Digital and Analog Resistive Switching Behavior in Si-NCs Embedded in a Si/SiO₂ Multilayer Structure for Neuromorphic...

Area of Science:

• Materials Science
• Nanotechnology
• Electrical Engineering

Background:

• Resistive-switching (RS) memristors are crucial for next-generation computing.
• Silicon nanocrystals (Si-NCs) offer potential for novel memristor applications.
• Integrating Si-NCs into complementary metal-oxide-semiconductor (CMOS) structures is key for scalable neuromorphic systems.

Purpose of the Study:

• To investigate the digital and analog resistive-switching characteristics of memristors based on silicon nanocrystals.
• To explore the potential of Si-NCs in CMOS structures for mimicking biological synapse functions.
• To evaluate the performance metrics including endurance, retention, and synaptic plasticity.

Main Methods:

• Fabrication of a memristor device using a SiO2/Si-NCs/SiO2 multilayer structure within a CMOS framework.
• Characterization of bipolar resistive-switching behavior, including SET and RESET processes.
• Evaluation of endurance (~70 DC cycles), retention time (>10^4 s), and analog synaptic plasticity (LTP/LTD) using pulse voltage stimuli.

Main Results:

• The Si-NCs memristor exhibited bipolar RS with an intermediate resistance step attributed to charge trapping.
• Achieved a high ON/OFF ratio of ~10^6 and retention time exceeding 10^4 seconds.
• Demonstrated long-term potentiation and depression (LTP/LTD) with analog characteristics, showing current modulation with pulse number and retention of states.

Conclusions:

• Si-NCs integrated into CMOS structures show promising resistive-switching properties for neuromorphic computing.
• The observed analog RS behavior and synaptic plasticity mimic biological synapse functions.
• These findings pave the way for developing advanced, brain-inspired computing architectures using Si-NCs-based memristors.