SiC LED

An important reliability issue in nitride LEDs is electrical static discharge (ESD) survivability. Both commercial DH LED products on sapphire include a warning of the sensitivity of these devices. The ESD survivability of the DH GaN SiC LED chip has recently been improved and is far superior to that... 

The very rich ODMR observed for Ti in SiC led to another experiment called cross relaxation [11]. A change in emission occurs when the Zeeman splitting of two Ti-exciton states equals the splitting for a nearby, coupled defect. No microwaves are necessary but the sensitivity and resolution are very high when magnetic-field modulation is employed. In as-grown materials, cross relaxation reveals the N-donor in 4H and 6H SiC and the Al acceptor in 4H SiC (see TABLES 1 and 2). The linewidths are smaller than those observed by the ODMR of donor-acceptor pairs. 

Both methods were employed by these researchers to fabricate SiC LEDs. The efficiencies of the blue LEDs produced by the overcompensation technique were nearly ten times higher than those of the double-epitaxy devices. The highest quantum efficiency that has been obtained was 4x 10 5 at a current density of 2 A cm 2. 

Figure 16 Output power (P) of the GaN / -n junction LED and the conventional 8-mcd SiC LED as a function of the forward current (/p). m is an exponent of /p when it is assumed that P is proportional to...

Table 1 shows a comparison of commercially available red, green, and blue LEDs in terms of luminous intensity, output power, and external quantum efficiencies. From this table, the peak wavelength of green and blue IIl-V nitride LEDs is much shorter than that of conventional green GaP and blue SiC LEDs. Also, the output power and the external quantum efficiencies of III-V nitride LEDs are much higher than those of conventional green and blue LEDs. 

Figure C2.16.2 shows tire gap-lattice constant plots for tire III-V nitrides. These compounds can have eitlier tire WTirtzite or zincblende stmctures, witli tire wurtzite polytype having tire most interesting device applications. The large gaps of tliese materials make tliem particularly useful in tire preparation of LEDs and diode lasers emitting in tire blue part of tire visible spectmm. Unlike tire smaller-gap III-V compounds illustrated in figure C2.16.3 single crystals of tire nitride binaries of AIN, GaN and InN can be prepared only in very small sizes, too small for epitaxial growtli of device stmctures. Substrate materials such as sapphire and SiC are used instead.

The first artificial abrasive, siUcon carbide [409-21 -2] SiC, was produced by Edward Acheson in 1891 (19). This invention led to the formation of the Carbomndum Company. The registered trademark, Carbomndum, is essentially synonymous with siUcon carbide. 

Silicon carbide, SiC [1] and silicon nitride, Si3N4 [2], have been known for some time. Their properties, especially high thermal and chemical stability, hardness, high strength, and a variety of other properties have led to useful applications for both of these materials. 

The industrial application of SiC began with the blue light emitting diode (LED), which was very weak due to the indirect bandgap of SiC but was the only commercial blue electroluminescent light source at the time (the late 1980s). The SiC blue LED was soon surpassed in intensity by the gallium nitride (GaN)-based... 

LED due to the direct bandgap of the Ill-nitrides. However, due to the lack of a native substrate for GaN, sapphire or SiC substrates were and are still used. The biggest use of semiconductor-grade SiC is still for LEDs, but now it serves the role as the substrate for the active GaN layer rather than both the substrate and the active layer. Today there are high-freqnency metal-semiconductor-field effect transistors (MES-EETs) offered commercially, as well as an emerging market for Schottky diodes made from SiC. We are still at the beginning of the SiC revolution, however, and the material s full potential has yet to be realized. 

The electronic properties of SiC were investigated shortly afterward and in 1907 the first LED was produced from SiC [19]. However, the extraction of crystals was a cumbersome process that required patience, and the purity of the crystals was not controllable. As a consequence, in 1955 another crystal growth invention of significant proportions was made by J. A. Lely [20]. 

The commercial potential of SiC has to date mainly been in the materials arena. SiC is used as substrate for LEDs made from GaN, which is the largest market of semiconductor-grade SiC. Recently Infineon launched its Schottky diode product... 

However, such comparison is a meaningless exercise. A comparison of silicon, sapphire, GaAs, or other established materials, rather than other SiC growth techniques, should occur instead. The blue and white LEDs can, for instance, be made on either sapphire or SiC substrates. From a performance point of view, there appear to be advantages with the SiC substrate but the substrate cost is lower on sapphire, therefore making it the most-used substrate. Granted, once larger substrates of SiC are made available, it is doubtful that the LED cost will be lower on sapphire. 

Silicon carbide has a clear place in society today, yet it has only one major commercial application today, which basically is materials or, more specifically, substrates for LED applications. This is perhaps not such a glamorous place to be in the commercial market but it does represent a significant starting point for SiC. 

---