Epitaxial growth of antiferromagnetic MnBi₂Te₄/CdTe heterostructures on GaAs(001) using molecular beam epitaxy: structure and electronic properties

Abstract

MnBi₂Te₄ is one of the most recent materials that integrates the class of topological quantum materials exhibiting topological insulating properties and magnetic ordering, thus providing the opportunity to investigate particular topological quantum states and design novel spintronic devices. The samples were grown on GaAs(100) substrates using molecular beam epitaxy with Bi₂Te₃, Te and Mn as sources. The growth was characterized through X-ray diffraction, atomic force microscopy and transmission electron microscopy. Topological insulator properties were probed through scanning tunneling microscopy and scanning tunneling spectroscopy, while antiferromagnetic order was investigated via magnetotransport measurements. Furthermore, we propose a high-quality CdTe thin film as a cap layer to protect topological surface states, thus opening the possibility of integrating topological quantum materials with group II–VI semiconductors via van der Waals epitaxy.

---

Leonarde N. Rodrigues

Leonarde do Nascimento Rodrigues received his doctoral degree from Federal University of Juiz de Fora, Juiz de Fora, Brazil, in 2016. He is currently Professor in the Physics Department at the Federal University of Viçosa, Brazil. His main current research area is the conventional epitaxial growth of semiconductor heterostructures and thin films, van der Waals epitaxial growth of chalcogenide materials, fabrication of semiconductor nanomembranes and devices and their characterization via high resolution X-ray diffraction and optical spectroscopy for inelastic light scattering and photoluminescence applications.

---

Introduction

Novel materials exhibiting exotic states of matter, such as topological insulators (TIs), have attracted considerable attention in solid-state physics in the last decades.¹ The quantum spin Hall effect observed in HgTe quantum wells² grown via molecular beam epitaxy (MBE) confirmed typical TI behavior with protected conductor surface states due to spin–orbit coupling and time-reversal symmetry. Such characteristics make this material a promising candidate for sophisticated spintronic devices and quantum computation applications.³ In recent years, the combination of topological insulation properties with magnetic ordering has enabled the emergence of a new class of topological quantum materials.⁴ The magnetic moment provides mass to the otherwise massless Dirac fermions, thus opening an energy gap and allowing topologically driven quantum states to emerge in the system.⁵ Exotic states of matter such as quantum anomalous Hall effect (QAHE)⁶ insulators and axion insulators⁷ have been observed as a response to the coexistence between magnetism and topological insulator state.

Magnetically doped topological insulators, such as Cr- and V-doped (Bi,Sb)₂Te₃, present the QAHE.⁸ However, the temperature to measure the effect is <1 K owing to structural defects in the Bi₂Te₃ crystal introduced by dopants. Magnetic topological insulators such as MnBi₂Te₄ with antiferromagnetic order⁹ are an alternative to magnetically doped TIs since they are expected to exhibit the QAHE at relatively high temperatures, thus enabling device applications. The QAHE has also been predicted for thin films/flakes of MnBi₂Te₄-family compounds.¹⁰

Quantum transport properties of MnBi₂Te₄ have been exploited to achieve the QAHE at zero magnetic field in thin film samples grown via MBE.⁹ Recently, a study addressed QAHE results¹¹ to clarify the controversial data available. Although the QAHE has been reported in thin flakes⁶ of MnBi₂Te₄ and giant nonlocal transport signals in the axion insulator state,¹² a thin film grown via MBE does not present limitations with the control in the fabrication of the film shape and does not have the size limitation as in the case of exfoliated flakes. The reproducibility and control of composition and thickness positions the MBE growth technique as an appropriate choice to achieve the fabrication of large-area and high-quality topological magnetic devices.^9,11 Previous works in this field have been performed using the MBE growth of MnBi₂Te₄ on different substrates such as BaF₂(111),¹³ sapphire(0001),¹¹ Si(111),¹⁴ Al₂O₃(0001),¹⁵ SrTiO₃(111)¹⁶ and GaAs(111)B.¹⁷ A common link between these works is the use of two effusion cells containing Bi₂Te₃ and MnTe or three effusion cells containing Te, Bi and Mn. The former¹³ basically controls the beam equivalent pressure between Bi₂Te₃ and MnTe. Other works,^15,17 since MnBi₂Te₄ is sensitive to the Bi : Mn flux ratio, provide the fine stoichiometry tuning of MnBi₂Te₄, controlling each source individually_. Some approaches may follow a more complex routine involving growth and annealing of a Bi₂Te₃ buffer layer.¹⁷

