Atomically thin magnetic device could lead to new memory technologies

Enabling data storage at a greater density.

A depiction of the crystal structure of chromium triiodide (CrI3), with chromium atoms shown in purple and iodine atoms in yellow. The black arrows represent the electron “spins,” which are analogous to tiny bar magnets.Tiancheng Song
A depiction of the crystal structure of chromium triiodide (CrI3), with chromium atoms shown in purple and iodine atoms in yellow. The black arrows represent the electron “spins,” which are analogous to tiny bar magnets.Tiancheng Song

Magnetic materials offer a very rich versatility for modern technologies; applications cover e.g. hard-disk drives in information technology, nanoscale field sensors, anti-theft devices and renewable energy harvesting in generators. Now, scientists at the University of Washington push this technology one step forward by encoding data with the help of magnets that are just a few layers of atoms in thickness.

Scientists specifically used stacks of ultrathin materials to apply phenomenal control over the stream of electrons in view of the course of their twists — where the electron “turns” are practically equivalent to modest, subatomic magnets. The materials that they utilized includes sheets of chromium tri-iodide (CrI3), a material portrayed in 2017 as the main ever 2-D attractive protector. Four sheets — each only atoms thick — made the most slender framework yet that can piece electrons in light of their twists while applying in excess of 10 times more grounded control than different techniques.

According to scientists, this breakthrough may revolutionize both distributed computing advancements and consumer electronics by empowering information storage at a more prominent density and enhanced energy proficiency.

Co-lead author Tiancheng Song, a UW doctoral student in physics said, “Our work reveals the possibility to push information storage based on magnetic technologies to the atomically thin limit.”

In the experiment, the researchers sandwiched two atomic layers of CrI3 between graphene contacts and measured the electron flow through the CrI3.Tiancheng Song
In the experiment, the researchers sandwiched two atomic layers of CrI3 between graphene contacts and measured the electron flow through the CrI3.Tiancheng Song

Researchers sandwiched two layers of CrI3 between conducting sheets of graphene. They showed that, depending on how the spins are aligned between each of the CrI­­3 sheets, the electrons can either flow unimpeded between the two graphene sheets or were largely blocked from flowing. These two different configurations could act as the bits — the zeroes and ones of binary code in everyday computing — to encode information.

With up to four layers of CrI3, the group found the potential for “multi-bit” data stockpiling. In two layers of CrI3, the twists between each layer are either adjusted a similar way or opposite directions, prompting two unique rates that the electrons can course through the attractive door. Be that as it may, with three and four layers, there are more mixes for turns between each layer, prompting numerous, unmistakable rates at which the electrons can course through the attractive material from one graphene sheet to the next.

Co-lead author Xinghan Cai, a UW postdoctoral researcher in physics said, “The functional units of this type of memory are magnetic tunnel junctions, or MTJ, which are magnetic ‘gates’ that can suppress or let through electrical current depending on how the spins align in the junction. Such a gate is central to realizing this type of small-scale data storage.”

Co-author Bevin Huang, a UW doctoral student in physics said, “Instead of your computer having just two choices to store a piece of data in, it can have a choice A, B, C, even D and beyond. So not only would storage devices using CrI3 junctions be more efficient, but they would intrinsically store more data.”

Corresponding author Xiaodong Xu, a UW professor of physics and of materials science and engineering said, “Although our current device requires modest magnetic fields and is only functional at low temperature, infeasible for use in current technologies, the device concept and operational principle are novel and groundbreaking. We hope that with developed electrical control of magnetism and some ingenuity, these tunnel junctions can operate with reduced or even without the need for a magnetic field at high temperature, which could be a game changer for new memory technology.”

Additional co-authors are Nathan Wilson, Kyle Seyler, Lin Zhu and David Cobden at the UW; co-corresponding author Wang Yao and Matisse Wei-Yuan Tu at the University of Hong Kong; co-corresponding author Di Xiao and Xiao-Ou Zhang at Carnegie Mellon University; Takashi Taniguchi and Kenji Watanabe at the National Institute for Materials Science in Tsukuba, Japan; and Michael McGuire at the Oak Ridge National Laboratory in Tennessee.

The study is published online May 3 in the journal Science.