The antiferromagnetic kagome semimetals Mn3X (X = Ge, Sn, Ga) are of great interest due to properties arising from their Berry curvature, such as large anomalous Nernst and anomalous Hall coefficients, and spin to charge conversion efficiencies at ambient temperatures. However, the synthesis of epitaxial thin films of Mn3Ge in the desired hexagonal phase has been challenging because they do not wet insulating substrates, necessitating the use of a metallic buffer layer. Furthermore, a ferrimagnetic tetragonal phase also forms readily under typical growth conditions, interfering with hexagonal phase properties. We have synthesized atomically smooth and continuous epitaxial thin films of hexagonal Mn3Ge directly on insulating LaAlO3 (111) substrates using electron beam assisted molecular beam epitaxy, using a three-step process that mitigates the formation of the tetragonal phase. The anomalous Nernst coefficient is found to be more than six times larger in our films than in sputtered thin films of Mn3Ge and significantly larger than that of Fe. Our approach can be used to grow thin layers of kagome materials, without interference from a buffer layer in transport properties, and may be applicable to a broader range of materials with large surface energies that do not grow readily on insulating substrates.

Topological materials have been studied extensively over the past decade and a half, with discoveries of the quantum spin Hall effect in 2D1 and 3D topological insulators2 and more recently a broad range of topological semimetals.3–5 Thin films and heterostructures have played a central role in this endeavor, including seminal discoveries such as the quantum spin Hall1 and quantum anomalous Hall6 effects. Thin films are ideal platforms for exploring properties such as those that arise due to the quantum confinement,7,8 strain tuning,9 electric field-effect gating,10 mesoscopic non-local effects,11 charge-spin conversion,12,13 and superconducting proximity effects,14 motivated by both fundamental questions and applications. In order to realize the full potential of topological materials, we need to synthesize films with well controlled surfaces and interfaces, high crystal quality, and correct stoichiometry to avoid artifacts that may arise from defects. While the growth of thin films and heterostructures has been pursued right from the birth of topological materials, realizing conditions for the growth of high-quality films can be time-consuming. Many of the recently discovered topological materials are not line compounds, and thus are at best metastable in the desired composition under typical conditions for films growth. Furthermore, their lattice often does not match to a readily available substrate, and the conditions that favor the growth of thin films such as wetting, diffusivity, and stability of adatoms on the surface are not assured. These issues are also relevant for the synthesis of noble and refractory metal films15 and 2D materials16–19 on insulating substrates, where challenges in growing high-quality large-area films constrain the development of both science and technology.

Hexagonal Mn3X (X = Ge, Sn, Ga) are antiferromagnetic Weyl semimetals.20–22 Structurally, they are composed of kagome layers stacked along the c-axis in the sequence ABAB. Each kagome layer is formed by Mn atoms with X atoms sitting in the center of hexagons [Fig. 1(a)]. Due to the 120° anti-chiral alignment of the Mn moments in the kagome plane, multiple Weyl points related by a combination of mirror, time-reversal, and nonsymmorphic glide mirror plane symmetries are predicted to occur in reciprocal space, several in the vicinity of the Fermi surface. These symmetries also allow the existence of a non-zero Berry curvature21 in the kagome plane, which can be quite substantial due to the presence of the Weyl points. As a result, large anomalous Hall effect (AHE) and anomalous Nernst effect (ANE) have been proposed and detected in experiments, comparable or better than the best ferromagnetic metals, even though the hexagonal Mn3X compounds remain antiferromagnetic with only a very small net magnetic moment.23–25 Large magneto-optic Kerr rotation26 and magnetic spin Hall effects27 have also been measured in these antiferromagnetic materials. We note that while Mn3Ge retains its anti-chiral magnetic structure down to the lowest temperatures, Mn3Sn form a spin glass below about 50 K28 and, thus, loses its topological properties. Thus, despite their similarities, Mn3Ge and Mn3Sn are quite distinct and Mn3Ge presents the opportunity to study topological properties down to low temperatures. For example, the study of topological band structures by either angle resolved photoemission spectroscopy (ARPES) or scanning tunneling microscopy (usually carried out at low temperatures) requires high quality surfaces that may be possible to obtain in MBE grown films. Films are also candidates for exploring low temperature mesoscale transport in patterned structures where again Mn3Ge is a better choice as its topological properties are preserved to low temperatures. Furthermore, the Berry curvature related effects such as the ANE in single crystals of Mn3Ge29 have been found to be substantially larger than in Mn3Sn25 at room temperature, making it more attractive for some applications. Although much of the experimental work mentioned above has been realized in bulk single crystals or polycrystalline films, work on the growth of high quality Mn3Ge and Mn3Sn thin films on insulating substrates is still lacking.

