The engineering of the surface morphology and the structure of the thin film is one of the essential technological assets for regulating the physical properties and functionalities of thin film-based devices. This study presents an easy and handy approach to tailor the surface structure of epitaxial thin films utilizing low-energy ion beam. Here, we investigate the evolution of the surface structure and magnetic anisotropy (MA) in epitaxial Fe/MgO (001) model systems subjected to multiple cycles of ion beam erosion (IBE) after thin film growth. The growth of Fe film occurs in the form of three–dimensional islands and exhibits intrinsic biaxial MA. Following a few cycles of IBE, an induced uniaxial magnetic anisotropy leads to a split in the hysteresis loop, and the film displays almost uniaxial magnetic switching behavior. More distinctly, we present a clear and conclusive evidence of (2 × 2) reconstruction of the Fe surface due to the atomic rearrangement by IBE. Furthermore, 57Fe isotope sensitive nuclear resonance scattering measurement provides insight into the depth-resolved magnetic information due to the modified surface topography. We also demonstrate that thermal annealing can reversibly tune the surface reconstruction and induced UMA. The feasibility of the IBE technique by adequately selecting IBE parameters for surface structure modification has been highlighted apart from conventional tailoring of the morphology for the tuning of UMA and introduces a new dimension to our understanding of self-assembled surface morphology evolution by IBE.

Magnetic thin films are an area of intense study that has expanded tremendously in the previous decade. The atomic-scale investigation of the link between an ultrathin epitaxial film and magnetic properties is a challenging problem due to several effects, such as the surface/interface morphology,1 interface diffusion,2 hybridization,3 film structure,4 and strain.5 On the other hand, these dependencies open up the possibilities of the fine tailoring of magnetic properties according to the desired functionalities. Magnetic anisotropy (MA) is a crucial characteristic of ferromagnetic materials from a fundamental perspective to study the interlayer exchange coupling,6 exchange bias,7 magneto-optical properties for magneto-plasmonics8 to the applications in magnetic memory devices, and sensors.9 The fundamental processes behind MA include dipolar and spin–orbit interactions;10 therefore, the external shape and the local lattice symmetry of the magnetic films are important factors in MA. Low-energy ion beam erosion (IBE) is a relatively new, versatile, and handy tool for creating a wide range of nano-morphologies on the thin film and substrate surface arising from the interplay between curvature-dependent ion beam sputtering and ion or temperature-dependent viscous flow of the material based on the ion energy and angle of incidence.11–14 Since the structure and morphology of ultrathin films directly affect their MA, researchers are utilizing the IBE technique for fine and flexible tailoring of MA.

The magnetic film deposited over the ion beam eroded patterned substrate (bottom-up approach) induces uniaxial magnetic anisotropy (UMA) in polycrystalline15–17 and epitaxial thin films18,19 due to shape anisotropy and step edge anisotropy by reduced atomic coordination.7 On the other hand, direct erosion of the film surface by ion beam (top-down approach) also induces UMA due to the formation of periodic ripple-like self-organized, quasi-sinusoidal ordered roughness on the film surface.12,20–22 In both approaches, the wavelength and amplitude of the ripple pattern are the two main factors influencing the magnetization dynamics. In particular, the Schlömann equation is useful for determining the magnetostatic contribution to the overall strength of induced UMA based on morphological factors for a magnetic film with anisotropic surface roughness.23 Exciting recent investigations demonstrate that, due to intrinsic crystalline anisotropy in the epitaxial thin film, a competition between shape and crystalline anisotropy gives rise to the multistep switching process, splitting the hysteresis loop,1,24 etc. Furthermore, a crossover in surface anisotropy contribution has also been observed in the polycrystalline thin film during direct erosion of the film surface.25 It has also been reported that in ultrathin single crystalline magnetic films grown on randomly corrugated surfaces where the magnetization is in an undulating state following the local surface profile caused by surface roughness, an additional MA contributes to the net MA.26 Even the amplitude and defects in the patterns significantly affect the distribution of the anisotropy axis or energy.27 These advancements show that the field is quite rich in physics and evolving day by day with the new findings. However, apart from these significant progresses, the impact of low-energy ion beam on the surface structure of the film is quite unexplored. It is important to note that the interaction of the film surface with low-energy ions causes mass redistribution (MR) due to the momentum transfer from the incoming ions.28–30 Therefore, the surface structure can also be altered with the low-energy ion beam, which can influence diffusion, nucleation, interface quality, and UMA.31 Furthermore, surface properties such as chemical reactivity, work function, vibration, and electronic state are influenced by the surface structure. However, studies exploring the corresponding ion beam interaction-induced structural changes in the polycrystalline and epitaxial thin film are missing. We have recently shown through in situ investigation that the IBE changes the surface crystallographic texture in the polycrystalline thin film.25,32 This motivates us to explore further the effect of IBE on MA and structural modification in the case of epitaxial thin films.

