Investigation of Nanoscale Spin Dynamics with Scanning Probes

Carnegie Mellon University, Pittsburgh PA

Investigators

Abstract

"This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5)." This proposal will investigate the spin torque induced switching and ferromagnetic resonance dynamics of small patterned magnetic tunnel junction devices using scanning probe microscopy in several modalities. Scanning probes will be used to assist in the patterning of devices between 10 nm and 1 μm in size. Additionally, as previously demonstrated under NSF funding, scanning probes will be used to make electrical measurements on devices fabricated in this way. Custom probes developed for the delivery of the high currents needed for switching and for making measurements of FMR up to 10 GHz will be used and further developed. Samples for this work will be provided through collaboration with Everspin Technologies. The goal will be to quantify the role of size and shape of devices in the sub-100 nm range in determining critical current for switching and the onset of spin-torque induced resonance. These results will have application to memory as well as spin-torque oscillator devices in aggressively scaled technologies. The approach proposed in this work offers several advantages for the fabrication and characterization of spin-torque driven devices. Most importantly, testing with probes dramatically simplifies the fabrication challenges, reducing the patterning aspect ratio that is required, as well as eliminating leads and the need for planarization. Secondly, patterning with probes allows for very small features to be made with equipment of modest cost and complexity, amplifying the activities of the one graduate student who would be funded by this work. Finally, the scanned probe geometry can quickly test many devices to assess distributions and statistics, which will be central to success of spin torque driven devices both for memory and oscillator applications. Technical Merit: The technical merit of this work is that it offers the ability to assess both DC and RF interactions among patterned spin torque devices, and the spread of these values. Clearer understanding and quantification of these distributions will enable the design of dense arrays for memory without unacceptable interactions, or, conversely, arrays of oscillators with interactions sufficiently strong to facilitate the coupling of many devices together and large power outputs. This will move us along on the development of a spin torque gain medium, or ?swaser?, as coined by Luc Berger, with enormous application to RF communications. We are taking a novel approach where we attempt to couple discrete oscillating elements with magnetostatic interactions, rather than continuous films to attempt to further quantize the allowed modes of the system. Broader Impact: The broader impact of this work has two aspects. First is our strong industrial interaction. Everspin is following this work closely as it bears directly on future generations of technology they might develop. A second dimension of this broader impact of this work is the planned outreach activities that we will undertake. The PIs in this team have a proven track record working with undergraduates building artifacts. We will focus on implementing a first generation CMOS-MEMS scanned probe microscope on a chip. To do this we will enlist a small team of undergrads to design and prototype a CMOS-MEMS scanned probe microscope suitable for deployment in a classroom. While ambitious, we believe this is quite tractable for modest cost and complexity and will present an excellent learning opportunity for ECE and Physics undergraduates.