Magnetic tunnel junction and neuron nonlinear response device

Abstract

The invention discloses a magnetic tunnel junction, which comprises a top electrode, a bottom electrode and a tunnel junction, and is characterized in that the bottom electrode is a heavy metal layer with spin Hall angles at two ports; the tunnel junction comprises a magnetic free layer, an insulating barrier layer, a magnetic pinning layer, an antiferromagnetic coupling spacer layer, an antiferromagnetic coupling magnetic layer and an antiferromagnetic layer which are sequentially arranged on the bottom electrode; wherein the magnetic free layer, the magnetic pinning layer and the antiferromagnetic coupling magnetic layer all have in-plane magnetic moments, easy axes of the magnetic free layer, the magnetic pinning layer and the antiferromagnetic coupling magnetic layer are along the width direction of the bottom electrode, the magnetic moment directions of the magnetic pinning layer and the antiferromagnetic coupling magnetic layer are opposite, and the antiferromagnetic sequence direction of the antiferromagnetic layer is the easy axis direction of the antiferromagnetic coupling magnetic layer. The invention also provides a neuron nonlinear response device comprising the magnetic tunnel junction. According to the magnetic tunnel junction and the corresponding nonlinear response device provided by the invention, the nonlinear response characteristic of the device is excellent, the energy consumption is low, and the stability is high.

Description

technical field [0001] The invention belongs to the technical field of magnetic elements, and furthermore, relates to a magnetic tunnel junction and a neuron nonlinear response device including the magnetic tunnel junction. Background technique [0002] The von Neumann system adopted by modern computer systems is a system that separates operation and storage. The speed of data transmission between the computing core (cpu) and the cache area severely restricts the increase in the overall computing rate, and the cache area will also bring The substantial increase in energy consumption has limited the improvement of computing efficiency. The system of separating operation and storage has been increasingly unable to meet people's needs for high-performance computing, especially the realization of recognition functions. [0003] After the brain-like artificial neural network was proposed, it has been regarded as a strong competitor of the von Neumann architecture in recent years...

AI Technical Summary

Problems solved by technology

[0004] At present, the artificial neuron nonlinear response system mainly uses a system composed of CMOS technology or directly uses a traditional computer for simulation. The traditional switching transistor is still used as the core component of the operation. The nonlinear response system formed is very complicated and cannot be separated from the high energy consumption. Features of low efficiency

This limits the development of artificial neural networks in the direction of greater integration, high efficiency, and small size

Images