FuSe: Ultra-Low-Energy Logic-in-Memory Computing using Multiferroic Spintronics

William Marsh Rice University, Houston TX

Investigators

Abstract

Non-technical Description The energy consumption of computing is a significant global challenge, as the demand for computing explodes. If current trends persist, computing will soon become the dominant energy consumer. Limits to energy production and storage will limit availability of critical applications and hinder development of new technologies. Spintronics, which uses an electron’s spin as well as its charge, offer a new paradigm for computing that can meet this challenge. This FuSe project aims to enable a new generation of energy-efficient computing devices by integrating materials research, device physics and ultimately circuit design and architectures. These devices will be based on materials with electric and magnetic properties that can be controlled by external fields, called multiferroics. The team is committed to educating the next generation semiconductor workforce. A diverse group of graduate and undergraduate students will be trained in interdisciplinary research, with a focus on those from underrepresented groups in STEM. The PIs will also develop educational materials and participate in K-12 outreach events to inspire a broad audience. Technical Description This project explores electrically driven and detected spin transport in a voltage-switchable multiferroic insulator as the foundation for ultra-low-energy logic-in-memory computing. By exploiting the correlation and non-volatility in multiferroic materials, the team aims to greatly reduce the operating voltage of computers substantially below what is achievable by today's complementary metal oxide semiconductor (CMOS) technology and enable transformative logic-in-memory computing architectures with significantly alleviated communication costs between memory and logic. The project seeks to obtain fundamental understanding and transformative innovations by probing multiferroic materials and devices at unprecedented dimension, time, and energy scales. The team aims to address engineering challenges by integrating bottom-up research on materials synthesis, fabrication, and junction physics, and top-down from systems and circuit requirements. The project involves significant efforts to develop advanced characterization techniques including optical spectroscopy, electron microscopy, and magnetotransport for multiferroic materials and heterostructures. In addition, the project develops a circuit simulation framework and reference circuit designs, to realistically evaluate multiferroic spintronics at the system level and facilitate top-down research. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.