This is the html version of the file https://arxiv.org/abs/2004.12348.
Google automatically generates html versions of documents as we crawl the web.

Location via proxy:   [ UP ]   [Report a bug]   [Manage cookies]    No cookies    No scripts    No ads    No referrer    Show this form

Page 1

Spin Current Generation by a Surface Plasmon Polariton

Theodorus Jonathan Wijaya,1 Daigo Oue,2, 3 Mamoru Matsuo,3, 4 Yasutoshi Ito,5

Kelvin Elphick,6 Hironaga Uchida,5 Mitsuteru Inoue,5 and Atsufumi Hirohata6

1Department of Electrical and Electronic Engineering,

University of Tokyo, Hongo 7-3-1, Tokyo 113-8656, Japan

2Department of Physics, Imperial College London, London SW7 2BW, United Kingdom

3Kavli Institute for Theoretical Sciences, University of Chinese Academy of Sciences, Beijing 100190, P. R. China

4CAS Center for Excellence in Topological Quantum Computation,

University of Chinese Academy of Sciences, Beijing 100190, P. R. China

5Department of Electrical and Electronic Information Engineering,

Toyohashi University of Technology, Toyohashi 441-8580, Japan

6Department of Electronic Engineering, University of York, York YO10 5DD, United Kingdom

Surface plasmon polariton (SPP) is an electromagnetic wave which is tightly localised beyond

the diffraction limit at metallic surfaces. Recently, it is theoretically proposed that the angular

momentum conversion between the SPP and electrons. In this work, we have successfully measured

SPP induced spin currents, which proves the fact that the angular momenta are interconverted.

Such conversion from light to a spin current can be used as a coupler in a next generation spintronic

computing with optical data transfer or storage.

Introduction.— Spin electronics is at the verge of

becoming a mainstream technology in microelectronics.

The launch of magnetic memory production in 2018 at

major microelectronic foundries marks the adoption of

spintronics by microelectronics industry. This decisive

step has now been passed and lot of further developments

should yield to new applications of spintronic phenomena

and devices [1]. For further improvements or new direc-

tions in spintronic device applications, coupling of pho-

tonics and spintronics/nanomagnetism needs to be de-

veloped as an interdisciplinary research field in relation

with the development of optical interconnects in electron-

ics and all optical writing in storage technology.

Conventionally, photoexcitation using circularly po-

larised light has been commonly used to generate a spin

polarised electrical current mainly in a semiconductor [2].

In this method, the frequency of light needs to be compa-

rable to the semiconductor band-gap. Thus, the method

works in the limited frequency range. To exploit a wider

range of frequency of light, an alternative mechanism of

angular momentum conversion from optical transverse

spin in surface plasmon polaritons (SPPs) to conduction

electron spin has been proposed very recently [3, 4].

Surface plasmon polaritons (SPPs) are electromagnetic

waves localised at a metal/insulator interface and have

been known to be excited optically with the Otto and

Kretschman-Raether arrangements (or their mixture),

consisting of a prism placed on a metal/insulator bilayer

with and without a gap [5]. SPP can generate a trans-

versely spinning electric field at the frequency of a few

PHz according to the Drude-Lorentz model. Conduction

electrons in the metal can follow the SPP field [6–13],

creating their orbital motion and the corresponding in-

homogeneous magnetic moments in the metal [12, 13].

An earlier experimental attempt to link spintronics and

photonics has been made to utilise SPP resonance to gen-

erate a spin-polarised electrical current in Au nanoparti-

cles and films [14]. The spin current is induced by a lo-

calised SPP, which does not have an angular momentum.

Recently, by solving the diffusion equation under the in-

homogeneous moments, Oue and Matsuo have recently

predicted the generation of a spin-polarised electrical cur-

rent, proposing that angular momenta are interconverted

between the propagating SPP and the electronic system

[3, 4]. They estimated a spin current of 105 A · m−2 can

be generated in Au by SPP at 1.25 PHz. For a 20 nm

thick Au film with a laser spot size of 1mmφ the resis-

tance can be estimated as 6.1 × 10−12 Ω (resistivity of

Au: 2.44 × 10−8 Ω · m), which can be measured using an

inverse spin Hall effect.

