# Fileset

[APL25-AR-01605.pdf](https://mdr.nims.go.jp/filesets/e45f63b9-6a7a-43f3-bd54-759f5a00d56c/download)

## Creator

K. Ishihara, [S. Ichinokura](https://orcid.org/0000-0002-7968-4016), S. V. Eremeev, [T. T. Sasaki](https://orcid.org/0000-0002-5952-7638), R. Takada, H. Nishimichi, R. Akiyama, E. V. Chulkov, T. Hirahara

## Rights

This article may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing. This article appeared in K. Ishihara, S. Ichinokura, S. V. Eremeev, T. T. Sasaki, R. Takada, H. Nishimichi, R. Akiyama, E. V. Chulkov, T. Hirahara; Manipulation of the anomalous Hall effect in magnetic topological insulator heterostructure MnBi2Te4/Bi2Te3 by Si substrate surface engineering. Appl. Phys. Lett. 24 November 2025; 127 (21): 211601 and may be found at https://doi.org/10.1063/5.0266580.
[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[Manipulation of the anomalous Hall effect in magnetic topological insulator heterostructure MnBi2Te4/Bi2Te3 by Si substrate surface engineering](https://mdr.nims.go.jp/datasets/956df3fd-f329-4747-b144-86fc97a2d568)

## Fulltext

Manipulation of the anomalous Hall effect inmagnetic topological insulator heterostructure MnBi2Te4/Bi2Te3 by Si substrate surface engineeringManipulation of the anomalous Hall effect in magnetic topologicalinsulator heterostructure MnBi2Te4/Bi2Te3 by Si substrate surfaceengineeringK. Ishihara,1 S. Ichinokura,1, a) S. V. Eremeev,2 T. T. Sasaki,3 R. Takada,1 H. Nishimichi,1 R. Akiyama,1 E. V.Chulkov,4, 5, 6 and T. Hirahara11)Department of Physics, Institute of Science Tokyo, Tokyo 152-8551, Japan2)Institute of Strength Physics and Materials Science, Tomsk, 634055, Russia3)Research Center for Magnetic and Spintronic Materials, National Institute for Materials Science, Tsukuba 305-0047,Japan4)Donostia International Physics Center (DIPC), Paseo de Manuel Lardizabal, 4, 20018 San Sebastián/Donostia, Basque Country,Spain5)Tomsk State University, Tomsk, 634050, Russia6)Saint Petersburg State University, Saint Petersburg, 198504, Russia(*Electronic mail: hirahara@phys.sci.isct.ac.jp)(Dated: 14 September 2025)We developed an in situ Hall measurement setup and measured the anomalous Hall effect (AHE) in magnetic topolog-ical insulator heterostructures MnBi2Te4/Bi2Te3 grown on different Si(111) substrate surfaces. For the sample grownon the Si(111)-7×7 surface, the AHE signal appears at 15 K and becomes larger by further cooling, showing that theCurie temperature Tc is 15 K. In contrast, although the Tc is the same, the AHE signal shows a local maximum at 10 Kfor the sample grown on the β -Bi/Si(111)-√3×√3 surface. A plausible explanation for this peculiar behavior is theenhanced skew scattering caused by the Bi layer, or the presence of the states localized at the interfacial Bi layer, whichwill affect the Berry curvature of the system. Our results demonstrate the possibility to artificially control the propertyof a two-dimensional magnet by modification of the substrate surface with a single monatomic layer.The Hall effect, which is a voltage generated transverse toan electrical current due to the application of a magnetic fieldor magnetization, is one of the key parameters to discuss quan-tum properties of materials1,2. Among various types of Halleffect, the anomalous Hall effect (AHE) has been regarded asan important physical quantity not only to characterize mag-netic systems3 but also to discuss the Berry curvature in non-magnets4.While experiments to detect the Hall effect have been per-formed extensively, there are only a few reports that haveaddressed Hall effect measurements in ultrahigh vacuum(UHV). Even then, the Hall coefficient for non-magnetic sys-tems has only been discussed5–8, and no AHE measurementshave been performed up to now. Due to the rise of varioustwo-dimensional magnets which can be derived from van derWaals layered crystals9,10 or intrinsic magnetic topologicalinsulators11,12, the need for a Hall effect measurement sys-tem in UHV environment has increased since some systemsare not stable in air and capping the surface to prevent oxi-dation may change the sample magnetization. In this respect,in situ characterization is desired