Fe/MgO system has recently received much attention due to its substantial interfacial perpendicular magnetic anisotropy33 and applicability in tunnel magnetoresistance devices.34 Therefore, we have selected the Fe/MgO as a prototype system for the present study. An epitaxial Fe film was grown on a MgO substrate under UHV conditions, and the Fe film surface was etched by Ar ion in several steps. The evolution of the surface structure and UMA was studied in situ simultaneously. The film exhibits an additional UMA that alters its magnetization-switching behavior. More importantly, IBE-induced a 2 × 2 surface reconstruction (SR) has been obtained. As surfaces of clean metal generally do not reconstruct because the range of atomic movement is restricted by atom jamming against each other, this study provides a pathway for tailoring clean surface structure and exploring deeper into the IBE-induced self-assembling phenomena.

Present experiments have been performed inside a UHV chamber with a base pressure of ∼5 × 10−10 mbar. The chamber has the facilities for thin film deposition by electron beam evaporation and in situ characterization using the magneto-optic Kerr effect (MOKE) and reflection high energy electron diffraction (RHEED technique). The chamber is also equipped with an ion gun for IBE. Before Fe deposition, the cleaned and well-ordered MgO(001) surface was prepared by cycles of Ar+ ion beam etching and subsequent annealing at 600 °C until a sharp RHEED pattern was observed. To prevent any growth-induced anisotropy, Fe was deposited by keeping the deposition flux normal to the substrate. A calibrated quartz crystal monitor monitored film thickness during deposition. After 2.5 nm deposition, the film surface was eroded with an Ar+ ion of energy 1 keV at an angle 50° from the surface normal. The azimuthal angle-dependent (ϕ) magnetic hysteresis loop was recorded using the longitudinal MOKE measurement. We have taken the projection of the IBE direction on the film surface as ϕ = 0°. MOKE and RHEED measurements were conducted by rotating the sample with respect to the IBE direction in order to correlate structural, morphological, and UMA in conjunction with the direction of IBE. To study depth-resolved magnetism (hyperfine field, spin orientation) using the synchrotron-based nuclear resonance scattering (NRS) technique, a separate film was prepared with Fe isotope (57Fe) on MgO (001) in identical conditions. NRS measurements were performed at beamline P01 at PETRA III, DESY, Hamburg, using energy 14.4 keV x ray, which corresponds to the 57Fe Mössbauer transition to determine the magnitude and direction of magnetic hyperfine fields (Bhf) at the bulk and interface regions of the Fe film. The synchrotron was operated in the 40-bunch mode with a bunch separation of 192 ns for these measurements. An avalanche photodiode detector was used for NRS measurement, having a time resolution of ∼1 ns.

