Detection of ultralow magnetic field requires a magnetic sensor with high sensitivity and a low noise level. In this work, we used the Co20Fe60B20/Ti/Co20Fe60B20 magnetically coupled multilayer as the core structure of an anomalous Hall sensor. We adjusted the thickness of the Ti interlayer to modify its perpendicular magnetic anisotropy and interlayer magnetic coupling, thereby improving the sensitivity of the anomalous Hall sensor. Through the investigation of magnetic field response and noise properties of devices with different Ti thicknesses, the highest sensitivity of 34 803 Ω/T and the best magnetic field detectivity of 4.6 nT/Hz at 1 Hz were achieved with a Ti thickness of 2.0 nm at room temperature. This anomalous Hall sensor has both ultrahigh sensitivity and magnetic field detectivity, making it a good candidate for applications in detecting weak magnetic fields.

Magnetic sensors play an important role in production and life applications. They are widely used in the aerospace,1 automotive industry,2 power grid detection,1 geological exploration, transportation,3 environmental monitoring, consumer electronics,3 and other fields. The semiconductor Hall sensors, such as those made from GaAs/AlGaAs4 and GaAs/InSb,5 have been widely used to detect magnetic fields. However, they have the limitations of low sensitivity and low operating frequency; therefore, they are not suitable for detecting a weak or a high-frequency magnetic field. Moreover, due to their semiconductor nature, the semiconductor Hall sensors are also susceptible to temperature variation. With the discovery of the anomalous Hall effect (AHE) in magnetic metals and alloys,6 magnetic sensors based on the AHE have also been developed in recent years.7–9 Analogous to the semiconductor Hall sensors, the AHE sensors can also detect a magnetic field in the vertical direction via the detection of the transverse Hall voltage. The Hall voltage of the AHE sensor normally is much larger than that of the semiconductor Hall sensor and can be saturated just in a small magnetic field range. Hence, the AHE magnetic sensors can possess a higher sensitivity,7,10 allowing them to detect a weak magnetic field. Moreover, they also have the advantages of being able to work in a large frequency range11 and with low intrinsic noise.12,13

At present, highly sensitive magnetic sensors, which are capable of detecting a very weak magnetic field, are highly desired in emerging application scenarios such as biomedicine and environmental monitoring.14–16 For instance, they have been extensively used to detect nanomagnetically labeled tiny biomolecules in the applications of cancer diagnostics and prognostics, genotyping DNA, and point-of-care devices.17 In addition to the sensitive magnetoresistive sensors,18–20 the AHE sensors can be another promising option for the above-mentioned applications due to their advantages of simple manufacturing process, easy integration, low cost, as well as high sensitivity.

The sensitivity of the AHE sensor can be further improved by either increasing the anomalous Hall resistance21–23 or by reducing the saturation magnetic field.24 In this work, we designed a dual Ti/Co20Fe60B20 interface to enhance the anomalous Hall resistance. Meanwhile, we also adjusted the thickness of the Ti interlayer to control the saturation field of the sensor device. As a consequence, the sensitivity of the anomalous Hall sensor can be improved significantly. The highest sensitivity reaches ∼34 803 Ω/T, which is about 30 times higher than that of the semiconductor Hall sensors.4,5,10 In addition to the sensitivity, it is also important to reduce the intrinsic noise of the magnetic sensor. Here, we characterized the noise of a series of samples and obtained sensor devices with low noise of ∼100 nV/Hz and high field detectivity of ∼4.6 nT/Hz. The high-performance AHE sensors are promising for application scenarios in detecting weak magnetic fields.

