Abstract:Artificial neural networks (ANNs) have had an enormous impact on a multitude of sectors, from research to industry, generating an unprecedented demand for tailor-suited hardware platforms. Their training and execution is highly memory-intensive, clearly evidencing the limitations affecting the currently available hardware based on the von Neumann architecture, which requires frequent data shuttling due to the physical separation of logic and memory units. This does not only limit the achievable performances but also greatly increases the energy consumption, hindering the integration of ANNs into low-power platforms. New Logic in Memory (LiM) architectures, able to unify memory and logic functionalities into a single component, are highly promising for overcoming these limitations, by drastically reducing the need of data transfers. Recently, it has been shown that a very flexible platform for logic applications can be realized recurring to a multi-gated Schottky-Barrier Field Effect Transistor (SBFET). If equipped with memory capabilities, this architecture could represent an ideal building block for versatile LiM hardware. To reach this goal, here we investigate the integration of a ferroelectric Hf$_{0.5}$Zr$_{0.5}$O$_2$ (HZO) layer onto Dual Top Gated SBFETs. We demonstrate that HZO polarization charges can be successfully employed to tune the height of the two Schottky barriers, influencing the injection behavior, thus defining the transistor mode, switching it between n and p-type transport. The modulation strength is strongly dependent on the polarization pulse height, allowing for the selection of multiple current levels. All these achievable states can be well retained over time, thanks to the HZO stability. The presented result show how ferroelectric-enhanced SBFETs are promising for the realization of novel LiM hardware, enabling low-power circuits for ANNs execution.

Reconfigurable logic-in-memory circuits with ferroelectric nanosheet field-effect transistors

Computing-In-Memory Using Ferroelectrics: From Single- to Multi-Input Logic

Ferroelectric FETs-Based Nonvolatile Logic-in-Memory Circuits

Reconfigurable Multifunctional van der Waals Ferroelectric Devices and Logic Circuits

Reconfigurable aJ-Level Ferroelectric Transistor-Based Boolean Logic for Logic-in-Memory

Exploiting Ferroelectric FETs for Low-Power Non-Volatile Logic-in-memory Circuits

Ferroelectric Field-Effect Transistors for Logic and In-Situ Memory Applications

Reconfigurable Quasi-Nonvolatile Memory/Subthermionic FET Functions in Ferroelectric-2D Semiconductor vdW Architectures

A Polarization-Switching, Charge-Trapping, Modulated Arithmetic Logic Unit for In-Memory Computing Based on Ferroelectric Fin Field-Effect Transistors

A Ferroelectric Reconfigurable Memory with Excitatory and Inhibitory Synaptic Responses

Nonvolatile Logic Inverters Based on 2D CuInP2S6 Ferroelectric Field Effect Transistors

Integration of Neuromorphic and Reconfigurable Logic-in-Memory Operations in an Electrolyte-Manipulated Ferroelectric Organic Neuristor.

Logic Compatible High-Performance Ferroelectric Transistor Memory

Ferroelectrically-enhanced Schottky barrier transistors for Logic-in-Memory applications

Ferroelectric‐Nanocrack Switches for Memory and Complementary Logic with Zero Off‐current and Low Operating Voltage

Ferroelectric Memory Based on Nanostructures

Logic and in-memory computing achieved in a single ferroelectric semiconductor transistor

A Reconfigurable Ferroelectric Transistor as An Ultra‐Scaled Cell for Low‐Power In‐Memory Data Processing

Experimental Demonstration of Non-volatile Boolean Logic with Field Configurable 1Fefet-1Rram Technology

Reconfigurable SWCNT ferroelectric field-effect transistor arrays