FIG. 1.

(a) Atomic structure of hexagonal Mn3Ge: two unit-cells along the c axis (left) and kagome structure for each layer (right). (b) Atomic structure of LaAlO3 substrate with (111) termination, which shows the kagome structure.

FIG. 1.

(a) Atomic structure of hexagonal Mn3Ge: two unit-cells along the c axis (left) and kagome structure for each layer (right). (b) Atomic structure of LaAlO3 substrate with (111) termination, which shows the kagome structure.

Close modal

Recently, several groups have reported on the synthesis of Mn3X thin films by magnetron sputtering. Markou et al. grew Mn3Sn on Ru buffered Y:ZrO2 (111), which is heteroepitaxial to the substrate.30 However, due to wetting issues, their Mn3Sn films were discontinuous, which limits the study of the electrical transport properties. Similar effects were also observed in the MBE grown films of the kagome antiferromagnet FeSn.31 Continuous and high mobility films32,33 of FeSn on SrTiO3 were grown using a procedure of capping and post-growth annealing, where the capping layer covers the surface of the film. Several groups34–37 have grown continuous polycrystalline Mn3X film by magnetron sputtering. By introducing a Ru underlayer that helps with wetting of Mn3X on perovskite substrates, Hong et al.38 and Taylor et al.39 reported epitaxial growth of continuous thin films of Mn3Ge and Mn3Sn, respectively. The large anomalous Nernst effect and anomalous Hall effect have been reported in these films, but these anomalies are significantly smaller than that in single crystals, presumably due to defects such as structural imperfections or compositional variations.

Molecular beam epitaxy (MBE) has been used for growing a variety of topological materials, including Bi2(Se,Te)3 based materials, Cd3As2,40 Na3Bi,41,42 Fe3Sn2 (on Pt underlayers),43 and two-dimensional stanene.44 During MBE synthesis, films can be grown relatively slowly, ideally in an atomically layer-by-layer manner, where the atomic species have thermal energies that depend on the temperature of the evaporated atoms and substrate (in this instance under 60 meV). Under such conditions, thin films grown by MBE can have high crystalline quality, comparable to or better than in bulk single crystals.1,45–47 In this work, we show that epitaxial and continuous Mn3Ge thin films can be directly grown on insulating perovskite substrate LaAlO3 (111), without any buffer layer. Both the crystal structure and electrical transport properties are comparable to bulk single crystals. Compared with sputtered Mn3Ge thin films, our MBE films have an anomalous Nernst coefficient that is more than six times larger and is comparable to that in single crystals.23 While Mn3X crystals do not cleave well, which hinders surface studies, our atomically smooth films are suitable for characterization using angle-resolved photoelectron spectroscopy (ARPES) and scanning tunneling microscopy (STM).

Figure 1(a) shows the crystal structure of Mn3Ge. Two kagome layers formed by Mn atoms are stacked along the c-axis, with Ge atoms sitting in the center of the hexagons. These two layers can be transformed into each other by a glide mirror symmetry with a c/2 translation. According to calculations, Weyl nodes of opposite chirality are protected by this symmetry in21 Mn3Ge. Considering that the in-plane lattice constants are 5.32 Å for both a and b in Mn3Ge, we choose LaAlO3 (111) as the substrate to optimize the lattice match (3.79 Å × 2=5.36 Å). Figure 1(b) shows the atomic structure of LaAlO3 substrate where there is a similar triangular structure as Mn3Ge when viewing along the [111] direction (the O atoms in the LaO3 layer are on a kagome lattice).

Before growth, the substrates are sonicated in acetone and isopropanol for 5 min each and then dried with N2. The substrate is further degassed at 400 °C in a preparation chamber and then cooled down to room temperature when the pressure is lower than 5 × 10−9 Torr. After transferring the substrate to the growth chamber (base pressure ∼1 × 10−10 Torr), we heat the substrate up to 870 °C and hold it there for 30 min to improve the surface quality. This process is monitored in situ using the reflection high energy electron diffraction (RHEED) from a 10 keV electron beam. During the entire process mentioned above, the heating and cooling rates are fixed at 20 °C/min. Before film growth, the growth rates of Mn and Ge are calibrated using a quartz crystal micro-balance (QCM). The growth is carried out using co-evaporation of Mn and Ge in an overflux of Mn, at a growth rate of ∼0.25 unit cell/min, while keeping the substrate at 570 °C. The films’ stoichiometry is analyzed by the Rutherford backscattering, which show an Mn to Ge ratio of 0.74:0.26.