Figures 1(a) and 1(b) show the RHEED images of the MgO substrate taken along the [110]MgO and [100]MgO directions, respectively, after the cleaning process. The presence of sharp streaks confirms the cleaned and single crystalline nature of the substrate. The corresponding streaky RHEED images obtained after the deposition of a 2.5 nm thick Fe film are presented in Figs. 1(c) and 1(d), which confirms the epitaxial growth of the Fe film with its well-established crystallographic orientations (001)[100]Fe‖(001)[110]MgO (see Fig. S1 in the supplementary material). It may be noted that the streaks are not continuous but somewhat spotty.35 This indicates that the Fe film surface is not atomically flat. The magnetic characterization of the sample, e.g., magnetization reversal process, coercivity, remanence, strength, and symmetry of magnetic anisotropy was done in situ by MOKE measurement. The MOKE system attached to the UHV chamber for in situ study can detect the magnetic signal from a few monolayers of the thin film covering from the superparamagnetic region to the ferromagnetic region.13 The MOKE hysteresis loops are recorded along the different in-plane directions with respect to the projection of the IBE direction (ϕ = 0°) on the film plane over 0°–180° angular range at 5° intervals. Three exemplary MOKE hysteresis loops taken by applying a magnetic field (H) along ϕ = 0°, 45°, and 90°, i.e., [100]Fe, [110]Fe, and [010]Fe directions are presented in Figs. 1(e)1(g), respectively. It is observed that MOKE loops along [100]Fe and [010]Fe directions exhibit high squareness with normalized magnetization (Mr/Ms) close to 1, whereas MOKE loops along the [110] direction have Mr/Ms close to 0.7. The polar plot of the remanence (Mr) value normalized to its saturation (Ms) value is plotted in Fig. 1(h). The shape of the graph indicates that the film exhibits well-defined biaxial MA.

FIG. 1.

In situ RHEED images taken along (a) [110]MgO and (b) [100]MgO directions. (c) and (d) represent RHEED images along [100] and [110] directions of the 2.5 nm thick Fe film. In situ MOKE hysteresis loop taken along (e) [100]Fe, (f) [110]Fe, and (g) [010]Fe directions, respectively. (h) The corresponding polar plot of the Mr/Ms value.

FIG. 1.

In situ RHEED images taken along (a) [110]MgO and (b) [100]MgO directions. (c) and (d) represent RHEED images along [100] and [110] directions of the 2.5 nm thick Fe film. In situ MOKE hysteresis loop taken along (e) [100]Fe, (f) [110]Fe, and (g) [010]Fe directions, respectively. (h) The corresponding polar plot of the Mr/Ms value.

Close modal

After initial characterization, the film surface was eroded with Ar+ ion step by step. Each step consists of 2.5 min IBE by 1 keV Ar+ ion. The projection of ions on the film surface was along the [100]Fe direction, corresponding to the thermodynamically preferred step orientation direction. The evolution of the film surface structure and MA was investigated as a function of IBE time to establish their correlation.

The evolution of RHEED images taken along the [100]Fe and [110]Fe direction at an interval of 2.5 min IBE is presented in Fig. 2. It is observed that after 7.5 min of IBE, an intermediate spot in between integer streaks appears, which becomes more prominent after the fourth cycle (10 min) of IBE. For quantitative information about the position of new streaks, the line profiles have been extracted from the RHEED images as a function of IBE time and are plotted in Figs. 2(i) and 2(j). We found that the new streaks are positioned between the integer streaks. A similar symmetric diffraction pattern is also recorded along equivalent crystallographic directions, confirming the (2 × 2) SR. Such changes in the RHEED patterns after erosion can be understood in terms of the surface morphology or structural changes on the surface. Generally, the physical origin of SR involves the displacement of the surface atom, increment or decrement of the surface density due to the addition or removal of the surface atom that leads to the minimization of the number of surface atom dangling bonds or improves the compactness of the outermost surface layer, etc.36–38 It is established that IBE at an oblique angle creates a periodical self-assembled morphology (nano-patterns) on the single crystal surface. Such nano-patterning on the single crystal surface can be explained theoretically by several models based on nonlinear and stochastic extensions to the Sigmund39 model-based Bradley–Harper theory.14 For example, Cuerno's model40 is given by
h t = v o + γ h x + ν x 2 h x 2 + ν y 2 h y 2 + λ x 2 ( h x ) 2 + λ y 2 ( h y ) 2 K 2 h + η .
(1)
FIG. 2.

In situ RHEED images taken along the (a)–(d) [100]Fe direction and (e)–(h) [110]Fe direction after successive 2.5 min IBE. Line profile of RHEED intensity distribution pattern extracted from RHEED image taken along (i) the [100]Fe and (j) [110]Fe directions, respectively.

FIG. 2.

In situ RHEED images taken along the (a)–(d) [100]Fe direction and (e)–(h) [110]Fe direction after successive 2.5 min IBE. Line profile of RHEED intensity distribution pattern extracted from RHEED image taken along (i) the [100]Fe and (j) [110]Fe directions, respectively.

Close modal

The readers are referred to Ref. 40 for a detailed description. Sputter-erosion-induced surface instability, mass transport, or local atomic rearrangement along the surface have been proposed for kinetic micro-roughening or smoothening and height evolution of the surface. In the present case, the energies of the incoming noble gas ions are just a few hundred electron volts. Therefore, both the sputtering/erosion process28,41 (although negligible) and redistributive/displacement mechanism of mass transport on the film surface concurrently exist. Moreover, this low-energy regime has also revealed the formation of nanostructures in the form of ripples and disordered arrays of spikes.28,42 In particular, crystalline metal does not get amorphized by the low-energy and low fluence IBE.43,44 Furthermore, it has been observed that in epitaxial thin film systems, erosion mechanism is much more dominant compared to the redistribution of atoms for the modification of the morphology or ripple pattern formation.43,44 However, it is easier to displace the atom within the cell of the amorphous material rather than removing it by sputtering and the former mechanism is pronounced for pattern formation in the amorphous material.43,44 Additionally, the surface response to low-energy ion bombardment in a close-packed crystalline structure such as Al and Fe is slower due to the need for the accumulation and propagation of permanent defects to change the surface morphology, while amorphous environments require less energy to permanently displace atoms. Nonetheless, in this low-energy domain, mass redistribution (MR) caused by the inelastic displacement of atoms significantly increases due to the transfer of momentum from the incident ions to those atoms residing near the solid surface.29,30 Atoms within the crystalline Fe structure are captured more frequently at positions of the nearest neighbors from their initial point of origin as defined by the BCC structure, and less frequently at sites further away. The ion flux used for the present experiment ranges from 1011 to 1012 ions/cm2, approximately 3–5 orders of magnitude less than the conventional ion fluxes used for surface patterning. Therefore, a complex interplay of ion-induced MR with low-energy and low flux ions causes SR, whereas erosion mechanism is responsible for the evolution of the surface morphology. Earlier studies have reported the SR of Fe surfaces due to unavoidable sulfur adsorption,45 migration of carbon atoms from the substrate, or contamination from carbon and oxygen atoms.46 Note that we performed the present experiment under UHV conditions. Moreover, we have repeated the experiment with an epitaxial Fe/Ag(001) system and obtained a similar 2 × 2 SR (see Fig. S2 in the supplementary material), nullifying any contamination or substrate-induced contribution to SR. Nonetheless, the ion-film physical interaction-based SR is advantageous compared to thermal annealing or impurity adsorption-induced SR, because the latter mechanisms can cause interface alloying, separate phase formation, etc. and show the novelty of the present approach.

The evolution of the hysteresis loop along (ϕ = 0°) and across (ϕ = 90°) to the IBE direction, i.e., along [100]Fe and [010]Fe directions after the second to fourth cycles, is reported in Fig. 3(a). One can observe that the height of the MOKE signal decreases with the IBE cycle due to the decrement in the film thickness associated with the sputter removal of the film.25 A closer inspection reveals that after 10 min of IBE, the loop along the ϕ = 0° direction has split into two semi-loops. The magnetic remanence also decreases in the as-deposited state from ∼1 to ∼0.15 after the fourth cycle of IBE. However, hysteresis loops along the ϕ = 90° direction retain their shape similar to the as-prepared state. It may be noted that both [100]Fe and [010]Fe directions are crystallographically and magnetically equivalent directions. Therefore, this change in the shape of the loop is associated with the change in the magnetization-switching process due to the onset of an extra UMA.47–49 The normalized remanence behavior after the fourth cycle of IBE is plotted in Fig. 3(b). A clear difference in the shape of the polar plot in comparison with that observed in Fig. 1(h) can be noted. It suggests that the symmetry of the MA of the film has been converted from purely biaxial to uniaxial or biaxial–uniaxial mixed state. Furthermore, it shows the utility of the IBE technique for inducing UMA and switching the anisotropy axis from one symmetry to the other. A schematic of the experimental geometry along with the crystallographic axis and anisotropy axis is shown in Fig. 3(c) for better visualization.

FIG. 3.

(a) Evolution of the MOKE hysteresis loop along 0° and 90° with the IBE cycle. The inset shows the relevant crystallographic orientations, symmetry of the anisotropy axis, IBE directions, and the in-plane azimuthal angle ϕ. (b) Polar plot of the normalized remanence value after the fourth cycle of IBE. (c) Schematic of IBE and crystallographic directions.

FIG. 3.

(a) Evolution of the MOKE hysteresis loop along 0° and 90° with the IBE cycle. The inset shows the relevant crystallographic orientations, symmetry of the anisotropy axis, IBE directions, and the in-plane azimuthal angle ϕ. (b) Polar plot of the normalized remanence value after the fourth cycle of IBE. (c) Schematic of IBE and crystallographic directions.

Close modal
In the present case, the UMA easy axis is oriented perpendicular to the IBE direction, i.e., along the [010]Fe direction. Therefore, the [010]Fe direction is the energetically easiest direction, whereas the [100]Fe direction is an intermediate direction due to the cubic easy and uniaxial hard axis. A quantitative measurement of the strength of induced UMA can be obtained by measuring the shift field Hs, which is proportional to the split in the loop. It is defined as the difference between the zero fields and the center of one shifted loop, i.e.,
H S = ( H S 2 H S 1 ) / 2.
(2)
Hs is pointed out in Fig. 3(a) and directly connected to the UMA constant,19,50
K U = M S H S .
(3)

Here, MS is the saturation magnetization. KU = 3.13 × 104erg/cm3 is obtained and is lower than the intrinsic biaxial anisotropy (KC).

The energy of anisotropic magnetization related to KC and KU via the following relation:
E = K C 4 si n 2 ( 2 ϕ ) + K U si n 2 ( ϕ ) M . H .
(4)
Since the film is epitaxial, an intrinsic magnetocrystalline anisotropy (MCA) is present. Therefore, the origin of induced UMA in the film can be devoted to combined contribution coming from the MCA due to the unbalance between the number of steps oriented parallel and perpendicular to the IBE direction, and low coordinated atoms present at step edge or surface defects sites, as well as shape anisotropy due to the ripple structure itself.20,51,24 The formation of an unidirectional correlated morphology, as shown schematically in Fig. 3(c), breaks the fourfold symmetry of the Fe surface, and induces a UMA oriented parallel to the ripple direction. Since the ion sculpting process is mainly limited to the surface and subsurface layers, MA with strength proportional to the inverse of film thickness (d) is expected. In such case, the strength of UMA due to anisotropic correlated morphology is expressed by Schlömann's formula,23 
K U ( shape ) = 2 π M S 2 π σ 2 ξ d .
(5)

Here, σ and ξ are the root mean square roughness and the average length of the ordered domain, respectively. Thus, the appearance of UMA after 10 min of IBE results from the gradual buildup of the density of atomic steps along the [100]Fe directions and gradually increases magnetostatic contributions.20 To draw the correlation between the surface structure on the induced UMA, its structural evolution was investigated by RHEED as a function of temperature. The RHEED images taken along the [100]Fe direction at different temperatures are presented in Figs. 4(a)4(d). The corresponding intensity profiles drawn at different temperatures are presented in the insets of each image. We observe that the intensity of the extra streaks starts to disappear close to 300 °C and completely disappears at 320 °C. The MOKE hysteresis loop taken at 320 °C along ϕ = 0°, 45°, and 90° directions, i.e., [100]Fe, [110]Fe, and [010]Fe directions are displayed in Fig. 4(e). We observe that all the hysteresis loop exhibits similar magnetization reversal behavior without any two-step switching as observed before annealing along the ϕ = 0° direction, suggesting the disappearance of the induced UMA.

FIG. 4.

(a)–(d) Temperature-dependent evolution of the RHEED pattern along the [100]Fe direction with the corresponding line profile of the RHEED intensity distribution pattern (inset). (e) In situ MOKE hysteresis loop taken at room temperature after annealing at 320 °C along ϕ = 0°, 45°, and 90° directions with respect to the IBE direction.

FIG. 4.

(a)–(d) Temperature-dependent evolution of the RHEED pattern along the [100]Fe direction with the corresponding line profile of the RHEED intensity distribution pattern (inset). (e) In situ MOKE hysteresis loop taken at room temperature after annealing at 320 °C along ϕ = 0°, 45°, and 90° directions with respect to the IBE direction.

Close modal

To further understand the interfacial magnetism (Bhf, spin orientation) due to SR, a 57Fe epitaxial thin film of 25 nm thickness was deposited on the same MgO(001) substrate. The film surface was eroded with identical ion fluence and ion energy for 10 min. The corresponding MOKE hysteresis loop, RHEED images, and x-ray diffuse scattering measurement are presented in the Figs. S3–S5 in the supplementary material. Similar to earlier observation, a 2 × 2 SR was also observed.

We conducted nuclear resonance reflectivity (NRR), x-ray fluorescence (XRF), and x-ray reflectivity (XRR) studies52,53 to determine the suitable incident angle to enhance nuclear resonance scattering (NRS) spectra. These measurements were carried out at the nuclear resonance beamline P01 at PETRA III, DESY, using a 14.4 keV (57Fe Mössbauer transition) x-ray photon. The nuclear (NRR) and electronic (XRR) parts of the signal were separated by making use of the fact that nuclear transitions are delayed in time due to the finite lifetime of the Mössbauer excited state (140 ns in the case of 57Fe isotope).54,55 Thus, photons detected within a few nanoseconds of the incident x-ray pulse constitute the XRR signal due to electronic scattering, while those detected in an interval of 10–160 ns after the incident x-ray pulse are used to get NRR and NRS patterns. Synchrotron radiation-based NRS technique is capable of determining the magnitude and direction of magnetic hyperfine fields at the bulk and surface/interface regions of the Fe film. The experimental XRF, XRR, and NRS plots along with their best fitting are shown in Fig. 5. The XRF and NRR peaks (marked by a dashed line) exactly match the critical angle (Qz = 0.057 Å−1) of Fe. Therefore, performing the NRS measurement at incident angle Qz = 0.057 Å−1 will enhance the nuclear resonance counts due to the confinement of the electric field intensity.52,53,56 Considering the above fact, the NRS data (time spectra) are collected at an incidence angle of Qz = 0.057 Å−1, as displayed in Fig. 5(c).

FIG. 5.

(a) XRR, XRF, and NRR patterns of the epitaxial 57Fe thin film deposited on the MgO (001) substrate. (b) Schematic of the experimental setup for NRS measurements. (c) The corresponding NRS time spectra with the best possible fitting. The red continuous line is the best fit to the experimental data.

FIG. 5.

(a) XRR, XRF, and NRR patterns of the epitaxial 57Fe thin film deposited on the MgO (001) substrate. (b) Schematic of the experimental setup for NRS measurements. (c) The corresponding NRS time spectra with the best possible fitting. The red continuous line is the best fit to the experimental data.

Close modal

The NRS spectrum (time spectra) is fitted using the REFTIM57 software with film parameters obtained from simultaneous XRR, NRR, and XRF fitting. The best match to the time spectra was achieved by splitting the 57Fe layer into three layers with three different Bhf components, namely, 32.7, 33.3, and 30.5 T (with an error of ±0.5 T), which corresponds to the top layer (2 nm), bulk layer (20.5 nm), and interface layer, respectively. After fitting, it is observed that the angles β and γ are about 90° and 0°, respectively. As shown in Fig. 5(b), β is the direction of Bhf with respect to the surface normal, and γ is the azimuthal angle with respect to the x-ray polarization direction. It indicates that the magnetic moment lies within the plane of the film surface. The slightly reduced hyperfine field of 32.7 T of the top layer may be due to the SR. The hyperfine field 33.3 T of the bulk layer matches well with the bulk Fe.3 However, the reduced hyperfine field of 30.5 T at the interface does not match any of the known oxides of Fe. Therefore, in the present case, the reduced Bhf at the interface might be related to the hybridization of O-2p and Fe-3d levels as claimed earlier3 and rules out the possible oxidation of Fe in the interfacial region.

In conclusion, our study establishes a comprehensive understanding of the effects of IBE on ultrathin epitaxial Fe films. We successfully demonstrated the ability to induce UMA in ultrathin Fe films through controlled morphology modification. Furthermore, our investigation revealed that IBE alters the surface structure and induces (2 × 2) SR in the Fe films. Thus, the present study shows that IBE holds excellent promise to simultaneously engineer the morphology, structure, and magnetic properties of thin films. In situ compatible surface-sensitive techniques such as RHEED must be utilized to explore structural details further and resolve questions related to self-assembled nanostructures, as demonstrated in this study. The induced UMA and SR hold great promise for various technological applications in thin film-based nanodevices. Additionally, controlled SR can be utilized to enhance the performance of magnetic devices through improved surface-to-volume ratios, increased interfacial interactions, and tailored magnetic domain structures. Future research can explore the feasibility of integrating these findings into practical applications and modeling to comprehend these phenomena.

See the supplementary material for the schematic of crystallographic orientations of the epitaxial Fe film with respect to the MgO(001) substrate, the RHEED image of the 2 × 2 reconstructed epitaxial Fe surface grown on the Ag (001) surface; and the MOKE hysteresis loop and RHEED images of the 2 × 2 reconstructed 25 nm thick epitaxial Fe film grown on the MgO(001) substrate.

The authors thank Dr. H. C. Wille and Dr. Olaf Leupold for the NFS measurement at the P01 beamline, PETRA III, Desy, Germany. The authors also acknowledge the Department of Science and Technology (Government of India) for providing financial support within the framework of the India@DESY collaboration.

The authors have no conflicts to disclose.

Anup Kumar Bera: Conceptualization (lead); Data curation (lead); Investigation (lead); Methodology (lead); Writing – original draft (lead); Writing – review & editing (lead). Md. Shahid Jamal: Conceptualization (supporting); Formal analysis (supporting); Investigation (supporting); Visualization (supporting); Writing – review & editing (supporting). Avinash Ganesh Khanderao: Formal analysis (supporting); Methodology (supporting); Software (supporting); Validation (supporting). Sharanjeet Singh: Formal analysis (supporting); Investigation (supporting). Dileep Kumar: Conceptualization (supporting); Project administration (lead); Resources (lead); Supervision (supporting); Writing – review & editing (supporting).