In this study, we have experimentally demonstrated

the propagating SPP induced spin current by measuring

the inverse spin Hall effect to prove the interconvertibil-

ity between the propagating SPP and a spin current for

the first time. In order to confirm the presence of the

SPP induced spin current, the other parasitic effects in-

duced by the local heating of the laser introduction need

to be removed, such as the spin current generation from

spin caloritronics. This has been achieved by three mea-

surements; (i) reverse alignment of the inverse spin Hall

effect, (ii) s- and p-polarisation introduction and (iii) in-

cident angular dependence of the inverse spin Hall effect.

The demonstrated results can be useful for the develop-

ment of a SPP based opto-spintronic coupler as an inter-

face between a spintronic device and optical data transfer

or storage.

Experimental setup.— A high vacuum sputtering sys-

tem (PlasmaQuest, HiTUS) was used to deposit 20 nm

thick films of non-magnets, Ag, Pt and W. Both a ther-

mally oxidised Si substrate and single crystal MgO(001)

substrates were used to induce SPP at the interfaces

with the above metallic films. The base pressure was

arXiv:2004.12348v1 [cond-mat.mes-hall] 26 Apr 2020

2

5.0 × 10−7 Pa, while the Ar pressure for sputtering was

11 Pa. The films were also patterned into a Hall bar with

the width of 1.2 mm and the length of 11 mm, which was

designed to fit the laser spot as estimated below.

By placing a prism (ThorLabs, N-BK7) on the sur-

face of the Hall bar, the Kretschmann arrangement was

achieved for SPP introduction by a semiconductor laser

(Thorlabs, LDM635 with the wavelength of 635 nm and

the power of 0.4mW). As shown in FIG. 1, a linear po-

lariser and a half-wave plate (Thorlabs, SM05 mounted

zero-order 633 nm) were used to generate both s- and

p-polarisations. The spot size of the incident beam was

then controlled using a pair of an objective lens (Mitu-

toyo, 2 × /0.055/f = 200 mm M Plan) and a convex

lens (Comar, Doublet 50DQ25/f=50 mm). The mea-

sured laser spot size was 1.73 mmφ. The incident angle

θ of the beam to the prism was controlled by the sample

stand.

The generated spin current was detected using an in-

verse spin Hall effect as schematically shown in FIG. 1.

The source meter (Keithley, 2400) was used to detect the

spin Hall voltage by connecting electrical contacts at the

diagonal corners of the continuous film and the ends of

the Hall bars. The incident angle θ was adjusted to show

the largest Hall voltage. The measurements were carried

out with 1, 000 repetitions with 0.1 ms interval, provid-

ing an average value with a standard deviation. A source

current was set to be 0.0 and 0.1 mA for the Hall voltage

and resistance measurements, respectively.

FIG. 1. Schematic diagram of the spin current generation

by SPP. Angle α indicates sample rotation with respect to its

centre.

The continuous films were measured to demonstrate

the spin current generation by SPP as shown in FIG. 1.

Ag, Pt and W films were used for the inverse spin Hall

measurement with and without a prism. In order to sub-

tract any parasitic effects, e.g., spin caloritronic and pho-

tovoltaic effects induced by the laser introduction a spin

current generated by the electro-motive force Iemf was

first estimated as

Iemf = Iemf

with prism − Iemf

without prism

(1)

where Iemf

with prism and Iemf

without prism represent the

current generated by the electro-motive force with and

without the prism, respectively. To offset loss caused by

the prism, the transmittivity of 0.920 [15] and the at-

tenuation factor κ = 1.2164 × 10−8 [16] were taken into

account in estimating jemf

with prism.

Experimental results.— The spin current generated

by the electro-motive force (EMF) Iemf was measured via

a series of measurements as shown in FIGs. 2a and 2b,

the average of which was calculated as listed in TABLE I.

