| Summary: | 碩士 === 明新科技大學 === 電子工程系碩士班 === 105 === Insulated Gate Bipolar Transistor (IGBT) is a power discrete with both benefits; one of them is high input impedance of MOSFET and the other is high current gain of BJT (Bipolar Junction Transistor). Successfully associated with the merits of high input impedance gate-controlling MOSFET and internal driven high current gain of Bipolar, Insulated Gate Bipolar Transistor (IGBT) performs well on controlling high power convey. IGBT may also work in the circumstance with the high reversed bias, which is quite impressive. However, to test the electrical performances, it might not be definitely due to the use of extremely high power tester. Authors thus expect to explore certain process-related or structure related properties of fabricated IGBT using Agilent 4156C. With the fixed gate bias (5V) and the uniform threshold voltage (identified to be 4.1V), the VDS raise at the constant current increase is proposed to understand the influence subjected to the corresponding various process conditions and pre-designed lengths of source and gate.
FinFET devices are fabricated on SOI wafer (Silicon on Insulator) with 3-D fin-like body wrapped with poly-silicon as gate electrode. The whole body is easily depleted as applied with a bias on gate. The depleted region, of course, may demonstrate high resistive path reducing and lowering the leakage current even if the channel length is shortened. In thesis, there are three devices with two kinds of channel lengths, 100nm and 500nm. Those devices are put to measurements and the current-voltage data are therefore figured and compared with the basic conventional formula. For one thing, ID-VG characteristic curves may show turn-off leakage current addressing Drain Induced Barrier Lowering (DIBL) and Punch-through issues. On the other hand, ID-VD characteristic curves are fitted with the conventional formula to abstract the threshold voltages, which are used to analyze the electric performance and the implicit or underlying physics. The fitting is done by introducing the deviation which is defined to be the summation of the square of subtraction of the fitting values from the measured values. As the deviation reaches the minimum, the fitting is supposed to be as successful as possible.
|
|---|
