November 22, 2024

Wonderfully Weird: How Hafnia Is Paving the Way for Neuromorphic Computing

This image shows an artistic impression of the hafnia atomic structure. Credit: Reproduced with authorization from Springer Nature
The properties of hafnium dioxide, frequently referred to as hafina, may appear average on the surface. When this material is rendered into ultrathin layers, it displays remarkable qualities: the layers can be utilized as non-volatile computer memory through the changing of dipoles with an electric field.
In addition, since the strength of these dipoles is influenced by the past electric field it has actually experienced, they are ideal for building memristors for brain-like computer architectures.
Beatriz Noheda, Professor of Functional Nanomaterials at the University of Groningen, has studied the product and recently wrote a Perspective post on its residential or commercial properties for the journal Nature Materials. She mentions, “It is already utilized in devices, although we do not understand all of the physics.”

This is Beatriz Noheda, Professor of Functional Nanomaterials and scientific director of the Groningen Cognitive Systems and Materials Center at the University of Groningen (the Netherlands). She is the lead author of the Perspective paper on Hafnium Oxide ferroelectrics, released in Nature Materials. Credit: University of Groningen
Ferroelectric materials appeared to be great prospects. These materials are made up of units with dipoles that will switch collectively utilizing an electric field.
Oxygen jobs
It did, nevertheless, bring in the attention of microchip manufacturers when amorphous hafnia turned out to be an extremely efficient gate insulator in transistors. Noheda: “By replacing the conventional silicon oxide with hafnia, transistors could be made smaller sized.”
Nohedas interest in the material stems from her work for the Groningen Cognitive Systems and Materials Center (CogniGron), of which she is the scientific director. The goal of CogniGron is to develop a neuromorphic computing architecture. Hafnia is among the products that are studied there. “In a paper released by Science in 2021, we explain how switching does not simply occur simply through dipoles. We found that the motion of oxygen vacancies also contributes,” says Noheda. Based upon her experience, she was welcomed to go over the lessons gained from hafnia in a Perspective article for Nature Materials.
Sustainable
Hafnia acts like a ferroelectric, however it maintains its residential or commercial properties only at the nanometer scale. That said, it appears that hafnia does not exactly behave like a ferroelectric. “We need to understand this to completely utilize its potential, states Noheda.
Noheda likewise points to another idea that must be taken into consideration: the surface energy in nanoparticles. “The stage diagram shows that the relatively large area of these particles produces the equivalent of an incredibly high pressure in hafnium dioxide, which appears to play a role in the homes of this material.” This type of understanding is very important in the search for other materials that act like hafnium. “Hafnium is not the most sustainable option for microchip production considering that the around the world supplies are too small. By looking for materials with comparable properties, we might discover a better prospect.” One alternative could be zirconium.
Neuromorphic chips
Discovering a sustainable alternative for hafnium could speed up using ferroelectrics in RAM memory. And given that the dipole strength depends upon the history of the electrical field that produces the dipoles, it would be a perfect product to produce memristors, which allows intermediate values in between de timeless binary values of 0 and 1. Such analog gadgets could behave like the neurons in our brain and would be prospects for a neuromorphic computer system architecture. “We are working towards such neuromorphic chips. But initially, we should totally comprehend the physics of hafnium dioxide and comparable materials.”
Referral: “Lessons from hafnium dioxide-based ferroelectrics” by Beatriz Noheda, Pavan Nukala and Mónica Acuautla, 3 May 2023, Nature Materials.DOI: 10.1038/ s41563-023-01507-2.
The Perspective article has actually been written in partnership with Monica Acuautla from the Engineering and Technology institute Groningen (ENTEG) at the University of Groningen (The Netherlands) and Pavan Nukala from the IISC Bengaluru (India).

She is the lead author of the Perspective paper on Hafnium Oxide ferroelectrics, published in Nature Materials. Noheda: “By changing the traditional silicon oxide with hafnia, transistors could be made smaller.”
Based on her experience, she was invited to go over the lessons learned from hafnia in a Perspective post for Nature Materials.
Finding a sustainable alternative for hafnium could accelerate the usage of ferroelectrics in RAM memory. First, we need to fully understand the physics of hafnium dioxide and similar products.”