Similar to the results by Markou et al.,30 we find that the Mn3Ge films tend to be discontinuous after growth. As shown in Fig. 2(a), atomic force microscope (AFM) measurements show that the film is composed of islands of ∼100 nm size with varied crystalline orientations, faceted surfaces, and with gaps in between the islands. The x rays from these regions (not shown) show a poly-crystalline structure. The overall root mean square (rms) roughness is over 10 nm, indicating a large variation in island heights. However, we found that the region of the film under illumination from the RHEED beam shows a totally different morphology, as shown in Fig. 2(b), where a sharp RHEED pattern develops. Although the film is still discontinuous, islands in this area show atomically flat surfaces, similar to the MBE growth of another kagome semimetal FeSn.31 Unlike plasma-assisted MBE growth using activated gaseous species (such as in nitrides48,49), we hypothesize that the electron beam illumination introduces atomic level defects on the substrate and in the growing film, which create nucleation sites that promote epitaxial growth.

FIG. 2.

AFM images of the Mn3Ge film without (a) and with (b) electron beam illumination. An optical image of the film is shown in the inset, where an obvious contrast can be seen. Sharp RHEED streaks in the inset of (b) indicate a smooth surface forming with the assistance of electron irradiation.

FIG. 2.

AFM images of the Mn3Ge film without (a) and with (b) electron beam illumination. An optical image of the film is shown in the inset, where an obvious contrast can be seen. Sharp RHEED streaks in the inset of (b) indicate a smooth surface forming with the assistance of electron irradiation.

Close modal

Based on this finding, we pursued a path toward obtaining continuous atomically smooth Mn3Ge films and arrived at a three-step growth process. First, we grow a discontinuous but crystalline and epitaxial seed layer, with assistance from electron beam illumination; second, we grow a “connection” layer at a lower growth temperature; and third, we grow a final layer at a high temperature to generate single phase films and anneal away defects in the connecting layer grown in the second step.

As shown in Figs. 3(a)3(d), we stopped the growth after each step and the morphology evolution during the three growth steps is captured by AFM. The seed layer is grown at 570 °C with the electron beam density fixed at ∼0.5 μA/mm2 to assist the nucleation process. After the growth of ∼20 nm thick film, the c-axis oriented islands are about 100 nm in lateral dimension. At the second step, we cool down the substrate to 475 °C and keep growing for an additional 10 nm. During this “connecting” layer growth, a large increase in the RHEED streaks’ intensity is seen [Fig. 3(c)], indicating significant improvement of the surface quality. Interestingly, we see the RHEED intensity oscillations during this period as well, indicating a layer-by-layer growth mode for the kagome layers. We note here that the hexagonal phase (ABAB stacking of the kagome layers) and the tetragonal phase (ABCABC stacking of the kagome layers) are close in energy. Considering that the tetragonal phase of Mn3Ge50 is more stable than the hexagonal phase at lower temperatures, at the end of second step, we heat up the substrate up to 570 °C again for the third and final step where we grow an additional 15–20 nm of Mn3Ge. During this period, the RHEED oscillations disappear. However, the RHEED spots become sharper and more intense and the background from diffuse scattering becomes dimmer, indicating a transition to step-flow growth, presumably due to more diffusive atoms at higher temperature and the formation of atomically smooth surfaces. At the end of this last step, we turn off the heater power abruptly and cool the substrate at the highest rate possible to reduce the probability of formation of the low temperature tetragonal phase. As shown in Fig. 3(d), the result is a percolating network of epitaxial single crystalline (see the next paragraph) micrometer-scale structures with very smooth surfaces.

FIG. 3.

Evolution of the film morphology during Mn3Ge growth using three-step technique. (a) Blank substrate, (b) seed layer growth; (c) connecting layer growth, and (d) final layer growth. For each period, a RHEED image is captured.

FIG. 3.