As clearly seen Iemf is found to be almost proportional

to ω as expected. By reversing the propagating direction

of SPP, only the magnetic field gradient can be reversed

with maintaining the thermal gradient, i.e., topological

SPP effect by spin-momentum locking [9]. This is con-

firmed by reversing the measurement configuration of the

Hall voltage. Note that the amplitude of Iemf before and

after the 180◦ rotation is different but in the same order,

which is due to the minor misalignment in the optics.

(a)

(b)

FIG. 2. Measured emf signals of chopper-like laser profile

introduction to the Ag samples: (a) the continuous film and

(b) the Hall bar. The values indicate averaged signals and

standard deviations of each block of voltages under laser in-

troduction, which is much larger than the background signals

shown as dips.

Additionally, by introducing both s- and p-

polarisations using a linear polariser and a half

wave plate, the Hall voltage was measured on the Ag

film by the spin caloritronics (i.e., spin Nernst effect by

local Joule heating) and the SPP, respectively. The spin

current generated by SPP ISPP was then estimated by

taking the difference between those generated by s- and

p-polarisations since only the p-polarised light generates

SPP.

ISPP = Iemf

p − Iemf

s.

(2)

By subtracting the measured inverse spin Hall signals

under the introduction of the s- and p- polarisations, the

3

TABLE I. List of estimated Iemf in the Ag, Pt and W continuous films. Individual histograms can be found in Supplemental

Information.

Samples Plasma frequency [PHz] Damping constant [THz]

Iemf [µA]

Iemf with 180◦ rotation [µA]

Ag

2.18 [17]

4.353 [17]

1.77 ± 0.0453

−2.65 ± 0.0601

Pt

1.244 [17]

16.73 [17]

0.619 ± 0.00783

W

0.0798 ∼ 0.967 [18]

0.0511 ± 0.00476

magnitude of the SPP induced signals were estimated as

listed in TABLE II.

TABLE II. List of estimated Iemf and ISPP in the Ag contin-

uous film and Hall bar. Individual histograms can be found

in Supplemental Information.

Ag samples

Iemf

p [µA]

Iemf

s [µA]

ISPP [µA]

Continuous film 22.7 ± 0.201 20.6 ± 0.197 2.13 ± 0.281

Hall bar

19.2 ± 0.0496 12.6 ± 0.162 6.65 ± 0.169

As a further proof of the SPP induced signals, incident

angle dependence of these voltages was also be measured

as shown in Figs. 3(a) and (b). The optimum incident

angle to maximise emf is measured to be approximately

32.5o. On the other hand, by solving the wave-number

matching condition between the laser and SPP, the op-

timum incident angle is calculated to be as follows using

the reflective indices of the prism: ˜n = n + iκ (n : 1.515

and κ = 1.2164 × 10−8) [16] and SiO2 : 1.457 [19], and

the plasma frequency of Ag: 2.18 PHz [17, 20],

θ = arcsin

( kSPP(ωlaser)

ωlasernprism/c

)

≈ 42.6◦,

(3)

where kSPP

denotes the satisfying wave-number-

matching condition between the wave-number of laser at

the frequency ωlaser and that of SPP as elaborated in

Refs. [21]. The departure may be due to several rea-

sons in our experimental setup. The additional changes

in the spot size of the laser at varied incident angles can

change the area to generate SPP and the resulting in-

verse spin Hall voltages. The fact that the rotation axis

of the mirror is not on the mirror plane can also induce

non-negligible errors.

For the MgO(001)/Ag sample, n of MgO is 1.7345

[22], which is larger than that of the prism, 1.515, re-

sulting in no SPP induced spin current to be generated.

The measured values are 0.0612 ± 0.0247 µA, which is

almost comparable with the offset of 0.0665±0.0269 µA.

These figures are almost fifty times smaller than those

of Ag samples grown on SiO2. This fact also confirms

the validity of our setup to measure SPP induced spin

currents. As listed in TABLE II, ISPP is estimated to be

2.128±0.281 and 6.650±0.169 µA for the Ag continuous

film and Hall bar, respectively. According to the theo-

retical calculation, where the SPP induced spin current

(a)

(b)

FIG. 3. Angular dependence of the EMF signals for the Ag

samples: (a) the continuous film and (b) the Hall bar.

