Memristive artificial synapses based on Au–TiO2 composite thin film for neuromorphic computing

Haonan Zhu,Zhenxun Tang,Guoliang Wang,Yuan Fang,Jijie Huang,Yue Zheng

DOI: https://doi.org/10.1063/5.0149154

IF: 6.6351

2023-06-01

APL Materials

Abstract:Neuromorphic computing architecture is considered to be a highly desirable next-generation computing architecture as it simulates the way the brain processes information. The basic device supporting such an architecture is called an artificial synapse, which possesses synapse-like functionalities. Here in this work, an Au–TiO2 composite thin film (Au nanoparticles embedding into TiO2 matrix) based memristive artificial synapse has been fabricated with excellent interface-type resistive switching (RS) characteristics. The conductivity of the device can be continuously tuned by applying different sequences of pulses, which could be analogous to the weight change of synapses. Various synaptic behaviors have been emulated, such as long-term potentiation/depression, short-term/long-term memory, learning-forgetting process, and paired-pulse facilitation. Finally, an artificial neural network for hand-written digits recognition has been constructed with an accuracy level as high as ∼90%. The excellent performance of the Au–TiO2 based device demonstrates the availability of incorporating the second phase to tune RS properties and shows its potential in a memristor for artificial synapses and neuromorphic computing with enhanced performance.

materials science, multidisciplinary,physics, applied,nanoscience & nanotechnology

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to achieve efficient artificial synapses in neuromorphic computing. Specifically, the researchers have developed a memristor - based artificial synapse using a gold - titanium dioxide (Au - TiO₂) composite thin film, aiming to simulate the way the brain processes information and overcome the problems caused by the physical separation of computing and storage in the traditional von Neumann architecture. This new - type artificial synapse has the following characteristics:

1. **Excellent resistance - switching characteristics**: By applying different pulse sequences, the conductivity of the device can be continuously adjusted, which is similar to the change of synaptic weight.
2. **Simulation of multiple synaptic behaviors**: Including long - term potentiation/depression (LTP/LTD), short - term/long - term memory (STM/LTM), learning - forgetting process and paired - pulse facilitation (PPF), etc.
3. **High - performance handwritten digit recognition**: An artificial neural network was constructed using this device, achieving a handwritten digit recognition accuracy of up to approximately 90%.

Through these characteristics, the researchers have demonstrated the potential of Au - TiO₂ composite thin films in neuromorphic computing, especially in improving the performance of memristors. By introducing the metal phase to modulate the resistance - switching characteristics, a new method for constructing efficient artificial synapses is provided.