Although van der Waals (vdW) material growth conditions are challenging considering weak interlayer forces, good quality systems can be grown via MBE, despite the large lattice mismatch between the film and substrate.¹⁸ Careful control of growth conditions is necessary for the formation of correct stoichiometry and phase. In this work, MnBi₂Te₄ thin films were directly grown on a GaAs(001) substrate using molecular beam epitaxy. GaAs is a suitable substrate choice used in spintronics and can be a platform for integrating topological insulator devices.¹⁹ The growth recipe used here is similar to that reported previously²⁰ for the growth of Bi₂Te₃ using one additional Mn effusion cell. This versatile control can allow the growth of heterostructures and superstructures such as Bi₂Te₃/MnBi₂Te₄ with only three effusion cells. To avoid the exposure of topological insulators to air and prevent the degradation of surface states, we grew a CdTe cap layer on top of the MnBi₂Te₄ film, substituting the usually used Te cap because the metallic Te layer would work as a transport channel, masking the transport that occurs through the MnBi₂Te₄ film.²¹ This high-resistivity CdTe layer opens the possibility to integrate TIs with II-VI heterostructures. The morphological and structural properties of MnBi₂Te₄ are characterized using atomic force microscopy (AFM), high resolution X-ray diffraction (HRXRD) and high-resolution transmission electron microscopy (HRTEM). Topological surface state properties were studied using scanning tunneling spectroscopy (STS), thus providing information about the surface local density of states (LDOS). To verify the antiferromagnetic order and electronic properties of MnBi₂Te₄, magnetoresistance and van der Pauw measurements were conducted.

