HomeTechnologyComputingReview of ferroelectric devices for intelligent computing

Review of ferroelectric devices for intelligent computing

Intelligent computers (2022). DOI: 10.34133/2022/9859508″ width=”800″ height=”351″/>
Schematics of (a) challenges faced by modern computers using von Neumann architecture and (b) “heat wall”, “memory wall” and von Neumann bottleneck solutions based on ferroelectric devices. Credit: Intelligent computers (2022). DOI: 10.34133/2022/9859508

Transistors or “microchips” partly explain why our wafer-thin laptops can perform much more complicated tasks than their clumsy, gargantuan predecessors. To maximize computing capabilities, engineers try to make transistors in the smallest possible size and pack billions of them into a single computer chip.

However, despite the rapid evolution of manufacturing techniques, traditional transistors are approaching their physical limit – that one nanoscale devices can’t afford to shrink beyond a certain point – and that hinders the development of computing capabilities.

But as the data continues to pour in, the demand for computing capabilities continues to rise. New devices, especially new storage and logical devices with higher speed and lower power consumptionare needed to unleash new computing capabilities while removing major obstacles for existing computing systems.

Recently, a group of researchers from China pointed to ferroelectric devices as a promising solution and published a review article on emerging ferroelectric materials and devices for intelligent computing. The review was published in Intelligent computers.

Ferroelectric materials are quite versatile and are widely used as special-purpose memory, including in aerospace storage devices. They have special polarization characteristics, a magnetism-like property that can be maintained even after the external electric field is removed. But when the film thickness is reduced to less than 10 nm, most conventional ferroelectric materials lose their polarization characteristics at 25°C and thus are not adapted to the nanoscale integrated circuit (IC) fabrication process.

Must Read
I bought a Mac Studio: Here are the specs I chose and why

New ferroelectric materials with high scalability potential can solve these problems. “The Discovery of the Polarization Effect in High-κ Materials, the Most Commonly Used Gate-Oxide Materials for Nanoscale MOSFETs [metal-oxide-semiconductor field-effect transistors]is a breakthrough for the mass production of ferroelectric transistors,” said the researchers.

They reviewed two prominent examples of polycrystalline Hf-based and amorphous oxide-based ferroelectric materials, and briefly described some recently reported new materials and devices. All have been found to be compatible with the Complementary Metal Oxide Semiconductor (CMOS) fabrication process.

For the state-of-the-art ferroelectric devices, the researchers categorized them into low-power logic devices, high-performance memory cells, and neuromorphic devices, and summarized each in detail. The summaries covered the development of the devices and their capabilities to break the “heat wall”, the “memory wall”, and the von Neumann bottleneck, respectively.

Ferroelectric negative capacitor field-effect transistors (NCFETs) as low-power logic devices can break the “heat wall” that hinders the improvement of the main frequency of the processor due to the rising power density and the heating effect. Lowering the drive voltage of the chips is one possible method of breaking the ‘heat wall’, and its feasibility is highly dependent on the SS [subthreshold swing] of the transistor’, the researchers explained.

“Ferroelectric NCFETs, together with the voltage gain effect, can overcome Boltzmann’s tyranny and achieve SS less than 60 mV/dec. Thus, they are considered one of the most promising device architectures for ultra-low power applications and can support the rapid development of the IC industry.”

Ferroelectric capacitor based random access memory (FeRAM) and ferroelectric field effect transistor (FeFET) based memory, categorized as high power memory cells, show excellent performance in dynamic random access memory (DRAM) replacement and embedded applications.

Must Read
The journey to Entrepreneurship and a Healthy Life - Piyush Somani the Cloud expert debuts as an author with the launch of his 2 Books

The ferroelectric capacitor, unlike the conventional DRAM capacitor, can store information through the Pr charge, which is non-volatile and has a much higher charge density per area.

Therefore, replacing the dielectric material of a flash device with doped HfO2 ferroelectric material or amorphous oxide ferroelectric material to realize a FeFET is an alternative method to further reduce the power or delay of these memories. said the researchers. This helps bridge large performance or area gap between the logical device and the memory cell, overcoming the so-called “memory wall”.

In addition, FeFETs can be used as neuromorphic devices to break the von Neumann bottleneck. The von Neumann bottleneck refers to the delay and flow problems caused by the inefficient data transfer between the originally separate memory module and logical processor, for which neuromorphic computing – the imitation of the neuron system for information processing – is a possible solution.

In a neuromorphic system, artificial neurons and synapses are the key components, and FeFETs are said to be able to implement both. For applications in neurons, FeFETs have been used as pulsed neural networks; for artificial synapse applications with spike neural networks (SNNs) and convolutional neural networks (CNNs), FeFETs are applicable for their ability to simultaneously perform storage and processing functions.

In addition, ferroelectric tunnel junctions (FTJs) have attracted much attention for synaptic device applications due to their compact device structure, non-destructive readout scheme, and high write/read access speeds.

In conclusion, the researchers indicated that if the trade-off between process compatibility and device performance can be achieved, then NCFET, FeRAM or FeFET memory and ferroelectric synapse devices could be integrated into the same chip to build a multi-functional intelligent computing system.

Must Read
Despite Tech Weakness, Cloud Computing Investment Continues

“Based on the advances made in ferroelectric device processing technology, the integration of low-power logic, high-performance memories, and neuromorphic systems on a single chip seems feasible with continuous process improvement,” they emphasized. “This will help realize the development of powerful and highly efficient intelligent computing systems in the future.”

More information:
Genquan Han et al, Ferroelectric devices for intelligent computing, Intelligent computers (2022). DOI: 10.34133/2022/9859508

Provided by Intelligent Computing

Quote: Overview of Ferroelectric Devices for Intelligent Computing (2022, December 5) Retrieved December 5, 2022 from https://techxplore.com/news/2022-12-ferroelectric-devices-intelligent.html

This document is copyright protected. Other than fair dealing for private study or research, nothing may be reproduced without written permission. The content is provided for informational purposes only.



Please enter your comment!
Please enter your name here

Most Popular

Recent Comments