As shown in Fig. 1(a), multilayers of MgO (2 nm)/Co20Fe60B20 (1 nm)/Ti (tTi nm)/Co20Fe60B20 (0.7 nm)/MgO (2 nm)/SiO2 (4 nm) were deposited on the thermally oxidized Si wafer at room temperature using a high-vacuum magnetron sputtering system with a base pressure of 5 × 10−9 Torr (AJA, ATC-2200-UHV). A wedged-shaped Ti interlayer was deposited using an in situ moving shutter technique, with tTi ranging from 1 to 2 nm. Varying the Ti thickness allows us to gain insight into the performance of AHE sensors with different perpendicular magnetic anisotropy. After film growth, Hall-bar devices were fabricated by ultraviolet lithography and ion milling for transport measurements. The Ti/Cu/Pt electrodes were formed finally by direct current (DC) sputtering and lift-off technology. All samples and devices in this study were not annealed. Figure 1(b) displays the schematic diagram of the anomalous Hall measurement. The active cross area of the Hall cross is 20 × 20 μm2. For the anomalous Hall resistance measurements, we used a current source (Keithley 2450) to apply a DC current and a nanovoltmeter (Keithley 2182A) to measure the Hall voltage. At the same time, a series of control Hall-bar devices of Ti (2 nm)/Co20Fe60B20 (0.7 nm)/MgO (2 nm)/SiO2 (4 nm) with the same device size were fabricated and measured. The Hall resistance (RH) as a function of the external perpendicular magnetic field (Hz) is shown in Fig. 1(c). It is clearly observed that the double Ti/Co20Fe60B20 interfaces significantly amplify the anomalous Hall resistance. The sensing mode based on a Co20Fe60B20 single-layer structure was realized by tuning the thickness of the Co20Fe60B20 layer to make the magnetization near the spin reorientation transition.25 

FIG. 1.

(a) Schematic illustration of the multilayer structure. (b) The micrograph of the Hall-bar device with a schematic diagram of the anomalous Hall measurement. The center active area of the Hall cross is 20 × 20 μm2. (c) Hall resistance vs external perpendicular magnetic field for samples with a single-Ti (2 nm)/Co20Fe60B20 (0.7 nm) interface (blue) and with Co20Fe60B20 (1 nm)/Ti (2 nm)/Co20Fe60B20 (0.7 nm) double interfaces (red) in the as-grown state. (d) Hall resistance of MgO (2 nm)/Co20Fe60B20 (1 nm)/Ti (2 nm)/Co20Fe60B20 (0.7 nm)/MgO (2 nm) during a sweep of the perpendicular magnetic field (Hz).

FIG. 1.

(a) Schematic illustration of the multilayer structure. (b) The micrograph of the Hall-bar device with a schematic diagram of the anomalous Hall measurement. The center active area of the Hall cross is 20 × 20 μm2. (c) Hall resistance vs external perpendicular magnetic field for samples with a single-Ti (2 nm)/Co20Fe60B20 (0.7 nm) interface (blue) and with Co20Fe60B20 (1 nm)/Ti (2 nm)/Co20Fe60B20 (0.7 nm) double interfaces (red) in the as-grown state. (d) Hall resistance of MgO (2 nm)/Co20Fe60B20 (1 nm)/Ti (2 nm)/Co20Fe60B20 (0.7 nm)/MgO (2 nm) during a sweep of the perpendicular magnetic field (Hz).

Close modal

We have confirmed that the stack of Sub./Ti (2 nm)/Co20Fe60B20 (0.7 nm)/MgO (2 nm) is in a state of tilted magnetization, as shown in Fig. 1(c). Moreover, the Sub./MgO (2 nm)/Co20Fe60B20 (1 nm)/Ti (2 nm)/SiO2(4 nm) has an in-plane magnetic anisotropy (not shown). However, by combining the above-mentioned two parts, the whole stack of Sub./MgO (2 nm)/Co20Fe60B20 (1 nm)/Ti (2 nm)/Co20Fe60B20 (0.7 nm)/MgO (2 nm) exhibits a distinct behavior and gets a noticeably improved perpendicular magnetic anisotropy. Figure 1(d) shows the Hall resistance of the whole stack during a sweep of the perpendicular magnetic field up to 1.3 T. Note that the 1.3 T magnetic field is large enough to saturate all the magnetization. From our thin film optimization, we found that the dead layer in the lower Co20Fe60B20 layer (near the substrate) is about 0.3 nm thicker than that in the upper Co20Fe60B20 layer due to the Ti bombardment during the film preparation process.26 Therefore, the actual thickness of the upper and lower layers of Co20Fe60B20 is similar. As shown in Fig. 1(d), the Co20Fe60B20/Ti/Co20Fe60B20 sample has a very small amount of in-plane component magnetization, which is less than 15% of the whole magnetization. This indicates the existence of interlayer magnetic coupling between two Co20Fe60B20 layers. As reported previously, the Ruderman–Kittel–Kasuya–Yosida (RKKY) interlayer exchange coupling in a ferromagnet/Ti/ferromagnet trilayer is negligible;27,28 hence, the dominant magnetic coupling mechanism of our Co20Fe60B20/Ti/Co20Fe60B20 sample is expected to be magnetostatic coupling.

