Atomically precise graphene etch stops for three dimensional integrated systems from two dimensional material heterostructures

By Dr Jangyup Son and Prof Arend M. van der Zande from Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, USA

Thumb dsc 4922  tojpeg 1446202647502 x2
Jan 26, 2019

1. Could you briefly outline the key findings of your paper?

Arend van der Zande:

“We discovered a technique overcoming a technical challenge that has caused trouble in 2D materials research for the last few years- when you engineer a heterostructures from 2D materials, how do you separately access the individual layers? 

This paper explores the implications of a newly discovered fabrication technique, and what kinds of new devices it enables. Specifically, we discovered a highly selective etching technique where XeF2 etches most 2D materials except for graphene. We apply the selective etch to pattern heterostructures of 2D materials where buried graphene layers act as atomically precise etch stops. allowing us to directly contact layers of graphene at different depths within 2D heterostructures. Surprisingly, we found that the etch stops made excellent electrical one-dimensional contacts, allowing us to engineer high performance transistors with graphene that was fully encapsulated in hBN to contact resistances and mobilities close to the theoretical limits. To explore the potential for this technique, we took inspiration from technologies like integrated circuitry and nanoelectromechanical systems, which commonly use etch masks and etch stops, We demonstrated all-2D versions of interlayer vias, devices with multiple active layers, and dry etching to suspend electromechanical resonators.”

2. What is your role in this work?

Arend van der Zande:

“The main conception and execution of the idea came from our two co-corresponding PIs Arend van der Zande (left in first photo) and Gwan-Hyoung Lee (left in second photo), and two lead authors (origin story below). Dr. Jangyup Son (a postdoctoral researcher in the vdZ lab, right in first photo) and Mr. Junyoung Kwon (a PhD student in the Lee lab, right in second photo). This was a highly-collaborative work, with many authors contributing unique capabilities and would not have been possible without the newly awarded Illinois Materials Research Science and Engineering Center (MRSEC). Beyond our own expertise, the center brought expertise in atomic scale imaging (Huang Lab), quantum simulations (Ertekin Lab), and low temperature electronic transport (Mason Lab). The capabilities brought by these groups enabled us to build a much more sophisticated and impactful set of conclusions.”

3. What was the genesis of this paper? How did you come to this particular problem?

Arend van der Zande:

“While the underlying issues that we address in this paper have been around for a while, the key idea came about through serendipity. Shortly after Jangyup Son joined my lab, he was looking into the fluorination of graphene using XeF2 for a separate experiment, and I casually suggested that he just try putting in some other 2D materials (specifically hBN), just to see what would happen. A day later, he excitedly came back to me to show me that the hBN was completely removed. We were surprised by this result, but we realized that we had stumbled onto a potentially useful new technique. Once we had that core idea, the rest of the implications on engineering devices led naturally from our expertise. We brought in our collaborator Gwan-Hyoung Lee, who is a master at fabricating 2D heterostructures to engineer some of the devices we demonstrated and bring some additional ideas.”


Jangyup Son:

“Etching of 2D materials (such as hBN and TMDCs, besides of graphene) by XeF2 gas and graphene etch masks have been already reported before. However, all published papers just have utilized this for etching 2D devices or etching of sacrificial layer under graphene. In contrast, we reached a creative idea that these results (etching selectivity and graphene etch stops) can be used to make buried graphene contact even in hBN-encapsulated heterostructures, which can lead to 2D materials based integrated systems. Since then, we’ve discussed and designed various experiments to demonstrate many applications using this etching technique, and all that have not been possible to fabricate with conventional techniques so far have been easily fabricated with our technique. After many efforts, we were finally able to show the possibility on the fabrication of 3D integrated systems from 2D material heterostructures.”

4. What is the most empowering implication of your results?

Arend van der Zande:

“There have been many demonstrations of different kinds molecular scale electronics from heterostructures of 2D materials. However, a common challenge to many of these demonstrations is how to pattern and access the individual layers. This will become very important as we start to scale up from individual devices to integrated circuits and full wafers of devices. The techniques we have demonstrated give a strategy for patterning that is both scalable and self-limiting, allowing us to achieve atomic precision on the wafer scale. Moreover, we have demonstrated that the resulting devices have properties equivalent or superior to the state of the art.”


Jangyup Son:

“Our results have two important implications. First, fabrication process to make electrical contact with our technique is very simple and the contact resistance is lower than 100 Ω·mm, so that this technique can be an alternative way for the conventional contacting method in hBN–encapsulated system, (for example, 1D edge contact). Second, it allows realization of 3D integrated systems from 2D material heterostructures.”