Evolution of the film morphology during Mn3Ge growth using three-step technique. (a) Blank substrate, (b) seed layer growth; (c) connecting layer growth, and (d) final layer growth. For each period, a RHEED image is captured.

Close modal

We characterized the structure of our Mn3Ge films by x-ray diffraction (Cu Kα1), as shown in Fig. 4. For the 2θ ranging from 20° to 100°, only the substrate peaks and Mn3Ge (002) and Mn3Ge (004) peaks are observed in our 2θ ∼ ω scan [2θ = 41.86° for Mn3Ge (002)]. Using Bragg’s law, the lattice constant along the c-axis is inferred to be ∼4.312 Å, very close to those in single crystals. By doing a rocking curve around Mn3Ge (002), we get the FWHM ∼0.07°, indicating a small mosaicity of the film (for reference, the mosaicity38 in sputtered films is ∼0.59°). To reveal the epitaxial relation between the Mn3Ge film and the LaAlO3 substrate, we performed azimuthal XRD ϕ-scans choosing Mn3Ge {203} and LaAlO3{110} peaks. As shown in Figs. 4(b) and 4(a), a sixfold symmetry is found in Mn3Ge {203}, which aligns well with the threefold LaAlO3{110} peaks. This epitaxial relation can also be inferred from RHEED images captured before and after growth, where the RHEED streaks after film growth align well with the substrate RHEED spots before growth. By calculating the streaks’ spacing, we found the in-plane lattice constant of LAO and Mn3Ge are about 1% different, consistent with the expected mismatch of the two materials and indicating that the film is relaxed.

FIG. 4.

Crystal structure characterization using x-ray diffraction and TEM. (a) In 2θ – ω scan from 20° to 100°, only Mn3Ge (002), Mn3Ge (004), and substrate peaks are seen. Inset: rocking curve near Mn3Ge (002) peak. (b) In-plane φ scans of LAO {110} and Mn3Ge {203} peaks. The film peaks align with the substrate peaks indicating their epitaxial relation. Cross sections of Mn3Ge films grown by three-step technique (c) and two-step technique (d) demonstrating epitaxial growth. ABAB stacking is dominant for (c) while ABC stacking proliferates in (d). Insets in (c) and (d) indicate how the kagome layers stack with Mn/Ge atoms shown in green for each stacking.

FIG. 4.

Crystal structure characterization using x-ray diffraction and TEM. (a) In 2θ – ω scan from 20° to 100°, only Mn3Ge (002), Mn3Ge (004), and substrate peaks are seen. Inset: rocking curve near Mn3Ge (002) peak. (b) In-plane φ scans of LAO {110} and Mn3Ge {203} peaks. The film peaks align with the substrate peaks indicating their epitaxial relation. Cross sections of Mn3Ge films grown by three-step technique (c) and two-step technique (d) demonstrating epitaxial growth. ABAB stacking is dominant for (c) while ABC stacking proliferates in (d). Insets in (c) and (d) indicate how the kagome layers stack with Mn/Ge atoms shown in green for each stacking.

Close modal

We further characterized the atomic structure of the films using high resolution transmission electron microscopy (HRTEM). Cross-sectional TEM specimens were prepared using a focused ion beam (FIB) lift-out technique. HRTEM images were acquired using the Argonne Chromatic Aberration-corrected TEM (ACAT) (200 kV) equipped with an image corrector to correct both spherical and chromatic aberrations. As shown in Fig. 4(c) for a film grown using the three-step process, a sharp interface between the LAO substrate and Mn3Ge film is observed, and the film is well ordered in with an ABAB stacking sequence for the kagome layers, indicating a predominantly hexagonal phase. On the other hand, for the film grown using only a two-step process (without final layer growth at high temperature), a mix of ABAB and ABC type stacking is observed. The ABAB stacking is observed at the initial growth of the film followed by the ABC stacking, indicating the formation of the low temperature tetragonal phase. This stacking difference has a significant impact on transport properties, as we discuss below.

