RES EARCH

SPINTRONICS abled widespread sensing applications in fields

ranging from condensed-matter science to ge-
Observation and control of hybrid ology and biophysics (20–22). Here we used
the sensitivity of the NV spins to microwave
spin-wave–Meissner-current transport modes magnetic fields to image the spin waves in
our YIG film (17, 18, 23, 24). When resonant
M. Borst, P. H. Vree, A. Lowther, A. Teepe, S. Kurdi, I. Bertelli, B. G. Simon, with an NV electron spin resonance (ESR) fre-
Y. M. Blanter, T. van der Sar* quency, the stray field of the spin waves drives
transitions between the NV spin states, which
Superconductors are materials with zero electrical resistivity and the ability to expel magnetic we detected through the spin-dependent NV
fields, which is known as the Meissner effect. Their dissipationless diamagnetic response is central photoluminescence under green-laser excitation.
to magnetic levitation and circuits such as quantum interference devices. In this work, we used We applied a magnetic bias field to tune the
superconducting diamagnetism to shape the magnetic environment governing the transport of NV ESR frequency into resonance with spin
spin waves—collective spin excitations in magnets that are promising on-chip signal carriers—in a waves of different spin wavelengths (Fig. 1B).
thin-film magnet. Using diamond-based magnetic imaging, we observed hybridized spin-wave– By orienting the field along one of the four
Meissner-current transport modes with strongly altered, temperature-tunable wavelengths and possible crystallographic NV orientations (Fig.
then demonstrated local control of spin-wave refraction using a focused laser. Our results demonstrate 1A), we split the ESR frequency of this “field-
the versatility of superconductor-manipulated spin-wave transport and have potential applications aligned” NV ensemble (f1) off from that of the
three other NV ensembles (f2) as shown in the

Downloaded from https://www.science.org at Columbia University on August 07, 2024

in spin-wave gratings, filters, crystals, and cavities.
optically detected resonance spectrum of Fig.

