Over the past few years, more and more research and development has been devoted to gallium nitride (GaN). Many see GaN as the driver for next-generation high-frequency, high-power transistors— especially because it can operate at high temperatures. In the microwave market, many predict that GaN will be an almost ideal fit for mobile-infrastructure and WiMAX applications.

GaN has a 3.4-eV band gap. Because of this wide band gap, GaN-transistor performance promises to remain stable at higher temperatures than silicon transistors. This performance advantage could potentially help GaN satisfy the markets for RF power amplifiers—particularly in high-speed wireless data transmission. In fact, new research from ABI Research (New York, NY) notes that GaN is now entering the mainstream of process technologies for RF power semiconductor devices. According to ABI's new research brief, however, the infrastructure sector may actually be ill-suited for GaN because of this sector's price sensitivity.

Simply put, GaN is more costly than alternative power-transistor process technologies. When it is compared with silicon LDMOS for amplifiers, the device cost is substantially higher on a one-to-one basis. Of course, developments could reduce the price gap. But performance also must be considered when looking at price. GaN's fanbase may be surprised to hear ABI's conclusion that GaN and LDMOS are basically equal when it comes to price and performance. As a result, the research firm suggests that GaN will not completely dominate the wireless-infrastructure power-amplifier market.

Once it is beyond the performance range of silicon LDMOS, however, GaN is expected to flourish. At frequencies above 4 GHz, ABI predicts that GaN may dominate practically all of the high-power markets. Eudyna Devices and Toshiba already have footholds in this segment, as they have targeted the microwave (>4-GHz) markets for much of their participation with GaN.

Whether GaN wins the infrastructure market or not, the material-will certainly succeed in other areas. GaN emits exceptionally brilliant light, which makes it suitable for applications like light-emitting diodes (LEDs). Some even predict that GaN-based RF transistors could replace those old vacuum-sealed magnetrons as the microwave source for microwave ovens. Over the past few years, GaN nanotubes also have been produced. They could satisfy various applications including biochemical sensing—a field that is currently expanding due to homeland-security and military needs. No matter which markets it ends up dominating, GaN seems fated to become at least as important as silicon—if not more so.