An Unconventional Ultra-Sub-Wavelength Receiving Nano-Antenna Activated by ac Spin Pumping and the ac Inverse Spin Hall Effect

Erdem Topsakal; Michael Suche; Raisa Fabiha; Supriyo Bandyopadhyay

REVIEW 4 major objections 4 minor 20 references

An Unconventional Ultra-Sub-Wavelength Receiving Nano-Antenna Activated by ac Spin Pumping and the ac Inverse Spin Hall Effect

T0 review · 4 major / 4 minor · reviewed 2026-08-08 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A 160-square-micrometer array of cobalt nanomagnets on platinum strips is claimed to act as a receiving antenna at 2.4 GHz with a gain 4,000 times larger than the conventional limit for its size.

desk verdict The detection effect is likely real, but the -9 dB gain at 2.4 GHz is an artifact of a misprinted Friis formula and a frequency mismatch, so the central quantitative claim does not hold. read the letter →

arxiv 2502.07162 v1 pith:CAYA4HBU submitted 2025-02-11 cond-mat.mes-hall eess.SP

Raisa Fabiha , Michael Suche , Erdem Topsakal , Supriyo Bandyopadhyay This is my paper · ORCID

classification cond-mat.mes-halleess.SP PACS 72.25.Mk75.30.Ds85.75.-d

keywords spinpumpinginverseHalleffectwavesreceivingnano-antennasub-wavelengthantennamagnoniccrystalcobaltnanomagnetsWi-Fifrequency

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper claims that an array of 285,000 ledged cobalt nanomagnets on platinum nanostrips, covering only about 160 square micrometers, can act as a receiving antenna at 1.5 GHz and 2.4 GHz, the latter being the Wi-Fi and Bluetooth band. Incident electromagnetic radiation excites spin waves in the nanomagnets, which pump spin into the platinum and generate an alternating voltage through the ac inverse spin Hall effect. At 2.4 GHz the authors measure a receiving gain of about −9 dB, which they say is more than 4,000 times larger than the theoretical limit for a conventional antenna of the same area at the same wavelength. A control sample without nanomagnets shows only ordinary electromagnetic pickup, supporting the spintronic origin of the response. If correct, this would make ultra-miniaturized spintronic receivers feasible for on-chip and wearable communication.

What carries the argument

The central object is a magnonic-crystal receiver made of ledged 15-nm-thick cobalt nanomagnets (285,000 of them, arranged in 3,000 linear arrays) contacting 5-nm-thick, roughly 300-nm-wide platinum nanostrips that connect to two output pads. The ledged geometry leaves most of each magnet unclamped so the magnetostrictive cobalt can expand and contract with its magnetization, sustaining spin waves. The mechanism that carries the argument is resonant spin-wave excitation by the EM field, spin pumping of the precessing magnetization into the platinum, and conversion of the resulting spin current into an alternating voltage by the ac inverse spin Hall effect. The paper also identifies intrinsic spin-wave modes of the array (at 750 MHz and 2.5 GHz) as the frequency-determining elements of the output, with an extrinsic mode at the excitation frequency appearing at 1.5 GHz but not cleanly at 2.4 GHz.

What would settle it

Substitute the measured voltage ratio and the stated horn gain into the standard Friis equation $P_r/P_t = G_tG_r (\lambda/4\pi R)^2$; with the reported 6-inch data this yields a receiving gain greater than unity, which a passive antenna cannot have, indicating the formula printed in the paper omits a factor and the frequency-translated signal is being misattributed.

Watch

Extended reading notes

Core claim

The central claim is that an ultra-sub-wavelength receiving antenna can be built from a two-dimensional array of ledged cobalt nanomagnets placed on platinum nanostrips, operating through a two-step transduction: incident EM radiation excites intrinsic and extrinsic spin-wave modes in the nanomagnets, and these modes pump spin into the platinum, where the ac inverse spin Hall effect converts the injected spin into a polychromatic alternating voltage. The authors demonstrate this at 1.5 GHz and 2.4 GHz and report that the output contains intrinsic modes at 750 MHz and 2.5 GHz that are independent of excitation frequency and of transmitter-to-sample separation. For 2.4 GHz excitation, the dominant output component is the 750 MHz intrinsic mode rather than a 2.4 GHz component. Using the Friis transmission formula, they calculate a receiving gain of about −9 dB, which exceeds by roughly 4,000 times the theoretical limit they quote for a conventional antenna of the same area and wavelength. They conclude that the unconventional operating principle—spin-wave excitation, spin pumping, and the ac inverse spin Hall effect—bypasses the size constraint of ordinary antennas and, together with their earlier transmitting antenna on the same sample, enables a monolithic spintronic transceiver.