T
1C. Alternatively, we applied the field in plane
he ability to control the transport of spins teraction with Abrikosov vortices (14). Recently, along z^ to enable measurements at different
and charges with metal electrodes is fun- it was proposed to harness the diamagnetism frequencies at a given magnetic field. Because
damental to information-processing of a superconductor to create the spin-wave the applied magnetic fields are much smaller
devices and an indispensable tool in equivalents of optical mirrors and cavities (15). than the YIG saturation magnetization, the
quantum and condensed-matter physics. Being able to image and control spin waves as YIG magnetization lies predominantly in-plane
Although devices such as spin valves and tran- they travel underneath superconducting elec- along z^ for both field orientations (Damon-
sistors are based on the transport of uncorre- trodes would enable insight into the nature of Eshbach geometry).
lated particles (1), the excitations of magnetic the spin wave–superconductor interaction and To demonstrate the spin-wave modulation
materials, known as spin waves, are emerging unlock opportunities to control the propaga- capabilities of our superconductor, we imaged
as promising alternative information carriers (2). tion, dispersion, and refraction of spin waves. the spin-wave transport above and below the
These collective spin excitations provide new In this work, we developed, imaged, and molybdenum-rhenium (MoRe) superconduct-
opportunities for realizing analog or binary studied temperature-, field-, and laser-tunable ing transition temperature Tc = 8.7 K (Fig. 2, A
device functionality based on their wave nature, spin-wave transport enabled by a supercon- to D). We generated NV-resonant spin waves
nonreciprocal transport properties, and low ducting strip on a thin-film magnetic insu- with wavevector k ¼ ky ^ by applying a micro-
intrinsic damping (3). lator (Fig. 1A). We used magnetic resonance wave current at NV frequency f1 to the gold
Control of spin-wave transport is possible by imaging based on nitrogen-vacancy (NV) spins microstrip located just left outside of the im-
heavy-metal electrodes that enable modulation in diamond (16–18) to study the spin waves aging area. The interference between the mi-
by means of the spin-Hall effect (4–6) or by as they travelled underneath the optically crowave magnetic stray field generated by
auxiliary magnetic materials that modify the opaque superconductor. these spin waves and the direct microstrip
spin-wave spectrum (7, 8). However, metallic field leads to a spatial standing-wave modula-
gates can also introduce additional spin-wave Imaging hybrid spin-wave–Meissner-current tion of the NV ESR contrast (17, 18). Crucial for
damping because of uncontrolled spin pumping transport modes using spins in diamond our measurements, this interference effect en-
or spin-wave–induced eddy currents (6, 9, 10). Our system consisted of a thin film of yttrium ables a straightforward extraction of the spin
Furthermore, the diamagnetic response of iron garnet (YIG), a magnetic insulator with wavelength. The spatial map of the ESR con-
normal metals is dominated by ohmic resistance, low spin-wave damping (2), equipped with gold trast C1 at T = 10.7 K (above Tc) shows spin
precluding effective stray-field control of the microstrips for spin-wave excitation and a waves traveling toward and then underneath
spin-wave spectrum. molybdenum-rhenium superconducting strip the MoRe strip without a change in wavelength
An attractive approach for strong, low-damping for spin-wave modulation (Fig. 1A). To image (Fig. 2B). By contrast, the spin-wavelength in-
spin-wave modulation is to use superconduct- the spin waves, we placed a diamond mem- creased almost twofold when the strip was
ing electrodes. Superconductors are materials brane that contained a thin layer of NV sensor cooled into its superconducting state at T =
with zero electrical resistivity and a strong spins on top of the sample (Fig. 1A and fig. S1) 5.5 K (Fig. 2C). Averaging the maps along z^
diamagnetic response that enables the crea- (18, 19). These spins detect the spin waves by (Fig. 2E) highlights the spatial homogeneity of
tion of magnetic shields, magnetic lenses, and their microwave magnetic stray fields, which the wavelength change.
circuits such as quantum bits and quantum enables imaging through optically opaque We explained the superconductor-induced
interference devices (11, 12). Spin-wave spec- materials (10). The sample was embedded in a change of the spin wavelength by developing
troscopy measurements have demonstrated variable temperature cryostat with a base tem- an analytical expression for the spin-wave dis-
that superconducting strips on magnetic films perature of 5.5 K and free-space optical access persion in a magnet-superconductor thin-film
can alter the spin-wave spectrum through the to read out the NV sensor spins. hybrid. In this model, building on the formalism
backaction of induced currents (13) or the in- NV centers are atomic defects in the dia- developed in (15), the spin waves induce AC
mond carbon lattice with an S = 1 electron Meissner currents that are governed by the
Department of Quantum Nanoscience, Kavli Institute of spin (16). The sensitivity of the NV spin to London penetration depth lL of the super-
Nanoscience, Delft University of Technology, 2628 CJ Delft,
Netherlands. magnetic fields, combined with its optical spin conductor. These currents, in turn, generate
*Corresponding author. Email: [email protected] readout and excellent spin coherence, has en- a magnetic field that acts back on the spin waves.

Borst et al., Science 382, 430–434 (2023) 27 October 2023 1 of 5

RES EARCH | R E S E A R C H A R T I C L E

A B f2
3

f (GHz)
X 2
f1

1
10 -2 10 -1 1

0
0 10 20 30
Bz (mT)

C f1 f2
1

C2

PL / PL 0
C1

0.98

2.6 2.8 3

Downloaded from https://www.science.org at Columbia University on August 07, 2024

fsw (GHz)

Fig. 1. Magnetic resonance imaging of hybridized spin-wave–Meissner- and directional Damon-Eshbach spin-wave excitation. GGG, gadolinium gallium
current transport modes. (A) Overview of the experiment. A gold (Au) garnet, SW, spin wave. (Inset) Optical micrograph of NV diamond and Au and
microstrip excites spin waves with wavevector k ¼ k^ y in a 245-nm-thick film MoRe strips on the YIG. Scale bar, 30 mm. Au thickness, 200 nm. (B) YIG
of yttrium iron garnet (YIG). The spin waves travel toward a molybdenum dispersion (color map) and NV electron spin resonance (ESR) frequencies
rhenium (MoRe) superconducting strip (width W = 30 mm, thickness t = (red lines) as a function of the in-plane field component Bz = BNVcosq. f1 and
140 nm) where their stray fields induce Meissner currents that act back on f2 denote the ESR frequencies of NV spins with zero-field quantization axis
the spin waves, shifting their wave number to kYIG/SC < kYIG (SC, superconductor). aligned or misaligned with BNV, respectively. The intersection of the ESR
We imaged the waves underneath and next to the superconductor by their frequencies with the spin-wave dispersion sets the detectable spin-wave
microwave magnetic stray fields using a ~20-nm-thick layer of nitrogen- numbers k. (C) Optically detected NV ESR spectrum at Bz = 10 mT, at a location
vacancy (NV) spins implanted at ~0.07 mm below the bottom surface of a denoted by the red cross in the inset of (A). The ESR contrast C1 or C2,
100 × 100 × 5 mm3 diamond membrane placed on top of the sample. A where the subscript identifies the field-aligned or misaligned NV ensemble,
magnetic field BNV ¼ Bx ^x þ Bz ^z applied at q = 54° with respect to the respectively, results from interference between the microstrip field and spin-wave
x axis yields an in-plane YIG magnetization along ^z for the small fields applied field, enabling spatial mapping of the spin-wave fronts.