to discuss the intrinsic mag-netic properties. However, there are hardly any facilities thatare capable of performing such experiments. Furthermore, itis desired to simultaneously measure the electronic structuretogether with macroscopic magnetic properties since the dif-ference in the sample quality as well as the degree of oxidationmay also cause confusion in the data interpretation.a)Present Address: Center for Basic Research on Materials, National Institutefor Materials Science, Tsukuba 305-0003, JapanTherefore in the present study, we modified our all-in-onemultimodal UHV system (Fig. S1) enabling sample fabrica-tion and in situ angle-resolved photoemission spectroscopy(ARPES) as well as transport measurements13,14 and mea-sured the AHE of an intrinsic magnetic topological insula-tor heterostructure MnBi2Te4/Bi2Te3 (MBT/BT) grown onSi(111) substrates15 with different surface structures. TheCurie temperature was determined as 15 K for all the samples,but we found that the AHE shows a peculiar behavior whenthe heterostructure was grown on the β -Bi/Si(111)-√3×√3surface. This likely originates from the skew scattering fromthe Bi monolayer, or the change in the Berry curvature ofthe system since the states stemming from this interfacial Bilayer are present near the Fermi level (EF) and show a cross-ing with the Dirac cone of MBT/BT. Our results show thata single monatomic layer decoration at the substrate surfacecan significantly alter the magnetotransport property of a two-dimensional magnet, which is a potentially important point indeveloping atomic-layer spintronic devices.The heterostructure samples were prepared by molecularbeam epitaxy equipped with a reflection-high-energy electrondiffraction (RHEED) system. First, a clean Si(111)-7×7 sur-face was prepared on an n-type substrate (1.5-5 Ωcm resis-tivity at room temperature) by a cycle of resistive heat treat-ments. The β -Bi/Si(111)-√3×√3 (hereafter β√3-Bi) sur-face was formed by 1 ML (7.83×1014 cm−2) of Bi depositionon the 7×7 surface at 620 K monitored by RHEED. Then Biwas deposited either directly on the 7× 7 surface or after theformation of the β√3-Bi surface at ∼250 ◦C in a Te-rich con-dition to grow quintuple layer (QL) structured Bi2Te3 filmsand further annealed at ∼250 ◦C for 5 minutes. This resultsin a QL by QL growth and the thickness is monitored withThis is the author’s peer reviewed, accepted manuscript. However, the online version of record will be different from this version once it has been copyedited and typeset.PLEASE CITE THIS ARTICLE AS DOI: 10.1063/5.02665802-2-1012Ra(H)-Ra(-H) []-0.4 -0.2 0.0 0.2 0.4Magnetic field [T]-0.3 -0.2 -0.1 0 0.1 0.2 0.3k (/A)-0.5-0.4-0.3-0.2-0.100.1Energy (eV)-0.3 -0.2 -0.1 0 0.1 0.2 0.3k (/A)-0.5-0.4-0.3-0.2-0.100.1Energy (eV)MBT / 6 QL BT / Si(111)-7×7MBT / 6QL BT / β-Bi/Si(111)-√3×√3-2-1012RAHE(H)[]-0.4 -0.2 0.0 0.2 0.4Magnetic field [T] 3.5K 12K 15K 28K4.854.804.754.704.654.60Ryx(0) []20151050Temperature [K]29.429.229.028.8Ryx(0) []20151050Temperature [K]EF0.20.4Binding energy (eV)EF0.20.4Binding energy (eV)0.2-0.2 00.2-0.2 0k (Å-1)k (Å-1)(a)(b)0.2-0.2 0Magnetic field (T)(d)0.4-0.40.2-0.2 0Magnetic field (T)0.4-0.4Rxy (Ω)(c)01.0-1.0Rxy (Ω)01.0-1.0 3.2K 8K 15K 22K-0.4 -0.2 0.0 0.2 0.4Magnetic field [T]155 10Temperature (K)200-0.4 -0.2 0.0 0.2 0.4Magnetic field [T]155 10Temperature (K)20R(B=0) (Ω) 2.42.3R(B=0) (Ω)12.612.4(f)(e)IVB0FIG. 1. (a, b) Band dispersion (second derivative) of MBT/ 6 QL BT / Si(111)-7×7 (a), and that of MBT / 6 QL BT / β -Bi/Si(111)-√3×√3(b), respectively, along the Γ̄-M̄ direction measured at room temperature. (c, d) The asymmetrized Hall resistance Rxy as a function of theapplied magnetic field measured at various temperatures for MBT / 6 QL BT / Si(111)-7×7 (c) and MBT / 6 QL BT / β -Bi/Si(111)-√3×√3(d), respectively. (e, f) Temperature dependence of the measured resistance R at zero field after magnetizing the sample for MBT / 6 QL BT /Si(111)-7×7 (e) and MBT / 6 QL BT / β -Bi/Si(111)-√3×√3 (f), respectively. The inset in (e) shows the measurement configuration.RHEED oscillations. Finally, Mn was deposited on Bi2Te3film in a Te-rich condition at ∼260 ◦C. The 1× 1 periodicitywith the same