Load-bearing premise

The central claim assumes that the standard Friis receiving-gain formula can be evaluated at 2.4 GHz using a voltage ratio whose dominant measured output component is at 750 MHz, even though the paper's own spectrum shows no 2.4 GHz component.

Editorial extensions

If this is right

A receiving antenna with a footprint of $160\,\mu\mathrm{m}^2$ can respond to radiation whose wavelength is about 12.5 cm, shrinking the area-to-wavelength-squared ratio to below $10^{-8}$.
The same sample that transmits via the spin Hall effect can also receive via spin pumping and the inverse spin Hall effect, so a complete transceiver can be fabricated in one process flow for on-chip communication.
Because the output frequency is set by intrinsic spin-wave modes rather than by the carrier frequency, the detector can be made frequency-agile by tuning the nanomagnet dimensions and array pitch.
The measured gain exceeding the conventional small-antenna limit by more than three orders of magnitude implies that the spin-pumping transduction channel is not constrained by the usual radiation-resistance limit of electrically small antennas.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The absence of a measurable 2.4 GHz component in the output at 2.4 GHz excitation suggests the device should be characterized as a frequency-translating detector rather than a conventional receiving antenna; its 'gain at 2.4 GHz' is better defined as a conversion gain from 2.4 GHz input to 750 MHz output.
If a conversion-gain definition were adopted, the comparison with the conventional antenna limit would need to account for the bandwidth and the frequency offset, and the 4,000-fold advantage might shrink; a direct measurement of the power at exactly the carrier frequency would settle this.
The intrinsic-mode frequencies (750 MHz and 2.5 GHz) are set by nanomagnet shape and array pitch, so the receiver could be engineered to match particular channels by lithography, potentially enabling spectrum-selective detection without external filters.
A useful extension would be to measure the output power as a function of incident power to test whether the spin-pumping transduction is linear, which would inform how the device behaves as a receiver in realistic multipath environments.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Here's the short version: the device plausibly detects microwaves via spin pumping and the inverse spin Hall effect, and the control experiment is decent. But the headline claim of -9 dB receiving gain at 2.4 GHz is not supported. The output spectrum at 2.4 GHz excitation is dominated by a 750 MHz peak with a 2.5 GHz satellite, not a 2.4 GHz component. And Eq. (1) as printed is off by a factor of 4π relative to the standard Friis formula. Fixing either issue destroys the specific gain number; fixing both shows the device behaves as a frequency-converting detector, not as an antenna with gain at the carrier frequency.

What's genuinely new: an array of 285,000 ledged cobalt nanomagnets on Pt strips used as a receiver is a new configuration. The measured difference between real and control samples, especially at 100 cm where the control sees nothing, is good evidence that the voltage is magnetically mediated. The intrinsic modes at 750 MHz and 2.5 GHz that persist at both 1.5 and 2.4 GHz excitation are a plausible signature of array-defined spin-wave modes. So the qualitative detection story is credible.

The soft spots are in the gain analysis. The printed Eq. (1) has Gt Gr λ^2/(4πR^2); the standard Friis expression is Gt Gr λ^2/(16π^2R^2). That missing factor 4π alone changes the reported -8.9 dB to roughly +2 dB (and the 100 cm value to above unity). A passive reciprocal antenna cannot have gain above unity, but that is moot because the output is not at 2.4 GHz. Fig. 5 shows no resolvable 2.4 GHz component; V_out in their ratio is the 750 MHz signal. So "receiving gain at 2.4 GHz" is undefined. The authors actually acknowledge the frequency mismatch in Sec. 4.1, yet proceed with the Friis calculation anyway. The Supporting Information shows a 1.5 GHz component under 1.5 GHz excitation, so the gain there might be better defined, but the paper sells the 2.4 GHz number.

