Nov 9 2018
A group of multi-disciplinary engineers and researchers from the University of Illinois at Urbana-Champaign has identified an innovative and more accurate way for developing nanoscale-size electromechanical devices. The results of the study have been reported in Nature Communications.
This is University of Illinois Assistant Professor of Mechanical Science and Engineering, Arend van der Zande, and Postdoctoral Researcher, Jangyup Son. (Image credit: University of Illinois Department of Materials Science and Engineering)
In the last five years, there has been a huge gold rush where researchers figured out we could make 2D materials that are naturally only one molecule thick but can have many different electronic properties, and by stacking them on top of each other, we could engineer nearly any electronic device at molecular sizes The challenge was, though we could make these structures down to a few molecules thick, we couldn’t pattern them.
Arend van der Zande, Professor of Mechanical Science and Engineering, University of Illinois College of Engineering.
At any scale of electronic devices, layers are etched away in accurate patterns to regulate the way the current flows.
“This concept underlies many technologies, like integrated circuits. However, the smaller you go, the harder this is to do,” said van der Zande. “For example, how do you make electrical contact on molecular layer three and five, but not on layer four at the atomic level?”
Later, an unexpected discovery resulted in an approach for doing just that.
Jangyup Son, being a new postdoctoral researcher in van der Zande’s laboratory, was performing a number of experiments on single graphene layers with the help of xenon difluoride (XeF2), and he happened to “throw in” an extra material available on hand—hexagonal Boron Nitride (hBN), which is an electrical insulator.
“Jangyup shoved both materials into the etching chamber at the same time, and what he saw was that a single layer of graphene was still there, but a thick piece of hBN was completely etched away by the Xenon difluoride.” said van der Zande
This unexpected discovery made the researchers see where they can apply the graphene’s potential to tolerate the etching agent.
This discovery allowed us to pattern two-dimensional structures by placing layers of graphene between other materials, such as hexagonal boron nitride (hBN), transition metal dichalcogenides (TMDCs), and black phosphorus (BP), to selectively and precisely etch one layer without etching the layer underneath.
Arend van der Zande, Professor of Mechanical Science and Engineering, University of Illinois College of Engineering.
Upon exposure to the etching agent XeF2, graphene was able to retain its molecular structure and conceal, or safeguard, the layer underneath and essentially halted the etching process.
“What we’ve discovered is a way to pattern complicated structures down to a molecular and atomic scale,” van der Zande said.
In order to study the strengths of the latest method, the researchers developed a simple graphene transistor to analyze its performance with regards to conventionally developed graphene transistors, which are being patterned in a way that not only causes disorder in the material but also degrades their performance.
Because these molecules are all surface, if you have it sitting on anything with any disorder at all, it messes up the ability for the electrons to move through the material and thus the electronic performance. In order to make the best device possible, you need to encapsulate the graphene molecule in another two-dimensional material such as insulating hBN to keep it super flat and clean.
Arend van der Zande, Professor of Mechanical Science and Engineering, University of Illinois College of Engineering.
This is where the novel method proves to be very useful. The graphene molecule can stay pristine and encapsulated, and at the same time, it can tolerate the etching required to come into contact with the material, thus maintaining the properties of the material.
As proof of concept, the transistors developed with the help of the latest method were found to outperform all the other transistors, “making them the best graphene transistors so far demonstrated in the literature,” van der Zande said.
He said that the subsequent steps would be to check how scalable the method is and whether it will actually allow formerly impossible devices. Is it possible to leverage the self-arresting nature of this method to develop a million analogous transistors instead of only one? Is it possible to pattern devices down to the nanoscale in all three dimensions simultaneously to develop nanoribbons without any disorder?
Now that we have a way of minimizing the disorder within the material, we are exploring ways to make smaller features because we can do encapsulation and patterning at the same time. Normally, when you try to make smaller features like nanoribbons of 2D materials the disorder begins to dominate, so the devices do not work properly. The graphene etch stop, as the technique is called, will make the entire process of building devices easier.
Arend van der Zande, Professor of Mechanical Science and Engineering, University of Illinois College of Engineering.
The study involved a multi-disciplinary teamwork of people and shared facilities equipment from the Materials Research Laboratory and the Micro & Nanotechnology Lab. Expert faculty include Associate Professor of Physics and Director of the Illinois Materials Research Science and Engineering Center (MRSEC), Nadya Mason, for electronic transport; Associate Professor of Mechanical Science and Engineering, Elif Ertekin, for modeling interfaces; and Assistant Professor of Materials Science and Engineering, Pinshane Huang, for electron microscopy. MRSEC provided the primary funding for this research.