We measured transport properties of Mn3Ge, as shown in Fig. 5(a). The resistivity at 300 K is about 475 μΩ·cm, substantially larger than that found in bulk single crystals24,51 (150 μΩ·cm), possibly due to remnant gaps between islands in our seed layer. However, the resistivity ratio R300K/R2K ∼ 4.8 is larger than that reported in single crystals, consistent with a high degree of metallicity in our films. From Hall measurements over a macroscopic region of one of our three-step grown films, we inferred a carrier density of 4.9 × 1021/cm3 and a mobility of 2.66 cm2/V s in one of our films. However, these values may be affected by the film morphology and scattering from boundaries. We would need to measure these properties within one of the micrometer-scale single crystalline regions seen in Fig. 3(d) to obtain intrinsic values. Although the hexagonal Mn3Ge is a noncollinear (anti-chiral) antiferromagnet, there is a small net ferromagnetic moment originating from the off-stoichiometry of Mn3+xGe, which is thought to be important for stabilizing the correct structure in bulk samples.52 This net ferromagnetic moment is coupled with the anti-chiral Mn spins on the kagome sites, and by applying an external magnetic field to this moment, the antiferromagnetic structure can be rotated in-plane due to its small in-plane anisotropy, which in turn can be used to tune the Berry curvature. As shown in the inset of Fig. 5(a), we measured the field dependence of the in-plane magnetization in Mn3Ge film. By subtracting a linear background, which is mainly due to the diamagnetic contribution from LAO substrate, we obtain a ferromagnetic moment ∼0.02 μB/Mn at 300 K, comparable with that in bulk24,51 single crystals.

FIG. 5.

(a) Temperature dependence of resistance of Mn3Ge film. Inset: Magnetization curve at 300 K. (b) Schematics of the Nernst device with in-plane magnetic field and out-of-plane thermal gradient. (c) Nernst measurement for Mn3Ge films (grown by both MBE and sputtering) and traditional Ferromagnetic material Fe. (d) Comparison of the Nernst signal for MBE grown Mn3Ge films using three-step and two-step techniques. Inset: A comparison of Mn3Ge (203) peaks for films grown by MBE using three-step and two-step techniques.

FIG. 5.

(a) Temperature dependence of resistance of Mn3Ge film. Inset: Magnetization curve at 300 K. (b) Schematics of the Nernst device with in-plane magnetic field and out-of-plane thermal gradient. (c) Nernst measurement for Mn3Ge films (grown by both MBE and sputtering) and traditional Ferromagnetic material Fe. (d) Comparison of the Nernst signal for MBE grown Mn3Ge films using three-step and two-step techniques. Inset: A comparison of Mn3Ge (203) peaks for films grown by MBE using three-step and two-step techniques.

Close modal

The signature properties of Mn3Ge are its giant transverse transport coefficients such as anomalous Hall and anomalous Nernst effects (AHE and ANE), while having a very small net magnetization. These properties are believed to arise as a result of the Berry curvature from Weyl nodes at various points in the Brillouin zone. While AHE is sensitive to the Berry curvature integrated over all occupied states in a band, ANE is sensitive only to the Berry curvature near the Fermi energy. Considering that the Berry curvature in Mn3Ge lies in the kagome planes, which are also in the plane of our films, our samples are not ideal for measuring the AHE for in-plane currents. However, they are well suited for ANE measurements using an out-of-plane thermal gradient. A schematic of the structure used for measuring the Nernst effect is shown in Fig. 5(b), where a heater wire is integrated with our film, separated by an insulating layer. By applying a low frequency (3 Hz) sinusoidal current through the heater wire, a thermal gradient perpendicular to the film is generated. With the magnetic field applied in the film plane and perpendicular to the length of the device, a Nernst signal is obtained at the second harmonic (6 Hz), which is detected using a lock-in technique. By using Sij = Ei/∇jT, where Ei is the electric field created by ANE in the i direction, ∇jT is the thermal gradient along the j direction, which is governed by Fourier’s law q = −κT (q is the thermal flux density, and κ is the thermal conductivity). Details of the measurement technique can also be found in our previous work.38 As shown in Fig. 5(c), the Nernst coefficient Sxz (x along 101̄0) at 300 K in our MBE grown film is about 0.75 μV/K, comparable to values reported for single crystals of Mn3Ge (∼0.45 μV/K at 300 K, 1.6 μV/K at 100 K)23 and more than six times larger than that measured in sputtered Mn3Ge films (∼0.11 μV/K)38 and also larger than in sputtered Fe films (∼0.42 μV/K).45 However, for a film grown using the first two steps only (seed layer and connecting layer), the overall Nernst signal is smaller, and the coercive field needed for ANE reversal increases, as shown in Fig. 5(d). As we observe in our TEM measurements, without the third high-temperature growth step, there is a higher concentration of defects in our films where ABAB stacking does not persist over long range and regions of ABC stacking (tetragonal phase) proliferate. Besides decreasing the Nernst signal, these defects also pin the antiferromagnetic domains, which then require stronger magnetic fields to switch. Besides the TEM characterizations in Figs. 4(c) and 4(d), the increase of defects can also be inferred from the 2θ ∼ ω scan of the hexagonal Mn3Ge (203) peak. As shown in the inset of Fig. 5(d), the peak intensity is much lower for the film grown using the three-step technique. This indicates a greater degree of disorder, consistent with our TEM measurements.