lattice constant is maintained during this pro-cess for the samples we have fabricated15,16.The ARPES measurements were performed after the sam-ple preparation with a commercial hemispherical photoelec-tron spectrometer equipped with angle and energy multidetec-tions (ScientaOmicron R4000) with He Iα radiation (21.2 eV)at room temperature. Some additional ARPES measurementswere performed in a different UHV chamber (ScientaOmicronDA20).Then, the transport measurements were performed in ourcustom-made system13,14 which has four probes that can moveindependently and by touching the surface of the sample, four-point probe (4PP) resistance measurements can be performedin UHV (Figs. S2, S3). To measure the Hall effect, the probeswere aligned in a cross geometry as shown in the inset ofFig. 1(e). The samples were cooled down to ∼3 K in thepresent study and a magnetic field as large as 0.4 T was ap-plied perpendicular to the sample surface. Although it is notpossible to deduce physical parameters quantitatively sincewe have not fabricated a Hall bar structure, it is still possi-ble to make a qualitative discussion as we will show in thefollowing.The samples were capped with ∼10 nm of Te before takingthem out of the UHV chamber for scanning transmission elec-tron microscopy (STEM) measurements. Electron transparentspecimens for STEM observations were prepared by the stan-dard lift-out technique using an FEI Helios G4-UX dual-beamsystem. Probe abberation corrected STEM, FEI Titan™G280-200 microscope, was used. Chemical compositions weremeasured by energy-dispersive X-ray spectroscopy (EDS).For structural optimization and electronic band calcula-tions we used the Vienna Ab Initio Simulation Package17,18with generalized gradient approximation (GGA-PBE)19 to theexchange-correlation potential. The interaction between theion cores and valence electrons was described by the projectoraugmented-wave method20,21. Relativistic effects were takeninto account, including the spin-orbit interaction. In order todescribe the vdW interactions we made use of the DFT-D3functional with Becke-Johnson dumping scheme22. The ge-ometry optimization was performed until the residual forceson atoms became smaller than 1 meV/Å. To correctly describethe highly correlated Mn-d electrons of MBT we include thecorrelation effects within the GGA+U method23 within theDudarev scheme24. The Ueff = U − J value for the Mn 3d-states was chosen to be equal to 5.34 eV11,25. For analy-sis of interactions between atoms at the interfaces we usethe projected crystal orbital Hamilton population (pCOHP)method26,27 implemented within Local-orbital basis suite to-This is the author’s peer reviewed, accepted manuscript. However, the online version of record will be different from this version once it has been copyedited and typeset.PLEASE CITE THIS ARTICLE AS DOI: 10.1063/5.02665803wards electronic structure reconstruction (LOBSTER) code28.Figures 1(a) and (b) show the band dispersion (secondderivative) of MBT / 6 QL BT / 7×7 (a), and that of MBT / 6QL BT / β√3-Bi (b), respectively, along the Γ̄-M̄ direction29.One can find the linearly dispersion Dirac cone near the Fermilevel (EF) together with the bulk bands, showing that the filmsare n-doped. For the higher binding energy side, there are twobands dispersing away from the Γ̄ point. These features arebasically the same as what is shown in Ref. 15 and demon-strates that two samples apparently do not exhibit significantdifference in the surface electronic structure.However, some difference was observed in the transportdata for the two samples. Figures 1(c) and (d) show the Hallresistance Rxy as a function of the applied magnetic field mea-sured at various temperatures for MBT / 6 QL BT / 7 × 7(c) and MBT / 6 QL BT / β√3-Bi (d), respectively. Thesedata were asymmerized from the measured raw data R as ex-plained in Fig. S5 to deduce the Hall resistance. It can benoticed that above 20 K, both samples exhibit the normal Halleffect, showing a linear dependence. The negative Hall coef-ficient shows that the carriers are electrons, consistent withthe band dispersion measured with ARPES (Figs. 1(a) and(b)). Upon lowering the temperature, a hysteresis behaviorcan be observed, demonstrating the detection of the AHE. InFig. 