Experimental

Bi₂Te₃ and MnBi₂Te₄ were grown using a custom-built molecular beam epitaxy (MBE) system on undoped GaAs(001) substrates. Before growth, the substrate was annealed at 580 °C for 5 min to completely remove native oxide. Standard effusion cells containing high-purity powders of Bi₂Te₃, Te, Mn, and CdTe were used for growth. Growth parameters were the same as those already reported²⁰ for the growth of Bi₂Te₃, with the addition of a Mn source. Unlike previously mentioned works on the growth of MnBi₂Te₄ on different substrates, our growth recipe uses independent sources of Bi₂Te₃, Te and Mn. The same beam fluxes of Bi₂Te₃ and Te were maintained by adjusting the same deposition rate grown at room temperature and measuring the thickness of pure Bi₂Te₃ and Te films. The MnBi₂Te₄ structure (deposition rate of approximately 0.5 nm min⁻¹) was reached using pure Mn evaporated from an effusion cell with an optimized temperature of 840 °C. The substrate temperature used was 380 °C for MnBi₂Te₄ and CdTe thin films. High-purity polycrystalline CdTe was evaporated from single effusion cells with a deposition rate of approximately 3 nm min⁻¹. The base pressure never exceeded 10⁻⁸ Torr. Bi₂Te₃ samples were grown for comparison. To avoid the exposure of the topological insulator to air, we grew an undoped 60 nm-thick CdTe cap layer on top of the MnBi₂Te₄ film. The crystalline structure of the Bi₂Te₃ and MnBi₂Te₄ films was investigated through high-resolution X-ray diffraction (HRXRD) using a four-circle Bruker D8-discover diffractometer equipped with Cu Kα (λ = 1.54056 Å) radiation source and a 2-bounce Ge (220) monochromator. The deposition rate and interface roughness were estimated using X-ray reflectometry (XRR) measurements. XRR fits using the Parratt formalism²² or the available algorithm²³ allows the evaluation of the layer thickness and interface roughness of the deposited layer. AFM measurements were carried out using a FlexAFM Nanosurf microscope operating in a tapping (semi-contact) mode with controlled humidity maintained below 30%. Cross-sectional focused ion beam (FIB) thin lamellae were prepared using an FEI Helios Nanolab 650 DualBeam system, with final polishing conducted at 2 keV and a 7° tilt to minimize amorphization caused by the Ga FIB. Scanning transmission electron microscopy (STEM) was performed using a Cs probe-corrected Titan 80-300 (FEI Company) transmission electron microscope operating at 300 kV. The microscope was equipped with a conventional high-angle annular dark-field (HAADF) detector and an Oxford Aztec Energy TEM advanced microanalysis system. STM and STS measurements were performed under ultra-high vacuum conditions in an Omicron-VT STM system operating at room temperature, with a base pressure of 1.0 × 10⁻¹⁰ mbar. All STM images were acquired using electrochemically etched polycrystalline W tips in a constant current mode, and for STS measurements, a lock-in amplifier (operating a 3000 Hz) was used to obtain differential conductance (dI/dV) curves. All STS curves considered here were an average of at least 30 curves taken at the same point. For STS measurements, the samples were capped with Te and immediately exfoliated in vacuum before STS measurements. For magnetotransport measurements, electrical contacts in the van der Pauw geometry were soldered using indium droplets. This procedure was performed using a hot plate at 180 °C for 2 min to guarantee that the diffusion occurs through the layers. The measurements were carried out using the physical properties measurement system (PPMS) from Quantum Design, equipped with a helium-cooled superconducting system with a magnetic field of up to 9 T that operates over a broad temperature range from 1.9 to 400 K. A standard four-probe AC lock-in technique with a constant excitation current of 10 μA was employed.

Results and discussion

Fig. 1(a) and (e) show the XRR data for Bi₂Te₃ and MnBi₂Te₄ thin films without the CdTe cap layer, with experimental data shown as a black circle and the fitting curve as a blue line. Both samples exhibited well-defined fringes, indicating a uniform film thickness and smooth surface and interface across the substrate area. The model parameters yielded a Bi₂Te₃ film thickness of 47.3 nm and surface roughness of 2.6 nm. The MnBi₂Te₄ film had a similar thickness of 48.2 nm with a surface roughness of approximately 10% smaller than Bi₂Te₃. The comparable film thickness resulted from using the same growth time and parameters for the Bi₂Te₃ and Te effusion cells. The fitting curves were obtained using the bulk material density, and results were in good agreement with experimental results, suggesting that a stoichiometric phase was achieved.

Fig. 1 Structural comparison of Bi₂Te₃ and MnBi₂Te₄ TI. (a) and (e): XRR spectra for Bi₂Te₃ and MnBi₂Te₄. Black circles represent the experimental data, while blue lines represent the fitting curve. (b) and (f): HRXRD measurements for the [0001] direction for Bi₂Te₃ and MnBi₂Te₄. (c) and (g): reciprocal space mapping around Bi₂T₃(006) and MnBi₂Te₄(009) reflections and GaAs(002) as a reference indicated by a circle. (d) and (h) Show the RSM around GaAs(002) from (c) and (g), respectively.

