Spin-Controlled Light-Emitting Diodes and Lasers

For decades, improving computers and processing information simply relied on putting more transistor in a single microchip. However, with an explosive growth of information and communication technology, involving artificial intelligence, high-performance computing, and big data, this approach faces fundamental obstacles. Modern computers rely on so-called multicore architecture with a complex network of interconnects, establishing communication between different computer parts, including their logic and memory, for processing and storing information.

The key concern in computers is their power consumption, dominated not by the transistors and information processing, but rather by interconnects and information transfer. In a simple analogy, transistors can be viewed as cars and interconnects as highways. Replacing an older car with a luxury vehicle in a traffic jam will not make much difference.

Instead, it is crucial to design better (information) highways. Lasers and optical interconnects are closely intertwined, with lasers acting as the light source for high-speed data transmission and communication (interconnects) in fiber optics and other systems. They enable faster and more efficient data transfer in areas like data centers and other applications needing high bandwidth.

Thus, this transformative research seeks to develop new principles for their ultrafast and energy-efficient operation. Specifically, instead of conventional lasers which transfer information by the corresponding changes in the intensity of the emitted light, this research proposes to take advantage of the much faster changes in the polarization of the emitted light.

With the preliminary experimental demonstration of this principle, completed by the recent breakthrough that the polarization of the emitted light is electrically controlled at room temperature in light-emitting diodes, the proposed research envisions superior performance by seamless integration of processing, transferring, and storing information.

Controlling the intensity of emitted light and charge current is the basis of transferring and processing information. In contrast, robust information storage and magnetic random-access memories are implemented using the electrons’ spins and the associated magnetization in ferromagnets. An unequal number of spins along or against the magnetization axis represents the binary information “0” and “1” which, as the magnet attached to the fridge door, can be preserved without external power.

While commercial spintronic devices rely on the change of the resistance with the spin orientation--magnetoresistance, by taking electrons out of ferromagnets the spin information is quickly lost (within nanosecond) and cannot travel far (typically, up to a micrometer). However, since the light also has spin through its circular polarization or helicity, the spin information transferred from electron to light, could be carried much faster and farther.

The proposed research builds on the breakthrough of switching the magnetization of a ferromagnet in light-emitting diodes (LEDs) at room temperature and no applied magnetic field. Through the conservation of the total angular momentum, switching the magnetization of the ferromagnet reverses the orientation of the injected spin and the corresponding helicity or handedness (left vs right) of the emitted light.

This principle is used to predict the operation of spin-controlled LEDs and lasers, with potentially transformative implications for processing, transferring, and storing information. The device modeling will combine first-principles studies, generalized rate-equation description, and microscopic optical gain calculations. This effort will be complemented by implementation of these devices, pursued by the experimental collaborators and using commercial materials for magnetic memories.

The proposed research will be supported by closely integrated educational and outreach efforts, as well as by developing resources for newcomers interested in spintronic devices beyond magnetoresistance.

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.