Predictive Materials Design of Magnetic Random-Access Memory Based on Nanoscale Atomic Structure and Element...

Area of Science:

• Materials Science
• Condensed Matter Physics
• Device Physics

Background:

• Magnetic tunnel junctions (MTJs) are crucial for nonvolatile memory and logic due to their electrical read/write capabilities.
• The underlying causes for key experimental properties like tunneling magnetoresistance (TMR), perpendicular magnetic anisotropy (PMA), and voltage-controlled magnetic anisotropy (VCMA) in MTJs are not fully understood.

Purpose of the Study:

• To elucidate the correlations between material properties and the performance of MTJs.
• To establish a predictive materials design framework for optimizing MTJ performance.

Main Methods:

• High-resolution transmission electron microscopy (HRTEM) and energy-dispersive X-ray spectroscopy (EDS) were employed to analyze MTJ stacks.
• Formation enthalpy and thermal diffusion analysis were used to understand element distribution.
• Experimental validation of the predictive framework was performed.

Main Results:

• MTJ crystallization quality, boron diffusion, and fixed layer oxidation correlate with TMR.
• Seed layer diffusion into the free layer/MgO interface negatively impacts interfacial PMA.
• Metal-oxide concentrations in the free layer correlate with VCMA.
• High PMA (1.74 mJ/m²) and VCMA (115 fJ/V·m⁻¹) were achieved with annealing stability above 400 °C.

Conclusions:

• A predictive materials design framework was established, linking material characteristics to MTJ performance.
• The framework facilitates the exploration of design spaces for high-performance MTJs.
• The findings provide insights into optimizing MTJ properties for advanced memory and logic applications.