PORTLAND, Ore. -- Every semiconductor research organization in the world is trying to fabricate single monolayers of graphene and pattern them in such a way as to make superior circuitry to silicon. Now, however, some of the IBM researchers who have been growing monolayers of graphene have found another advantageous use for them- -- one that greatly lowers the cost of fabricating blue light emitting diodes (LEDs) from gallium nitride (GaN).
"We grew a single-crystalline gallium nitride [GaN] film on wafer-scale graphene formed on a silicon carbide (SiC) wafer," the self-described "Master Inventor" and Research Staff Member at IBM's T.J. Watson Research Center (Yorktown Heights, New York), Jeehwan Kim, told EE Times.
"An entire GaN film was then transferred to a silicon [Si] substrate and the graphene which remained on the SiC wafer was reused multiple times to grow and transfer multiple GaN films. In principle, this method can be much more cost-effective compared to conventional methods where an expensive SiC wafer or sapphire wafer is used only one time to grow GaN films without reusability. Also we found that the film quality is higher on graphene than other substrates (less defect density) Therefore, our technique offers extreme cost-efficiency and better overlaying film quality."
By growing graphene on a silicon carbide (SiC) wafer, gallium nitride (GaN) can be overgrown, lifted off and put on a cheaper silicon substrate then overgrown again on the same graphene covered wafer, greatly lowering the costs of GaN LEDs which today throw away the expensive SiC wafer after each time.
It has been extremely challenging to grow graphene monolayer films on a wafer scale, but boiling off the silicon from a silicon carbide wafer has been found to be one of the most reliable methods. For fabricating circuitry using graphene as the semiconductor many different methods have been tried in the past, most only partially successful. In fact, Kim has also demonstrated that graphene films obtained by boiling off the Si from SiC wafers can be reliably transferred to silicon substrates (see illustration) as well as demonstrated that their quality was higher than trying to grow graphene in-place on the wafer where it will be used.
But the growth of other types of films on top of the graphene opens a new door to the seemingly miracle material. Kim calls his GaN films "direct van der Waals epitaxy of high-quality single-crystalline GaN films on epitaxial graphene." He claims that the GaN and other films, which can be transferred to arbitrary substrates, produce ideal films for manufacturing -- in the case of GaN for manufacturing blue LEDs. In fact, Kim's lab has already created what they claim are superior blue LEDs by repeatedly growing stacks of GaN on reused graphene layers atop boiled SiC wafers.
"We have introduced a completely new way of using graphene to reduce the cost of semiconductor manufacturing. We for the first time demonstrate the growth of wafer-scale single-crystalline films on graphene and are able to recycle this graphene for multiple growth/transfer of the films. Our work also offers a general principle of growing high quality single-crystalline semiconductors on graphene."
Graphene monolayers can be lifted from a silicon carbide (SiC) wafer, after boiling off the silicon) then transferred to any substrate -- here a crystalline silicon substrate.
Because the perfect graphene layers are not damaged by lifting films off of them, Kim also proposes to grow other types of semiconductors on his perfect crystalline graphene substrates and deposit them onto other, possibly flexible substrates.
Kim says that his techniques could be the "next big thing" in high-frequency transistors, photodetectors, biosensors and other "post-silicon era" devices, for which IBM is investing $3 billion over the next five years. He described graphene as the world's "first two-dimensional material."
He also predicts that his monolayers of graphene will be used for transparent electrodes on touch screens, rollable e-paper, foldable LEDs, logic transistors, thin-film transistors, and high-frequency transistors.
Details are available in Principle of direct van der Waals epitaxy of single-crystalline films on epitaxial graphene by Jeehwan Kim, Can Bayram, Hongsik Park, Cheng-Wei Cheng, Christos Dimitrakopoulos, John Ott, Kathleen Reuter, Stephen Bedell, and Devendra Sadana.
— R. Colin Johnson, Advanced Technology Editor, EE Times