Fig. 1(b) and (f) show symmetrical HRXRD scans for 47 nm Bi₂Te₃ and 48 nm MnBi₂Te₄ thick films, respectively. In addition to the peaks belonging to the GaAs substrate, it is possible to observe only the (0 0 3n) peaks as labeled in the figure, indicating that both films grew epitaxially on the GaAs substrate. Fig. 1(c) and (g) show the reciprocal space maps (RSMs) acquired near the symmetric Bi₂Te₃ (006) and MnBi₂Te₄ (009) diffraction peak, respectively, and the (002) diffraction peak of the GaAs as a reference (indicated by the circle). Note that the (002) diffraction peak of GaAs is a nearly perfect single crystal, which reflects the discrepancy between the width of the Bi₂Te₃ and MnBi₂Te₄ films and the GaAs substrate as can be observed on the RSM around GaAs (002) in Fig. 1(d) and (h). The vertical alignment between the Bi₂Te₃ and MnB₂Te₄ films and GaAs indicates well-oriented growth with reflections in the (003n) parallel plane. The distortion of GaAs reflection on the RSM originates from the resolution limit of the equipment. The out-of-plane direction reveals symmetric oscillation patterns that confirm the smooth surfaces and a well-ordered structure. In-plane direction has an intense diffuse scattering, indicating the presence of defects in the structure, such as mosaicity, due to the large lattice mismatch. The out-of-plane direction of the MnBi₂Te₄(009) RSM, shown in Fig. 1(g), is asymmetric with a more pronounced broadening at lower values of reciprocal space. The in-plane diffraction has a FWHM that is twice as large as that of Bi₂Te₃, indicating a larger defect density. Stacking errors may be responsible for the asymmetric behavior and the broadening of the diffraction peaks.

Fig. 2(a, b) and (e, f) shows AFM images (5 × 5 μm²) and (1 × 1 μm²) for the Bi₂Te₃ and MnBi₂Te₄ thin films, respectively. The images are presented using a shading effect that highlights step-like features, such as the triangular terraces that are typical of Bi₂Te₃ samples,²⁰ as shown in Fig. 2(b). These triangular structures are also observed on the surface of the MnBi₂Te₄ film, as shown in Fig. 2(c), but lack distinct steps and reach a height of up to 20 nm. The presence of triangular shapes on the surface indicates regions where adatoms have less mobilities and thus form clusters, already observed in others van der Waals materials.²⁴ The van der Waals growth mode allows growth with good out-of-plane orientation even in a large mismatch substrate such as cubic GaAs. However, the in-plane rotation responsible for the defects observed in the RSM can be easily identified by the misalignment of the triangular features shown in Fig. 2(a, b) and (e, f) for Bi₂Te₃ and MnBi₂Te₄, respectively. The red dotted lines in HRTEM images of Fig. 2(c) and (g) indicate the quintuple layer (QL) of Bi₂Te₃ and the septuple layer (SL) of MnBi₂Te₄. Fig. 2(d) and (h) are higher resolution images showing details of the QL (Te–Bi–Te–Bi–Te) and SL (Te–Bi–Te–Mn–Te–Bi–Te) separated by the van der Waals gap, which form the Bi₂Te₃ and MnBi₂Te₄ structures, respectively. From reflection (003)-Bi₂Te₃ and (006)-MnBi₂Te₄ obtained from HRXRD data and estimates using data from HRTEM, the value estimated for the QL thickness was approximately 1.0 nm, while that for the SL was approximately 1.4 nm. These results were in good agreement with other experimental results.^20,25

Fig. 2 (a and b) AFM surface and (c and d) HRTEM images of Bi₂Te₃ thin films. The triangular structures with flat terraces and observable steps corresponding to the quintuple layers of Bi₂Te₃ are clearly shown in (b). (e and f) AFM surface and (g and h) HRTEM images of MnBi₂Te₄ thin films. MnBi₂Te₄ exhibits a smoother surface with grains and flat triangle structures. The maximum peak height S_p (which corresponds to the maximum height with respect to the mean height) for AFM images (a), (b), (e), and (f) are 24 nm, 16 nm, 110nm and 75 nm, respectively. The red dotted line corresponds to the (c) Bi₂Te₃ quintuple layers and (g) MnBi₂Te₄ septuple layers.

STM and STS measurements were conducted to investigate the surface electronic properties of the Bi₂Te₃ and MnBi₂Te₄ thin films. In a typical STS curve, the differential tunneling conductance (dI/dV) is proportional to the local density of states (LDOS) at the STM tip position;²⁶ therefore, it allows us to probe the presence of topological surface states on the MBE grown thin film surface.