Fig. 2. Magnetic resonance imaging of spin waves PL 0 (10 7 s -1 )

above and below the superconducting transition A B C 1 (%)
C C 1 (%)
0 1 2 0 2 4 0 2 4
temperature. (A) Spatial map of the NV photo- T = 10.7 K T = 10.7 K T = 5.5 K
luminescence PL0 in the absence of microwaves, showing YIG YIG/MoRe YIG
the MoRe strip (between the vertical dashes). Scale
bar, 10 mm. (B and C) Spatial maps of the NV electron
spin resonance contrast C1 above (B) and below (C) the
superconducting transition temperature of Tc = 8.7 K
at T = 10.7 and 5.5 K, respectively. The Au, which
excites spin waves, is located just outside the left
edge of the imaged area. Above Tc, the wavelength is
unaffected by the MoRe strip; below Tc, it is lengthened. z z z
(D) DC resistance R of the MoRe strip as a function of
temperature T, with markers indicating the resistance y y y
of the film during the measurements of (B) triangle and
(C) square. (E) Data from (B) and (C) averaged over D E 4
F
200 +0.3 10.7 K 3
the z direction, with the MoRe strip indicated by yellow f1
5.5 K
C 1avg (%)

shading. (F) Calculated spin-wave dispersion fYIG(k)

f (GHz)
R( )

for bare YIG, and fYIG/SC(k, lL) for YIG, covered by a 2 k (1) k (1)
2 YIG/SC YIG
superconducting film with London penetration depth f YIG
lL = 400 nm. The superconductor shifts the dispersion 1
f YIG/SC
0
upwards by fSC(k, lL), which manifests as a reduction in the 0
ð1Þ ð1Þ 5 9 13 -20 0 20 40 60 0 0.1 0.2 0.3 0.4
wave number at NV frequency f1 from kYIG to kYIG=SC ,
indicated by the dashed lines. T (K)

Borst et al., Science 382, 430–434 (2023) 27 October 2023 2 of 5

RES EARCH | R E S E A R C H A R T I C L E

By integrating this field self-consistently into 1  e2h=lL ductor thickness, g = 28 GHz/T is the electron
the Landau-Lifshitz-Gilbert equation, we found fSC ðk; lL Þ ≈gm0 Ms ktr gyromagnetic ratio, m0 is the vacuum perme-
ðklL þ 1Þ  ðklL  1Þ2 e2h=lL
2
that the spin-wave dispersion shifts upwards ability, and r is a dimensionless factor asso-
in frequency as ð2Þ ciated with the YIG thickness and spin-wave
ellipticity. The approximation holds when the
fYIG=SC ðk; lL Þ ¼ fYIG ðkÞ þ fSC ðk; lL Þ ð1Þ is the superconductor-induced shift [supple- kinetic inductance dominates the impedance, as
mentary text, section S1 (19)]. Here, Ms is the is the case for our superconducting strip [sup-
where fYIG(k) is the bare-YIG spin-wave dis- YIG saturation magnetization, t = 245 nm is plementary text, section S1 (19)], and when
persion (SI) and the YIG thickness, h = 140 nm is the supercon- k2 l2L ≪ 1. A more general expression is given
in the supplementary text (19). The dispersion
shift fSC(k,lL) is maximal when lL → 0, in which
A YIG YIG/MoRe
3 C 0.35 0.35
case the superconductor perfectly screens the
fit Ms k (1)
YIG/SC
spin-wave stray field. This limit was analyzed
Bz (mT)

12 0.3 0.3
C 1 (%)
fit L k (2)
YIG/SC in (25, 26) by considering a magnetic film cov-
8
0.25 k (1)
YIG
0.25 ered by a “perfect metal” as defined by a per-
4 ) k (2) fect magnetic field screening. Indeed, when we

