Multi-level control of conductive nano-filament evolution in HfO2 ReRAM by pulse-train operations

Area of Science:

• Materials Science
• Electrical Engineering
• Nanotechnology

Background:

• Resistive random-access memory (ReRAM) requires precise control of nanoscale defects for multi-level data storage.
• Vacancy nano-filaments are critical defects influencing ReRAM performance.
• Current methods for controlling filament evolution lack sufficient reliability for advanced applications.

Purpose of the Study:

• To investigate the efficacy of a pulse-train operation scheme for reliable multi-level control of conductive filament evolution in ReRAM.
• To enhance the stability and precision of resistance states in ReRAM devices.
• To provide insights into the physical mechanisms governing filament rupture.

Main Methods:

• Systematic investigation of pulse-train operation parameters.
• Application of the pulse-train scheme to Hafnium Oxide (HfO2) ReRAM devices.
• Characterization of resistance levels and their statistical variations.
• Analysis of the relationship between pulse amplitude and saturated resistance.

Main Results:

• The pulse-train scheme significantly improves the reliability of multi-level control in ReRAM.
• Relative standard deviations of resistance levels were reduced by up to 80% compared to single-pulse schemes.
• An exponential relationship was observed between saturated resistance and pulse amplitude during filament rupture.
• Evidence supporting the gap-formation model for filament rupture was established.

Conclusions:

• The pulse-train operation scheme offers a robust method for reliable multi-level control of ReRAM.
• This technique enhances data storage density and reliability by precisely managing conductive filament evolution.
• The findings contribute to a deeper understanding of filament dynamics and support the gap-formation model in ReRAM.