Low-cost and nanoscale non-volatile memory concept for future silicon chips

Area of Science:

• Materials Science
• Electrical Engineering
• Computer Engineering

Background:

• Non-volatile flash memories are crucial for integrated circuits, retaining data without power.
• The semiconductor industry seeks advanced non-volatile memories with enhanced performance and scalability.
• Current flash memory technologies face limitations in speed and cell size reduction.

Purpose of the Study:

• To demonstrate a novel semiconductor memory concept utilizing phase-change materials.
• To explore the potential of nanoscale phase-change lines for improved memory performance.
• To assess the feasibility of integrating this new memory into existing CMOS processes.

Main Methods:

• Fabrication of individual memory cells based on narrow lines of phase-change material.
• Programming the phase-change material between two resistive states using low-power current pulses.
• Utilizing a doped-SbTe phase-change material for enhanced performance characteristics.

Main Results:

• Reversible programming of the phase-change material between two states on a nanosecond timescale.
• Demonstrated improved speed and reduced power consumption by scaling phase-change lines to the nanometre scale.
• Confirmed the viability of the doped-SbTe material for high-performance memory applications.

Conclusions:

• The demonstrated phase-change memory concept offers a promising alternative to current non-volatile memory technologies.
• Nanoscale engineering of phase-change materials significantly enhances memory speed and energy efficiency.
• The simplicity of the proposed design suggests potential for straightforward integration into standard CMOS manufacturing flows.