In summary, we have successfully synthesized an epitaxial Mn3Ge thin film directly on insulating oxide substrates LAO (111) using a three-step technique with the assistance of electron beam irradiation. With only a small ferromagnetic moment (0.02 μB/Mn), the film shows a large ANE signal at 300 K, vastly improved over sputtered Mn3Ge films and comparable with that in bulk single crystals. We relate this improvement to the mitigation of defects—both morphological and structural—via our three-step process. Without any metallic buffer layers, our films will enable intrinsic spin and charge transport measurements for probing mesoscale and non-local phenomena in this class of materials. Due to the difficulties in cleaving Mn3Ge bulk samples, our films with an atomic smooth surface also provide a path to surface sensitive characterizations such as angle-resolved photoelectron spectroscopy (ARPES) and scanning tunneling microscopy (STM) for probing the electronic structure of these topological materials. Our approach may be applicable to a range of kagome, 2D, and other materials that are challenging to grow on insulating substrates due to wetting issues.

The research presented here, including work done by D.H., C.L., Q.D., J.S.J., and A.B., was supported by the Center for the Advancement of Topological Semimetals, an Energy Frontier Research Center funded by the U.S. DOE, Office of Basic Energy Sciences, through Ames Laboratory under its Contract No.. DE-AC02-07CH11358. J.E.P. acknowledges the Materials Science and Engineering Division, U.S. DOE, Office of Basic Energy Sciences, for assisting with synthesis and characterization. Use of the Center for Nanoscale Materials, a DOE Office of Science User Facility, was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No.. DE-AC02-06CH11357. The authors thank Mr. Kenny Huynh for TEM sample preparation using FIB.

The authors have no conflicts to disclose.

Synthesis of Mn3Ge films using MBE, including all structural characterization, was carried out by D.H. with assistance from J.P., Q.D., B.F. and A.B. All transport and thermogalvanic measurements were carried out by D.H. with assistance from C.L. and B.F. The TEM measurements were carried out by J.W. Sputtered Mn3Ge films were prepared by J.S.J.. The project was supervised by A.B. This paper was written by D.H. and A.B., and all authors contributed to discussions regarding the manuscript.

Deshun Hong: Conceptualization (equal); Data curation (lead); Formal analysis (lead); Investigation (lead); Methodology (lead); Validation (lead). Changjiang Liu: Formal analysis (supporting); Investigation (supporting); Visualization (supporting); Writing – review & editing (supporting). Jianguo Wen: Data curation (equal); Investigation (supporting); Validation (equal); Writing – review & editing (supporting). Qianheng Du: Investigation (supporting); Writing – review & editing (supporting). Brandon L Fisher: Investigation (supporting); Writing – review & editing (supporting). Jidong Jiang: Investigation (supporting); Writing – review & editing (supporting). John E. Pearson: Investigation (supporting); Writing – review & editing (supporting). Anand Bhattacharya: Conceptualization (equal); Funding acquisition (lead); Investigation (supporting); Project administration (lead); Resources (lead); Supervision (lead); Writing – original draft (supporting); Writing – review & editing (equal).

