Related Experiment Video
Updated: Nov 5, 2025
10:40
A Fabrication and Measurement Method for a Flexible Ferroelectric Element Based on Van Der Waals Heteroepitaxy
Published on: April 8, 2018
8.4K
Flexible Ta/TiO/TaO/Ru memristive synaptic devices on polyimide substrates
Jiacheng Li1, Chenyang Hao1, Shuqin Guo1
1Tianjin Key Laboratory of Film Electronic & Communication Devices, School of Electrical and Electronic Engineering, Tianjin University of Technology, No. 391, Bin Shui Xi Dao Road, Xiqing District, Tianjin, 300384, People's Republic of China.
Nanotechnology
|May 13, 2021
Summary
Researchers developed electroforming-free memristive synapses using interface engineering. These devices mimic biological synapses, showing promise for neuromorphic computing and flexible electronics.
Area of Science:
- Materials Science
- Neuroscience
- Electrical Engineering
Background:
- Memristive synapses are crucial for neuromorphic computing and simulating biological functions.
- Linear conductance modulation is key for analog memristors in neural networks.
- Wearable devices require robust and flexible synaptic components.
Purpose of the Study:
- To develop electroforming-free memristive synapses with tunable plasticity.
- To investigate the effect of interface engineering on memristive behavior.
- To assess the performance of memristive synapses in flexible electronic devices.
Main Methods:
- Fabrication of Ta/TiO/TaO/Ru memristor devices with varying TiOthickness.
- Characterization of synaptic behaviors including potentiation, depression, paired-pulse facilitation, and spike-timing dependent plasticity.
- Evaluation of device performance on polyimide substrates under bending conditions.
Main Results:
- All fabricated devices exhibited electroforming-free properties.
- Stable short-term and long-term plasticity (potentiation and depression) were mimicked at a TiOthickness of 25 nm.
- Improved linearity and symmetry in potentiation/depression with increasing TiOthickness.
- Memristive synapses on polyimide maintained performance down to a 6 mm bending radius.
Conclusions:
- Optimized interface engineering enables electroforming-free memristive synapses with excellent synaptic functions.
- The developed memristive synapses are suitable for artificial neural network applications.
- The flexible memristive devices show potential for wearable neuromorphic systems.
