1887
banner image
No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Thermal spin pumping and magnon-phonon-mediated spin-Seebeck effect
Rent:
Rent this article for
USD
10.1063/1.4716012
/content/aip/journal/jap/111/10/10.1063/1.4716012
http://aip.metastore.ingenta.com/content/aip/journal/jap/111/10/10.1063/1.4716012
View: Figures

Figures

Image of FIG. 1.
FIG. 1.

(a) A schematic illustration of the conventional Seebeck effect. When a temperature gradient is applied to a conductor, an electric field (electric voltage V) is generated along the direction. (b) A schematic illustration of the SSE. When is applied to a ferromagnet, a spin voltage is generated via magnetization () dynamics, which pumps a spin current into an attached paramagnetic metal. In the paramagnetic metal, this spin current is converted into an electric field due to the ISHE. (c) Difference between the Seebeck effect and the SSE. The SSE appears even in insulators. (d) Experimental reports on the SSEs.

Image of FIG. 2.
FIG. 2.

[(a), (b)] Schematic illustrations of the longitudinal configuration (a) and transverse configuration (b) for measuring the SSE. (c) A schematic illustration of the ISHE.

Image of FIG. 3.
FIG. 3.

A photograph (a) and a schematic illustration ((b): side view, (c): top view) of the measurement system for the longitudinal SSE experiments. The Cu plate above the sample (“upper Cu plate” in the main text) is thermally well connected to a heat bath through thick (M3) molybdenum (Mo) screws with high thermal conductivity (), while the Cu plate just below the sample (“lower Cu plate” in the main text) is thermally insulated from the heat bath by thin (M2) ceramic screws with low thermal conductivity (). Since the diameter of the tip of the tungsten (W) needles is very small (), the heat flow from the needles to the sample is negligibly small.

Image of FIG. 4.
FIG. 4.

(a) A photograph of the YIG/Pt sample in the longitudinal configuration. (b) A schematic illustration of the longitudinal SSE and the ISHE in the YIG/Pt sample. (c) dependence of V in the single-crystalline YIG/Pt sample at , measured when was applied along the (upward) or (downward) direction. The magnetic field was applied along the x direction () or the y direction (). (d) H dependence of V in the single-crystalline YIG/Pt sample for various values of at , measured when was along the direction. (e) dependence of V in the polycrystalline YIG/Pt sample at and , measured when was along the direction. The inset to (e) shows the H dependence of V in the polycrystalline YIG/Pt sample at . All the data shown in this figure were measured at room temperature.

Image of FIG. 5.
FIG. 5.

T dependence of in the single-crystalline and polycrystalline YIG/Pt samples at , measured when and were applied along the and x directions, respectively. denotes the ISHE voltage induced by the longitudinal SSE per unit temperature difference: . The vertical axis is normalized by at 290 K.

Image of FIG. 6.
FIG. 6.

[(a), (b)] A photograph (a) and a schematic illustration (b) of the measurement system for the transverse SSE experiments. (c) A temperature image and profile along the x direction of a La:YIG/Pt sample for , measured with an infrared camera. Temperatures of the metallic films cannot be measured due to very low infrared emittance. (d) dependence of the thermoelectric voltage in a film placed on a sapphire substrate for , where denotes the distance between the electrodes attached to the film. The position of one electrode was fixed at the lower-temperature end of the film.

Image of FIG. 7.
FIG. 7.

(a) A photograph of the La:YIG/Pt sample in the transverse configuration. The experimental data in this figure were measured for a La:YIG film with two Pt wires, although the sample in this photograph has ten wires. (b) A schematic illustration of the transverse SSE and the ISHE in the La:YIG/Pt sample. (c) dependence of V in the La:YIG/Pt sample at , measured when the Pt wires were attached near the lower temperature (300 K) and higher temperature () ends of the La:YIG layer. (d) H dependence of V in the La:YIG/Pt sample for various values of . (e) H dependence of V in the La:YIG/Pt sample at , measured when was applied at an angle to the x direction. (f) H dependence of V in a La:YIG/Cu sample at , measured when was along the x direction. The experimental data shown in (e) and (f) were measured at the higher-temperature end of the sample.

Image of FIG. 8.
FIG. 8.

(a) Dependence of on , the displacement of the Pt wire from the center of the La:YIG layer along the x direction, in the La:YIG/Pt sample for various values of T at (solid circles). The temperatures of the lower and higher temperature ends of the sample were stabilized to T and , respectively. The solid curves are the fitting results using a hyperbolic sine function , where and are adjustable parameters. (b) T dependence of λ. (c) T dependence of at , measured when the Pt wires were placed at , , and .

Image of FIG. 9.
FIG. 9.

(a) A photograph of the /Pt sample in the transverse configuration. (b) A schematic illustration of the measurement setup for the transverse SSE in the /Pt sample. (c) dependence of V in the /Pt sample at , measured when the Pt wire was attached to the lower temperature (300 K) and higher temperature () ends of the layer. (d) H dependence of V in the /Pt sample for various values of . (e) H dependence of V in a plain film for various values of . (f) dependence of V in the /Pt samples for various values of at .

Image of FIG. 10.
FIG. 10.

(a) A photograph of the sapphire/[Ni81Fe19/Pt-wire] sample for the measurement of the acoustic SSE. (b) A schematic illustration of the acoustic SSE and the ISHE in the sapphire/[Ni81Fe19/Pt-wire] sample. The double lines, bold lines, and dotted lines represent electron spin-density propagators, magnon propagators, and phonon propagators (see also Fig. 13), respectively. (c) dependence of V in the sapphire/[Ni81Fe19/Pt-wire] and sapphire/[Ni81Fe19/Cu-wire] samples at , measured when the Ni81Fe19/Pt and Ni81Fe19/Cu wires were respectively placed near the lower-temperature (300 K) and higher-temperature () ends of the sapphire substrate. (d) dependence of V in the sapphire/[Ni81Fe19-wire array] and glass/[Ni81Fe19/Pt-wire array] samples at and , where denotes the displacement of the Ni81Fe19/Pt wire from the center of the substrate along the x direction. [(e), (f)] Temperature profiles along the x direction of the sapphire/[Ni81Fe19/Pt-wire array] (e) and glass/[Ni81Fe19/Pt-wire array] (f) samples at .

Image of FIG. 11.
FIG. 11.

T dependence of in the sapphire/[Ni81Fe19/Pt-wire] sample at , measured when the Ni81Fe19/Pt wire was placed near the higher-temperature end of the sapphire substrate.

Image of FIG. 12.
FIG. 12.

Mechanism of the spin-current generation by the SSE at the F/PM interface. and denote the thermal spin-pumping current from F to PM proportional to an effective magnon temperature in F and the Johnson-Nyquist spin-current noise from PM to F proportional to an effective electron temperature in the PM, respectively. The dimension of and is Joule, i.e., the flow of per unit time.

Image of FIG. 13.
FIG. 13.

Feynman diagrams representing the dominant contributions to the longitudinal and transverse SSEs in the samples used in the present study.5,7,8,10,15 The double lines, bold lines, and dotted lines represent electron spin-density propagators, magnon propagators, and phonon propagators, respectively.

Loading

Article metrics loading...

/content/aip/journal/jap/111/10/10.1063/1.4716012
2012-05-17
2014-04-16
Loading

Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Thermal spin pumping and magnon-phonon-mediated spin-Seebeck effect
http://aip.metastore.ingenta.com/content/aip/journal/jap/111/10/10.1063/1.4716012
10.1063/1.4716012
SEARCH_EXPAND_ITEM