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

1.
M.
Konig
,
S.
Wiedmann
,
C.
Brune
,
A.
Roth
,
H.
Buhmann
,
L. W.
Molenkamp
,
X. L.
Qi
, and
S. C.
Zhang
,
Science
318
,
766
(
2007
).
2.
D.
Hsieh
,
D.
Qian
,
L.
Wray
,
Y.
Xia
,
Y. S.
Hor
,
R. J.
Cava
, and
M. Z.
Hasan
,
Nature
452
,
970
(
2008
).
3.
Z.
Wang
,
Y.
Sun
,
X.-Q.
Chen
,
C.
Franchini
,
G.
Xu
,
H.
Weng
,
X.
Dai
, and
Z.
Fang
,
Phys. Rev. B
85
,
195320
(
2012
).
4.
X.
Wan
,
A. M.
Turner
,
A.
Vishwanath
, and
S. Y.
Savrasov
,
Phys. Rev. B
83
,
205101
(
2011
).
6.
C. Z.
Chang
 et al,
Science
340
,
167
(
2013
).
7.
M. S.
Bahramy
 et al,
Nat. Commun.
3
,
1159
(
2012
).
8.
M.
Goyal
,
L.
Galletti
,
S.
Salmani-Rezaie
,
T.
Schumann
,
D. A.
Kealhofer
, and
S.
Stemmer
,
APL Mater.
6
,
026105
(
2018
).
9.
C. Z.
Xu
 et al,
Phys. Rev. Lett.
118
,
146402
(
2017
).
10.
Y.
Deng
,
Y.
Yu
,
M. Z.
Shi
,
Z.
Guo
,
Z.
Xu
,
J.
Wang
,
X. H.
Chen
, and
Y.
Zhang
,
Science
367
,
895
(
2020
).
11.
S. A.
Parameswaran
,
T.
Grover
,
D. A.
Abanin
,
D. A.
Pesin
, and
A.
Vishwanath
,
Phys. Rev. X
4
,
031035
(
2014
).
12.
Y.
Wang
,
P.
Deorani
,
K.
Banerjee
,
N.
Koirala
,
M.
Brahlek
,
S.
Oh
, and
H.
Yang
,
Phys. Rev. Lett.
114
,
257202
(
2015
).
13.
J. H.
Han
,
A.
Richardella
,
S. A.
Siddiqui
,
J.
Finley
,
N.
Samarth
, and
L. Q.
Liu
,
Phys. Rev. Lett.
119
,
077702
(
2017
).
14.
K.-R.
Jeon
,
B. K.
Hazra
,
K.
Cho
,
A.
Chakraborty
,
J.-C.
Jeon
,
H.
Han
,
H. L.
Meyerheim
,
T.
Kontos
, and
S. S. P.
Parkin
,
Nat. Mater.
20
,
1358
(
2021
).
16.
17.
M.
Nakano
,
Y.
Wang
,
Y.
Kashiwabara
,
H.
Matsuoka
, and
Y.
Iwasa
,
Nano Lett.
17
,
5595
(
2017
).
18.
19.
M.
Chubarov
 et al,
ACS Nano
15
,
2532
(
2021
).
20.
Y.
Zhang
,
Y.
Sun
,
H.
Yang
,
J.
Zelezny
,
S. P. P.
Parkin
,
C.
Felser
, and
B. H.
Yan
,
Phys. Rev. B
95
,
075128
(
2017
).
21.
H.
Yang
,
Y.
Sun
,
Y.
Zhang
,
W.-J.
Shi
,
S. S. P.
Parkin
, and
B.
Yan
,
New J. Phys.
19
,
015008
(
2017
).
22.
K.
Kuroda
 et al,
Nat. Mater.
16
,
1090
(
2017
).
23.
C.
Wuttke
 et al,
Phys. Rev. B
100
,
085111
(
2019
).
24.
A. K.
Nayak
 et al,
Sci. Adv.
2
,
1501870
(
2016
).
25.
M.
Ikhlas
,
T.
Tomita
,
T.
Koretsune
,
M.-T.
Suzuki
,
D.
Nishio-Hamane
,
R.
Arita
,
Y.
Otani
, and
S.
Nakatsuji
,
Nat. Phys.
13
,
1085
(
2017
).
26.
28.
S.
Nakatsuji
,
N.
Kiyohara
, and
T.
Higo
,
Nature
527
,
212
(
2015
).
29.
T. S.
Chen
 et al,
Nat. Commun.
12
,
4562
(
2021
).
30.
A.
Markou
,
J. M.
Taylor
,
A.
Kalache
,
P.
Werner
,
S. S. P.
Parkin
, and
C.
Felser
,
Phys. Rev. Mater.