( js) can be obtained as

js = σ0∇δµ,

(4)

where σ0 and δµ are the conductivity of the metal and

the induced spin accumulation, respectively [3]. Here δµ

under an equilibrium state can be written as

δµ =

¯hM0

m

f(ω)(2κ2λs)2

(2κ2λs)2 − 1

e2κ2x.

(5)

Here, M0 = (|E0|2e)/2mc, f(ω) = {2g(1 − ϵ)√(−ϵ)}/ϵ2

(g: Gaussian unit factor and ϵ: permittivity), and κ2: de-

cay coefficient in the metal and λs: spin diffusion length

in the metal. The upper bound of ISPP is theoretically

estimated to be 100 µA. Due to the unperfect coupling

from the laser beam to the SPP mode and the deviation

from the surface plasmon resonance, the experimentally

measured value can be lower than this upper bound.

In our experiment, the coupling rate from the laser

beam to the SPP is indeed anticipated to be lower than

100%, and the frequency of the laser is below the surface

plasmon resonance frequency. By simply compare the

ratio between the above theoretically estimated current

and the measured one, we can estimate the conversion

ratio to be 2 ∼ 7%. To date the selection rule gov-

erns the optoelectrical conversion especially in a direct

bandgap semiconductor. On the other hand, the conver-

sion efficiency in the conventional semiconductor devices

governed by the selection rule is limited to the spin polar-

isation inducible in GaAs of ∼ 40% [23]. The conversion

ratio can further be controlled by changing the size of

4

the Hall bar. A periodic array has been reported to form

a wave-guide theoretically [24] and experimentally [25],

which can be implemented in our devices to control the

spin generation to maximise the efficiency.

Summary.— In summary, we have successfully mea-

sured SPP induced spin currents unambiguously with the

conversion ratio of up to 7%, which can be further im-

proved by optimising the device dimensions. Our findings

prove the fact that angular momenta are interconverted

between the propagating SPP and the electronic system

in our setup. Such conversion from light to a spin current

can be used as a coupler in a next generation comput-

ing with optical data transfer or storage. In addition,

SPPs [21] can be tightly localised beyond the diffraction

limit at metallic surfaces and are compatible with device

miniaturisation. The band structures of SPPs, for ex-

ample, can be controlled by implementing an artificial

structure in a metal or insulator thanks to the recent de-

velopment in plasmonics [24–26]. Such phenomena can

hence revolutionise the field of optoelectronics, or rather

“opto-spintronics,” and can open a new pathway for the

spintronic devices to replace the current Si based com-

puting devices.

This work is financially supported by EPSRC-JSPS

Core-to-Core programme (EP/M02458X/1) and JST

CREST (JPMJCR17J5). T.J.W. acknowledges the Go

Global Scholarship from the University of Tokyo. D.O.

and M.M. are supported by the Priority Program of Chi-

nese Academy of Sciences, Grant No. XDB28000000.

Y.I. thanks the Toyohashi University of Technology for

their internship programme and the financial support of

the Toyoaki Scholarship Foundation.

[1] A. Hirohata, K. Yamada, Y. Nakatani, L. Prejbeanu,

B. Diény, P. Pirro, and B. Hillebrands, Review on spin-

tronics: Principles and device applications, Journal of

Magnetism and Magnetic Materials 509, 166711 (2020).

[2] A. Hirohata and J. Kim, Optically induced and detected

spin current (London, UK: Oxford Univ. Press, 2012).

[3] D. Oue and M. Matsuo, Electron spin transport driven

by surface plasmon polaritons, Physical Review B 101,

161404 (2020).

[4] D. Oue and M. Matsuo, Effects of surface plasmons

on spin currents in a thin film system, New Journal of

Physics 22, 033040 (2020).

[5] J. Sambles, G. Bradbery, and F. Yang, Optical excita-

tion of surface plasmons: an introduction, Contemporary

physics 32, 173 (1991).