5. How have 2D materials been uniquely instrumental to enabling these results?

Arend van der Zande:

“2D materials are one of the most promising candidates for next generation nanoelectronic systems. However, there are still many technical barriers that need to be overcome to best utilize these materials. This paper is all about strategies for overcoming some of these important technical barriers and move 2D materials from a scientific curiosity into a viable technology.”


6. Can you describe the main challenges associated to the preparation of this manuscript? Any anecdotes you’d like to share with us?

Arend van der Zande:

“The main challenge we had in this paper was to figure out how to unite the many different discoveries. The use of graphene etch stops had several different implications – scalability, improved contact resistance and device properties, and useful new device applications. We had trouble figuring out how to write one paper which included all the ideas. In the end, I think it makes for a stronger paper, because it shows how all these improvements work together, but it was a challenge to bring everything under one theme while still being readable, and we considered writing several papers rather than one.”


7. Anything that stroke you as particularly surprising, unexpectedly pleasant/unpleasant during the peer review process?

Arend van der Zande:

“Most of the comments were positive, but all of the comments were very constructive. The critical comments forced us to rewrite the paper to clarify how our results were unique from what came before. In particular, there had been several papers using graphene as an etch mask for various materials, but none that had used the graphene as an etch stop in a heterostructure or as a method for engineering contacts. These comments allowed us to change our title, and refocus the introduction and add contextualization throughout to highlight the uniqueness of graphene etch stops. The end result is that the paper is much more clearly written and focused, and the reviewers were happy with our changes. This is a case where the peer review process really worked.”


8. What is your favorite 2D paper published in 2016/2017, and why?

Arend van der Zande:

“One of my favorite papers from last year is:

Layer-by-layer assembly of two-dimensional materials into wafer-scale heterostructures

Kibum Kang, Kan-Heng Lee, and Jiwoong Park, Nature, 550, 229-233 (2017)

In this paper, they showed a simple technique to transfer large area 2D materials and fabricate arbitrary multilayer heterostructures without getting any bubbles or defects at the interface. While there are many papers demonstrating growth of monolayer 2D materials, or one-off demonstrations of a particular device concept made from 2D heterostructures, there are relatively few thinking about scalably-manufacturing devices beyond monolayers.”


9. Which is the development in the field of 2D materials that you would like to see in the next 10 years?

Arend van der Zande:

“2D materials represent the ultimate limit of both a mechanical, atomically-thin membrane and of molecular electronics. One of the most promising areas of research right now is trying to unite these two important characteristics to engineer mechanically active electronics that cannot be made with conventional materials. This brings many fundamental questions like how the electronic properties of 2D materials change under large deformations and how interlayer slip affects energy transfer across 2D interfaces. Answering those fundamental questions will allow us to explore applications with 3D devices like origami nanomachines, stretchable electronics and conformable sensors. As someone who started graduate school right when graphene was “discovered” I am frequently astounded at how much progress we have made in the last 15 years, first in graphene then in the wide diversity of the of 2D materials and sophisticated heterostructures that followed. I am excited to see what the next 15 years brings.”


Jangyup Son:

“It is the realization of all-2D-materials-based integrated systems such as practical electronic devices or creative nano-architectures. Equally, we want to see great improvements for growth technique of 2D materials because wafer scale production is essential to make commercialized 2D materials devices.”


10. And now, what’s next?

Arend van der Zande:

“For this project, next questions are: Can we scale up, can we scale down, what other things do the tools of chemical functionalization and selective etching allow? We demonstrated several devices on exfoliated heterostructures in this paper, but can we integrate them with large area 2D heterostructures to engineer integrated circuits? Our results show that the patterning works down to nanometer length scales, so a natural question is how small of a device or graphene nanoribbon can we make? Can we use this technique to pattern the chemical functionalization and get atomically precise circuitry along all three dimensions?”


Jangyup Son:

“We are trying to understand more why the contact resistance in our devices is extremely low and how we can control it. Moreover, we are currently working on fabrication of all-2D-materials-based integrated systems and on making new nanostructures.

Medium dsc 4922  tojpeg 1446202647502 x2

Kanudha Sharda

Associate Editor, Nature Communications

Kanudha became a part of Condensed Matter Physics team at Nature Communications in November 2017. She graduated in physics from University of Delhi, India and was an Erasmus Mundus PhD scholar at Politecnico di Torino, Italy. She was a postdoc at Imperial College London where her research focused on the fabrication and characterization of devices based on layered transition metal dichalcogenide materials for electronic and energy storage applications. Kanudha handles manuscripts on synthesis and wider applications of two-dimensional materials and their heterostuctures.

No comments yet.