Get your freshly baked graphene in Delft

Delft-based 2D materials foundry Applied Nanolayers is seeing “a drastic increase in demand” for its wafer-sized graphene sheets, used by customers to manufacture next-generation sensing devices.

A little over a decade ago, in a bar in Leiden, the seed of a new company was sown. Over a couple of beers, two Leiden University scientists told a British semiconductor industry veteran about the work they’d been doing in the lab. Studying the growth process of graphene using scanning tunneling microscopy, the researchers had uncovered mechanistic details that allowed them to grow large chunks of high-quality graphene. Agreeing that the 2D semiconductor harbored a lot of potential as well as hitting it off on a personal level, the trio soon afterward decided to start a company and make large quantities of the material.

That was a rather bold move. At the time, scientists were still teasing out the unique electronic properties of the honeycombed network of carbon atoms, providing clues for future applications but no more than that. What little graphene was needed, researchers could make themselves by peeling off graphite layers with adhesive tape – just as its discoverers had done years earlier. And here, three men were thinking about commercially manufacturing the material.

The gamble seems to have paid off. Initially based in Nijmegen, but later moving into the Else Kooi Lab building on the Delft University of Technology campus, ‘2D material foundry’ Applied Nanolayers (ANL) has been patiently biding its time, waiting for the commercial interest in graphene and other 2D materials to gain momentum.

And that’s what seems to be happening now, according to ANL CEO and co-founder Paul Hedges. “We’ve been seeing a drastic increase in demand lately. We expect to expand rapidly over the next few years – as fast as limitations on floor space, availability of talent and money will allow us.”

Carbon powered

Hedges, who acquired “a bit of scar tissue in the semiconductor industry” after working for several Dutch companies since the 80s, is ANL’s commercial brain. He started the company together with Joost Frenken, back then professor of experimental physics at Leiden University and currently director of the ASML co-sponsored research institute ARCNL, and Frenken’s former PhD student Richard van Rijn, who currently serves as ANL’s CTO.

ANL manufactures graphene using chemical vapor deposition (CVD). This involves exposing a heated substrate to carbon-containing gaseous precursors in a low vacuum environment. As the gas decomposes on the hot surface, carbon atoms deposit on it, which through a series of steps arrange into the famous honeycombed network of graphene. Similar procedures yield other 2D materials, such as hexagonal boron nitride. Through carefully controlling the CVD process, ANL manages to grow high-quality sheets of these materials, covering 200-millimeter wafers. Ultimately, it plans to master 300-millimeter discs as well.

ANL keeps the details of its manufacturing process under lock and key, but Hedges doesn’t mind revealing that his engineers heavily modified commercially available CVD tools to make it happen. “In Nijmegen, we shared a building with a brokerage company for semiconductor equipment, which allowed us to acquire very good CVD tools. We basically ripped those apart, re-engineered the reaction chambers and replaced the electronics and software to optimize and automate our process.”

Once the 2D material has been synthesized, it needs to be separated from its substrate. ANL has developed a proprietary dry process to transfer the sheet to another carrier. Typically, this is a customer’s planarized wafer on which ICs and their wiring have already been fabricated. The graphene, another 2D material or even a stack of 2D materials is added as a sensing layer, which together with the read-out, preprocessing and I/O electronics underneath constitutes a graphene-based sensor chip. “Imaging, biosensing, pressure, magnetism: there are a lot of sensing applications where 2D materials can add performance or enable unique functionality. Photonic switching has a lot of potential too,” explains Hedges.

But wait a second, wasn’t graphene supposed to give us the next-generation transistors, once silicon runs out of steam? There’s nothing wrong with sensing, but it doesn’t quite capture the imagination as much as carbon-powered nanoswitches. “Transistors based on 2D materials are on international semiconductor roadmaps, but for the time being, device performance outside the lab is still unsatisfactory.”

“At ANL, we anticipated that early graphene applications would be at the back end of the line. Indeed, this has become our first source of revenue. As graphene applications gain momentum, the manufacturing infrastructure, defect management and so on will get a chance to mature. So, once the R&D has caught up and the silicon roadmap comes to an end, the world will be ready for front-end-of-the-line graphene applications as well.”

Intimate

Currently, ANL is supplying engineering volumes to customers that are in the process of raising the technology readiness level of their product. Once they’re ready to move to volume manufacturing, his company will scale along with them, Hedges says. “Either they keep sending us their half-finished wafers for integration of a 2D material or we supply them with the material on a carrier, so they can handle the integration themselves.”

Being an intimate partner of companies laboring to get the first graphene-based electronics on the market, surely Hedges can give an estimated guess on when we can expect those? “We are at this moment working on consortia to deliver these devices. In my view, these are still a couple of years out.”