A comparison between the LDOS of Bi₂Te₃ and MnBi₂Te₄ is presented in Fig. 3. Fig. 3(a) shows a series of dI/dV curves (shaded blue region) taken along a line scan with different positions on the Bi₂Te₃ surface at room temperature, superposed with the average dI/dV curve (solid blue line). In all measured spectra, the region of positive bias voltage denotes the bulk conduction band (BCB) and the region of negative bias voltage denotes the bulk valence band (BVB), thus confirming a homogenous behavior of Bi₂Te₃ LDOS throughout the sample surface. The presence of a finite density of states near the Fermi energy (E_F = 0) that varies linearly with the bias voltage is a signature of topologically protected surface states, confirming the topological properties of MBE-grown Bi₂Te₃ thin films.^26–29 Moreover, by extrapolating the linear region of the dI/dV curve to the zero conductance axis, it is possible to obtain a Dirac point energy (E_D) of −190 meV, which indicates a lightly n-type doping character of Bi₂Te₃ thin films. Fig. 3(b) shows an averaged dI/dV curve (solid blue line) taken along different points on the MnBi₂Te₄ surface (blue shaded region). In this case, the topological properties of MBE-grown MnBi₂Te₄ thin films can be confirmed by the presence of a finite density of states from −0.3 eV to the Fermi level (E_F = 0), where the E_D is obtained considering the minimum of the dI/dV curve as already been reported in STS measurements in MnBi₂Te₄ bulk single crystals and thin films.^27,30 Because the Bi₂Te₃ growth parameters are the starting point to obtain MnBi₂Te₄ with correct stoichiometry, the STS curve indicates that the n-type doping is still present in the MnBi₂Te₄ phase with the Dirac point located around −210 meV below the Fermi level. Indeed, since at room temperature, antiferromagnetic ordering is not expected to occur, time-reversal symmetry is preserved at the surface of MnBi₂Te₄, maintaining its topological properties.²⁷ The value for the QL (Bi₂Te₃) and SL (MnBi₂Te₄) measured via STM, as shown in the insets of Fig. 3(a) and (b), was 1.01 nm and 1.35 nm, respectively, in agreement with those mentioned previously.

Fig. 3 A series of dI/dV spectra (shaded blue region) and spatially averaged dI/dV curve (solid blue line) obtained along the Bi₂Te₃ (a) and MnBi₂Te₄ surface (b). In both graphs, the zero- bias voltage represents the Fermi level (E_F) and the red arrow denotes the Dirac point energy (E_D). In panel (a), the linear conductance region (dashed red line) indicates the presence of topological surface states (TSSs). Inset shows the STM topographic image where the STS curves were measured, evidencing the QL termination of Bi₂Te₃ and the SL termination of MnBi₂Te₄ with step heights of 1.01 and 1.35 nm, respectively. STM images were acquired using I = 400 pA and V = 0.8 V.

To prove the possibility of integrating MnBi₂Te₄ with group II–VI semiconductors with epitaxial quality, 60 nm thick CdTe was grown on the top of MnBi₂Te₄ as an insulating cap layer owing to its high resistivity. The HRXRD of this sample is shown in Fig. 4(a). With the exception of the CdTe film reflections known and labeled by CdTe(111) and CdTe(222), all others are identical to those depicted in Fig. 1(f). The film growth aligned with the 00L planes of MnBi₂Te₄ and showed high crystalline quality. The growth dynamics of the zinc blende crystal CdTe layer on top of the van der Waals material has not been studied; however, it is remarkable that a highly crystalline CdTe layer can be grown onto the MnBi₂Te₄ film with such a large lattice mismatch. The RSM around CdTe(111) and MnBi₂Te₄(009) was performed and is shown in Fig. 4(b). The Bragg diffraction peak displays a broader peak in the in-plane direction for the CdTe(111) and MnBi₂Te₄(009) peaks. The scattering along the crystal truncation rod of CdTe(111) is symmetrical, contrary to MnBi₂Te₄(009), which is larger for in-plane smaller values.

