Ab Initio Study of Novel Phase-Change Heterostructures

Area of Science:

• Materials Science
• Computational Materials Science
• Solid State Physics

Background:

• Neuromorphic computing aims to mimic the brain's unified processing and storage.
• Phase-change materials (PCMs) are suitable for non-volatile, scalable memory but suffer from variability and drift.
• Phase-change heterostructures with confinement materials can mitigate PCM issues.

Purpose of the Study:

• To investigate the potential of TiTe2-based superlattice heterostructures for neuromorphic computing.
• To evaluate the switching behavior and stability of GeTe and Ge2Sb2Te5 PCMs within TiTe2 nanolayers.
• To identify promising PCM candidates for overcoming limitations in current neuromorphic devices.

Main Methods:

• Ab initio molecular dynamics simulations were employed.
• The study focused on superlattice heterostructures of TiTe2 with GeTe and Ge2Sb2Te5.
• Analysis of atomic diffusion and structural integrity during PCM switching was performed.

Main Results:

• Switching of PCMs within the superlattice structure is possible without structural degradation.
• Atom diffusion across TiTe2 nanolayers was prevented, maintaining heterostructure integrity.
• The Ge2Sb2Te5/TiTe2 system exhibited weak interfacial coupling and high amorphous state stability.

Conclusions:

• TiTe2-based superlattices effectively mitigate variability and drift in PCMs.
• The Ge2Sb2Te5/TiTe2 heterostructure demonstrates significant potential for stable and reliable neuromorphic computing applications.
• This work provides a pathway for developing advanced memory devices for brain-inspired computing.