An ultralow power magnetoelectric nonvolatile memory

 · Invited

Abstract


Most modern computing is built upon the von Neumann architecture composed of the data storage and data operation in a central processing unit. However, as Moore’s law made exponential gains over the decades, a memory bottleneck emerged, also known as the von Neumann bottleneck. This memory bottleneck can often degrade computing by orders of magnitude below the maximum potential of the computer especially for applications with large data sets. Hence it is of great interest to develop new energy-efficient memory technologies. Spintronic memory based on spin-transfer torque has emerged as a potential on-chip memory but it suffers from high energy dissipation (due to the requirement for a large current) and high voltages (due to the use of tunnel junctions). In contrast, a magneto-electric memory can operate with a capacitive displacement charge and potentially reach a 1-10 aJ/switching operation. Here we show magneto-electric switching of a memory element with a giant magnetoresistance (GMR) readout, operating at ~200 mV. We utilize a combination of phase detuning via isovalent substitution, thickness scaling, and conductive oxide electrode choice to scale the switching energy density to 20 µJ/cm2. To the best of our knowledge, this is among the lowest voltage and energy density demonstrated in a spintronic memory element, thus presenting an attractive pathway to ultralow-power electronics.

*U.S. Department of Energy Advanced Manufacturing Office, and the support of Intel Corp. through the FEINMAN program

–

Presenters

  • Yen-Lin Huang

    • Lawrence Berkeley National Laboratory
    • Lawrence Berkeley National Laboratory, USA
    • Department of Materials Science and Engineering, University of California, Berkeley
    • University of California, Berkeley

Authors

  • Yen-Lin Huang

    • Lawrence Berkeley National Laboratory
    • Lawrence Berkeley National Laboratory, USA
    • Department of Materials Science and Engineering, University of California, Berkeley
    • University of California, Berkeley

  • Bhagwati Prasad

    • Department of Materials Science and Engineering, UC Berkeley

  • James Steffes

    • Department of Materials Science and Engineering, University of Connecticut

  • Chia-Ching Lin

    • Intel Corp.
    • Intel Corporation
    • Intel Corp

  • Tanay Gosavi

    • Intel Corp.
    • Intel Corporation
    • Intel Corp

  • Dmitri Nikonov

    • Intel Corp.

  • Mengmeng Yang

    • University of California, Berkeley
    • Department of Physics, UC Berkeley

  • Zi Q. Qiu

    • University of California, Berkeley
    • Department of Physics, UC Berkeley

  • Jorge Íñiguez

    • Materials Research and Technology Department, Luxembourg Institute of Science and Technology

  • Bryan Huey

    • Department of Materials Science and Engineering, University of Connecticut

  • Lane Wyatt Martin

    • DMSE, University of California, Berkeley
    • Department of Materials Science and Engineering, UC Berkeley
    • Department of Materials Science and Engineering, University of California, Berkeley
    • University of California, Berkeley

  • Ian Young

    • Intel Corp.
    • Intel Corporation

  • Sasikanth Manipatruni

    • Intel Corp.
    • Intel Corporation

  • Ramamoorthy Ramesh

    • Department of Materials Science and Engineering, UC Berkeley
    • University of California, Berkeley, USA
    • University of California, Berkeley
    • Materials Science and Engineering, University of California, Berkeley
    • Department of Materials Science and Engineering, University of California, Berkeley
    • Department of Materials Science and Engineering,, University of California, Berkeley, California 94720, USA