)
-1

-1
0.2 YIG 0.2
0 let lL → 0, the shift calculated by our model
B YIG YIG/MoRe
2.6
approached the shift predicted in (19, 25, 26).
k

k
0.15 0.15
12 The calculated bare YIG and hybridized YIG-

Downloaded from https://www.science.org at Columbia University on August 07, 2024

Bz (mT)

C 2 (%)

0.1 0.1 MoRe spin-wave dispersions are compared in

8
Fig. 2F. The upwards frequency shift under-
0.05 0.05 neath the superconductor manifested as a re-
4
0
-20 -10 0 10 20 0 10 2.4 2.6 2.8 duction in wave number of the NV-resonant spin
y Bz (mT) f (GHz) waves detected in our experiments (Fig. 2E).
We observed that the ESR contrast just to the
Fig. 3. Magnetic field dependence of the spin-wave dispersion in the magnet-superconductor hybrid. right of the MoRe strip was higher than that in
(A and B) Spatial line traces of the NV ESR contrasts C1 and C2 as a function of magnetic field Bz. Spin the MoRe-strip region (Fig. 2C), consistent with
waves excited in the bare YIG region (y < 0) by the left Au microstrip (outside the imaged area) travel toward the screening of the spin-wave field by the
and then underneath the superconducting strip (y > 0), changing their wavelength. Interference with Meissner currents. In addition, we observed that
secondary spin waves excited at the MoRe strip edge (at y = 0) due to inductive coupling between Au the ESR contrast just to the right of the MoRe
and MoRe strips yields a beating pattern along Bz for y > 0. T = 5.5 K. The drive frequency is adjusted at each strip exceeded that just to the left of it (Fig. 2E),
Bz to maintain resonance with the NV ESR transitions. (C) Spin-wave number as a function of field (left) which indicated the excitation of additional, sec-
and frequency (right), extracted from the data in (A) and (B) by means of Fourier transformation (fig. S2). ondary spin waves by the MoRe strip itself. This
The k(i) are the wave numbers measured with the field-aligned (i = 1) and misaligned (i = 2) NV ensembles. enhanced excitation of secondary spin waves is
The error bars (indicated by shading) are determined by the inverse of the spatial sampling range in the presumably caused by an additional microwave
y direction. We determined the saturation magnetization Ms by fitting the data in the bare YIG region, and the current in the MoRe strip that is excited through
London penetration depth lL by fitting the data in the YIG-MoRe region using our YIG/SC model the direct geometric inductive coupling with the
[supplementary text, section 2 (19)]. gold microstrip when the MoRe impedance
changes as it is cooled below Tc.
A C1* (a.u.) B C Temperature- and field dependence of the
10 0.3 0.8
data spin-wave dispersion and extraction of the
London penetration depth
T (K)

8 theory
0.2
0.6 We characterized the magnetic-field depen-
6 3.3 mT 4.9 mT dence of the spin-wave dispersion underneath
10 Tc the superconductor and used it to extract the
0.1 London penetration depth lL at the T = 5.5 K
0.4
T (K)

8 base temperature of our cryostat. Spatial line

0 5 10
Bz (mT)
traces of the NV ESR contrast across the strip
6 6.5 mT 9.8 mT shows the dependence of the spin wave-
0.2
0 10 20 0 10 20 6 8 10 6 7 8 length on the applied magnetic field for the
y y T (K) T (K) field-aligned (Fig. 3A) and misaligned (Fig. 3B)
NV ensembles. In both measurements, we ad-
Fig. 4. Temperature tunability of the hybrid spin-wave–Meissner-current dispersion and extraction justed the drive frequency at each magnetic
of the London penetration depth. (A) Spatial line traces of the NV ESR contrast C1 across the YIG-MoRe field to maintain resonance with the NV ESR
region as a function of temperature, showing the continuous change of the spin wavelength underneath frequency. We extracted the spin-wave numbers
the MoRe strip, for different in-plane magnetic fields Bz. The data are linearly detrended along y. Above in the bare YIG and YIG-MoRe regions sepa-
the superconducting phase transition there is no temperature dependence of the wavelength, indicating the rately by Fourier transformation (fig. S2) and
absence of the Meissner effect. A.u., arbitrary units. (B) Spin-wave numbers k extracted from data in (A) plotted these as a function of field and frequen-
and from additional data in fig. S4 as a function of temperature. The colors indicate the different magnetic cy in Fig. 3, C and D. A similar measurement
field values Bz. (C) London penetration depth lL of the MoRe film as a function of temperature, extracted with the bias field applied in plane along z^
from the data in (B) through our YIG/SC model. The black line represents the fit of the temperature shows that Meissner screening of the bias field
dependence of lL(T) from which we extracted Tc = 8.7 K and l0L = 380 nm. The colors indicate the different does not play a significant role in the wave-
values of Bz as in (B). length shift (fig. S3).