2
,
051001(R)
(
2018
).
31.
D.
Hong
,
C.
Liu
,
H.-W.
Hsiao
,
D.
Jin
,
J. E.
Pearson
,
J.-M.
Zuo
, and
A.
Bhattacharya
,
AIP Adv.
10
,
105017
(
2020
).
32.
H.
Inoue
,
M.
Han
,
L.
Ye
,
T.
Suzuki
, and
J. G.
Checkelsky
,
Appl. Phys. Lett.
115
,
072403
(
2019
).
33.
35.
T.
Ogasawara
,
J.-y.
Kim
,
Y.
Ando
, and
A.
Hirohata
,
J. Magn. Magn. Mater.
473
,
7
(
2019
).
36.
T.
Matsuda
,
N.
Kanda
,
T.
Higo
,
N. P.
Armitage
,
S.
Nakatsuji
, and
R.
Matsunaga
,
Nat. Commun.
11
,
909
(
2020
).
37.
T.
Higo
,
D.
Qu
,
Y.
Li
,
C. L.
Chien
,
Y.
Otani
, and
S.
Nakatsuji
,
Appl. Phys. Lett.
113
,
202402
(
2018
).
38.
D. S.
Hong
,
N.
Anand
,
C. J.
Liu
,
H. H.
Liu
,
I.
Arslan
,
J. E.
Pearson
,
A.
Bhattacharya
, and
J. S.
Jiang
,
Phys. Rev. Mater.
4
,
094201
(
2020
).
39.
J. M.
Taylor
,
A.
Markou
,
E.
Lesne
,
P. K.
Sivakumar
,
C.
Luo
,
F.
Radu
,
P.
Werner
,
C.
Felser
, and
S. S. P.
Parkin
,
Phys. Rev. B
101
,
094404
(
2020
).
40.
L.
Galletti
,
T.
Schumann
,
O. F.
Shoron
,
M.
Goyal
,
D. A.
Kealhofer
,
H.
Kim
, and
S.
Stemmer
,
Phys. Rev. B
97
,
115132
(
2018
).
41.
Y.
Zhang
,
Z. K.
Liu
,
B.
Zhou
,
Y.
Kim
,
Z.
Hussain
,
Z. X.
Shen
,
Y. L.
Chen
, and
S. K.
Mo
,
Appl. Phys. Lett.
105
,
031901
(
2014
).
42.
J.
Hellerstedt
,
M. T.
Edmonds
,
N.
Ramakrishnan
,
C.
Liu
,
B.
Weber
,
A.
Tadich
,
K. M.
O’Donnell
,
S.
Adam
, and
M. S.
Fuhrer
,
Nano Lett.
16
,
3210
(
2016
).
43.
I.
Lyalin
,
S.
Cheng
, and
R. K.
Kawakami
,
Nano Lett.
21
,
6975
(
2021
).
44.
F.-f.
Zhu
,
W.-j.
Chen
,
Y.
Xu
,
C.-l.
Gao
,
D.-d.
Guan
,
C.-h.
Liu
,
D.
Qian
,
S.-C.
Zhang
, and
J.-f.
Jia
,
Nat. Mater.
14
,
1020
(
2015
).
45.
C. J.
Liu
 et al,
Nat. Commun.
11
,
1402
(
2020
).
46.
D.
Hong
,
C.
Liu
,
J.
Pearson
, and
A.
Bhattacharya
,
Appl. Phys. Lett.
111
,
232408
(
2017
).
47.
B. A.
Assaf
,
T.
Phuphachong
,
V. V.
Volobuev
,
G.
Bauer
,
G.
Springholz
,
L. A.
de Vaulchier
, and
Y.
Guldner
,
npj Quantum Mater.
2
,
26
(
2017
).
48.
B.
Heying
,
R.
Averbeck
,
L. F.
Chen
,
E.
Haus
,
H.
Riechert
, and
J. S.
Speck
,
J. Appl. Phys.
88
,
1855
(
2000
).
49.
T.
Cheng
,
A.
Summerfield
,
C.
Mellor
,
A.
Khlobystov
,
L.
Eaves
,
C.
Foxon
,
P.
Beton
, and
S.
Novikov
,
Materials
11
,
1119
(
2018
).
50.
J. F.
Qian
,
A. K.
Nayak
,
G.
Kreiner
,
W.
Schnelle
, and
C.
Felser
,
J. Phys. D: Appl. Phys.
47
,
305001
(
2014
).
51.
N.
Kiyohara
,
T.
Tomita
, and
S.
Nakatsuji
,
Phys. Rev. Appl.
5
,
064009
(
2016
).
52.
N.
Yamada
,
H.
Sakai
,
H.
Mori
, and
T.
Ohoyama
,
Physica B&C
149
,
311
(
1988
).