Fig. 4 (a) HRXRD and (b) reciprocal map of the GaAs/MnBi₂Te₄/CdTe heterostructure.

In Fig. 5, we present the results of electrical and magnetotransport measurements for the MnBi₂Te₄ thin film with a the CdTe cap layer (Fig. 4). As CdTe is highly resistive (usually R_CdTe > 100 MΩ), it does not contribute to the measurements. A schematic of the measurement configuration is shown in the inset of Fig. 5(a). Longitudinal electric resistance (R_xx) as a function of temperature is shown in Fig. 5(a), showing the antiferromagnetic transition temperature that manifests as a peak in the R_xx around 19 K. As this value is very close to values reported in the literature for a few SL layers of MnBi₂Te₄ exfoliated flakes^16,25 and for thin films grown via MBE on Si(111),¹⁴ it is not clear whether the CdTe cap layer or defects could play some role in the critical temperature of this transition. At low temperatures (<10 K), one observes an increase in R_xx as the temperature is decreased. This is explained by localization effects caused by disorder in the crystal. Moreover, the metallic behavior of the R_xx(T) curve (Fig. 5(a)) shows that electrical transport occurs via the MnBi₂T₄ layer; otherwise, one would observe a semiconductor-like behavior instead of metallic behavior. Fig. 5(b) shows normalized electrical resistance as a function of temperature for different applied magnetic fields. For clarity, the curves were shifted. The applied magnetic field dislocates the transition to lower temperatures, as expected, because the applied magnetic field tends to suppress antiferromagnetic alignment. It is also possible to observe that the transition is suppressed for B > 3T. Fig. 5(c) presents the transversal electrical resistance (R_xy) as a function of the applied magnetic field. At high temperatures, 20 and 30 K, one observes the linear behavior expected for a single channel transport in this system.³¹ However, for lower temperatures, 2.1, 6 and 10 K, a dip in the R_xy curves around 3T is clearly observed. In this temperature region, the sample is already in the antiferromagnetic state. The dip in the R_xy curves indicates that for B > 3T, the sample suffers a transition from antiferromagnetic to ferromagnetic.

Fig. 5 (a) Electrical resistance as a function of temperature, where the AFM transition is clear around 19 K. The inset illustrates the diffusion of indium through the CdTe layer until it reaches the MnBi₂T₄ film, forming ohmic contact for the measurements. (b) Normalized electrical resistance as a function of temperature for different applied magnetic fields, where it is possible to observe the suppression of the AFM transition for higher fields. (c) The transversal electrical resistence showing a transition from AFM to FM for B > 3T.

Conclusions

In summary, we studied MnBi₂Te₃ thin layers with topological insulator and ordered magnetic properties. High-quality MnBi₂Te₃ layers were obtained on GaAs(001) substrates using a recently published Bi₂Te₃ growth recipe²⁰ by adding only Mn with an effusion cell to achieve the correct MnBi₂Te₄ stoichiometry. A CdTe layer was used as a cap layer with high epitaxial quality and integration potential with the II–VI semiconductor group through van der Waals growth. Moreover, the metallic behavior of the electrical resistance curve shows that the CdTe cap layer was successfully integrated, thus exhibiting a clean and noise-free signal. GaAs has consolidated industrial fabrication processes and is commonly utilized in existing optical-electronic devices. The integration of MnBi₂Te₄thin films on the GaAs substrate with II–VI semiconductors establishes a path to intrinsic magnetic topological insulator devices based on the integration between van der Waals materials.

Author contributions

Wesley F. Inoch, Sukarno O. Ferreira and Leonarde N. Rodrigues were responsible for the growth of MnBi₂Te₄via MBE and conducted HRXRD and reciprocal map measurements. Gilberto Rodrigues-Junior and Ângelo Malachias were responsible for STM and STS measurements. S. L. A. Mello carried out AFM measurements. S. de Castro and M. L. Peres carried out transport measurements. Bráulio S. Archanjo, Maybi F. Sampaio and Olavo Teixeira Neto conducted the HRTEM measurements.