1(c), the value of Rxy at zero field does not change somuch for the data at 12 K and 3.5 K, but the value of the mag-netic field that Rxy crosses zero Ω increases. This shows thatthe residual magnetization does not increase but the coercivefield increases below 12 K. This is a typical behavior of ferro-magnets and one can say that the measured AHE is reflectingthe magnetization of the MBT/BT heterostructure. Since thecontribution of the longitudinal resistance Rxx can be expectedto be very small in the measured R as shown in Fig. S5, wemeasured the temperature dependence of R(B = 0) after thefield sweep to deduce the Curie temperature Tc. As shown inFig. 1(e), one can find a sharp rise around 15 K by loweringthe temperature and this shows that Tc is ∼ 15 K. Although wecould not detect any signal of ferromagnetism for this samplewith ex situ XMCD measurements down to 6 K15, the presentmeasurements unambiguously suggest that the MBT/BT het-erostructure grown on Si(111)-7× 7 is ferromagnetic below15 K, which is similar to the report in Ref. 12. The origin ofthis discrepancy cannot be clearly elucidated at this time.A qualitatively different behavior was observed for theMBT/BT heterostructure grown on β√3-Bi. As shown inFig. 1(d), the clear detection of the hysteresis loop shows thatthis system is also ferromagnetic. However, although the coer-cive field is larger for the data at 3.2 K compared to that at 8 K,Rxy at zero field is smaller at 3.2 K. This characteristic can beseen more vividly in the temperature dependence of R(B = 0)as shown in Fig. 1(f). The signal shows a sharp rise at 15 K,but starts to decrease again below 10 K. This feature is clearlydifferent from the behavior shown in Fig. 1(e) and cannot beexplained by a simple effect of typical ferromagnetism.To elucidate the origin of this peculiar behavior found inAHE, we have performed STEM measurements. Figure 2(a)shows the STEM image for the MBT / 6 QL BT / 7× 7. Be-tween the Te capping layer and the Si substrate, the struc-MBT / 6 QL BT / Si(111)-7×7MBT / 6QL BT / β-Bi/Si(111)-√3×√3Si substrateTe capping layerBi layerdisordered (amorphous Te) layerMBT BT 1 nm(a)(b)FIG. 2. TEM image of MBT / 6 QL BT / Si(111)-7×7 (a), and thatof MBT / 6 QL BT /β -Bi/Si(111)-√3×√3 (b), respectively. TheMBT and BT layers are explicitly indicated and the difference in theinterface between the Si substrate and the film is emphasized.ture can be mostly identified as the designed one. It shouldbe emphasized that at the heterostructure/substrate interface,the image is somehow disturbed. This is due to the fact thatdisordered Te will build up as a wetting or a buffer layer tosaturate dangling bonds of the 7× 7 surface30,31 before theBT growth. In contrast, the heterostructure/substrate interfacefor the MBT / 6 QL BT / β√3-Bi shown in Fig. 2(b) hasa sharp appearance with a well-defined monolayer. This islikely due to the Bi-trimer layer which originally formed theβ√3-Bi surface reconstruction32, although we have no evi-dence whether the in-plane arrangement has changed or notupon the deposition of BT. But the EDS mapping shows thatthis is a Bi layer as shown in Fig. S6. Since there is hardly anydifference in the distribution of unintentionally displaced Mnatoms that can influence the magnetic property as measuredby EDS (Fig. S7)33–37, we conclude that the only clear differ-ence in the atomic structure between the two samples is foundat the BT/Si interface.It can be anticipated that since the top MBT surface layerswhich is probed by ARPES is not different for the two sys-tems, the band dispersion shown in Figs. 1(a) and (b) demon-strate no difference. However, because our films are n-dopedand the bulk carriers are also involved in transport, it can bespeculated that the difference in the interface structure can in-fluence the AHE signal. It is known that the band dispersionof BT films grown on Si(111)-7× 7 films can be reproducedby DFT calculations for free-standing BT slabs38. However,it seems that we need to take into account the Bi layer at theBT/Si interface to describe the electronic structure of BT /This is the author’s peer reviewed, accepted manuscript. However, the online version of record will be different from this version once it has been copyedited and typeset.PLEASE CITE THIS ARTICLE AS DOI: 10.1063/5.02665804β√3-Bi. To this end, we have constructed a model shown inFig. 