I recently visited the IMS event in Baltimore and before that I was over at the PCIM Power Electronics Exhibition in Nuremberg, Germany and GaN-related products were sprouting up everywhere at these two events. Nothing wrong with that when you look at what it brings to the technology table.
Gallium Nitride to give it its full title is a semiconductor material just like silicon but with a difference. It can provide devices with a much reduced on-resistance and it can work at much higher frequencies. Alex Lidow recently explained why in an edition of Engineering Radio. It's because of a couple of fundamental properties of gallium nitride crystal. (EngineeringRadio report http://electronicdesign.com/Blogs/EngineeringRadio/tabid/745/entryid/142/Default.aspx.)
Firstly, it can stand much higher voltages across smaller distances and the structural electrons are very mobile which is what creates the on-resistance reductions. Whatl this means is that devices created from GaN can be made smaller allowing those zippy electrons to cover the shorter distances very quickly. The pay-off on that is increased operating frequencies
So what's the problem with all that? Absolutely nothing except cost. Those GaN crystals have been very expensive to generate but that situation is improving. GaN has been known about for many years but it is only recently that pioneering companies have cracked the way in which GaN can be created on standard silicon wafers. This of course has reduced GaN prices. Nevertheless, nobody is saying that growing GaN crystals is cheap but costs are falling which is essential if customers are going to buy into devices using that technology. Standard wafer fabrication methods is a breakthrough and bring in the inevitable economy of scale factor as more GaN products get to market and it must mean that the GaN game ante will need less of a bankroll in future.