Borst et al., Science 382, 430–434 (2023) 27 October 2023 3 of 5

RES EARCH | R E S E A R C H A R T I C L E

well with static-field nano–superconducting quan- environment seen by the spin waves, leading
A laser
NV diamond tum interference device measurements (28). to local spin-wave refraction akin to defect-
The temperature dependence of the London controlled spin-wave scattering (30). The abil-
hot spot MoRe penetration depth provides a powerful tool for ity to optically induce spin-wave refraction at
x tuning the spin wavelength. To demonstrate target sites could be used to create devices
YIG
this, we imaged the spin waves in the YIG- such as gratings or magnonic crystals (31) and
y k YIG/SC k YIG k YIG/SC GGG MoRe region while sweeping through Tc at enable spin-wave manipulation through opti-
different magnetic fields (Fig. 4A and fig. S4). cal switching of flux-focusing regions in the
B C D
laser The extracted spin-wave number k is shown in superconducting strip.
Fig. 4B, with the color indicating the in-plane
component of the magnetic field. We observed Conclusions
z z z that k changes continuously with temperature We demonstrated local measurements of hy-
over the superconducting phase transition in bridized spin-wave–Meissner-current transport
y y y the YIG-MoRe region and remains unchanged modes in a magnetic thin film equipped with
in the bare YIG region (fig. S5). We did not a superconducting gate. The wavelength was
laser observe global heating of the superconductor tunable by temperature and field, enabling
due to our excitation laser (fig. S6). Using our efficient phase-shifting of the spin-wavefronts
model, we extracted the London penetration and a notable in situ visualization and quanti-
depth lL(T) for every observed value of k (Fig. 4C) tative extraction of the London penetration depth

Downloaded from https://www.science.org at Columbia University on August 07, 2024

[supplementary text, section S2 (19)]. We found as a function of temperature. Because MoRe
that almost all data collapse onto a single curve is a type-2 superconductor with an estimated
h
 i
described by lL ¼ l0L ½1  ðT = Tc Þ4 1=2 (29), F0 lL
laser lower critical field of Hc1 ¼ 4pl2 ln x þ 0:5 ≈
with Tc = 8.7 K and l0L ¼ 380nm. The excep- L