3(a), in which 5 QL of BT was placed on top of β√3-Bi. The in-plane lattice constants of BT(111) and Si(111)-√3×√3 differ approximately by factor 1.5 (4.38 vs. 6.69Å) and hence the better matching between them is realizedwith superlattices ratio of 3/2, i.e. BT(111)-3× 3 should beplaced on a β -Bi/Si(111) with 2√3× 2√3 periodicity with1.9% compression. To simulate the Si(111) substrate we useda slab of two bi-layers (BL) thickness and Hydrogen atomswere used to passivate the Si dangling bonds at the bottom(unreconstructed) surface of the slab. The atomic positions ofthe bottom BL were fixed at their bulk positions while atomsof the top Si BL as well as β -Bi layer and in the interfaceQL of BT were relaxed. The calculated band dispersion is de-picted in Fig. 3(b) and the different markers show the stateslocalized at different positions in the slab structure.The bonding energy between Bi-trimer atoms with the un-derlying Si atoms and that with the nearest Te atoms of theBT were calculated and it turns out that the interaction of BTwith the Bi-trimer layer is derived to be about 9 times weakercompared to that between the Bi-trimers and the Si(111) sub-strate. Due to this weak interaction at the BT/ β√3-Bi inter-face, the variation in the interface potential is relatively smalland limited to the first (closest to the interface) QL, as shownin Fig. 3(c). This makes the seemingly complicated band dis-persion of Fig. 3(b) easy to understand. The states markedby red circles in Fig. 3(b) have their origin as the Rashba-splitsurface states near the K̄ and M̄ points (Fig. S8(a)) of theβ√3-Bi surface32, but the energy position has shifted closer to EFdue to the change in the potential at the interface. It shouldalso be noted that due to the imposed 2√3×2√3 periodicity,the Rashba states of β√3-Bi are folded to the Γ̄ point. Sim-ilarly, the energy level of the Dirac cone at the top surface(grey circles in Fig. 3(b)) and that at the interface (blue cir-cles) is shifted due to the potential difference at the interfaceand at the vacuum side of BT.As discussed above, we are not sure if the above structuremodel is actually reflecting the situation in the experiment.However, the presence of the bands near the Fermi level lo-calized at the interface Bi layer in addition to the BT Diraccone seems to be universal irrespective of the presence of theMBT septuple layer at the top (Fig. S8(b)) or whether the BTfilm thickness is only 1 QL nor the change in stacking betweenthe BT and the β -Bi/Si(111)-2√3×2√3 surface (Figs. S8(c)-(e)). Therefore, we have experimentally tried to confirm thepresence of this interface state for the 1 QL BT /β√3-Bi sys-tem. As shown in Fig. S9, there are indeed some features nearthe Fermi level other than the Dirac cone, but since the inten-sity is quite weak, we cannot make a direct comparison to thecalculated band dispersion. Nevertheless, this shows that in-terface states localized at the β -Bi interface layer can affectthe film property for the MBT / BT grown on the β√3-Bisurface.The most simple explanation concerning the unique behav-ior of the AHE shown in Fig. 1(d) and (f) is that the carriersundergo skew scattering at the interfacial layer and this cangive rise to extrinsic AHE due to the strong SOC of Bi. How-ever, since it is well known that the extrinsic mechanism iszHSiBiBT(a) (b)(c)FIG. 3. (a) Atomic structure model (the presented structure was vi-sualized with VESTA39) and (b) calculated band dispersion near EFof the 5 QL BT / β -Bi/Si slab. The different markers show the stateslocalized at different positions within the model shown in (a). (c)In-plane (x,y) averaged electrostatic potentials V (z) within the slab.Since the interaction at the Bi/BT interface is weak, there is only aslight change compared to the vacuum/BT interface.only dominant for low-mobility (side jump) or high-mobility(skew scattering) systems and most materials