This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: ECE 654 Solid State Devices II Prof. S. Mohammadi - 98 - Chapter 6 Metal Semiconductor Field Effect Transistors (MESFETs) GaAs MESFETs were first introduced in 1966. Because of simple fabrication technology and excellent microwave characteristics, for over 20 years, MESFETs was the device of choice for high speed, low noise and high power applications. Modern MESFETs are still used in power amplifiers, however their applications are limited by the advent of HBTs, HEMTs and nanoscale MOSFETs. Basic Structure GaAs MESFET takes advantage of velocity saturation under high electric field. Fig. 6.1 shows the cross section and the top view of GaAs MESFET devices. Gate is formed by a Schottky- barrier contact formed between Al metal and N-doped GaAs region. Source and drain are formed by putting AuGe contact metallization on N+ GaAs contact regions. Gate Source Drain Source Drain Source Source Source Source Drain Drain Drain Drain L Gate (Al) a h- -- + + + Dipole Domain Source Drain o V DS Semi-insulating GaAs N-doped GaAs (N D ) N+ GaAs Gate Source Drain Source W/2 W/2 L (a) (b) (c) Gate Source Drain Source Drain Source Source Source Source Drain Drain Drain Drain Gate Source Drain Source Drain Source Source Source Source Drain Drain Drain Drain L Gate (Al) a h- -- + + + Dipole Domain Source Drain o V DS Semi-insulating GaAs N-doped GaAs (N D ) N+ GaAs Gate Source Drain Source W/2 W/2 L Gate Source Drain Source W/2 W/2 L (a) (b) (c) Fig. 6.1. (a) Cross section of a GaAs MESFET. (b) Top view of a two finger transistor. (c) Top view of a multi-finger transistor. Airbridge technology is used to connect all source and drain contacts. ECE 654 Solid State Devices II Prof. S. Mohammadi - 99 - An important rule of thumb for microwave FET design is that the distance from the gate feed point to the end of gate finger (W/2) be less than λ /16 where λ is the wavelength along the gate at the frequency of operation. The upper limits on W/2 range are 5mm for 1GHz operating frequency and 50μm for 100GHz operating frequency. For high power applications, many unit cells, each having a gate finger width less than λ /16 are placed close together and fed in parallel in a multi-finger power FET shown in Fig. 6.1(c). Non of the boundaries of this configuration may exceed λ /16. In addition to these size constraints, there are also impedance and thermal constraints on power FET designs. With common microwave techniques, it is difficult to match typical transmission line impedances (~50 Ω ) to FET input impedances that are below a few ohm. Because power FET impedance varies inversely with the number of cells connected in parallel, the impedance matching restriction may limit the number of fingers. The spacing between the fingers in a power FET may also have to be restricted to some minimum value to avoid excessive temperatures in each cell, thereby restricting the total number of fingers as well....
View Full Document
This note was uploaded on 02/19/2012 for the course ECE 654 taught by Professor Mohammadi during the Spring '08 term at Purdue University-West Lafayette.
- Spring '08