-power microwave devices and high power switching devices. To improve the breakdown voltage, the introduction of field plate is shown to be effective, but it may increase the parasitic capacitance, leading to degrading the high-frequency performance. In a previous work, as another method to improve the breakdown voltage, we proposed a structure including a high-k passivation layer, and showed that the breakdown voltage increased significantly. We assumed an undoped semi-insulating buffer layer where a deep donor compensates a deep acceptor. Recently, Fe- and C-doped semi-insulating buffer layers are often adopted and they acts as deep acceptors. Therefore, in this work, we analyze AlGaN/GaN HEMTs with a buffer layer including only a deep acceptor, and studied how the breakdown voltage is influenced by its density. Here, we adopt the Fe-doped semi-insulating buffer layer and the deep-acceptor density NDA is set to 1017 cm-3 and 2×1017 cm-3. In both cases, the breakdown voltage increases as the relative permittivity of the buffer layer er increases, because the electric field of the drain edge of the gate is reduced. When NDA is 1017 cm-3, the breakdown voltage is determined by impact ionization of carriers when er is low, but it saturates when er > 30 because the buffer leakage current becomes to determine the breakdown voltage. On the other hand, when NDA is 2×1017 cm-3, the breakdown voltage is determined by impact ionization of carriers until er = 60 and it reaches 500 V, which corresponding to an average electric field of 3.3 MV/cm between the gate and the drain.
Journal: TechConnect Briefs
Volume: 4, Informatics, Electronics and Microsystems: TechConnect Briefs 2018
Published: May 13, 2018
Pages: 24 - 27
Industry sectors: Advanced Materials & Manufacturing | Sensors, MEMS, Electronics
Topic: Informatics, Modeling & Simulation
ISBN: 978-0-9988782-1-8