can be describedby intrinsic effects due to the Berry curvature2,40, we specu-late that the AHE of MBT / 6 QL BT / β√3-Bi may also beaffected by the band structure. One can obviously find thatthe states originating from the β -Bi layer and the interfacialDirac cone cross each other (Fig. 3(b)). Such crossing maybe a source of Berry curvature, although the actual calculationis beyond the scope of the present work41. Nevertheless, weemphasize that the present results demonstrate a convenientway to manipulate the Berry curvature artificially using a sin-gle layer at the heterostructure/substrate interface, which canpotentially be used to develop spintronic devices using atom-ically thin magnets.In summary, by means of the developed in situHall measurement setup, we have shown that theAHE in magnetic topological insulator heterostructuresMnBi2Te4/Bi2Te3/Si(111) can be manipulated by controllingthe BT/Si interface. The formation of the β -Bi/Si(111)-This is the author’s peer reviewed, accepted manuscript. However, the online version of record will be different from this version once it has been copyedited and typeset.PLEASE CITE THIS ARTICLE AS DOI: 10.1063/5.02665805√3 ×√3 surface reconstruction before the heterostructuregrowth results in appearance a local maximum in the AHEsignal at 10 K. This behavior can stem from the skewscattering at the interface, or the effect of the states localizedat the Bi interfacial layer that resides at the same energy asthe Dirac state of the bottom side of the MBT/BT film, whichcan affect the Berry curvature of the system.See supplementary material for details on the instrumentalapparatus, linearity of the measured IV curves, raw ARPESdata of Figs. 1(a) and (b), data processing in the Hall mea-surements, EDS mapping of the samples, additional results ofDFT calculations, and ARPES data of 1QL BT / β√3-Bi.ACKNOWLEDGMENTSThe authors thank M. Uchida, Y. Fuseya, I. A. Shvets, andM. M. Otrokov for stimulating discussions and Y. Fukushimafor assisting in the data analyses. T.H. acknowledges the sup-port by Grants-In-Aid from Japan Society for the Promotionof Science (Grants No. 18H03877, No. 22H00293, and No.23H00268), the Murata Science Foundation (Grant No. H30-084), the Asahi Glass Foundation, the Iketani Science andTechnology Foundation (Grant No. 0321083-A), and Sup-port for Tokyo Tech Advanced Researchers. S.V.E. acknowl-edges support from the Government research assignment forISPMS SB RAS, project FWRW-2022-0001. E.V.C. acknowl-edges Saint-Petersburg State University for research Project116812735.AUTHOR DECLARATIONSCONFLICT OF INTERESTThe authors have no conflicts of interest to disclose.DATA AVAILABILITYThe data that support the findings of this study are availablefrom the corresponding authors upon reasonable request.1E. H. Hall, “On a new action of the magnet on electric currents,” AmericanJournal of Mathematics 2, 287–292 (1879).2N. Nagaosa, J. Sinova, S. Onoda, A. H. MacDonald, and N. P. Ong,“Anomalous Hall effect,” Rev. Mod. Phys. 82, 1539–1592 (2010).3H. Ohno, A. Shen, F. Matsukura, A. Oiwa, A. Endo, S. Katsumoto, andY. Iye, “(Ga,Mn)As: A new diluted magnetic semiconductor based onGaAs,” Applied Physics Letters 69, 363–365 (1996).4T. Liang, J. Lin, Q. Gibson, S. Kushwaha, M. Liu, W. Wang,H. Xiong, J. A. Sobota, M. Hashimoto, P. S. Kirchmann, Z.-X.Shen, R. J. Cava, and N. P. Ong, “Anomalous Hall effect in ZrTe5,”Nature Physics 14, 451–455 (2018).5I. Vilfan, M. Henzler, O. Pfennigstorf, and H. Pfnür, “Anomalous thick-ness dependence of the Hall effect in ultrathin Pb layers on Si(111),”Phys. Rev. B 66, 241306 (2002).6T. Hirahara, I. Matsuda, C. Liu, R. Hobara, S. Yoshimoto, and S. Hasegawa,“Direct measurement of the Hall effect in a free-electron-like surface state,”Phys. Rev. B 73, 235332 (2006).7D. Abdelbarey, J. Koch, Z. Mamiyev, C. Tegenkamp, and H. Pfnür,“Thickness-dependent electronic transport through epitaxial nontrivial Biquantum films,” Phys. Rev. B 102, 115409 (2020).8J. Koch, S. Sologub, C. Ghosal, T. Tschirner, A. Chatterjee, K. Pierz,H. W. Schumacher, and C. Tegenkamp, “Magnetotransport behaviorof epitaxial graphene inhomogeneously doped by Bi(110) islands,”Phys. Rev. B 109, 235107 (2024).9K. S. Burch, D. Mandrus, and J.