To evaluate the performance of AHE sensors, a three-dimensional Helmholtz coil and noise measurement system with four probes were adopted. X–Y–Z coils are orthogonally assembled and powered by three independent DC power sources. This setup is capable of providing a maximum magnetic field of 500 Oe with a high resolution of 0.05 Oe. Figure 2(a) shows the anomalous Hall resistance vs perpendicular magnetic field for the device of Sub./MgO (2 nm)/Co20Fe60B20 (1 nm)/Ti (tTi nm)/Co20Fe60B20 (0.7 nm)/MgO (2 nm)/SiO2 (4 nm) with tTi = 1.6 nm. The output signal of the device changes linearly with the magnetic field in the z direction within the dynamic range, and the AHE response can be regarded as hysteresis-free with a negligible coercive field, as illustrated in the inset of Fig. 2(a).29 Therefore, this device can be a good magnetic field sensing element. Similarly, RH vs Hz for other devices with different Ti thicknesses were measured using the same method and plotted in Fig. 2(b). The sensitivity is an essential parameter to characterize the performance of AHE sensors. It is defined as the first derivative of the sensor’s anomalous Hall resistance with respect to the perpendicular magnetic field and is written as S=dRHHzdHz.10 According to the results in Fig. 2(c), the sensitivity starts to increase as the Ti thickness reaches above 1.5 nm, and as the Ti interlayer thickness becomes 2.0 nm, the highest sensitivity of 34 803 Ω/T can be achieved. The sensitivity of our AHE magnetic sensor is much larger than that of semiconductor Hall sensors, which is usually about 1000 Ω/T.4,5,10,30 Meanwhile, there is another way to evaluate the sensitivity, and that is the alternating current “AC” sensitivity, which can be characterized under a sweeping perpendicular magnetic field Hz plus an alternating perpendicular magnetic field ΔHz with a small amplitude of 0.3 Oe.25 The “AC” sensitivity of the sample with Ti interlayer thickness of 2.0 nm was obtained with the value of ∼27 668.3 Ω/T. Figure 2(d) shows RH and dynamic range (DR) as a function of Ti thickness. The DR is defined as half of the whole range of Hz, within which RH shows a linear behavior regarding Hz. As the thickness of Ti increases, the perpendicular magnetic anisotropy increases and, therefore, the corresponding dynamic range decreases. Although RH decreases due to current shunting in the Ti layer, the dynamic range decreases significantly more than RH, especially in the devices with a thicker Ti interlayer. Consequently, the sensitivity increases [in Fig. 2(c)]. As is known, devices possessing different dynamic ranges and sensitivities can be selected to meet specific sensing requirements in practical applications.1 In our case, the device with tTi of 2.0 nm has larger perpendicular anisotropy and, therefore, a lower dynamic range of only ±3.5 Oe. Our AHE sensor devices with high sensitivity and a small working field range are suitable for measuring weak magnetic fields along the vertical direction.

FIG. 2.

(a) Anomalous Hall resistance (RH) vs perpendicular magnetic field (Hz) for devices with Ti interlayer thicknesses tTi of 1.6 nm. The inset is the magnified magnetization curve near zero magnetic field (±5 Oe). (b) RH vs Hz for devices with different Ti thicknesses. (c) The sensitivity of Hall sensor devices as a function of Ti thickness. (d) The RH and dynamic range (DR) as a function of Ti thickness.

FIG. 2.

(a) Anomalous Hall resistance (RH) vs perpendicular magnetic field (Hz) for devices with Ti interlayer thicknesses tTi of 1.6 nm. The inset is the magnified magnetization curve near zero magnetic field (±5 Oe). (b) RH vs Hz for devices with different Ti thicknesses. (c) The sensitivity of Hall sensor devices as a function of Ti thickness. (d) The RH and dynamic range (DR) as a function of Ti thickness.

