Two-dimensional (2D) materials have a huge potential for providing devices with much smaller size and extended functionalities with respect to what can be achieved with today’s silicon technologies. But to exploit this potential we must be able to integrate 2D materials into semiconductor manufacturing lines — a notoriously difficult step. A team of Graphene Flagship researchers in Sweden and Germany now reports a new method to make this work.
The integration of 2D materials with silicon or with a substrate with integrated electronics presents a number of challenges. “There’s always this critical step of transferring from a special growth substrate to the final substrate on which you build sensors or components,” said Arne Quellmalz, researcher at Graphene Flagship Associate Member KTH. “You might want to combine a graphene photodetector for optical on-chip communication with silicon read-out electronics, but the growth temperatures of those materials are too high, so you cannot do this directly on the device substrate.”
So far, most of the experimental methods for transferring 2D materials from their growth substrate to the desired electronics are either non-compatible with high-volume manufacturing or lead to a significant degradation of the 2D material and of its electronic properties. The beauty of the solution proposed by Quellmalz and co-workers is that it lies in the existing toolkits of semiconductor manufacturing: to use a standard dielectric material called bisbenzocyclobutene (BCB), along with conventional wafer bonding equipment.
“We basically glue the two wafers together with a resin made of BCB,” said Quellmalz. “We heat the resin, until it becomes viscous, like honey, and press the 2D material against it.” At room temperature, the resin becomes solid and forms a stable connection between the 2D material and the wafer. “To stack materials, we repeat the steps of heating and pressing. The resin becomes viscous again and behaves like a cushion, or a waterbed, which supports the layer stack and adapts to the surface of the new 2D material.”
The researchers demonstrated the transfer of graphene and molybdenum disulfide (MoS2), as a representative for transition metal dichalcogenides, and stacked graphene with hexagonal boron nitride (hBN) and MoS2 to heterostructures. All transferred layers and heterostructures were reportedly of high quality, that is, they featured uniform coverage over up to 100-millimeter sized silicon wafers and exhibited little strain in the transferred 2D materials.
According to the researchers, their transfer method is in principle applicable to any 2D material, independent of the size and the type of growth substrate. And, since it relies only on tools and methods that are already common in the semiconductor industry, it could substantially accelerate the appearance on the market of a new generation of devices where 2D materials are integrated on top of conventional integrated circuits or microsystems. This work is an important step towards that goal and, although many further challenges remain, the range of potential applications is large: from photonics, to sensing, to neuromorphic computing. The integration of 2D materials could be a real game-changer for the European high-tech industry.