Imaging Magnetization Structure and Dynamics in Ultrathin ${\mathrm{Y}}_{3}{\mathrm{Fe}}_{5}{\mathrm{O}}_{12}/\mathrm{Pt}$ Bilayers with High Sensitivity Using the Time-Resolved Longitudinal Spin Seebeck Effect

Imaging Magnetization Structure and Dynamics in Ultrathin $Y_{3} {Fe}_{5} O_{12} / Pt$ Bilayers with High Sensitivity Using the Time-Resolved Longitudinal Spin Seebeck Effect

Jason M. Bartell, Colin L. Jermain, Sriharsha V. Aradhya, Jack T. Brangham, Fengyuan Yang, Daniel C. Ralph, and Gregory D. Fuchs

Phys. Rev. Applied 7, 044004 – Published 6 April 2017

Abstract

We demonstrate an instrument for time-resolved magnetic imaging that is highly sensitive to the in-plane magnetization state and dynamics of thin-film bilayers of yttrium iron garnet ${[Y}_{3} {Fe}_{5} O_{12} (YIG)] / Pt$ : the time-resolved longitudinal spin Seebeck (TRLSSE) effect microscope. We detect the local in-plane magnetic orientation within the YIG by focusing a picosecond laser to generate thermally driven spin current from the YIG into the Pt by the spin Seebeck effect and then use the inverse spin Hall effect in the Pt to transduce this spin current to an output voltage. To establish the time resolution of TRLSSE, we show that pulsed optical heating of patterned YIG $(20 nm) / Pt (6 nm) / Ru (2 nm)$ wires generates a magnetization-dependent voltage pulse of less than 100 ps. We demonstrate TRLSSE microscopy to image both static magnetic structure and gigahertz-frequency magnetic resonance dynamics with submicron spatial resolution and a sensitivity to magnetic orientation below $0.3 ° / \sqrt{H z}$ in ultrathin YIG.

Received 17 January 2017

DOI:https://doi.org/10.1103/PhysRevApplied.7.044004

Physics Subject Headings (PhySH)

Inverse spin Hall effect Methods in magnetism Spin Seebeck effect Spintronics

Ferromagnetic resonance Imaging & optical processing Optical microscopy Thermal techniques

Condensed Matter, Materials & Applied Physics

Authors & Affiliations

Jason M. Bartell¹, Colin L. Jermain¹, Sriharsha V. Aradhya¹, Jack T. Brangham², Fengyuan Yang², Daniel C. Ralph^1,3, and Gregory D. Fuchs¹

¹Cornell University, Ithaca, New York 14853, USA
²Department of Physics, The Ohio State University, Columbus, Ohio 43016, USA
³Kavli Institute at Cornell for Nanoscale Science, Ithaca, New York 14853, USA

Article Text (Subscription Required)

Click to Expand

Supplemental Material (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 7, Iss. 4 — April 2017