Close modal

The schematic of the noise measurement system is illustrated in Fig. 3(a). The time domain noise signals from AHE sensors are amplified and filtered by a low-noise amplifier (SR560, Stanford Research System) and sampled by a spectrum analyzer (SR785, Stanford Research System). The noise power spectral density (shorted as noise thereinafter) SV is averaged 50 times. To calibrate the accuracy of the noise measurement system, the thermal noise of different standard resistors ranging from 270 to 5000 Ω was first measured, as displayed in Fig. 3(b). Note that we removed the noise peaks at 50 Hz and their harmonics arising from the power grids for clarity. The theoretical value of the thermal noise of resistors can be calculated via SV = 4 KBTR,31 where KB is the Boltzmann constant, T is the temperature, and R is the resistance. The solid lines in Fig. 3(b) represent the theoretical values of thermal noise generated by different resistors. The experimental SV is in good agreement with the theoretical calculation value (i.e., the deviation is less than 1% between the averaged Sv and the theoretical calculation value), which proves the reliability of the noise measurement system.

FIG. 3.

(a) Schematic illustration of the Hall sensor noise measurement system. (b) Thermal noise of resistors with different resistances ranging from 270 to 5000 Ω. The scattered points are experimental data, and the solid line represents the theoretical values of thermal noise generated by different resistors. (c) Noise of devices with different Ti thicknesses measured at zero magnetic field. (d) Extracted noise at 1 Hz as a function of Ti thickness.

FIG. 3.

(a) Schematic illustration of the Hall sensor noise measurement system. (b) Thermal noise of resistors with different resistances ranging from 270 to 5000 Ω. The scattered points are experimental data, and the solid line represents the theoretical values of thermal noise generated by different resistors. (c) Noise of devices with different Ti thicknesses measured at zero magnetic field. (d) Extracted noise at 1 Hz as a function of Ti thickness.

Close modal

We then characterized the noise of Hall sensors with different Ti thicknesses at room temperature and zero magnetic field. The noise measurements were performed under a constant current of 1 mA. Figure 3(c) shows the noise of different devices. In the low-frequency region, the 1/f noise dominates compared to other types of contributions to noise. While in the high-frequency region, the white noise dominates in the devices. Figure 3(d) shows the summarized noise values at 1 Hz corresponding to different Ti thicknesses. The data points represent the averaged values from three measurements, with error bars indicating the standard deviation. As the Ti thickness increases to 1.5 nm, the noise at 1 Hz starts to decrease and further reaches the minimum value of ∼100 nV/Hz when tTi is 2.0 nm. It is known that the 1/f noise mainly consists of electronic 1/f noise and magnetic 1/f noise.32 The electronic-originated noise SV is defined as SVI2R2=αNf,33 where R is the device resistance, N is the number of charge carriers, f is the frequency, and α is the Hooge parameter.34–36 Considering the metallic characteristics of our AHE sensors, the electronic noise generated by defect thermal fluctuations is expected to be low. Relatively, in our device consisting of Co20Fe60B20/Ti/Co20Fe60B20 magnetic structure, the main contribution to the 1/f noise is supposed to be a magnetic origin, which can be caused by the magnetic domain jumping between the metastable ripple states when the magnetization direction of the magnetic layer is switched.37 Our experimental results consolidate the above speculation. As tTi increases to the thick region, the Hall sensor has an improved perpendicular magnetic anisotropy and, therefore, the magnetization of the device tends to be stabilized (i.e., small magnetic fluctuation),25 leading to a smaller magnetic noise. Consequently, a notable decrease of the noise was observed in our experiments.

Another crucial figure of merit of the magnetic sensors is the magnetic field detectivity, which is closely correlated with both the sensitivity and the noise of the sensor. The magnetic field detectivity ST can be evaluated by ST0.5=SV0.5IS,33 where SV is the noise value, S is the sensitivity, and I is the applied current. Figure 4(a) plots the root square of the magnetic field detectivity of a series of devices with different Ti thicknesses. Note that the dimensions and applied current values for all the devices are the same. Due to the increased sensitivity and decreased noise of the devices at thicker Ti, the resultant magnetic field detectivity is improved significantly. The best magnetic field detectivity was obtained at a Ti thickness of 2.0 nm, with the value of 4.6(±1.9) nT/Hz [∼5.8(±2.3) nT/Hz evaluated using the “AC” sensitivity value], which is better than the detectivity of semiconductor Hall sensors (normally μT/Hz at 1 Hz) by 2–3 orders of magnitude.4,5,38–40

FIG. 4.

(a) Magnetic field detectability for devices with different Ti thicknesses measured at zero magnetic field. (b) Magnetic field detectability at 1 Hz as a function of Ti thickness.

FIG. 4.