I would also probe Eq. (2), the "theoretical limit" they compare against; it is not a standard Chu-Harrington type bound, and invoking reciprocity for a nonlinear frequency-converting device is a stretch. The comparison to a conventional antenna is therefore not meaningful.

Who benefits from this? Researchers working on spintronic microwave detectors and frequency conversion will find the qualitative observation useful. As a claim of an ultra-sub-wavelength antenna exceeding fundamental limits, it is not ready. I'd send it to peer review because the experimental core is real and the concept could be rehabilitated with a careful conversion-gain analysis. But I'd expect the authors to either produce a well-defined gain at the output frequency or reframe the paper entirely as a detector.

Referee Report

4 major / 4 minor

Summary. The manuscript reports a receiving nano-antenna based on an array of 285,000 ledged cobalt nanomagnets in contact with platinum nanostrips (total area about 160 µm²). Incident 2.4 GHz and 1.5 GHz radiation is claimed to excite intrinsic and extrinsic spin waves in the nanomagnets, which pump spin into the Pt and produce a polychromatic ac voltage through the (ac) inverse spin Hall effect. The central quantitative claims are a receiving gain of approximately -9 dB at 2.4 GHz, obtained by inserting measured oscilloscope voltage ratios into the Friis-type formula in Eq. (1), and that this gain exceeds the conventional small-antenna limit of Eq. (2) by more than a factor of 4,000. A control sample without nanomagnets shows only same-frequency, in-phase pickup, while the real sample produces an output dominated by intrinsic modes at 750 MHz and 2.5 GHz (Fig. 5 and Supporting Information).

Significance. If the quantitative claims were correct, this would be a significant advance in ultra-sub-wavelength antennas at Wi-Fi and Bluetooth frequencies. The experiment is structured carefully at the qualitative level: the control-versus-real-sample comparison, the waveform/phase arguments against direct electromagnetic pickup, the two-separation check of the gain value, and the explicit FFT analysis of the output (Fig. 5 and Supporting Information) are all to the paper's credit. The qualitative observation — that an array of nanomagnets on Pt transduces incident microwave radiation into an output voltage whose spectral content lies at intrinsic mode frequencies — is plausible and potentially interesting. However, the headline quantitative claims are not supported by the reported data: Eq. (1) is missing a factor of 4π, the output voltage used in it has no resolvable 2.4 GHz component, and Eq. (2), from which the "4,000 times" comparison is derived, is dimensionally inconsistent as printed. Because the abstract and conclusion are built on these numbers, the central claim as stated does not survive scrutiny.

major comments (4)

[§4.2, Eq. (1)] Equation (1) is not the standard Friis transmission formula: the printed denominator 4πR² is smaller by a factor of 4π than the correct (4πR)² = 16π²R² found in the cited Ref. [17]. Evaluated with the corrected denominator, the reported ratios yield Gr = 1.60 (about +2.1 dB) at the 6-inch separation and Gr = 1.46 (about +1.6 dB) at 100 cm, instead of the reported -8.9 dB and -9.3 dB. The headline figure of approximately -9 dB is therefore an artifact of the expression as printed, and the comparison with the theoretical limit in Eq. (2) must be recomputed against the corrected value, which changes the claim by about 11 dB.
[§4.1–§4.2, Fig. 5, Supporting Information] The FFT in Fig. 5 shows that, at 2.4 GHz excitation, the dominant component of the real-sample output is at 750 MHz, a satellite appears at 2.5 GHz, and there is no resolvable component at 2.4 GHz; Section 4.1 itself states that the 2.4 GHz EM wave does not spawn an extrinsic mode at its own frequency, and the Supporting Information confirms that the 750 MHz and 2.5 GHz peaks are intrinsic modes that persist when the excitation frequency is changed to 1.5 GHz. Hence the voltage V_out inserted into Eq. (1) is carried by a 750 MHz component, and evaluating Eq. (1) with the 2.4 GHz wavelength λ = 12.5 cm computes a ratio of powers at two different frequencies. This quantity is a frequency-conversion detection efficiency, not a receiving gain at 2.4 GHz, and the comparison with the conventional-antenna limit of Eq. (2), which presumes linear reception at the operating frequency, is undefined. The Supporting Information's statement that this issue is "somewhat academic" does not address the fact that the abstract's central quantitative claim depends on this evaluation.
[§4.2, Eq. (2)] Equation (2) is dimensionally inconsistent as printed: the first term A/(2πλ)² is dimensionless while the second term √(A/(πλ)) has units of (length)^{1/2}, so their sum cannot be a gain. In addition, inserting the stated A = 160 µm² and λ = 12.5 cm into the printed formula gives about 2.02 × 10⁻⁵ (about -47 dB), not the quoted 3.22 × 10⁻⁵ (-45 dB). The claimed "theoretical limit" and the factor of "more than 4,000" quoted in the abstract are therefore not well-defined, and a correct, consistent limiting formula (with proper derivation or an accurate citation of Refs. [18, 19]) must be provided before the claimed violation of the limit can be assessed.
[§4.1] The paper asserts that the intrinsic modes at 750 MHz and 2.5 GHz "will be absent without the EM field," but no measurement of the real-sample output with the microwave source turned off is reported. Since the qualitative claim that the output signals the presence of incident radiation is the experimental basis of the paper, the stimulus dependence should be demonstrated with a no-excitation (dark) spectrum of the real sample; the control-sample comparison is suggestive but does not by itself establish that the real sample is silent without illumination.