-G. Park, “Magnetism in two-dimensionalvan der Waals materials,” Nature 563, 47–52 (2018).10M. Gibertini, M. Koperski, A. F. Morpurgo, and K. S.Novoselov, “Magnetic 2D materials and heterostructures,”Nature Nanotechnology 14, 408–419 (2019).11M. M. Otrokov, I. I. Klimovskikh, H. Bentmann, D. Estyunin, A. Zeugner,Z. S. Aliev, S. Gaß, A. U. B. Wolter, A. V. Koroleva, A. M. Shikin,M. Blanco-Rey, M. Hoffmann, I. P. Rusinov, A. Y. Vyazovskaya, S. V.Eremeev, Y. M. Koroteev, V. M. Kuznetsov, F. Freyse, J. Sánchez-Barriga,I. R. Amiraslanov, M. B. Babanly, N. T. Mamedov, N. A. Abdullayev,V. N. Zverev, A. Alfonsov, V. Kataev, B. Büchner, E. F. Schwier, S. Ku-mar, A. Kimura, L. Petaccia, G. Di Santo, R. C. Vidal, S. Schatz,K. Kißner, M. Ünzelmann, C. H. Min, S. Moser, T. R. F. Peixoto, F. Rein-ert, A. Ernst, P. M. Echenique, A. Isaeva, and E. V. Chulkov, “Pre-diction and observation of an antiferromagnetic topological insulator,”Nature 576, 416–422 (2019).12P. Kagerer, C. I. Fornari, S. Buchberger, T. Tschirner, L. Veyrat, M. Kamp,A. V. Tcakaev, V. Zabolotnyy, S. L. Morelhão, B. Geldiyev, S. Müller,A. Fedorov, E. Rienks, P. Gargiani, M. Valvidares, L. C. Folkers, A. Isaeva,B. Büchner, V. Hinkov, R. Claessen, H. Bentmann, and F. Reinert,“Two-dimensional ferromagnetic extension of a topological insulator,”Phys. Rev. Res. 5, L022019 (2023).13A. K. Pedersen, S. Ichinokura, T. Tanaka, R. Shimizu, T. Hitosugi,and T. Hirahara, “Interfacial superconductivity in FeSe ultrathin films onSrTiO3 probed by in situ independently driven four-point-probe measure-ments,” Phys. Rev. Lett. 124, 227002 (2020).14K. Ide, T. Tanaka, A. Pedersen, S. Ichinokura, and T. Hirahara, “Tempera-ture dependence of the superconducting gap of single-layer FeSe/SrTiO3:Direct comparison between transport and spectroscopic measurements,”Phys. Rev. Mater. 6, 124801 (2022).15T. Fukasawa, S. Kusaka, K. Sumida, M. Hashizume, S. Ichinokura,Y. Takeda, S. Ideta, K. Tanaka, R. Shimizu, T. Hitosugi, and T. Hira-hara, “Absence of ferromagnetism in MnBi2Te4/Bi2Te3 down to 6 K,”Phys. Rev. B 103, 205405 (2021).16T. Hirahara, M. M. Otrokov, T. T. Sasaki, K. Sumida, Y. Tomohiro,S. Kusaka, Y. Okuyama, S. Ichinokura, M. Kobayashi, Y. Takeda,K. Amemiya, T. Shirasawa, S. Ideta, K. Miyamoto, K. Tanaka, S. Kuroda,T. Okuda, K. Hono, S. V. Eremeev, and E. V. Chulkov, “Fabrication of anovel magnetic topological heterostructure and temperature evolution of itsmassive Dirac cone,” Nature Communications 11, 4821 (2020).17G. Kresse and J. Hafner, “Ab initio molecular dynamics for open-shell tran-sition metals,” Phys. Rev. B 48, 13115–13118 (1993).18G. Kresse and J. Furthmüller, “Efficient iterative schemes forab initio total-energy calculations using a plane-wave basis set,”Phys. Rev. B 54, 11169–11186 (1996).19J. P. Perdew, K. Burke, and M. Ernzerhof, “Generalized gradient approxi-mation made simple,” Phys. Rev. Lett. 77, 3865–3868 (1996).20P. E. Blöchl, “Projector augmented-wave method,”Phys. Rev. B 50, 17953–17979 (1994).21G. Kresse and D. Joubert, “From ultrasoft pseudopotentials to the projectoraugmented-wave method,” Phys. Rev. B 59, 1758–1775 (1999).22S. Grimme, S. Ehrlich, and L. Goerigk, “Effect of the damp-ing function in dispersion corrected density functional theory,”Journal of Computational Chemistry 32, 1456–1465 (2011).23V. I. Anisimov, J. Zaanen, and O. K. Andersen, “Band theory and Mott in-sulators: Hubbard U instead of Stoner I,” Phys. Rev. B 44, 943–954 (1991).24S. L. Dudarev, G. A. Botton, S. Y. Savrasov, C. J. Humphreys, and A. P.Sutton, “Electron-energy-loss spectra and the structural stability of nickeloxide: An LSDA+U study,” Phys. Rev. B 57, 1505–1509 (1998).25S. V. Eremeev, M. M. Otrokov, and E. V. Chulkov, “Competing rhombo-hedral and monoclinic crystal structures in MnPn2Ch4 compounds: An ab-initio study,” Journal of Alloys and Compounds 709, 172–178 (2017).This is the author’s peer reviewed, accepted manuscript. However, the online version of record will be different from this version once it has been copyedited and typeset.PLEASE CITE THIS ARTICLE AS DOI: 10.1063/5.0266580626R. Dronskowski and P. E. Bloechl, “Crystal orbital