ΑΙhub.org
 

Deep learning-guided surface characterization for autonomous fabrication


by
13 May 2020



share this:
deep learning-assisted nanofabrication

The semiconductor industry as we know it is facing a critical roadblock that will lead to the end of Moore’s law. As transistors continue to shrink, quantum effects have a significant negative consequence on their operation. As such, the development of “beyond CMOS devices” has begun.

The push for devices that are cheaper, smaller, and faster has led to the use of scanning probe fabrication. One of the first examples of such a technique was IBM’s video “A boy and his atom”, where CO molecules were moved along a Cu surface using a sharp metallic tip. Limitations in the thermal stability of using metal surfaces for device fabrication lead to the development of hydrogen lithography. Using a scanning probe tip, hydrogen can be selectively removed from a silicon surface at atomic resolution (roughly 0.3 nm) providing thermal stability as well as compatibility with today’s enormous silicon-based transistor fabrication infrastructure.

Hydrogen lithography is used for the viable creation of such “beyond CMOS devices” as quantum computers, single atom transistors, and Binary Atomic Silicon Logic. The latter is pictured above where the absence of a hydrogen atom creates a quantum dot. These quantum dots are capable of representing binary information through the presence or absence of a single electron, drastically reducing the predicted energy consumption and spatial requirements of today’s modern transistors.

One of the leading issues with this fabrication technique is that at present, it is not scalable. The complexity of the surface requires constant monitoring to assess the quality of both the scanning probe tip and the silicon surface. By employing deep learning methods through the use of convolutional neutral networks (CNNs), we have been able to automate the entire fabrication process. Networks used for assessing tip quality must only distinguish between a “good” and “bad” tip which subsequently triggers an in situ tip reconditioning algorithm if needed. The surface quality is assessed using a greater number of factors, requiring a more complicated scheme.

A perfect surface consists of a perfectly flat silicon surface where each surface silicon atom bonds to a single hydrogen atom. In actuality there are numerous different configurations such as missing silicon atoms, additional hydrogens, or even residual water molecules, all of which make a specific lattice site unable to host a quantum dot. The formation of such defects can only be controlled to a certain degree, and ultimately depends on their thermodynamic probabilities of formation. By developing a CNN capable of semantic segmentation, we have been able to fully automate the surface quality assessment. Using seven different classes, the network is capable of distinguishing between defects that can alter the device behaviour, prevent the connection of additional devices, or simply prevent a single quantum dot from being fabricated. With a full understanding of where the defects exist on the surface, the most viable area of the surface can be calculated which allows for an automated hydrogen lithography process to fabricate these quantum dot devices in an effectively defect-free area.

Without the need for a single user to constantly monitor tip and surface quality, this allows for a fully scalable process where a single user could now monitor hundreds of machines providing a potential roadmap for the commercial fabrication of “beyond CMOS devices” using scanning probe systems.

Read the research articles

Deep learning-guided surface characterization for autonomous hydrogen lithography
Mohammad Rashidi, Jeremiah Croshaw, Kieran Mastel, Marcus Tamura, Hedieh Hosseinzadeh and Robert A Wolkow
Mach. Learn. Sci. Technol. 1, 025001 (2020)

Autonomous Scanning Probe Microscopy in situ Tip Conditioning through Machine Learning
Mohammad Rashidi and Robert A. Wolkow
ACS Nano 12, 5185–5189 (2018)




Jeremiah Croshaw is a PhD student at the University of Alberta specializing in condensed matter physics.
Jeremiah Croshaw is a PhD student at the University of Alberta specializing in condensed matter physics.




            AIhub is supported by:


Related posts :



monthly digest

AIhub monthly digest: May 2025 – materials design, object state classification, and real-time monitoring for healthcare data

  30 May 2025
Welcome to our monthly digest, where you can catch up with AI research, events and news from the month past.

Congratulations to the #AAMAS2025 best paper, best demo, and distinguished dissertation award winners

  29 May 2025
Find out who won the awards presented at the International Conference on Autonomous Agents and Multiagent Systems last week.

The Good Robot podcast: Transhumanist fantasies with Alexander Thomas

  28 May 2025
In this episode, Eleanor talks to Alexander Thomas, a filmmaker and academic, about the transhumanist narrative.

Congratulations to the #ICRA2025 best paper award winners

  27 May 2025
The winners and finalists in the different categories have been announced.

#ICRA2025 social media round-up

  23 May 2025
Find out what the participants got up to at the International Conference on Robotics & Automation.

Interview with Gillian Hadfield: Normative infrastructure for AI alignment

  22 May 2025
Kumar Kshitij Patel spoke to Gillian Hadfield about her interdisciplinary research, career trajectory, path into AI alignment, law, and general thoughts on AI systems.

PitcherNet helps researchers throw strikes with AI analysis

  21 May 2025
Baltimore Orioles tasks Waterloo Engineering researchers to develop AI tech that can monitor pitchers using low-resolution video captured by smartphones

Interview with Filippos Gouidis: Object state classification

  20 May 2025
Read the latest interview in our series featuring the AAAI/SIGAI Doctoral Consortium participants.



 

AIhub is supported by:






©2025.05 - Association for the Understanding of Artificial Intelligence


 












©2025.05 - Association for the Understanding of Artificial Intelligence