minor comments (4)

[Supporting Information] The 1.5 GHz gain estimate is internally inconsistent: the Fig. 6 caption states that the input amplitude is roughly 15 times the output amplitude, while the text uses V_out²/V_in² ≈ (30)²; neither ratio, when inserted into Eq. (1), reproduces the reported Gr = 0.07.
[Figs. 5 and 7] The FFT plots in Figs. 5 and 7 do not label their ordinate axes, so the relative amplitudes of the spectral components, and in particular the absence or presence of a 2.4 GHz component at small amplitude, cannot be assessed from the figures.
[Throughout] There are numerous typographical errors, including "db" for "dB", "Y et" in the abstract, "magnitue" in Section 5, "transmiter" in Section 4.1, "Wily" (for Wiley) in Ref. [17], and "21013" (for 2013) in Ref. [13]; the manuscript should be carefully proofread.
[§2, §4] The paper should define the "ac inverse spin Hall effect" at first use: the conventional ISHE converts a dc spin current into a dc charge current, and here the spin current pumped by the precessing magnetization is time-varying, so the resulting charge current is alternating; stating this explicitly would prevent confusion.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the reported gain is a measured quantity, not a fitted or self-referential prediction; the flaws in the gain calculation are mathematical and interpretive, not circular.

full rationale

The paper's central quantitative claims are extracted from measured oscilloscope ratios (V_in/V_out) and a manufacturer-supplied transmitter gain; no parameter is fitted to the target gain and no equation has an output that equals its input by construction. The receiving functionality is supported by a control-sample comparison and by FFT spectra showing frequency components that are independent of excitation frequency and separation, so the mechanism does not rest solely on the authors' prior work. Self-citations [6], [14], and [20] provide background and context, but they are not load-bearing: the nanomagnet/control contrast and the measured spectra carry the argument. The skeptic's concerns about Equation (1) missing a 4pi factor and about applying a 2.4 GHz Friis calculation to an output dominated by a 750 MHz component are correctness and interpretation issues, not circularity. A wrong formula or a mismatch between excitation and output frequency does not mean the paper's result was defined into existence or that a fitted parameter was relabeled as a prediction. Therefore the circularity burden is low, warranting a score of 1 rather than 0 only because several self-citations are present, though none are load-bearing.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No free parameters are fitted; all quantitative inputs are measured or taken from manufacturer data. The load-bearing assumptions are about how those measurements map to antenna gain definitions, rather than about new entities.

