Multi-Objective Optimization for Rapid Identification of Novel Compound Metals for Interconnect Applications

Area of Science:

• Materials Science
• Condensed Matter Physics
• Electrical Engineering

Background:

• Interconnect materials are crucial for integrated circuits, but traditional conductors like copper face limitations at nanoscale dimensions (<10 nm).
• The poor performance of scaled-down conductors hinders the further miniaturization and scalability of logic devices.

Purpose of the Study:

• To discover novel interconnect materials capable of overcoming the limitations of current technologies.
• To identify materials that optimize bulk electronic conductivity, surface scattering time, and chemical stability for advanced integrated circuits.

Main Methods:

• A multi-objective search strategy was employed, integrating first-principles calculations to screen over 15,000 potential materials.
• Physically motivated surrogate properties were utilized to simultaneously optimize key performance metrics from accessible materials databases.

Main Results:

• The screening identified promising local interconnect candidates that could outperform ruthenium, the current standard.
• Potential semi-global interconnects with significant skin depths at GHz frequencies were also discovered.
• The material Cobalt Platinum (CoPt) was validated as a promising candidate through ab initio and experimental transport studies.

Conclusions:

• The developed multi-objective search approach effectively identifies next-generation interconnect materials.
• The identified materials, including CoPt, have the potential to replace ruthenium and copper in future local interconnect applications.
• This work paves the way for overcoming scalability limitations in advanced integrated circuits.