Next-Generation Computing Using Spin-Based and 2D Materials
Abstract: We are at a time where the electronics industry is feeling pressure from two sides. On the small scale we are facing the limits of silicon. On the large scale, new applications such as big-data analytics and the internet of things present new challenges to the traditional computing paradigm. The future of computing will require us to look to new materials and physics for both energy-efficient electronics and data-driven, application-specific designs. We already see photonics making an impact in data centers. The next revolution may be in magnetics, and integration of magnetics with ever-present electronics. Spin-based materials such as ferromagnets are a promising candidate for future electronics, due to their low voltage operation, nonvolatility, and low fabrication temperature, which can open up new energy-efficient, normally-off, memory-in-computing architectures. Additionally, the emerging class of 2D materials, such as graphene and transition metal dichalcogenides (TMDs), are amenable to scaling and have new physics we can use. Thus, magnetic and 2D materials are very important classes of materials to explore for beyond-CMOS devices and systems.
In this talk, I will show how we can design magnetic-based switches to be practical logic devices. I will present my results on device design and fabrication, and I will show the devices satisfy the requirements for energy-efficient logic. I show a single device acts as an inverter and potentially as a functionally-complete NAND. We can propagate bits between the spin switches to build up circuits, showing they are cascadable. I will also show results on controlling the spin and valley Hall effect in monolayer WSe2 TMD transistors, which could be used for future 2D-magnetic devices that combine the benefits of both material classes. I will discuss the future directions of this work, including building energy-efficient systems by 3D monolithically integrating magnetic devices. I will also look beyond classical computing to how we can apply magnetics to quantum computing.
More about the Speaker: Jean Anne C. Incorvia is a postdoctoral research fellow at Stanford University in electrical engineering and a visiting scholar at UC Berkeley EECS. She is the 2017 IEEE Magnetics Society representative to the IEEE Nanotechnology Council. She received her Ph.D. in physics from Harvard University in 2015, cross-registered at MIT, where she was a Department of Energy Graduate Student Fellow. She received her bachelor's in physics from UC Berkeley in 2008. Her research focuses on emerging materials and devices for nanoelectronics. She has presented at over 25 conferences and meetings around the world, including the International Electron Devices Meeting.
This lecture is part of the Young Investigators Lecture Series sponsored by the Caltech Division of Engineering & Applied Science.