The G2REC project leader explains the role GaN can play in solving a major energy efficiency challenge.
Under the aegis of the €30 million European G2REC (Grand Gap Rectifier) project, Jean-Baptiste Quoirin of STMicroelectronics, is leading a consortium of companies and labs tackling the problem of rectifiers for things like computer server power supplies and motor control in large appliances.
Rectifiers convert AC to DC and ensure that electricity flows in just one direction. For applications under 250V that require fast switching, this is handled by silicon-based Schottky diodes. For higher voltage, fast-switching applications, however, the choice is currently limited to a range of much more complex and expensive components, most of which are notorious wasters of energy.
ASN had a chance to talk to Dr. Quoirin about the project, and the role wide-gap materials can play in creating a new generation of cost- and energy-efficient rectifiers
ASN: Could you provide a quick overview of your vision for G²REC?
Jean-Baptiste Quoirin: G²REC is a project that unites a consortium of industrial players and public research labs. The goal is to create a new industrial network that enables production of electronic power components in wide gap semiconductor materials: silicon carbide (SiC) and gallium nitride (GaN). The entire value chain – from the starting material to the device – is represented in this consortium.
ASN: How far into the project are you now? What are the objectives for the end?
JBQ: This December will mark the end of the first year of the project. Project completion is scheduled for December 2011.
There are two final objectives:
- To qualify an industrial network for thick epitaxy (7µm) of GaN on a 6-inch silicon substrate, and of SiC on a 4-inch silicon substrate.
- To qualify a manufacturing network for Schottky diodes serving the 600V to 1200V markets.
ASN: Why did you choose a GaN approach? Which new or existing markets will the GaN solutions address?
JBQ: We want to be able to epitaxially deposit GaN on a silicon substrate, and take advantage of the low-cost, wide-diameter wafers that silicon enables. This is not possible with SiC.
However, despite its high cost, SiC’s high thermal conductivity (which is equivalent to copper) makes it the choice for high-current, high-voltage industrial applications (50A/1200V).
GaN addresses the “power management” market for any piece of equipment that plugs into the regular (230V) power mains.
ASN: What are the challenges you face with the GaN solution ?
JBQ: The current state-of-the-art in GaN epitaxy is 1µm on a 4-inch silicon substrate. The objective of G²REC is to enable at least 7µm on a 6-inch silicon wafer. The challenge is that there is a 20% difference between the crystalline structure of silicon and GaN. Also, the ratio of their thermal dilation coefficients is a factor of two. This makes for some impressive materials engineering problems to resolve.