tions occurred when the spin wavelength 4 mT at T ≈ 5 K (32, 33), where x ≈ 0.01 mm is
lSW = 2p/k became comparable to the width the coherence length (34) and F0 the super-
of the MoRe strip. Here, our approximation conducting flux quantum, Abrikosov vortices
of the superconducting strip by an infinite film were expected in measurements such as those
breaks down. These results highlight that in Fig. 3B with a few-mT out-of-plane field com-
2 3 4 1 2 3 -1 0 1 imaging the hybridized spin-wave–Meissner- ponent. However, we did not identify vortex-
6 -1 current transport modes is a powerful tool for related effects in these measurements, which
PL 0 (10 s ) C 1on (%) C 1on - C 1off (%)
extracting the temperature dependence of the look qualitatively similar to those with purely
London penetration depth. in-plane field (fig. S3). Presumably, the pres-
Fig. 5. Laser-induced spin-wave refraction at
ence and location of vortices is strongly in-
target locations. (A) Schematic illustration of a Local control of spin-wave transport by fluenced by our focused excitation laser, as
laser-induced scattering spot. By shining an laser-induced spin-wave refraction highlighted by recent magneto-optical (35) and
auxiliary 594-nm laser on the sample, we created
Thus far, we have demonstrated dispersion wide-field NV-imaging experiments (36). In
a hot spot in the MoRe strip that locally altered
engineering through global control of temper- particular, (35) showed that vortices can be
the effective refractive index governing the
ature and magnetic field. We now show that annihilated by the laser through local heating
spin-wave propagation. (B) Scanning confocal
the creation of a hot spot in the superconduc- above Tc or pushed by the laser to new pinning
microscope images of the NV photoluminescence
tor with a focused laser enables local manip- sites or border regions of the superconductor.
at T = 5.5 K, with the auxiliary laser focused onto
ulation of the spin-wave transport by tuning The presented microwave magnetic imaging
the MoRe strip at three different locations indicated
the effective refractive index (Fig. 5A). To do of the spin-wave transport modes in a YIG-
by the arrows. Scale bar, 10 mm. (C) Spatial maps
so, we coupled an auxiliary, orange laser into MoRe heterostructure shows the versatility of
of the NV ESR contrast with auxiliary laser turned
our setup and focused it at target sites on the superconducting gates for spin-wave manipu-
on (Con
1 ) showing spin waves in the YIG-MoRe region superconductor (Fig. 5B). The laser spot is lation, enables determining the temperature-
that scatter on the laser spot. (D) Background- dependent London penetration depth, and
visible through the locally enhanced NV photo-
subtracted ESR contrast highlighting the laser- opens new opportunities for creating wave-
luminescence. Spatial measurements of the NV
induced spin-wave scattering obtained by subtracting
ESR contrast C1on with the auxiliary laser on based circuit elements such as filters, mirrors,
the ESR contrast with the auxiliary laser turned and cavities.
(Fig. 5C) show the spin-wave scattering pat-
off Coff
1 from the measurements in (C). Bz = 3.3 mT. terns induced by the local hot spot. The reduc-
tion in amplitude behind the hot spot indicates REFERENCES AND NOTES

destructive interference between the scattered 1. S. A. Wolf et al., Science 294, 1488–1495 (2001).
2. A. V. Chumak, V. I. Vasyuchka, A. A. Serga, B. Hillebrands,
From the field dependence of the extracted and incident spin waves. Subtracting a refer- Nat. Phys. 11, 453–461 (2015).
spin-wave numbers in the bare YIG region, we ence measurement with the auxiliary laser 3. A. Barman et al., J. Phys. Condens. Matter 33, 413001 (2021).
extracted the YIG saturation magnetization turned off (Fig. 5D) highlights the angular pro- 4. J. Sinova, S. O. Valenzuela, J. Wunderlich, C. H. Back,
T. Jungwirth, Rev. Mod. Phys. 87, 1213–1260 (2015).
Ms = 194(1) kA/m [supplementary text, sec- file of the scattered spin-wave patterns. 5. L. J. Cornelissen, J. Liu, R. A. Duine, J. Ben Youssef,
tion S2 (19)] in agreement with previous low- The characteristic “caustic” angles observed B. J. Van Wees, Nat. Phys. 11, 1022–1026 (2015).
temperature measurements (27). We then in these scattered patterns (dashed lines, Fig. 6. A. Hamadeh et al., Phys. Rev. Lett. 113, 197203 (2014).
7. T. Yu, C. Liu, H. Yu, Y. M. Blanter, G. E. W. Bauer, Phys. Rev. B
used Ms as a fixed parameter to fit the field 5D) result from the highly anisotropic dipolar 99, 134424 (2019).
dependence of the spin-wave numbers under- spin-wave dispersion (30). Tracing the patterns 8. J. Chen et al., Phys. Rev. Lett. 120, 217202 (2018).
neath the superconducting strip using the hy- to their origin shows that the scattering site is 9. Y. Sun et al., Phys. Rev. Lett. 111, 106601 (2013).
10. I. Bertelli et al., Adv. Quantum Technol. 4, 2100094 (2021).
bridized YIG-MoRe spin-wave dispersion (Eq. 1). tightly confined to the laser location. Presum-
11. Z. Y. Zhang et al., Supercond. Sci. Technol. 25, 025009 (2012).
From this fit, we extracted a London penetration ably, the laser locally breaks the superconduc- 12. A. Blais, A. L. Grimsmo, S. M. Girvin, A. Wallraff, Rev. Mod. Phys. 93,
depth lL = 405(10) nm at T = 5.5 K, which agrees tivity, inducing a local change in the magnetic 025005 (2021).