Data availability

Data are available upon request from the authors.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We acknowledge the financial support by the Brazilian funding agencies CAPES, CNPq (under Grant 408093/2022-6), and FAPEMIG (under Grant APQ-02500-22, APQ-00388-22 and RED-00223-23).

References

1. M. Z. Hasan and C. L. Kane, Rev. Mod. Phys., 2010, 82, 3045–3067.
2. M. König, S. Wiedmann, C. Brüne, A. Roth, H. Buhmann, L. W. Molenkamp, X.-L. Qi and S.-C. Zhang, Science, 2007, 318, 766–770.
3. O. Breunig and Y. Ando, Nat. Rev. Phys., 2021, 4, 184–193.
4. B. A. Bernevig, C. Felser and H. Beidenkopf, Nature, 2022, 603, 41–51.
5. Y. Tokura, K. Yasuda and A. Tsukazaki, Nat. Rev. Phys., 2019, 1, 126–143.
6. Y. Deng, Y. Yu, M. Z. Shi, Z. Guo, Z. Xu, J. Wang, X. H. Chen and Y. Zhang, Science, 2020, 367, 895–900.
7. D. Zhuo, Z.-J. Yan, Z.-T. Sun, L.-J. Zhou, Y.-F. Zhao, R. Zhang, R. Mei, H. Yi, K. Wang, M. H. W. Chan, C.-X. Liu, K. T. Law and C.-Z. Chang, Nat. Commun., 2023, 14, 7596.
8. C.-Z. Chang, J. Zhang, X. Feng, J. Shen, Z. Zhang, M. Guo, K. Li, Y. Ou, P. Wei, L.-L. Wang, Z.-Q. Ji, Y. Feng, S. Ji, X. Chen, J. Jia, X. Dai, Z. Fang, S.-C. Zhang, K. He, Y. Wang, L. Lu, X.-C. Ma and Q.-K. Xue, Science, 2013, 340, 167–170.
9. Q. Li, S.-K. Mo and M. T. Edmonds, Nanoscale, 2024, 16, 14247–14260.
10. M. M. Otrokov, I. P. Rusinov, M. Blanco-Rey, M. Hoffmann, A. Yu Vyazovskaya, S. V. Eremeev, A. Ernst, P. M. Echenique, A. Arnau and E. V. Chulkov, Phys. Rev. Lett., 2019, 122, 107202.
11. Y. Bai, Y. Li, J. Luan, R. Liu, W. Song, Y. Chen, P.-F. Ji, Q. Zhang, F. Meng, B. Tong, L. Li, Y. Jiang, Z. Gao, L. Gu, J. Zhang, Y. Wang, Q.-K. Xue, K. He, Y. Feng and X. Feng, Natl. Sci. Rev., 2023, 11, 1–9.
12. Y. Li, C. Liu, Y. Wang, Z. Lian, S. Li, H. Li, Y. Wu, H.-Z. Lu, J. Zhang and Y. Wang, Sci. Bull., 2023, 68, 1252–1258.
13. P. Kagerer, C. I. Fornari, S. Buchberger, S. L. Morelhão, R. C. Vidal, A. Tcakaev, V. Zabolotnyy, E. Weschke, V. Hinkov, M. Kamp, B. Büchner, A. Isaeva, H. Bentmann and F. Reinert, J. Appl. Phys., 2020, 128, 135303.
14. N. Liu, S. Schreyeck, K. M. Fijalkowski, M. Kamp, K. Brunner, C. Gould and L. W. Molenkamp, J. Cryst. Growth, 2022, 591, 126677.
15. J. Lapano, L. Nuckols, A. R. Mazza, Y.-Y. Pai, J. Zhang, B. Lawrie, R. G. Moore, G. Eres, H. N. Lee, M.-H. Du, T. Z. Ward, J. S. Lee, W. J. Weber, Y. Zhang and M. Brahlek, Phys. Rev. Mater., 2020, 4, 111201.