[6] K. Y. Bliokh and F. Nori, Transverse spin of a surface

polariton, Physical Review A 85, 061801 (2012).

[7] F. J. Rodrıguez-Fortu˜no, G. Marino, P. Ginzburg,

D. O’Connor, A. Martınez, G. A. Wurtz, and A. V. Zay-

ats, Near-field interference for the unidirectional excita-

tion of electromagnetic guided modes, Science 340, 328

(2013).

[8] A. Canaguier-Durand and C. Genet, Transverse spinning

of a sphere in a plasmonic field, Physical Review A 89,

033841 (2014).

[9] K. Y. Bliokh, D. Smirnova, and F. Nori, Quantum spin

hall effect of light, Science 348, 1448 (2015).

[10] C. Triolo, A. Cacciola, S. Patan`e, R. Saija, S. Savasta,

and F. Nori, Spin-momentum locking in the near field of

metal nanoparticles, ACS Photonics 4, 2242 (2017).

[11] Y. Dai, M. Dabrowski, V. A. Apkarian, and H. Petek,

Ultrafast microscopy of spin-momentum-locked surface

plasmon polaritons, ACS nano 12, 6588 (2018).

[12] K. Y. Bliokh, A. Y. Bekshaev, and F. Nori, Optical mo-

mentum, spin, and angular momentum in dispersive me-

dia, Physical review letters 119, 073901 (2017).

[13] K. Y. Bliokh, A. Y. Bekshaev, and F. Nori, Optical mo-

mentum and angular momentum in complex media: from

the abraham–minkowski debate to unusual properties of

surface plasmon-polaritons, New Journal of Physics 19,

123014 (2017).

[14] K. Uchida, H. Adachi, D. Kikuchi, S. Ito, Z. Qiu,

S. Maekawa, and E. Saitoh, Generation of spin currents

by surface plasmon resonance, Nature communications 6,

1 (2015).

[15] Thorlabs, N-BK7 product details.

[16] Schott, Optical glass data sheets May 2019.

[17] M. A. Ordal, R. J. Bell, R. W. Alexander, L. L. Long,

and M. R. Querry, Optical properties of fourteen metals

in the infrared and far infrared: Al, co, cu, au, fe, pb,

mo, ni, pd, pt, ag, ti, v, and w., Applied optics 24, 4493

(1985).

[18] M. KffiiLLOVA, L. Nomerovannaya, and M. Noskov, Op-

tical properties of molybdenum single crystals, Soviet

Physics JETP 33 (1971).

[19] I. H. Malitson, Interspecimen comparison of the refrac-

tive index of fused silica, Josa 55, 1205 (1965).

[20] E. J. Zeman and G. C. Schatz, An accurate electromag-

netic theory study of surface enhancement factors for sil-

ver, gold, copper, lithium, sodium, aluminum, gallium,

indium, zinc, and cadmium, Journal of Physical Chem-

istry 91, 634 (1987).

[21] S. A. Maier, Plasmonics: fundamentals and applications

(Springer Science & Business Media, Berlin, 2007).

[22] R. E. Stephens and I. H. Malitson, Index of refraction

of magnesium oxide, J. Res. Natl. Bur. Stand. 49, 249

(1952).

[23] D. T. Pierce and F. Meier, Photoemission of spin-

polarized electrons from gaas, Physical Review B 13,

5484 (1976).

[24] Z. Gao, H. Xu, F. Gao, Y. Zhang, Y. Luo, and B. Zhang,

Surface-wave pulse routing around sharp right angles,

Physical Review Applied 9, 044019 (2018).

[25] S. I. Bozhevolnyi, J. Erland, K. Leosson, P. M. Skov-

gaard, and J. M. Hvam, Waveguiding in surface plasmon

polariton band gap structures, Physical review letters 86,

3008 (2001).

[26] M. I. Stockman, Nanofocusing of optical energy in ta-

pered plasmonic waveguides, Physical review letters 93,

137404 (2004).