New Nanoscale Memory Cell for Minimizing the Size of Superconducting Computers

From the 1950s, considerable research and development initiatives have aimed to design a superconducting computer with the ability to perform high-speed computations without heat dissipation.

Professor Alexey Bezryadin works in the lab with Graduate Student Andrew Murphy, at the Loomis Laboratory of Physics in Urbana. CREDIT: Siv Schwink, Physics Illinois.

Besides requiring only a fraction of the energy consumed by the existing supercomputers, such a superconducting computer will be much faster and highly powerful. Although positive developments have been observed in this area for the past 65 years, there are still considerable hurdles to cross, such as developing miniaturized low-dissipation memory.

A research team from the University of Illinois at Urbana-Champaign has synthesized an innovative nanoscale memory cell that can be successfully incorporated into superconducting processors. Alexey Bezryadin, Professor of Physics and Andrew Murphy, Graduate Student, work in cooperation with Professor of Theoretical Physics Dmitri Averin from the State University of New York at Stony Brook to develop the new technology, which yields a stable memory whose size is small when compared to other intended memory devices.

The new device includes two superconducting nanowires that are connected to two unevenly spaced electrodes “written” by means of electron-beam lithography. A closed, asymmetric superconducting loop—known as a nanowire superconducting quantum interference device (SQUID)—is formed by electrodes and the nanowires. The direction in which current flows, clockwise or counterclockwise, through the loop corresponds to the “0” or “1” of binary code.

An oscillating current of specific magnitude at a particular magnetic field is applied for writing the memory state. For reading the memory state, the current is ramped up and the value of current at which superconductivity is lost is noted. It can be observed that such a loss of superconductivity or critical current varies for the two memory states, namely, “0” or “1.” In order to investigate memory stability, the Researchers delayed reading of the memory state, and discovered that there were no occurrences of memory loss. The Researchers employed a technique known as molecular templating to carry out the experiments on two nanowire SQUIDS formed of the superconductor Mo75Ge25. The outcomes of the research were published in New Journal of Physics on 13th June 2017.

This is very exciting. Such superconducting memory cells can be scaled down in size to the range of few tens of nanometers, and are not subject to the same performance issues as other proposed solutions.

Alexey Bezryadin, Professor of Physics, University of Illinois

Murphy further stated that “Other efforts to create a scaled-down superconducting memory cell weren’t able to reach the scale we have. A superconducting memory device needs to be cheaper to manufacture than standard memory now, and it needs to be dense, small, and fast.”

To date, working of the highly propitious supercomputing memory devices known as ‘single-flux quanta’ devices is dependent on manipulating circuits that comprise of Josephson junctions and inductive elements. The size of these devices is of the order of micrometers, and further miniaturization of the devices is restricted due to the size of the integrated Josephson junctions and their geometric inductances. For encoding information, a few of these devices even mandate ferromagnetic barriers. In contrast, the device developed by Bezryadin and Murphy does not need any ferromagnetic components and avoids magnetic-field cross-talk.

Because the kinetic inductance increases with decreasing cross-sectional dimensions of the wire, nanowire SQUID memory elements could be reduced further, into the range of tens of nanometers.

Alexey Bezryadin, Professor of Physics, University of Illinois

According to the research team, the new device causes a very low dissipation of energy during operation when the energies of two binary states are almost or wholly equal. The theoretical model for similar operations was created in partnership with Averin. The switching between the equal energy states can be accomplished through quantum tunneling or by performing adiabatic processes that involve multitude of jumps between the states.

In the future, Bezryadin aims to focus on measuring the switching time and analyzing greater arrays of the nanowire SQUIDs operating as arrays of memory elements. In addition, the Researchers will investigate the superconductors at higher critical temperatures to synthesize a memory circuit that can function at a temperature of 4 K. High-speed operations can be accomplished by making use of microwave pulses.

The National Science Foundation, Division of Electrical, Communications and Cyber Systems supported this research.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback