2D Materials LETTER • OPEN ACCESS You may also like - Magnetoresistance studies of two- Large-area van der Waals epitaxy and magnetic dimensional Fe3GeTe2 nano-flake Xiangyu Zeng, Ge Ye, Shuyi Huang et al. characterization of Fe3GeTe2 films on graphene - Antiferromagnetic coupling of van der Waals ferromagnetic Fe3GeTe2 Dongseuk Kim, Sijin Park, Jinhwan Lee et al. To cite this article: J Marcelo J Lopes et al 2021 2D Mater. 8 041001 - Emerging intrinsic magnetism in two- dimensional materials: theory and applications Songrui Wei, Xiaoqi Liao, Cong Wang et View the article online for updates and enhancements. al. This content was downloaded from IP address 158.109.48.66 on 07/12/2022 at 11:02 2D Mater. 8 (2021) 041001 https://doi.org/10.1088/2053-1583/ac171d LETTER Large-area van der Waals epitaxy and magnetic characterization OPEN ACCESS of Fe3GeTe2 films on graphene RECEIVED 29 April 2021 J Marcelo J Lopes1,∗, Dietmar Czubak1, Eugenio Zallo1,7, Adriana I Figueroa2, Charles Guillemard3, REVISED Manuel Valvidares3, Juan Rubio-Zuazo4,5, Jesús López-Sanchéz4,5, Sergio O Valenzuela2,6, Michael Hanke1 23 June 2021 and Manfred Ramsteiner1,∗ ACCEPTED FOR PUBLICATION 1 22 July 2021 Paul-Drude-Institut für Festkörperelektronik, Leibniz-Institut im Forschungsverbund Berlin e.V., Hausvogteiplatz 5-7, 10117 Berlin, Germany PUBLISHED 2 11 August 2021 Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and BIST, Campus UAB, Bellaterra, 08193 Barcelona, Spain 3 ALBA Synchrotron Light Source, Barcelona 08290, Spain 4 Spanish CRG BM25-SpLine at The ESRF—The European Synchrotron, 38000 Grenoble, France Original content from 5 Instituto de Ciencia de Materiales de Madrid (ICMM), CSIC, 28049 Madrid, Spain this work may be used 6 under the terms of the Institució Catalana de Recerca i Estudis Avançats (ICREA), 08010 Barcelona, Spain 7 Creative Commons Current address: Walter Schottky Institut and Physik Department, Technische Universität München, Am Coulombwall 4, Attribution 4.0 licence. 85748 Garching, Germany. ∗ Any further distribution Authors to whom any correspondence should be addressed. of this work must maintain attribution to E-mail: lopes@pdi-berlin.de and ramsteiner@pdi-berlin.de the author(s) and the title of the work, journal Keywords: magnetic 2D materials, van der Waals heterostructures, van der Waals epitaxy, molecular beam epitaxy, ferromagnetism citation and DOI. Supplementary material for this article is available online Abstract Scalable fabrication of magnetic 2D materials and heterostructures constitutes a crucial step for scaling down current spintronic devices and the development of novel spintronic applications. Here, we report on van der Waals (vdW) epitaxy of the layered magnetic metal Fe3 GeTe2 (FGT)—a 2D crystal with highly tunable properties and a high prospect for room temperature ferromagnetism (FM)—directly on graphene by employing molecular beam epitaxy. Morphological and structural characterization confirmed the realization of large-area, continuous FGT/graphene heterostructure films with stable interfaces and good crystalline quality. Furthermore, magneto-transport and x-ray magnetic circular dichroism investigations confirmed a robust out-of-plane FM in the layers, comparable to state-of-the-art exfoliated flakes from bulk crystals. These results are highly relevant for further research on wafer-scale growth of vdW heterostructures combining FGT with other layered crystals such as transition metal dichalcogenides for the realization of multifunctional, atomically thin devices. 1. Introduction potential to tailor magnetism via proximity-induced phenomena [6, 8–10]. Layered magnetic materials such as Cr2 Ge2 Te6 [1], 2D magnets and related vdW heterostructures CrI3 [2], CrTe2 [3], and Fe3 GeTe2 (FGT) [4] are con- promise a large impact in future applications if they sidered to be promising building blocks for the devel- can sustain magnetic order up to room temperat- opment of ultra-compact spintronic devices with ure [11]. Within the current catalogue of available faster response and low-power dissipation. Moreover, materials, one of the most promising candidates to their study is expected to open new perspectives fulfill this requirement is FGT [4, 12]. FGT is a fer- on a more versatile modulation of magnetic prop- romagnetic conductor with a robust out-of-plane erties, beyond what can be achieved in traditional anisotropy [13] and, most remarkably, highly tunable 3D magnetic thin films [5–7]. As an example, com- properties. It has been shown that the Curie temperat- bining these layered magnets with other 2D crys- ure (T C ) in few-layer thick FGT can be boosted (from tals such as graphene, transition metal dichalcogen- around 100 K) to room temperature, or even beyond ides (TMDCs), or metal monochalcogenides to form that (up to 400 K), via electrostatic doping [4], or van der Waals (vdW) heterostructures offers great when grown onto a topological insulator [12]. Nearly © 2021 The Author(s). Published by IOP Publishing Ltd 2D Mater. 8 (2021) 041001 room temperature ferromagnetism (FM) could also In this work, we assess the feasibility of grow- be achieved by modulating the composition dur- ing FGT thin films directly on graphene surfaces by ing growth to realize the Fe-rich phases Fe4 GeTe2 using MBE. We employed different characterization and Fe5−x GeTe2 [14, 15]. Besides these possibilit- tools to investigate the structure and magnetic prop- ies, strain engineering and substitutional doping have erties of the FGT/graphene heterostructures. Large- also been identified as effective routes to exert con- scale, homogeneous FGT films exhibiting a smooth trol over the magnetic properties of FGT [16–18]. surface morphology could be realized on graphene Finally, although a native surface oxide forms upon with a preferential epitaxial orientation and low in- air exposure [19], degradation in FGT is not as severe plane mosaicity. Additionally, no structural damage as in chromium halides [20], which should facilitate or modification took place in graphene due to FGT its realistic exploration in devices. growth, indicating the formation of a stable vdW As for other magnetic 2D materials, experi- interface between the materials. Magnetism in the mental research on FGT has mainly concentrated on samples were probed macro- and microscopically, micrometer-sized flakes exfoliated from bulk crys- revealing a ferromagnetic behavior for the FGT films tals. This type of sample allows for exploring fun- that is similar to state-of-the art exfoliated flakes as damental properties as well as device concepts based well as thin films grown on 3D substrates. Overall, on vdW heterostacks [21]. However, it is not com- our results demonstrate that vdW epitaxy of FGT on patible with standard device fabrication, which usu- graphene is a powerful approach for the wafer-scale ally requires a wafer-scale, uniform form of material. fabrication of magnetic vdW heterostructures com- Recent progress has been made in this direction, bining dissimilar 2D crystals for spintronics. with a few studies reporting on molecular beam epi- taxy (MBE) of FGT films on Al2 O3 (0001) [12, 22], GaAs(111) [22], and Ge(111) [23]. MBE growth of 2. Experimental section FGT/MnTe [22] and FGT/Bi2 Te3 [12] heterostruc- tures, both on Al2 O3 (0001) wafers, has also been FGT films were synthesized by MBE (base pressure demonstrated. In general, the collection of data con- around 5 × 10−11 mbar) using elemental Fe, Ge, and tained in these studies suggests—as usual in conven- Te evaporated from Knudsen cells. For all growth tional epitaxy—that lattice-matching between FGT experiments the flux ratio of Fe, Ge, and Te was kept and its growth template is crucial to obtain large-area at ∼2:1:20. The flux for each element was obtained films with good crystalline quality. However, such by measuring the respective beam equivalent pres- requirement also reduces the number of choices in sure employing a pressure gauge. Note that the flux terms of suitable materials that can be employed as ratio was chosen as it resulted in the epitaxial 2D substrate for FGT, or combined with it to realize mag- growth of FGT films exhibiting structural and mag- netic superlattices. Interestingly, this seems to hold netic properties similar to those reported for stoi- true even for Bi2 Te3 -buffered Al2 O3 (0001) templates chiometric FGT bulk crystals and thin films, as it will [12], on which FGT growth should in principle pro- be shown in the next section. Differently from bulk ceed via vdW epitaxy [24]. The detection of strain crystal growth, for which a stoichiometric mixture of in the FGT film has been associated with the lattice Fe, Ge, and Te powders is usually employed as start- mismatch between both materials (the (a = b) in- ing material [13], the difference in sticking coeffi- plane lattice constant of Bi2 Te3 and FGT is 4.38 and cient for each element on the graphene surface at the 3.99 Å, respectively) [25, 26], indicating that inter- growth temperature is the main reason for the utiliz- layer interactions beyond vdW-bonding take place at ation of a ratio differing from the nominal compos- the FGT/Bi2 Te3 interface. In spite of this observa- ition of 3:1:2. A similar growth strategy was adop- tion, the weak nature of vdW interactions is gener- ted for MBE growth of FGT on the 3D substrates ally believed to alleviate the need for lattice match- Al2 O3 and Ge(111) [12, 23]. Continuous films with ing conditions [27], so that vdW epitaxy of FGT on thicknesses around 10 and 20 nm (i.e. from 12 to top of 2D crystals such as graphene, hexagonal BN 25 quintuple layers (QLs); 1 QL ∼ 0.8 nm formed (h-BN), and TMDCs should enable the transfer-free, by sequential Te/Fe/FeGe/Fe/Te slabs) were obtained scalable realization of 2D heterostructures exhibiting using growth times of 1 and 2 h, respectively. As sub- atomically smooth vdW interfaces. Nevertheless, the strates, 1 cm × 1 cm large epitaxial graphene on semi- intrinsically weak bonding between 2D materials also insulating, on-axis 4H-SiC(0001) were utilized. They creates difficulties namely uncontrolled nucleation were fabricated via SiC surface graphitization at high on morphological defects (e.g. wrinkles) [28, 29], temperatures in an Ar atmosphere, following well- which can in turn result in non-uniform growth established synthesis protocols [30, 31]. FGT films of low-quality material. Hence, exploiting epitaxial were also synthesized on 1 cm × 1 cm Al2 O3 (0001) growth of FGT on other 2D crystals is greatly deman- pieces using the same conditions for comparison. ded for the development of atomically thin mag- The graphene/SiC(0001) substrates were outgassed at netic vdW materials to be implemented into future 450 ◦ C (550 ◦ C for sapphire) for at least 1 h and then applications. cooled to 300 ◦ C for FGT growth. All substrates were 2 2D Mater. 8 (2021) 041001 coated with ∼1 µm of Ti on the backside via elec- 3. Results and discussion tron beam evaporation to allow non-contact heat- ing by radiation. In-situ growth monitoring was per- AFM analyses revealed the formation of continu- formed by reflection high energy electron diffrac- ous FGT films on graphene/SiC(0001) without strong tion (RHEED). Finally, in some cases the FGT films thickness inhomogeneity due to island formation were capped in-situ with a ∼5 nm thick amorph- or localized out-of-plane-growth. Figure 1(a) shows ous Te layer deposited after sample cooling to room a typical AFM height image of a ∼10 nm thick temperature. This procedure was adopted in order FGT film which demonstrates its coverage over the to reduce FGT surface oxidation upon prolonged air micrometer-large graphene/SiC surface terraces. Very exposure [23], which could hinder the successful use similar results were obtained for 20 nm thick FGT. of the surface-sensitive techniques x-ray absorption Discontinuities (holes) in the film are observed spectroscopy (XAS), x-ray magnetic circular dichro- mostly close to step edges (indicated by red arrows), ism (XMCD), and grazing incidence x-ray diffraction resulting in relatively small areas with exposed (GID) (see below). graphene (see figure 1(b)). The absence of FGT in Atomic force microscopy (AFM) in tapping mode these regions is confirmed by phase-contrast imaging was employed to probe the surface morphology of (not shown), as well as by the depth profile shown in the FGT films. Raman spectroscopy in back scat- figure 1(b) (inset). tering configuration was performed with the aim of Interestingly, a small step of ∼0.4 nm is visible in obtaining information about the structure of FGT the valley region of the line profile, which is known and graphene. Raman measurements were also car- to form at the border between n and n + 1 graphene ried out on commercially available FGT bulk crys- layers as a result of the SiC decomposition process tals (HQ graphene) for comparison. The Raman spec- during graphene formation [33]. The appearance of tra were excited at a wavelength of 473 nm with such morphological irregularities in the FGT film is the laser beam focused onto the sample surface to a probably related to an inability of FGT to grow— 1 µm diameter spot. In order to probe the azimuthal under the employed synthesis conditions—over the dependence of various in-plane lattice parameters, bi- to few-layer thick graphene ribbons and patches we have used synchrotron-based GID. This experi- located on step edge regions [30, 31, 33, 34]. They ment was performed at the SpLine beamline BM25 at exhibit a much lower chemical reactivity in compar- the European Synchrotron Radiation Facility (ESRF) ison to that of monolayer graphene covering surface in Grenoble. An incidence angle of the illuminating terraces. This variation is known to affect the over- x-rays of 0.2◦ sufficiently suppresses strong scattering growth of 2D materials [35]. Further optimization of by the substrate, and thus makes this method highly FGT growth parameters and graphene substrate pre- surface sensitive. A primary x-ray photon energy of paration are anticipated to mitigate this problem. The 17 keV (λ = 0.7293 Å) enables the inspection of a root-mean-square (RMS) roughness of the FGT film comparatively large area in reciprocal space, which was measured to be ∼1 nm in 1 µm × 1 µm on a sur- is important to probe also multiple reflections of the face terrace, which is higher than that measured for an same lattice plane family. area of same size in graphene (RMS ∼ 0.2 nm). Note Magneto-transport characterization was per- that the AFM images depicted here were acquired in formed within the 4 K–300 K temperature range non-capped FGT films exposed to air for several days. and with magnetic field up to 8 T, using a large-area Thus, natural oxidation of the topmost few layers of van der Pauw (vdP) geometry with Al contact wires the FGT film [19], as well as water and other sur- bonded at the edges of the 1 cm × 1 cm FGT film. face contaminants, are expected to affect the surface XAS and XMCD measurements were employed to roughness. investigate the magnetic properties microscopically Figure 1(c) exhibits a typical RHEED pattern [32]. XAS spectra at the Fe L3 and L2 edges were obtained during FGT growth (bottom panel), where recorded in the beam line BL-29 (BOREAS) at the the narrow streaks indicate epitaxial growth of a 2D ALBA synchrotron (Spain), which provides an ultra- film with smooth surface and good structural qual- high-vacuum sample environment with a base tem- ity. In addition, the shift in their location with respect perature of ∼3 K and magnetic field up to 6 T. Meas- to the pattern obtained for the graphene/SiC tem- urements used total-electron-yield detection, where plate (upper panel) reveals the evolution to a larger the drain current was measured from the sample to in-plane lattice constant, as expected for FGT growth the ground. The magnetic field was applied along the (see GID later). x-ray beam at normal incidence relative to the sample Raman spectroscopy (see figure 1(d)) confirmed plane. XMCD was obtained by subtracting XAS spec- that the graphene layer did not suffer structural tra with the photon helicity vector antiparallel and changes due to FGT synthesis. Besides the SiC back- parallel to the magnetic field (see supplementary ground signal, the graphene-related 2D and G peaks materials (available online at stacks.iop.org/2DM/8/ remain mostly unaltered. The small variation in the 041001/mmedia)). position and width of the 2D peak can be associated 3 2D Mater. 8 (2021) 041001 Figure 1. (a) AFM height image of a ∼10 nm thick FGT grown on graphene/SiC(0001). The red arrows indicate the position of surface steps in graphene/SiC(0001). (b) AFM image of a surface area close to a step edge in graphene/SiC(0001) where a discontinuity in FGT is evident. The inset shows the profile corresponding to the red line in the main figure. (c) In-situ RHEED patterns taken perpendicular to the SiC<100> direction, before (upper panel) and after FGT growth (lower panel). (d) Raman spectra showing the graphene and SiC components for samples with (blue spectrum) and without (green spectrum) a ∼20 nm thick FGT film on top. The inset depicts the A1g and A1g + E2g Raman components associated with the FGT film (blue spectrum). For comparison, the Raman spectrum collected from a bulk FGT crystal is also plotted (green spectrum). with excitation of surface areas containing a differ- ∼155 cm−1 are associated with the A1g and the sum ent number of graphene layers, e.g. close to step of the A1g + E2g phonon modes in FGT [37]. The A1g edge regions [30, 31, 35]. Additionally, other effects mode for the FGT is slightly blue-shifted with respect that might take place at the FGT/graphene interface to that of bulk FGT. This might be associated with such as charge transfer, hybridization, as well as local the strain in the FGT film (see GID results below), or strain fluctuations, are also anticipated to promote due to different degrees of degradation of the mater- variations in the graphene Raman modes [35, 36]. ials due to surface oxidation and/or laser exposure In order to disentangle the potential contribution [37]. Investigating these issues in detail requires ded- of each of them, future investigations are required. icated experiments. The strong peak located around Importantly, the intensity of the defect-related D peak 205 cm−1 originates from the SiC substrate [38]. does not increase after growth. This confirms the Further information on the structure of structural integrity of graphene and indicates that no FGT/graphene heterostacks was obtained by GID. rehybridization takes place in graphene, e.g. due to Figure 2(a) shows an in-plane reciprocal space map the formation of covalent bonds with FGT. Finally, of a Te-capped, ∼10 nm thick FGT film, in which the Raman measurements also served to confirm the the color-coded scattered intensity in reciprocal formation of FGT. The inset in figure 1(d) shows space is plotted as a function of reciprocal lattice the spectral region that is relevant for this mater- units (rlu), referring to the hexagonal lattice of the ial. The two components appearing at ∼120 and SiC(0001) substrate. The corresponding absolute 4 2D Mater. 8 (2021) 041001 Figure 2. (a) GID in-plane reciprocal space map of a Te-capped 10 nm thick FGT on graphene/SiC(0001), reflecting the azimuthal dependence of various in-plane lattice parameters. The white line shows a linear scan along the SiC[21̄.0] direction. (b) Two azimuthal intensity profiles intersecting the maxima caused by the FGT lattice. length (in units of Å−1 ) of the scattering vector is their maxima at K = 0, which fully confirms the additionally provided along the H hex axis. There previous statement. Moreover, there are additional are highly localized substrate reflections, namely low-intensity maxima symmetric to the FGT(22.0) the SiC(21̄.0) reflection and its higher orders. Over at 0◦ and ±60◦ , which are about 10◦ off the (21̄.0) imposed to the map itself there is a line scan crossing direction. These highly symmetric sub-features are the SiC(21̄.0) reflection. Note that we use the four- most probably due to a coincidence lattice between component vector notation for hexagonal symmetry, the FGT layer and the underlying graphene and/or i.e. (hklm) = (hk.m) = (hk−(h + k)m). As a guide for SiC substrate [34]. The very same coincidence lattice the eye we have drawn different radii corresponding might be responsible for the additional arcs next to to net planes distances from graphene, G(10.0) and the SiC(21̄.0) and (42̄.0), which contain pronounced G(11.0), and FGT, i.e. the FGT(n0.0) with n = [2..4] maxima azimuthally 20◦ off the [21̄.0] direction, or and FGT(nn.0) with [n = 1..3]. From the radial scan in other words: they are 10◦ off the [10.0] direction. one can extract quantitative information on the vari- The magnetic and electrical characteristics of ous in-plane lattice spacing values. Based on the three FGT/graphene heterostacks were investigated by most intense contributions from FGT [i.e. (11.0), transverse resistance (RXY ) measurements. In fer- (30.0) and (22.0)] one can deduce an average in- romagnetic materials, RXY under the application of plane lattice parameter for FGT of 4.011 Å, which an out-of-plane magnetic field H is given by the indicates a slight tensile strain of about 0.5% (taking superposition of the ordinary (OHE) and anomal- as reference value dFGT = 3.991 Å for bulk crystals ous (AHE) Hall effects and can be expressed as: [26]). This tensile strain is much smaller than what has been reported for FGT grown on Bi2 Te3 (1.7% RXY = ROHE + RAHE with ROHE = µ0 RO H, tensile strain) [12], in agreement with the anticipated weak interaction between FGT and graphene. We and RAHE = µ0 RA Mz (1) speculate that the observed strain might be related to structural and morphological deformations in the where µ0 is the vacuum permeability and M z the FGT layers, such as bending taking place around sur- out-of-plane component of the magnetization. RO face irregularities (graphene wrinkles) and the SiC and RA are the ordinary and anomalous Hall coef- surface steps. ficients, respectively [39]. Whereas the OHE (ROHE ) Following the intensity distribution along the contains information about the electrical character- FGT arcs [e.g. the FGT(22.0)], one recognizes a pro- istics, the AHE (RAHE ) can be utilized to investigate nounced preferential azimuthal orientation of the the magnetic order and the magnetization reversal FGT layer in a way that the FGT[11.0] net planes are in ferromagnetic films. Figure 3(a) displays RXY for parallel to SiC[11.0]. Two of those angular profiles a 20 nm thick FGT on graphene measured at differ- crossing the FGT(11.0) and (22.0) reflection are plot- ent temperatures during subsequent downward and ted in figure 2(b). They prove a full width at half max- upward sweeps of an external magnetic field. At low imum of about 4.7◦ , indicating a FGT lattice well- temperature (5 K), the AHE is clearly detected as aligned with respect to the substrate. Additionally, a square shape hysteresis loop superimposed on an there are weaker maxima 30◦ off revealing a smal- OHE contribution with its linear dependence on the ler fraction of the FGT layer which appears twinned external field (see equation 1) [23, 40]. The occur- to the major variant. Complementary to that, arcs rence of the AHE reflects the ferromagnetic order in corresponding to FGT(n0.0) (see figure 2(a)) have the FGT film, with the observed remanence providing 5 2D Mater. 8 (2021) 041001 Figure 3. (a) Transverse resistance RXY of a 1 cm × 1 cm FGT/graphene vdP structure during downward and upward magnetic field sweeps at different temperatures. The RXY curves for temperatures above 5 K are successively upshifted by 2.5 Ω for clarity. (b) Coercive field H C and saturation resistance ∆RAHE extracted from the AHE contribution as a function of temperature. The solid lines are guides to the eye. (c) Transverse resistance RXY of the same FGT/graphene vdP structure as a function of magnetic field in the range from −4 to −8 T. (d) Landau-level index of the quantum Hall effect (QHE) plateaus in (c) as a function of the inverse magnetic field. The solid line represents the result of a linear fitting. evidence for a strong perpendicular magnetic aniso- stacks.iop.org/2DM/8/041001/mmedia)), which is in tropy. In agreement with [22] and [23], both the agreement with [22]. The sheet resistivity of the coercive fields H C as well as the saturation resist- FGT/graphene heterostructure (150 Ω sq−1 ) is found ance ∆RAHE decrease with increasing temperature. to be almost one order of magnitude smaller than the Thereby, ∆RAHE is defined as the value of RAHE (see one commonly observed for our epitaxial graphene equation (1)) at magnetization saturation. The tem- films (∼2 kΩ sq−1 ). These findings are explained con- perature dependencies of H C and ∆RAHE are shown sistently by considering parallel conduction through in figure 3(b). Note that one has to be cautious in two transport channels for the analysis of the OHE regarding the absolute value of H C as a physical prop- [41, 42]. In this case, the sign as well as the abso- erty of FGT due to its possible dependence on domain lute value of the ROHE slope depend on the relative wall pinning effects. Nevertheless, the extrapolation conductivities of the FGT and graphene films. There- of both quantities to zero indicates in average a Curie fore in general they do not reflect the carrier type and temperature of about 220 K that agrees well with what density of only one of the two transport channels [42]. has been reported for FGT nanoflakes [40]. Figure 3(c) reveals that the transverse resistance From the slope of the OHE contribution to the RXY actually exhibits plateaus which can be explained Hall curves shown in figure 3(a) a low-temperature by the QHE in graphene or precursors of the QHE carrier density of 6.8 × 1013 cm−2 is extracted which, occurring at magnetic fields with magnitudes above in principle, can be ascribed to highly n-type doped 4 T [43]. Please note that the RXY data shown in graphene on SiC(0001). In contrast, FGT films are figure 3(c) contain a certain contribution of the expected to exhibit a hole-like transport character magnetic-field dependent longitudinal resistance due with an opposite ROHE slope and have a much lar- the utilized vdP geometry. The positions of the plat- ger carrier density (on the order of 2 × 1015 cm−2 ), eaus can be associated with Landau-level indices and as determined for films on sapphire substrates exhibit the expected linear dependence on the inverse (see supplementary materials (available online at field (µ0 H)−1 [43, 44], as can be seen in figure 3(d). 6 2D Mater. 8 (2021) 041001 paramagnetic response even though long range ferromagnetic order has already vanished. Applic- ation of the magneto-optical sum rules for XAS and XMCD [46–48] allows us to extract the orbital, spin and total moments (morb , mspin and mTotal ) of Fe 3d electrons in the ground state of FGT. The number of 3d holes was set to 4, as reported for the Fe 3d6 configuration in FGT [49], and the value of the mag- netic dipole operator term was neglected. From the experimental Fe L2,3 edges in the spectra at 3 K and 6 T we obtain morb = 0.047 ± 0.004 µB/Fe; mspin = 1.13 ± 0.09 µB/Fe and mTotal = 1.18 ± 0.10 µB/Fe. The orbital-to-spin moment ratio is 0.042 ± 0.008, which is comparable to that of ele- mental Fe(0.043) [48], somewhat larger than what is reported for FGT(0.03) [49], but lower than that of Fe5 GeTe2 (0.056) [50]. In summary, vdW epitaxy of FGT magnetic films on graphene was investigated by employing MBE. The realization of large-area FGT/graphene heterostruc- ture films with high structural and interface quality was demonstrated via a thorough structural charac- terization, as well as magneto-transport and XMCD investigations, which revealed a robust perpendicular magnetic anisotropy in FGT and the observation of plateaus associated with the QHE effect in epitaxial Figure 4. (a) Hysteresis loop recorded at the Fe L3 edge for a graphene. These results are highly relevant for further Te-capped, ∼20 nm thick FGT film on graphene/SiC(0001) at 3 K. The XMCD signal is normalized to its value in research on wafer-scale growth of atomically thin, all- saturation. (b) Temperature dependence of the Fe L3 epitaxial vdW heterostructures with multifunctional XMCD maximum in saturation (6 T, blue circles) and remanence (red triangles). Inset: XMCD signal, expressed properties. By combining FGT with other selected in percentage of the average XAS, recorded at 3 K under an 2D crystals including TMDCs, metal monochalco- out-of-plane magnetic field of 6 T (blue) and at remanence genides, and h-BN, which can be grown by precise (red). methods such as MBE [51–55], we envision the real- ization of spintronic devices with dimensions and performance not achievable with conventional bulk The occurrence of the QHE plateaus underlines the materials. high quality and carrier mobility of the graphene film underneath FGT over the large-area vdP structure. Data availability statement XMCD/XAS results obtained for a Te-capped, 20 nm thick FGT film on graphene/SiC(0001) are The data that support the findings of this study are shown in figure 4. The magnetic hysteresis loop available upon reasonable request from the authors. shown in figure 4(a) was recorded at base temperat- ure of 3 K with the beam energy fixed at the max- imum of the Fe L3 XMCD signal (706.4 eV) (see inset Acknowledgments figure 4(b)). The signal is normalized to an energy point at the pre-edge region (703 eV) [45], measured The authors would like to thank H-P Schönherr, on the fly with a magnetic field ramp of 1.5 T min−1 at C Hermann, and C Stemmler for their dedicated high B (from ±6 to ±2 T) and of 1.0 T min−1 at low maintenance of the MBE system, as well as A Riedel, B (from ±2 to ∓2 T). In accordance with the mag- W Seidel, S, Rauwerdink, and A Tahraoui for tech- netotransport results, the square shape of the loop nical support. Furthermore, the authors appreciate reveals the strong out-of-plane ferromagnetic charac- the critical reading of the manuscript by Yukihiko ter of the Fe atoms in the FGT film. Takagaki. They also acknowledge the provision The temperature dependence of the maximum of beamtime under the project HC-4068 at the of the Fe L3 XMCD signal in saturation (6 T) and European Synchrotron Radiation Facility (ESRF), remanence (0 T) is plotted in figure 4(b). The reman- located in Grenoble (France). ICN2 researchers ent magnetization extrapolates to zero at about acknowledge support from the European Union ∼220 K, which agrees closely with the T C found Horizon 2020 research and innovation programme by transport measurements. The 6 T XMCD signal under Grant Agreement No. 881603 (Graphene persists at room temperature, demonstrating a strong Flagship). 7 2D Mater. 8 (2021) 041001 ORCID iDs [25] Hosokawa Y, Tomita K and Takashiri M 2019 Sci. Rep. 9 10790 [26] Deiseroth H-J, Aleksandrov K, Reiner C, Kienle L and J Marcelo J Lopes  https://orcid.org/0000-0001- Kremer R K 2006 Eur. J. Inorg. Chem. 2006 1561 5268-1862 [27] Koma A 1992 Thin Solid Films 216 72 Eugenio Zallo  https://orcid.org/0000-0003-1796- [28] Heilmann M, Bashouti M, Riechert H and Lopes J M J 2018 9777 2D Mater. 5 025004 [29] Lin Y-C et al 2014 ACS Nano 8 3715 Charles Guillemard  https://orcid.org/0000-0002- [30] Emtsev K V et al 2009 Nat. Mater. 8 203 3347-0047 [31] Oliveira M H, Schumann T, Ramsteiner M, Lopes J M J and Manuel Valvidares  https://orcid.org/0000-0003- Riechert H 2011 Appl. Phys. Lett. 99 111901 4895-8114 [32] Figueroa A I and Van Der Laan G 2014 Coord. Chem. Rev. 277–8 95 Jesús López-Sanchéz  https://orcid.org/0000-0002- [33] Galves L A, Wofford J M, Soares G V, Jahn U, Pfüller C, 2670-9347 Riechert H and Lopes J M J 2017 Carbon 115 162 Manfred Ramsteiner  https://orcid.org/0000-0001- [34] Boschker J, Galves L A, Flissikowski T, Lopes J M J, 5455-2001 Riechert H and Calarco R 2015 Sci. Rep. 5 18079 [35] Heilmann M, Prikhodko A S, Hanke M, Sabelfeld A, Borgardt N I and Lopes J M J 2020 ACS Appl. Mater. References Interfaces 12 8897 [36] Froehlicher G, Lorchat E and Berciaud S 2018 Phys. Rev. X [1] Gong C et al 2017 Nature 546 265 8 011007 [2] Huang B et al 2017 Nature 546 270 [37] Du L et al 2019 Adv. Funct. Mater. 29 1904734 [3] Purbawati A et al 2020 ACS Appl. Mater. Interfaces 12 30702 [38] Burton J C, Sun L, Pophristic M, Lukacs S J, Long F H, [4] Deng Y et al 2018 Nature 563 94 Feng Z C and Ferguson I T 1998 J. Appl. Phys. 84 6268 [5] Gibertini M, Koperski M, Morpurgo A F and Novoselov K S [39] Checkelsky J G, Lee M, Morosan E, Cava R J and Ong N P 2019 Nat. Nanotechnol. 14 408 2008 Phys. Rev. B 77 014433 [6] Gong C and Zhang X 2019 Science 363 eaav4450 [40] Tan C, Lee J, Jung S-G, Park T, Albarakati S, Partridge J, [7] Zutic I, Fabian J and Sarma S D 2004 Rev. Mod. Phys. Field M R, McCulloch D G, Wang L and Lee C 2018 Nat. 76 323 Commun. 9 1554 [8] Chen S, Huang C, Sun H, Ding J, Jena P and Kan E 2019 J. [41] Farshchi R, Bruski P, Ramsteiner M, Herfort J, Brandt O, Phys Chem. C 123 17987 Manzke Y, Friedland K-J and Grahn H T 2011 Solid State [9] Dolui K, Petrovic M D, Zollner K, Plechac P, Fabian J and Commun. 151 436 Nikolic B K 2020 Nano Lett. 20 2288 [42] Harshman D R, Dow J D and Fiory A T 2011 Phil. Mag. [10] Sierra J F , Fabian J, Kawakami R K , Roche S and Valenzuela 91 818 S O 2021 Nat. Nanotechnol. (https://doi.org/10.1038/s41565- [43] Jobst J, Waldmann D, Speck F, Hirner R, Maude D K, 021-00936-x) Seyller T and Weber H B 2010 Phys. Rev. B 81 195434 [11] Valenzuela S O and Roche S 2019 Nat. Nanotechnol. 14 1088 [44] Zhang Y, Tan Y-W, Stormer H L and Kim P 2005 Nature [12] Wang H et al 2020 ACS Nano 14 10045 438 201 [13] Fei Z et al 2018 Nat. Mater. 17 778 [45] Goering E, Fuss A, Wber W, Will J and Schütz G 2000 [14] May A F, Ovchinnikov D, Zheng Q, Hermann R, Calder S, J. Appl. Phys. 88 5920 Huang B, Fei Z, Liu Y, Xu X and McGuire M A 2019 ACS [46] Thole B T, Carra P, Sette F and van der Laan G 1992 Phys. Nano 13 4436 Rev. Lett. 68 1943 [15] Seo J et al 2020 Sci. Adv. 6 eaay8912 [47] Carra P, Thole B T, Altarelli M and Wang X 1993 Phys. Rev. [16] Zhuang H L, Kent P R C and Hennig R G 2016 Phys Rev. B Lett. 70 694 93 134407 [48] Chen C T, Idzerda Y U, Lin H-J, Smith N V, Meigs G, [17] Hu X, Zhao Y, Shen X, Krasheninnikov A V, Chen Z and Chaban E, Ho G H, Pellegrin E and Sette F 1995 Phys. Rev. Sun L 2020 ACS Appl. Mater. Interfaces 12 26367 Lett. 75 152 [18] Tian C, Pan F, Xu S, Ai K, Xia T and Cheng P 2020 Appl. [49] Zhu J-X et al 2016 Phys. Rev. B 93 144404 Phys. Lett. 116 202402 [50] Yamagami K et al 2021 Phys. Rev. B 103 L060403 [19] Kim D et al 2019 Nanotechnology 30 245701 [51] Dau M T et al 2018 ACS Nano 12 2319 [20] Shcherbakov D et al 2018 Nano Lett. 18 4214 [52] Nakano M, Wang Y, Kashiwabara Y, Matsuoka H and [21] Lin H, Yan F, Hu C, Lv Q, Zhu W, Wang Z, Wei Z, Chang K Iwasa Y 2017 Nano Lett. 17 5595 and Wang K 2020 ACS Appl. Mater. Interfaces 12 43921 [53] Chang K and Parkin S S P 2020 J. Appl. Phys. 127 220902 [22] Liu S et al 2017 NPJ 2D Mater. Appl. 1 30 [54] Mortelmans W, de Gendt S, Heyns M and Merckling C 2021 [23] Roemer R, Liu C and Zou K 2020 NPJ 2D Mater. Appl. Appl. Mater. Today 22 100975 4 33 [55] Lopes J M J 2021 Prog. Cryst. Growth Charact. Mater. [24] Walsh L A and Hinkle C L 2017 Appl. Mater. Today 9 504 67 100522 8