Hamiltonpopulations (COHP): energy-resolved visualization of chemi-cal bonding in solids based on density-functional calculations,”The Journal of Physical Chemistry 97, 8617–8624 (1993).27V. L. Deringer, A. L. Tchougréeff, and R. Dronskowski, “Crystal orbitalHamilton population (COHP) analysis as projected from plane-wave basissets,” The Journal of Physical Chemistry A 115, 5461–5466 (2011).28S. Maintz, V. L. Deringer, A. L. Tchougréeff, and R. Dronskowski, “LOB-STER: A tool to extract chemical bonding from plane-wave based DFT,”Journal of Computational Chemistry 37, 1030–1035 (2016).29The raw data can be found in Fig. S4.30T. Shirasawa, J. Tsunoda, T. Hirahara, and T. Takahashi, “Structure of aBi/Bi2Te3 heteroepitaxial film studied by x-ray crystal truncation rod scat-tering,” Phys. Rev. B 87, 075449 (2013).31Y. Liu, G. Bian, T. Miller, M. Bissen, and T.-C. Chiang, “Topologi-cal limit of ultrathin quasi-free-standing Bi2Te3 films grown on Si(111),”Phys. Rev. B 85, 195442 (2012).32K. Sakamoto, H. Kakuta, K. Sugawara, K. Miyamoto, A. Kimura, T. Kuzu-maki, N. Ueno, E. Annese, J. Fujii, A. Kodama, T. Shishidou, H. Na-matame, M. Taniguchi, T. Sato, T. Takahashi, and T. Oguchi, “PeculiarRashba splitting originating from the two-dimensional symmetry of the sur-face,” Phys. Rev. Lett. 103, 156801 (2009).33Y. Liu, L.-L. Wang, Q. Zheng, Z. Huang, X. Wang, M. Chi, Y. Wu,B. C. Chakoumakos, M. A. McGuire, B. C. Sales, W. Wu, andJ. Yan, “Site mixing for engineering magnetic topological insulators,”Phys. Rev. X 11, 021033 (2021).34Y. Lai, L. Ke, J. Yan, R. D. McDonald, and R. J. McQueeney,“Defect-driven ferrimagnetism and hidden magnetization in MnBi2Te4,”Phys. Rev. B 103, 184429 (2021).35S. X. M. Riberolles, Q. Zhang, E. Gordon, N. P. Butch, L. Ke, J.-Q. Yan,and R. J. McQueeney, “Evolution of magnetic interactions in Sb-substitutedMnBi2Te4,” Phys. Rev. B 104, 064401 (2021).36M. Sahoo, I. J. Onuorah, L. C. Folkers, E. Kochetkova, E. V. Chulkov,M. M. Otrokov, Z. S. Aliev, I. R. Amiraslanov, A. U. B. Wolter, B. Büch-ner, L. T. Corredor, C. Wang, Z. Salman, A. Isaeva, R. De Renzi,and G. Allodi, “Ubiquitous order-disorder transition in the Mn antisitesublattice of the (MnBi2Te4)(Bi2Te3)n magnetic topological insulators,”Advanced Science 11, 2402753 (2024).37R. Fukushima, V. N. Antonov, M. M. Otrokov, T. T. Sasaki, R. Akiyama,K. Sumida, K. Ishihara, S. Ichinokura, K. Tanaka, Y. Takeda, D. P. Sali-nas, S. V. Eremeev, E. V. Chulkov, A. Ernst, and T. Hirahara, “Direct evi-dence of induced magnetic moment in Se and the role of misplaced Mn inMnBi2Se4-based intrinsic magnetic topological insulator heterostructures,”Phys. Rev. Mater. 8, 084202 (2024).38Y.-Y. Li, G. Wang, X.-G. Zhu, M.-H. Liu, C. Ye, X. Chen, Y.-Y. Wang,K. He, L.-L. Wang, X.-C. Ma, H.-J. Zhang, X. Dai, Z. Fang, X.-C.Xie, Y. Liu, X.-L. Qi, J.-F. Jia, S.-C. Zhang, and Q.-K. Xue, “Intrinsictopological insulator Bi2Te3 thin films on Si and their thickness limit,”Advanced Materials 22, 4002–4007 (2010).39K. Momma and F. Izumi, “VESTA 3 for three-dimensionalvisualization of crystal, volumetric and morphology data,”Journal of Applied Crystallography 44, 1272–1276 (2011).40T. Miyasato, N. Abe, T. Fujii, A. Asamitsu, S. Onoda, Y. Onose,N. Nagaosa, and Y. Tokura, “Crossover behavior of the anomalousHall effect and anomalous Nernst effect in itinerant ferromagnets,”Phys. Rev. Lett. 99, 086602 (2007).41To perform such calculations it is necessary to construct the localized Wan-nier function basis from the Bloch function basis derived from the DFTcalculations. However, for the BT/β -Bi/Si(111) interface model which con-tains huge number of atoms and which spectrum is very complex, the con-struction of large number of Wannier functions, accurately describing thespectrum, is highly problematic. On the other hand, the Berry’s curvatureis very sensitive to the spectrum accuracy. Moreover, one needs to considerthe temperature effect to understand why the AHE shows a maximum at10 K.This is the author’s peer reviewed, accepted manuscript. However, the online version of record will be different from this version once it has been copyedited and typeset.PLEASE CITE THIS ARTICLE AS DOI: 10.1063/5.0266580