Complex electrical spiking activity in resistive switching nanostructured Au two-terminal devices

Area of Science:

• Materials Science
• Nanotechnology
• Computational Neuroscience

Background:

• Networks of nanoscale objects are gaining interest for resistive switching in neuromorphic computing.
• Nanostructured gold (Au) cluster films display non-ohmic behavior and resistive switching.

Purpose of the Study:

• To systematically characterize temporal correlations and power spectrum of spiking in nanostructured Au devices.
• To investigate how film thickness affects switching dynamics and electrical properties.

Main Methods:

• Fabrication of two-terminal devices with cluster-assembled Au films of varying thickness.
• Electrical characterization including resistance measurements and analysis of current spike temporal correlations.
• Analysis of the spiking rate power spectrum and resistance scaling behavior.

Main Results:

• Devices showed distinct switching dynamics based on initial resistance: power law for low resistance, bi-exponential for high resistance.
• Measured resistance exhibited a power law scaling behavior across analyzed frequencies.
• The temporal organization of current spikes was irregular, indicating stochastic properties.

Conclusions:

• Cluster-assembled Au films demonstrate tunable resistive switching dynamics suitable for neuromorphic applications.
• The observed stochasticity in switching behavior can be leveraged for specific neuromorphic system requirements.
• These findings highlight the potential of nanostructured gold films in next-generation computing architectures.