Current-induced crystallisation in Heusler alloy films for memory potentiation in neuromorphic computation

Abstract

The current information technology has been developed based on von Neumann type computation. In order to sustain the rate of development, it is essential to investigate alternative technologies. In a next-generation computation, an important feature is memory potentiation, which has been overlooked to date. In this study, potentiation functionality is demonstrated in a giant magnetoresistive (GMR) junction consisting of a half-metallic Heusler alloy which can be a candidate of an artificial synapse while still achieving a low resistance-area product for low power consumption. Here the Heusler alloy films are grown on a (110) surface to promote layer-by-layer growth to reduce their crystallisation energy, which is comparable with Joule heating induced by a controlled current introduction. The current-induced crystallisation leads to the reduction in the corresponding resistivity, which acts as memory potentiation for an artificial GMR synaptic junction.

Wireless Multiferroic Memristor with Coupled Giant Impedance and Artificial Synapse Application

“…The Internet of Things (IoT) [1] utilizes numerous ultralow-cost miniaturized devices called IoT nodes, which is developing as a massive ecosystem of technologies that enables ubiquitous information exchange between the physical world and IoT. It is have been studied according to physical mechanisms, such as phase change, [5] ferroelectric tunneling, [6][7][8] giant magnetoresistance (GMR), [9] and redox-based resistive switching, [10] etc. Among these candidates, the spintronic devices [11,12] (e.g., GMR and magnetic tunneling junction (MTJ) stacks) are the front runners with high technology maturity, whose resistance is determined by the relative directions of magnetizations between fixed and free layers.…”

“…Since the conduction of memristor varies depending on the history of electrical stimuli, it can be utilized to mimic the plasticity of synapses (i.e., synaptic weight) and modulate the strength of connection between neighboring neurons. Recently various memristor‐based synapse devices have been studied according to physical mechanisms, such as phase change, [ 5 ] ferroelectric tunneling, [ 6–8 ] giant magnetoresistance (GMR), [ 9 ] and redox‐based resistive switching, [ 10 ] etc.…”

“…However, the power consumption and computation speed of IoT nodes are still hindered by the von Neumann computing architecture with physically separated memory and processing units. Furthermore the current IoT nodes [1][2][3] usually implement the separated wireless communication modules (e.g., Bluetooth and RFID) to transmit and receive information of the memory and computation units, which impede the integration of chip and further increase the power consumption due to the large amount of data transferred among different units.In order to break the bottleneck of von Neumann computing architecture, various memristors [4][5][6][7][8][9][10][11][12] have been studied to construct the energy and computation speed efficient computing system (e.g., neuromorphic computing system) by integrating the memory and computation functions in a single device. Specifically the neuromorphic computing system is inspired by the human brain, which enables parallel signal storage and processing with much lower energy consumption.…”

“…In order to break the bottleneck of von Neumann computing architecture, various memristors [4][5][6][7][8][9][10][11][12] have been studied to construct the energy and computation speed efficient computing system (e.g., neuromorphic computing system) by integrating the memory and computation functions in a single device. Specifically the neuromorphic computing system is inspired by the human brain, which enables parallel signal storage and processing with much lower energy consumption.…”