Overcoming Volatility in Ion Gel via Ag Doping for Nonvolatile Memristive Switching

Area of Science:

• Materials Science
• Nanotechnology
• Neuroscience

Background:

• Ion-gel-based memory devices offer low-voltage resistance switching but suffer from volatility, limiting their use in nonvolatile memory and neuromorphic applications.
• Existing devices lack the stability and functionality required for advanced computing paradigms like artificial intelligence.

Purpose of the Study:

• To develop a novel ion-gel-based memristive device incorporating silver nanoparticles (AgNPs) for enhanced nonvolatile memory and analog data processing.
• To investigate the device's potential for both digital data storage and emulation of synaptic behaviors for neuromorphic computing.

Main Methods:

• Fabrication of an ion-gel memristive device integrated with silver nanoparticles (AgNPs).
• Characterization of resistive switching behavior, including forming-free and bipolar resistive switching (BRS) profiles.
• Evaluation of data retention at elevated temperatures and emulation of synaptic plasticity (LTP, LTD, STM-to-LTM transition).

Main Results:

• The AgNP-incorporated ion-gel device demonstrated reliable, forming-free bipolar resistive switching.
• Exceptional data retention exceeding 10^5 seconds at 85 °C was achieved, attributed to Ag+-ion pairing stability.
• Successful emulation of synaptic behaviors and high digit classification accuracy (96.4%) in learning and inference tasks.

Conclusions:

• The developed ion-gel memristive device with AgNPs overcomes the volatility limitations of traditional ion-gel devices.
• The device offers a promising platform for stable nonvolatile memory and efficient neuromorphic computing applications.
• The integration of AgNPs enhances charge storage stability and enables complex synaptic functions.