GTAT: Blazing Ambition
Crystal growth furnace manufacturer, GTAT, has a solution that could see the price of SiC substrates plummet, reports Rebecca Pool.
GTAT believes its process is ready to grow high yielding six-inch boules that will compete on quality, with four-inch boules.
Following decades of development, SiC devices are finally gaining traction. Across the board, manufacturers of power supplies, photovoltaic inverters, and now plug-in vehicles, are testing or prototyping blisteringly fast and highly efficient SiC devices.
And the latest market forecasts from France-based Yole Developpement estimate the total SiC device market will top US$ 1 billion come 2022, with compound annual growth rate reaching 40% from 2020 to 2022.
For US-based crystal growth and polysilicon technology supplier, GT Advanced Technologies, these market developments spell good news.
Emerging from bankruptcy following a fractured sapphire furnace supply contract with consumer electronics goliath, Apple, the reorganised GTAT is targeting the solar industry with its equipment, amongst other markets. And, right now, it is touting a repeatable process for producing high quality, six-inch SiC boules.
"For several years, we have been aggressively pursuing and have now established a stable, high-yield, robust six inch process," highlights GTAT's chief executive, Greg Knight. "This follows extensive thermal modelling, process development as well as switching from non-silicon carbide seeds to full silicon carbide seeds."
Six inch SiC boule grown in a SiClone 200 furnaces; the boule has been annealed and OD-shaped.
The process includes the company's 'SiClone' SiC sublimation furnace, related process technology and the all-important graphite hot zone, instrumental to reproducible, high quality crystal growth.
Crucially, Knight is certain his company's process is ready to grow high yielding six-inch boules that will compete on quality, with four-inch boules.
The chief executive won't be drawn on yield figures but as he puts it: "Getting the correct hot zone design and process were absolutely critical, and then the biggest challenge has been to reliably grow the boule over and over again, but this is what we are able to do."
"I'm not implying we have 100% yield but every time we open the furnace we're getting boules of the geometry and size that we expect to get," he adds.
Knight is confident that the time is right to market the company's SiC manufacturing solution, highlighting how much of the world's production remains captive, limiting supply and keeping prices high.
However, he hopes his company's approach will boost SiC semiconductor supply and drive costs down. And with a process that reproducibly produces quality boules, the priority is to get GTAT's technology and services into the market so manufacturers have access to an 'appropriately priced' six inch substrate.
"There are six-inch fabs out there that need this material and end up paying very high prices to get it," he says. "Our goal is to get the selling price of high-quality six-inch wafers down so that downstream, fabs get the economic benefit of going to the larger substrate, and with our process, yield and hot-zone cost, we're ready to reach this now."
"We're very good at developing the necessary equipment, including optimising the furnace hot zone so it yields a good boule," he adds. "We're able to do this and reduce hot zone costs, so the [overall] cost is going to come down astronomically."
Indeed, the cost of today's six-inch SiC wafer currently comes in at around $1500, while Knight believes, even now, this figure should be closer to $1000. As he highlights: "We have already identified a roadmap to get epi-ready wafer manufacturing costs down to around $300."
"Companies such as Cree are seeing a big uptake in the demand for SiC [technology], so when you see these businesses heavily investing in SiC, that is the biggest signal that the market growth is real," he adds.
Right now, Knight reckons GTAT is the only supplier of both the equipment and a robust process to produce SiC boules for six inch wafers, with the competition offering, as he puts it, a non-optimised baseline recipe. He also believes large markets for GTAT's equipment exist in Asia, including China, Taiwan and Korea, as well as the US, with key applications including PV inverters, electric vehicles alongside on-board electronics and charging stations.
"It has really been a tough transition for substrate providers; the move from four- to six-inch wafers is not so simple, especially for silicon carbide," he says. "But if the fabs have a six-inch substrate that is priced appropriately, six-inch production will increase, yields and demand will rise, costs will come down."