Subject Areas

Reuse & Permissions

Access Options

Article Available via CHORUS

Download Accepted Manuscript

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
Schematic of our TRLSSE measurement. A 780-nm 3-ps pulsed laser focused to a $0.606 − μ m − diameter$ spot is used to heat a YIG $(20 nm) / Pt (6 nm) / Ru (2 nm)$ film. The heating from the laser creates a temperature gradient $\nabla T_{z}$ . The pulsed heating drives a pulsed magnon flux $J_{s}$ from the YIG into the Pt where it is transduced into a pulsed voltage via the ISHE.
Reuse & Permissions
Figure 2
(a) Schematic of the LSSE detection circuit used for time-resolved voltage measurements. (b) Time-domain measurement of the LSSE-generated voltage in the $2 − μ m − wide$ wire. The time-varying LSSE signal is measured by electrically mixing the pulsed-laser-generated voltage with a 100-ps voltage pulse from the AWG. Comparing measurements of the YIG at $+ 414 Oe$ (filled blue circles) and $- 414 Oe$ (open orange circles) shows that the signal depends on the orientation of the magnetic moment. Here the dc level noise is removed. The data are acquired with a lock-in time constant of 500 ms and integration time of 2 s per point. (c) The solid blue circles show the difference between the two curves in (b). The orange line is a model normalized by the data amplitude of the signal determined by numerically convolving the calculated thermal gradient with the measured mixing pulse. (d) Difference signal of the temperature-dependent voltage $V_{J}$ measured using $+ / - 0.5 mA$ and a 600-ps mixing pulse. In (b)–(d), we report the voltage as detected at the lock-in after passing through the rf mixer, not the LSSE signal at the sample itself.
Reuse & Permissions
Figure 3
Stroboscopic detection of ferromagnetic resonance. (a) Schematic of measurement circuit for detection of magnetization dynamics in the $2 − μ m − wide$ wire. (b) TRLSSE-detected FMR for 4.1-GHz (blue closed circles) and 4.9-GHz (orange open circles) excitation. The solid lines are a fit to the data using a modified Lorentzian function. (c) Demonstration of stroboscopic FMR detection in which we measure the response of the YIG driven at phases that differ by 180°. The data are acquired with a lock-in time constant of 1 s and integration time of 5 s per point.
Reuse & Permissions
Figure 4
Measurement of YIG magnetization with LSSE measuring $V_{LSSE}$ versus external magnetic field for different laser powers and wire widths. For these curves, a dc background is subtracted. The inset shows the wire geometry. We define the signal size to be one-half of the difference in voltage when the magnetization is saturated in opposing directions. The data are acquired with a lock-in time constant of 500 ms and integration time of 2 s per point.
Reuse & Permissions
Figure 5
Images of the $4 − μ m − wide$ YIG/Pt wire. (a) Reflected light image of the YIG/Pt wire measured with a photodiode at the same time as the LSSE voltage. (b) Background-subtracted LSSE voltage at saturated magnetization and (c) remnant magnetization at 4 Oe after saturation. (d) Line cuts of the 2D scans. The normalized reflection signal is shown with black squares, blue circles represent the saturated magnetization, and the orange triangles represent the magnetization of the remnant state. Note that in the line cuts, the low-field line cut is normalized with respect to the saturation magnetization. The right side of the figure represents the raw images of the $4 − μ m$ wire at different fields around the resonance: (e) 896 Oe, (f) 1007 Oe, (g) 1025 Oe. Images (e)–(g) share the same color scale. Line cuts of the images are shown in (h); black squares, blue circles, and orange triangles correspond to the boxed regions of (e), (f), and (g), respectively. For (e)–(g), the data are acquired with a lock-in time constant of 200 ms and an integration time of 2 s.
Reuse & Permissions
Figure 6
Spatial maps of FMR fitting parameters for the $4 − μ m − wide$ wire. (a) Traces are the pixel values of three points on the sample as a function of magnetic field. (b)–(f) Spatial maps of the FMR fitting parameters made by fitting of the FMR curves at each pixel in the sequence of images measured with LSSE. Before fitting, we correct for image-to-image offset and use a $3 \times 3$ pixel moving average to smooth the data. (b) Resonance field: the symbols mark the pixels corresponding to the FMR spectra shown in (a). (c) Resonance amplitude, (d) resonance phase, (e) resonance linewidth (f) offset used in the fit.
Reuse & Permissions

Physical Review Applied

Imaging Magnetization Structure and Dynamics in Ultrathin Y3Fe5O12/Pt Bilayers with High Sensitivity Using the Time-Resolved Longitudinal Spin Seebeck Effect

Jason M. Bartell, Colin L. Jermain, Sriharsha V. Aradhya, Jack T. Brangham, Fengyuan Yang, Daniel C. Ralph, and Gregory D. Fuchs

Phys. Rev. Applied 7, 044004 – Published 6 April 2017

Abstract

Physics Subject Headings (PhySH)

Authors & Affiliations

Article Text (Subscription Required)

Supplemental Material (Subscription Required)

References (Subscription Required)

Issue

Subject Areas

Access Options

Authorization Required

Other Options

Download & Share

Images

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Log In

Search

Article Lookup

Paste a citation or DOI

Enter a citation

Imaging Magnetization Structure and Dynamics in Ultrathin $Y_{3} {Fe}_{5} O_{12} / Pt$ Bilayers with High Sensitivity Using the Time-Resolved Longitudinal Spin Seebeck Effect