(a) Magnetic field detectability for devices with different Ti thicknesses measured at zero magnetic field. (b) Magnetic field detectability at 1 Hz as a function of Ti thickness.

Close modal

We finally summarized the sensitivity and magnetic field detectivity for representative AHE sensors recently reported as well as semiconductor Hall sensors in Table I. It is found that the magnetic sensors with coupled magnetic bilayer structure in this work have outstanding sensitivity. More importantly, a much better magnetic field detectivity has been achieved, by which an nT-level magnetic field can be detected.

TABLE I.

Summary of the structure, sensitivity (DC and AC), and corresponding magnetic field detectivity of our AHE sensor and other reported AHE and semiconductor Hall sensors.

StructureSensitivity (Ω/T)ST0.5 at 1 Hz (nT/Hz)Sensor-typeRef. No.
Co20Fe60B20/Ti/Co20Fe60B20 34 803a 4.6 AHE This work 
27 668b 5.8 
Ta/Co40Fe40B20/MgO 2458 76 AHE 25  
MgO/FePt/MgO 644.1 237.5 AHE 41  
Pt/Co/Ta 31 196 27.7 AHE 10  
MgO/Co40Fe40B20/Ta/MgO 23 760 … AHE 42  
Ta/Co40Fe40B20/MgO 1845 81 AHE 43  
GaAs/AlGaAs 1100 1000 Hall 4  
GaAs/InSb 370 720 Hall 5  
StructureSensitivity (Ω/T)ST0.5 at 1 Hz (nT/Hz)Sensor-typeRef. No.
Co20Fe60B20/Ti/Co20Fe60B20 34 803a 4.6 AHE This work 
27 668b 5.8 
Ta/Co40Fe40B20/MgO 2458 76 AHE 25  
MgO/FePt/MgO 644.1 237.5 AHE 41  
Pt/Co/Ta 31 196 27.7 AHE 10  
MgO/Co40Fe40B20/Ta/MgO 23 760 … AHE 42  
Ta/Co40Fe40B20/MgO 1845 81 AHE 43  
GaAs/AlGaAs 1100 1000 Hall 4  
GaAs/InSb 370 720 Hall 5  
a

This is the “DC” sensitivity characterized under a sweeping perpendicular magnetic field Hz only.

b

This is the “AC” sensitivity characterized under a sweeping perpendicular magnetic field Hz plus an alternating perpendicular magnetic field ΔHz with a small amplitude of 0.3 Oe and a frequency of 5 Hz. The bias magnetic field HB is 0.2 Oe.

In summary, AHE magnetic field sensors based on the Co20Fe60B20/Ti/Co20Fe60B20 magnetostatically coupled multilayer have been fabricated and characterized. Through structural design and control of perpendicular magnetic anisotropy and interlayer magnetic coupling, the sensor sensitivity was increased to be 34 803 Ω/T, which is the highest sensitivity among AHE sensors in the same category to the best of our knowledge. More importantly, from low noise characteristics, a good performance of magnetic field detectivity of 4.6 nT/Hz was obtained at 1 Hz, which is an order of magnitude better than those in the existing reports on AHE sensors.10,25,43 Our AHE sensors with high sensitivity and outstanding field detectivity are expected to play an important role in weak magnetic field detection in emerging biology, medicine, and environmental applications.

The work is supported by the National Natural Science Foundation of China (Grant Nos. 12074052 and 12261131506), the Natural Science Foundation of Liaoning Province of China (2021-YQ-06), and the Fundamental Research Funds for the Central Universities (Grant Nos. DUT24GJ204 and DUT23BK036).

The authors have no conflicts to disclose.

