Introduction


What is a Wide Band Gap (WBG) semiconductor?

Generally speaking, a wide band gap semiconductor is a semiconductor with an energy band gap wider than about 2 eV. Examples of WBG semiconductors are gallium nitride (GaN, EG = 3.4 eV), aluminum nitride (AlN, EG = 6.2 eV), and silicon carbide (SiC, EG between 2.2 to 3.25 eV depending on polytype).


What is Silicon Carbide (SiC)?

Silicon carbide is a WBG semiconductor that possesses extremely high thermal, chemical, and mechanical stability. It is so thermally stable that dopant impurities cannot be diffused at any reasonable temperature; so chemically stable that it is impervious to any known chemical etchant; and so mechanically stable that it is used as a coating for drill bits and saw blades.

Single-crystal SiC forms in the hexagonal lattice, with alternating hexagonal planes of silicon and carbon atoms, as shown in Fig. 1 (below). Each silicon atom bonds to four nearest-neighbor carbon atoms, and each carbon atom bonds to four nearest-neighbor silicon atoms (bonds are not shown in the figure). Note that the atoms in the second silicon plane are offset with respect to the atoms in the first silicon plane. As successive planes are added, each plane must be offset with respect to the plane below. This stacking sequence leads to different flavors, or polytypes, of the basic SiC crystal. There are a large number of possible polytypes, but the most important are 3C, 4H, and 6H. The polytypes differ in band gap energy, carrier mobility, and breakdown field. For example, the EG = 2.2, 3.25, and 3.0 eV for 3C, 4H, and 6H-SiC respectively.

Figure 1. Primitive crystal structure of SiC. Silicon atoms (blue) and carbon atoms (red) form alternating hexagonal planes. The sticks in this rendering are intended to emphasize the hexagonal lattice arrangement, and do not represent bonds. Each carbon atom bonds to three silicon atoms in the plane below and to one silicon atom in the plane above (bonds not shown). Note that the silicon atoms in the top plane are offset with respect to the silicon atoms in the bottom plane. To pause the animation, hold down the mouse button. To halt the animation, click the "Stop Loading" button on your browser.


Why is SiC important?

Silicon carbide is the only WBG semiconductor that possesses a high-quality native oxide suitable for use as an MOS insulator in electronic devices. Thermal oxidation of SiC produces a layer of SiO2 on the surface, while the carbon atoms from the SiC form CO, which escapes as a gas. Thus it is possible to make all the devices found in silicon IC technology in SiC, including high quality, stable MOS transistors and MOS integrated circuits.

The breakdown field in SiC is about 8x higher than in silicon. This is important for high-voltage power switching transistors. For example, a device of a given size in SiC will have a blocking voltage 8x higher than the same device in silicon. More importantly, the on-resistance of the SiC device will be about 100x lower than the silicon device.

Because the bandgap of SiC is so much wider than silicon, thermal generation of electron-hole pairs is many orders of magintude lower at any given temperature. This makes it possible to build "dynamic" memories (DRAMs) in SiC that only need to be refreshed about once every 100 years at room temperature! This also makes it possible to operate SiC devices at temperatures as high as 650 °C without degradation in electrical performance.


Our WBG research group.

We began working on SiC devices in 1990 in collaboration with Cree Research of Durham, NC, the world's largest supplier of SiC wafers and blue LEDs. Our work has been supported primarily by the Office of Naval Research (ONR), the Ballistic Missile Defense Organization (BMDO), and the Semiconductor Research Corporation (SRC).

We are a world leader in characterizing and improving the MOS interface between SiO2 and SiC, and we have developed several novel devices in SiC for the first time. To our knowledge, we are the second university group to build MOS transistors in SiC and the first university group to build bipolar transistors. We originated the concept for the nonvolatile RAM in SiC, and worked with Cree Research to demonstrate storage times of >100 years with no power applied to the device. Our group also built the first monolithic digital integrated circuits in SiC (1993), the first charge-coupled devices (CCDs) in SiC (1995), the first planar double-implanted MOS (DMOS) power transistors in SiC (1996), the first p-well CMOS integrated circuits in SiC (1996), and the first lateral DMOS power transistors (1997) with blocking voltages of 2.6 kV.

Ours is thought to be the largest university-based WBG device research group in the United States, currently consisting of 15 graduate students, one post-doc, one senior research scientist, and four faculty. Our major emphases at this time are in the areas of power switching devices, SiC microwave devices, process development, and fundamental measurements.


Collaborating with our group.

We welcome collaboration with other WBG research groups. We currently have funded collaborations with Cree Research, Delco Electronics, Motorola PCRL, General Electric CR&D, the University of Texas at Austin, Howard University, Rensselaer Polytechnic Institute, and the University of California at Santa Barbara. We also have informal collaborations with a number of other industrial research groups, national laboratories, and universities.


How to join our group.

Unfortunately, at this time we are not able to accept new graduate students into our group. All students in the group are supported as Graduate Research Assistants through the academic year and summer periods. Persons interested in future positions may contact Profs. J. A. Cooper (cooperj@ecn.purdue.edu) or M. R. Melloch (melloch@ecn.purdue.edu) at 1285 EE Building, West Lafayette, IN 47907-1285 USA. Resumes will be kept on file pending the availablility of positions.

To probe further.

To obtain more information on the history and current status of silicon carbide crystal growth, epitaxy, and device fabrication (selective doping, ohmic contact formation, and thermal oxidation), along with references to the current literature, click on the link above.

Return to the Wide Band Gap Research homepage?

| About Purdue | | About Purdue ECE | | Introduction | | Device Research | | Basic Studies |
| People | | Accomplishments | | Facilities | | Publications | | Late News | | Related Links |