assumptions (4)

domain assumption The voltage ratio Vout/Vin between the two oscilloscope channels can be equated to the power ratio Pr/Pt in the Friis transmission formula.
No impedance matching or load calibration is given, and the Friis equation requires delivered powers, not raw oscilloscope voltage amplitudes, especially when input and output frequencies differ.
domain assumption The conventional antenna gain limit in Eq. (2) applies to this device by reciprocity and is the correct bound for its area.
The paper uses Eq. (2) to conclude the device exceeds the limit by 4,000x, but it applies a reciprocal linear-antenna bound to a nonlinear frequency-converting detector, and the printed formula gives a different value than the common Chu-Harrington form.
domain assumption The frequency components at 750 MHz and 2.5 GHz are intrinsic spin-wave modes of the nanomagnet array and are not artifacts of electromagnetic pickup.
The control sample supports pickup exclusion at 2.4 GHz, but at 1.5 GHz no control was tested, and the interpretation of the two frequencies as intrinsic modes is based on only two excitation frequencies.
domain assumption The ledged cobalt nanomagnet geometry supports the described spin pumping and ac inverse spin Hall voltage.
The paper relies on prior work for this and does not provide a self-contained calculation of the expected signal level from the measured geometry.

how reviews work

0 comments

Cite this review

Pith. "Pith review of An Unconventional Ultra-Sub-Wavelength Receiving Nano-Antenna Activated by ac Spin Pumping and the ac Inverse Spin Hall Effect." pith.science (2026). https://pith.science/paper/CAYA4HBU

@misc{pith2026250207162,
  author       = {Pith},
  title        = {Pith review of: An Unconventional Ultra-Sub-Wavelength Receiving Nano-Antenna Activated by ac Spin Pumping and the ac Inverse Spin Hall Effect},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CAYA4HBU}},
  note         = {Machine review of arXiv:2502.07162}
}

read the original abstract

We report an extreme sub-wavelength unconventional receiving antenna. It consists of an array of nanomagnets connected to heavy metal nanostrips. Incident electromagnetic (EM) radiation generates intrinsic and extrinsic spin waves in the nanomagnets, which pump spin into the heavy metal nanostrips at their own frequencies giving rise to a polychromatic alternating voltage across the latter owing to the ac inverse spin Hall effect. This implements a receiving nano-antenna. We demonstrate its operation at two different EM wave frequencies of 1.5 GHz and 2.4 GHz - the latter being the Bluetooth and Wi-Fi frequency. We measure the receiving gain at 2.4 GHz to be approximately -9 db. The free space radiated wavelength "lambda" at 2.4 GHz is 12.5 cm while the antenna area A is merely 160 micron^2, making the ratio A/lambda^2 = 0.97x10^-8. This antenna's receiving gain should be very poor because of the tiny size. Yet the measured gain is more than 4000 times larger than the theoretical limit for a conventional antenna of this size at this wavelength because of the unconventional operating principle.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

20 extracted references · 19 canonical work pages

[17]

W. A. Stutzman and G. A. Thiele, Antenna Theory and Design, 3rd ed., (Wily, New Y ork, 2012)

work page 2012
[1]

C. W. Sandweg, Y. Kajiwara, A. V. Chumak, A. A. Serga, V. I. Vasyuchka, M. B. Jungﬂeisch, E. Saitoh and B. Hillebrands, Phys. Rev. Lett., 2011, 106, 216601

work page 2011
[2]

A. K. Ganguly and D. C. Webb, IEEE Trans. Microwave Theory T ech., 1975, 23, 998

work page 1975
[3]

A. A. Serga, A.V. Chumak and B. Hillebrands, J. Phys. D: Appl. Phys., 2010, 43, 264002

work page 2010
[4]

A. G. Gurevich and G. A. Melkov, Magnetization Oscillations and Waves (CRC Press, New Y ork, 1996)

work page 1996
[5]

V. S. L ’vov, Wave Turbulence under Parametric Excitations: Applications to Magnetics (Springer, Berlin, 1994)

work page 1994
[6]

A. De, J. L. Drobitch, S. Majumder, S, Barman, S. Bandyopadhyay and A. Barman, Nanoscale 2021, 13, 10016-10023

work page 2021
[7]

J. Cho, S. Miwa, K. Y akushiji, S. T amaru, H. Kubota, A. Fukushima, S. Fujimoto, E. T amura, C-Y Y ou, S. Yuasa and Y. Suzuki, Phys. Rev. B, 2016, 94, 184411

work page 2016

Show all 20 references

[8]