Xinna Liu: Investigation (lead); Methodology (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Fanyu Meng: Methodology (equal); Software (equal). Meining Du: Writing – original draft (equal); Writing – review & editing (equal). Yankun Li: Data curation (equal); Writing – review & editing (equal). Pengzhen Li: Investigation (supporting); Validation (equal). Tuo Zhang: Formal analysis (equal). Ying Feng: Formal analysis (equal). Yi Wang: Conceptualization (lead); Formal analysis (lead); Funding acquisition (lead); Project administration (lead); Resources (lead); Supervision (lead); Writing – original draft (equal); Writing – review & editing (lead).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1
X.
Liu
,
K. H.
Lam
,
K.
Zhu
,
C.
Zheng
,
X.
Li
,
Y.
Du
,
C.
Liu
, and
P. W. T.
Pong
,
IEEE Trans. Magn.
55
,
1
(
2019
).
2
M. A.
Khan
,
J.
Sun
,
B.
Li
,
A.
Przybysz
, and
J.
Kosel
,
Eng. Res. Express
3
,
022005
(
2021
).
3
C.
Zheng
,
K.
Zhu
,
S.
Cardoso de Freitas
,
J.-Y.
Chang
,
J. E.
Davies
,
P.
Eames
,
P. P.
Freitas
,
O.
Kazakova
,
C.
Kim
,
C.-W.
Leung
et al,
IEEE Trans. Magn.
55
,
1
(
2019
).
4
K.
Vervaeke
,
E.
Simoen
,
G.
Borghs
, and
V. V.
Moshchalkov
,
Rev. Sci. Instrum.
80
,
074701
(
2009
).
5
A.
Sandhu
,
H.
Sanbonsugi
,
I.
Shibasaki
,
M.
Abe
, and
H.
Handa
,
Jpn. J. Appl. Phys.
43
,
L868
(
2004
).
6
N.
Nagaosa
,
J.
Sinova
,
S.
Onoda
,
A. H.
MacDonald
, and
N. P.
Ong
,
Rev. Mod. Phys.
82
,
1539
(
2010
).
7
Y. M.
Lu
,
J. W.
Cai
,
H. Y.
Pan
, and
L.
Sun
,
Appl. Phys. Lett.
100
,
022404
(
2012
).
8
W. L.
Peng
,
J. Y.
Zhang
,
L. S.
Luo
,
G. N.
Feng
, and
G. H.
Yu
,
J. Appl. Phys.
125
,
093906
(
2019
).
9
G.
Yang
,
Y.
Li
,
X.
Chen
,
J.
Zhang
, and
G.
Yu
,
Adv. Condens. Matter Phys.
2016
,
1
.
10
W.
Su
,
Z. Q.
Hu
,
Y. J.
Li
,
Y. L.
Han
,
Y. C.
Chen
,
C. Y.
Wang
,
Z. D.
Jiang
,
Z. X.
He
,
J. E.
Wu
,
Z. Y.
Zhou
et al,
Adv. Funct. Mater.
33
,
2211752
(
2023
).
11
I.
Fergen
,
K.
Seemann
,
A.
von der Weth
, and
A.
Schüppen
,
J. Magn. Magn. Mater.
242
,
146
(
2002
).
12
Y.
Zhang
,
Q.
Hao
, and
G.
Xiao
,
Sensors
19
,
3537
(
2019
).
13
J.
Shiogai
,
Z.
Jin
,
Y.
Satake
,
K.
Fujiwara
, and
A.
Tsukazaki
,
Appl. Phys. Express
12
,
123001
(
2019
).
14
C.
Dey
,
P.
Yari
, and
K.
Wu
,
Nano Futures
7
,
012002
(
2023
).
15
S. G.
Grancharov
,
H.
Zeng
,
S. H.
Sun
,
S. X.
Wang
,
S.
O'Brien
,
C. B.
Murray
,
J. R.
Kirtley
, and
G. A.
Held
,
J. Phys. Chem. B
109
,
13030
(
2005
).
16
Z.
Xiong
,
S.
Achavananthadith
,
S.
Lian
,
L. E.
Madden
,
Z. X.
Ong
,
W.
Chua
,
V.
Kalidasan
,
Z. P.
Li
,
Z.
Liu
,
P.
Singh
et al,
Sci. Adv.
7
,
1617
(
2021