Borst et al., Science 382, 430–434 (2023) 27 October 2023 4 of 5

RES EARCH | R E S E A R C H A R T I C L E

13. I. A. Golovchanskiy et al., J. Appl. Phys. 127, 093903 (2020). 29. M. Tinkham, Introduction to superconductivity (Dover Publications, placement: A.L. and S.K.; Constructing experimental setup: M.B. and
14. O. V. Dobrovolskiy et al., Nat. Phys. 15, 477–482 (2019). 2nd edition, 2004). A.L.; Measurements and software: M.B., P.H.V., A.L., and I.B.; Data
15. T. Yu, G. E. W. Bauer, Phys. Rev. Lett. 129, 117201 (2022). 30. R. Gieniusz et al., Appl. Phys. Lett. 102, 102409 (2013). analysis: M.B., P.H.V., and A.T.; Theory: T.S. and Y.M.B.; Figures: M.B.
16. L. Rondin et al., Rep. Prog. Phys. 77, 056503 (2014). 31. A. V. Chumak, A. A. Serga, B. Hillebrands, J. Phys. D Appl. Phys. and P.H.V.; Funding acquisition: T.S.; Project administration: M.B.
17. T. X. Zhou et al., Proc. Natl. Acad. Sci. U.S.A. 118, e2019473118 50, 244001 (2017). and T.S.; Supervision: T.S. and M.B.; Writing – original draft:
(2021). 32. W.-K. Kwok et al., Rep. Prog. Phys. 79, 116501 (2016). M.B., T.S., and P.H.V.; Writing – review and editing: M.B., T.S., P.H.V.,
18. I. Bertelli et al., Sci. Adv. 6, eabd3556 (2020). 33. V. E. Calado et al., Nat. Nanotechnol. 10, 761–764 (2015). and Y.M.B. Competing interests: The authors declare that they have
19. Materials and methods, and supplementary text, figures and 34. K. Makise et al., Mater. Res. Express 5, 096406 (2018). no competing interests. Data and materials availability: All
references are available as supplementary materials. 35. I. S. Veshchunov et al., Nat. Commun. 7, 12801 (2016). data presented in this work are available at Zenodo (37). License
20. F. Casola, T. Van Der Sar, A. Yacoby, Nat. Rev. Mater. 3, 17088 36. S. E. Lillie et al., Nano Lett. 20, 1855–1861 (2020). information: Copyright © 2023 the authors, some rights reserved;
(2018). 37. M. Borst et al., Data for: Observation and control of hybrid exclusive licensee American Association for the Advancement of
21. R. Schirhagl, K. Chang, M. Loretz, C. L. Degen, Annu. Rev. spin-wave–Meissner-current transport modes, version 1, Science. No claim to original US government works. https://www.
Phys. Chem. 65, 83–105 (2014). Zenodo (2023). science.org/about/science-licenses-journal-article-reuse
22. D. R. Glenn et al., Geochem. Geophys. Geosyst. 18, 3254–3267
(2017). ACKN OWLED GMEN TS
SUPPLEMENTARY MATERIALS
23. P. Andrich et al., npj Quantum Inf. 3, 28 (2017). We thank A. F. Otte for commenting on the manuscript and
24. C. S. Wolfe et al., Phys. Rev. B Condens. Matter Mater. Phys. 89, science.org/doi/10.1126/science.adj7576
T. Yu and G.E.W. Bauer for discussions. Funding: This work was
180406 (2014). Materials and Methods
supported by the Dutch Research Council (NWO) under awards
Supplementary Text
25. A. G. Gurevich, G. A. Melkov, Magnetization Oscillations and VI.Vidi.193.077, NGF.1582.22.018, and OCENW.XL21.XL21.058. This
Figs. S1 to S6
Waves (CRC Press, 1996). work was also supported by the Dutch National Growth Fund (NGF)
References (38–42)
26. S. R. Seshadri, Proc. IEEE 58, 506–507 (1970). as part of the Quantum Delta NL programme. Author contributions:
27. S. Knauer et al., J. Appl. Phys. 133, 143905 (2023). Conceptualization: T.S. and M.B.; Sample design and fabrication: Submitted 14 July 2023; accepted 18 September 2023
28. A. G. Shishkin et al., Supercond. Sci. Technol. 33, 065005 (2020). M.B.; Diamond membrane fabrication: B.S. and M.B.; Membrane 10.1126/science.adj7576

Downloaded from https://www.science.org at Columbia University on August 07, 2024

Borst et al., Science 382, 430–434 (2023) 27 October 2023 5 of 5