16. Y.-F. Zhao, L.-J. Zhou, F. Wang, G. Wang, T. Song, D. Ovchinnikov, H. Yi, R. Mei, K. Wang, M. H. W. Chan, C.-X. Liu, X. Xu and C.-Z. Chang, Nano Lett., 2021, 21, 7691–7698.
17. S.-K. Bac, K. Koller, F. Lux, J. Wang, L. Riney, K. Borisiak, W. Powers, M. Zhukovskyi, T. Orlova, M. Dobrowolska, J. K. Furdyna, N. R. Dilley, L. P. Rokhinson, Y. Mokrousov, R. J. McQueeney, O. Heinonen, X. Liu and B. A. Assaf, npj Quantum Mater., 2022, 7, 46.
18. A. Koma, Thin Solid Films, 1992, 216, 72–76.
19. M. Tanaka, J. Cryst. Growth, 2005, 278, 25–37.
20. L. N. Rodrigues, C. I. L. de Araujo, S. L. A. Mello, J. Laverock, J. M. Fonseca, W. Schwarzacher, W. F. Inoch and S. O. Ferreira, J. Appl. Phys., 2023, 134, 085301.
21. A. K. Ray, R. Swan and C. A. Hogarth, J. Non-Cryst. Solids, 1994, 168, 150–156.
22. L. G. Parratt, Phys. Rev., 1954, 95, 359–369.
23. J. Als-Nielsen and D. McMorrow, Elements of Modern X-ray Physics, Wiley, 2011.
24. M. He, Y. Fu, Y. Huang, H. Sun, T. Guo, W. Lin, Y. Zhu, Y. Zhang, Y. Liu, G. Yu and Q. L. He, J. Phys.: Condens. Matter, 2023, 35, 295701.
25. Y. Li, Y. Wang, Z. Lian, H. Li, Z. Gao, L. Xu, H. Wang, R. Lu, L. Li, Y. Feng, J. Zhu, L. Liu, Y. Wang, B. Fu, S. Yang, L. Yang, Y. Wang, T. Xia, C. Liu, S. Jia, Y. Wu, J. Zhang, Y. Wang and C. Liu, Nat. Commun., 2024, 15, 3399.
26. H. Beidenkopf, P. Roushan, J. Seo, L. Gorman, I. Drozdov, Y. S. Hor, R. J. Cava and A. Yazdani, Nat. Phys., 2011, 7, 939–943.
27. H. Kim, M. Liu, L. Frammolino, Y. Li, F. Zhang, W. Lee, C. Dong, Y.-F. Zhao, G.-Y. Chen, P.-J. Hsu, C.-Z. Chang, J. Robinson, J. Yan, X. Li, A. H. MacDonald and C.-K. Shih, ACS Appl. Nano Mater., 2024, 7, 21149–21159.
28. C. Parra, T. H. Rodrigues da Cunha, A. W. Contryman, D. Kong, F. Montero-Silva, P. H. Rezende Gonçalves, D. D. Dos Reis, P. Giraldo-Gallo, R. Segura, F. Olivares, F. Niestemski, Y. Cui, R. Magalhaes-Paniago and H. C. Manoharan, Nano Lett., 2017, 17, 97–103.
29. G. Rodrigues-Junior, T. Chagas, R. Reis, P. V. Sciammarella, C. I. Fornari, P. H. O. Rappl, E. Abramof, R. Magalhães-Paniago and Â. Malachias, Phys. Rev. B, 2023, 108, 035408.
30. J.-Q. Yan, Q. Zhang, T. Heitmann, Z. Huang, K. Y. Chen, J.-G. Cheng, W. Wu, D. Vaknin, B. C. Sales and R. J. McQueeney, Phys. Rev. Mater., 2019, 3, 064202.
31. J. Cui, M. Shi, H. Wang, F. Yu, T. Wu, X. Luo, J. Ying and X. Chen, Phys. Rev. B, 2019, 99, 155125.

---