Salikhov, I

R. Salikhov, I. Ilyakov, L. Körber, A. Kákay, R. A. Gallardo, A. Ponomaryov, J-C Deinert, T. V. A. G. de Oliveira, K. Lenz, J. Fassbender, S. Bonetti, O. Hellwig, J. Lindner and S. Kovalev, Nat. Phys., 2023, 19, 529

2023
[9]

Brataas, Y

A. Brataas, Y. T serkovnyak, G. E. W. Bauer and B. I. Halperin, Phys. Rev. B, 2002, 66, 060404

2002
[10]

Weiler, J

M. Weiler, J. M. Shaw, H, T. Nembach, and T. J. Silva, Phys. Rev. Lett., 2014, 113, 157204

2014
[11]

C. Hahn, G. de Loubens, M. Viret, O. Klein, V. V. Naletov and J. Ben Y oussef, Phys. Rev. Lett., 2013, 111, 217204

2013
[12]

D. Wei, M. Obstbaum, M. Ribow, C. H. Back and G. Woltersdorf, Nat. Commun., DOI: 10.1038/ncomms4768

doi
[13]

Jiao and G

H. Jiao and G. E. W. Bauer, Phys. Rev. Lett., 21013, 110, 217602
[14]

Fabiha, P

R. Fabiha, P . K. Pal, M. Suche, A. K. Mondal, E. T opsakal, A. Barman and S. Bandyopadhyay, arXiv preprint arXiv:2408.08368

arXiv
[15]

M. A. Abeed and S. Bandyopadhyay, Adv. Mater. T echnol., 2020, 5, 1901076

2020
[16]

H. Cui, X. Y ang, L. Ni, M. Zhang, J. Liu, B. Wei, Y. Chen and J. Yuan, AIP Advances, 2021, 11, 125314

2021
[18]

R. F . Harrington, J. Res. Natl. Bur. Stand. D - Radio Propagation, 1960, 64, 1

1960
[19]

Skrivervik, J.-F

A.K. Skrivervik, J.-F . Zurcher, O. Staub and J. R. Mosig, IEEE Ant. Prop. Mag., 2002, 43, 12-27

2002
[20]

Fabiha, J

R. Fabiha, J. Lundquist, S. Majumder, E. T opsakal, A. Barman, and S. Bandyopadhyay, Adv. Sci., 2022, 9, 2104644. 10 Fabiha, et al. SUPPORTING INFORMATION Results for 1.5 GHz frequency We obtained results when the excitation frequency was 1.5 GHz instead of 2.4 GHz. These resu...

2022

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.

[1] [17]

W. A. Stutzman and G. A. Thiele, Antenna Theory and Design, 3rd ed., (Wily, New Y ork, 2012)

work page 2012

[2] [1]

C. W. Sandweg, Y. Kajiwara, A. V. Chumak, A. A. Serga, V. I. Vasyuchka, M. B. Jungﬂeisch, E. Saitoh and B. Hillebrands, Phys. Rev. Lett., 2011, 106, 216601

work page 2011

[3] [2]

A. K. Ganguly and D. C. Webb, IEEE Trans. Microwave Theory T ech., 1975, 23, 998

work page 1975

[4] [3]

A. A. Serga, A.V. Chumak and B. Hillebrands, J. Phys. D: Appl. Phys., 2010, 43, 264002

work page 2010

[5] [4]

A. G. Gurevich and G. A. Melkov, Magnetization Oscillations and Waves (CRC Press, New Y ork, 1996)

work page 1996

[6] [5]

V. S. L ’vov, Wave Turbulence under Parametric Excitations: Applications to Magnetics (Springer, Berlin, 1994)

work page 1994

[7] [6]

A. De, J. L. Drobitch, S. Majumder, S, Barman, S. Bandyopadhyay and A. Barman, Nanoscale 2021, 13, 10016-10023

work page 2021

[8] [7]

J. Cho, S. Miwa, K. Y akushiji, S. T amaru, H. Kubota, A. Fukushima, S. Fujimoto, E. T amura, C-Y Y ou, S. Yuasa and Y. Suzuki, Phys. Rev. B, 2016, 94, 184411

work page 2016