).
17
K.
Wu
,
T.
Klein
,
V. D.
Krishna
,
D.
Su
,
A. M.
Perez
, and
J.-P.
Wang
,
ACS Sens.
2
,
1594
(
2017
).
18
T.
Nakano
,
K.
Fujiwara
,
S.
Kumagai
,
Y.
Ando
, and
M.
Oogane
,
Appl. Phys. Lett.
122
,
072405
(
2023
).
19
X.
Han
,
Y.
Zhang
,
Y.
Wang
,
L.
Huang
,
Q.
Ma
,
H.
Liu
,
C.
Wan
,
J.
Feng
,
L.
Yin
,
G.
Yu
et al,
Chin. Phys. Lett.
38
,
128501
(
2021
).
20
F.
Matos
,
R.
Macedo
,
P. P.
Freitas
, and
S.
Cardoso
,
AIP Adv.
13
,
025108
(
2023
).
21
Y.
Ni
,
Z.
Zhang
,
I. C.
Nlebedim
, and
D. C.
Jiles
,
IEEE Trans. Magn.
52
,
1
(
2016
).
22
Y.
Satake
,
K.
Fujiwara
,
J.
Shiogai
,
T.
Seki
, and
A.
Tsukazaki
,
Sci. Rep.
9
,
3282
(
2019
).
23
E.
Liu
,
Y.
Sun
,
N.
Kumar
,
L.
Muechler
,
A.
Sun
,
L.
Jiao
,
S.-Y.
Yang
,
D.
Liu
,
A.
Liang
,
Q.
Xu
et al,
Nat. Phys.
14
,
1125
(
2018
).
24
A.
Gerber
,
A.
Milner
,
M.
Karpovsky
,
B.
Lemke
,
H. U.
Habermeier
,
J.
Tuaillon-Combes
,
M.
Négrier
,
O.
Boisron
,
P.
Mélinon
, and
A.
Perez
,
J. Magn. Magn. Mater.
242
,
90
(
2002
).
25
X.
Zhang
,
Y.
Zhang
,
J.
Zhang
et al,
Appl. Phys. Lett.
116
,
212404
(
2020
).
26
J. H.
Kim
,
J. B.
Lee
,
G. G.
An
,
S. M.
Yang
,
W. S.
Chung
,
H. S.
Park
, and
J. P.
Hong
,
Sci. Rep.
5
,
16903
(
2015
).
27
M. C. H.
de Jong
,
C. O.
Avci
,
A.
Hrabec
, and
P.
Gambardella
,
Phys. Rev. B
102
,
174433
(
2020
).
28
29
Y.
Zhu
and
J. W.
Cai
,
Appl. Phys. Lett.
90
,
012104
(
2007
).
30
T.
Dai
,
C.
Chen
,
L.
Huang
,
J.
Jiang
,
L.-M.
Peng
, and
Z.
Zhang
,
ACS Nano
14
,
17606
(
2020
).
31
L.
Huang
,
Z. H.
Yuan
,
B. S.
Tao
,
C. H.
Wan
,
P.
Guo
,
Q. T.
Zhang
,
L.
Yin
,
J. F.
Feng
,
T.
Nakano
,
H.
Naganuma
et al,
J. Appl. Phys.
122
,
113903
(
2017
).
32
Z. Q.
Lei
,
G. J.
Li
,
W. F.
Egelhoff
,
P. T.
Lai
, and
P. W. T.
Pong
,
IEEE Trans. Magn.
47
,
602
(
2011
).
33
K.
Wang
,
Y.
Zhang
, and
G.
Xiao
,
Phys. Rev. Appl.
13
,
064009
(
2020
).
35
F. N.
Hooge
and
A. M.
Hoppenbrouwers
,
Physica
45
,
386
(
1969
).
37
C.
Ren
,
X. Y.
Liu
,
B. D.
Schrag
, and
G.
Xiao
,
Phys. Rev. B
69
,
104405
(
2004
).
38
D.
Collomb
,
P. L.
Li
, and
S.
Bending
,
J. Phys.: Condens. Matter
33
,
243002
(
2021
).
39
Y. J.
Min
,
C. K.
Kwon
,
H. K.
Kim
,
C.
Kim
, and
S. W.
Kim
,
IEEE Sens. J.
12
,
1195
(
2012
).
40
J.
Jankowski
,
S.
El-Ahmar
, and
M.
Oszwaldowski
,
Sensors
11
,
876
(
2011
).
41
K.
Wang
,
Y.
Zhang
,
S.
Zhou
, and
G.
Xiao
,
Nanomaterials
11
,
854
(
2021
).
42
T.
Zhu
,
P.
Chen
,
Q. H.
Zhang
,
R. C.
Yu
, and
B. G.
Liu
,
Appl. Phys. Lett.
104
,
202404
(
2014
).
43
R.
Li
,
S.
Zhang
,
S.
Luo
,
Z.
Guo
,
Y.
Xu
,
J.
Ouyang
,
M.
Song
,
Q.
Zou
,
L.
Xi
,
X.
Yang
et al,
Nat. Electron.
4
,
179
(
2021
).