Can you point to a journal or conference publication or website with more information? Starting with what "EVA" stands for - Ellipsoidal Void A????
The pictured material is probably useful for solar cells or LEDs. I can't imagine patterning repeatable transistors or circuitry if the voids were not aligned with the pattern. OTOH, perhaps they can be seeded from an etched or implanted pattern. Worth learning more about!
Keith Lofstrom, Beaverton Oregon
This is primarily a materials advance, but the applications are numerous and growing. LEDs are the most abundant application of GaN today, but the binary III/V material is being used to take on all sorts of high-power and high-frequency applications that the this new process could potentially improve, from the laser diodes used in Blu-Ray disks to the high-power MOSFETs used in electric cars to future spintronic devices.
I think this is truly a breakthrough of huge implications potentially. GaN materials quality is either challenging or terribly expensive.
Sapphire is a reasonable MOCVD substrate cost compromise, but with typically crappy films. The method described is an extension of variants of nanowire growth regimes, - the modestly well known wire to wine-glass ?denecking ( widening ) taken to extreme case, with HUGE materials growth quality results in the upper thicknesses.
There is some remote possibility this method might enable a debonding film transfer process akin to a Silicon SOI handle wafer process to possibly enable reuse of the expensive sapphire by a touch repolishing step, for now that is speculative.
I'd hazard that the statement "remains a laboratory curiosity" is out of place, aside from IP this method is likely to have very rapid uptake from potential licensees. The upper film quality looks orders of magnitude better than conventional films, likely leading to huge increases in potential power / efficiency with modest cost over crappy films.
This bears very close watching how fast industry uptake proceeds.
Interestingly incredible assertion based on a guess remains a laboratory curiosity. If the depo process yields LEDs and other devices with consistently reproducible results we can look forward to brighter solid-state lighting. It certainly bears following results of the new films.
Blog Doing Math in FPGAs Tom Burke 18 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...