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

1.
A.
Kumar Bera
,
S.
Singh
,
Md.
Shahid Jamal
,
Z.
Hussain
,
V. R.
Reddy
, and
D.
Kumar
,
J. Magn. Magn. Mater.
544
,
168679
(
2022
).
2.
N.
Pandey
,
S.
Pütter
,
S. M.
Amir
,
V. R.
Reddy
,
D. M.
Phase
,
J.
Stahn
,
A.
Gupta
, and
M.
Gupta
,
Phys. Rev. Mater.
3
,
114414
(
2019
).
3.
G.
Sharma
,
A.
Gupta
,
M.
Gupta
,
K.
Schlage
, and
H.-C.
Wille
,
Phys. Rev. B
92
,
224403
(
2015
).
4.
G.
Chen
,
S. T.
Howard
,
A. B.
Maghirang
,
K.
Nguyen Cong
,
R. A. B.
Villaos
,
L. Y.
Feng
,
K.
Cai
,
S. C.
Ganguli
,
W.
Swiech
,
E.
Morosan
,
I. I.
Oleynik
,
F. C.
Chuang
,
H.
Lin
, and
V.
Madhavan
,
Phys. Rev. B
102
,
115149
(
2020
).
5.
P.
Dey
,
T. K.
Nath
, and
A.
Taraphder
,
Appl. Phys. Lett.
91
,
012511
(
2007
).
6.
M.
Körner
,
K.
Lenz
,
M. O.
Liedke
,
T.
Strache
,
A.
Mücklich
,
A.
Keller
,
S.
Facsko
, and
J.
Fassbender
,
Phys. Rev. B
80
,
214401
(
2009
).
7.
M. O.
Liedke
,
B.
Liedke
,
A.
Keller
,
B.
Hillebrands
,
A.
Mücklich
,
S.
Facsko
, and
J.
Fassbender
,
Phys. Rev. B
75
,
220407
(
2007
).
8.
A.
Toma
,
D.
Chiappe
,
D.
Massabò
,
C.
Boragno
, and
F.
Buatier De Mongeot
,
Appl. Phys. Lett.
93
,
163104
(
2008
).
9.
J. M.
Iwata-Harms
,
G.
Jan
,
S.
Serrano-Guisan
,
L.
Thomas
,
H.
Liu
,
J.
Zhu
,
Y. J.
Lee
,
S.
Le
,
R. Y.
Tong
,
S.
Patel
,
V.
Sundar
,
D.
Shen
,
Y.
Yang
,
R.
He
,
J.
Haq
,
Z.
Teng
,
V.
Lam
,
P.
Liu
,
Y. J.
Wang
,
T.
Zhong
,
H.
Fukuzawa
, and
P. K.
Wang
,
Sci. Rep.
9
,
19407
(
2019
).
10.
B. D.
Cullity
and
C. D.
Graham
,
Introduction to Magnetic Materials
(John Wiley & Sons, Inc., Hoboken, New Jersey,
2008
).
11.
M. O.
Liedke
,
M.
Körner
,
K.
Lenz
,
M.
Fritzsche
,
M.
Ranjan
,
A.
Keller
,
E.
Čižmár
,
S. A.
Zvyagin
,
S.
Facsko
,
K.
Potzger
,
J.
Lindner
, and
J.
Fassbender
,
Phys. Rev. B
87
,
024424
(
2013
).
12.
H. L.
Liu
,
T.
Škereň
,
A.
Volodin
,
K.
Temst
,
A.
Vantomme
, and
C.
Van Haesendonck
,
Phys. Rev. B
91
,
104403
(
2015
).
13.
A. K.
Bera
,
P.
Gupta
,
D.
Garai
,
A.
Gupta
, and
D.
Kumar
,
Appl. Surf. Sci. Adv.
6
,
100124
(
2021
).
14.
R. M.
Bradley
and
J. M. E.
Harper
,
J. Vac. Sci. Technol. A
6
,
2390
(
1988
).
15.
A.
Keller
,
L.
Peverini
,
J.
Grenzer
,
G. J.
Kovacs
,
A.
Mücklich
, and
S.
Facsko
,
Phys. Rev. B
84
,
035423
(
2011
).
16.
K.
Bukharia
,
P.
Karmakar
,
D.
Kumar
,
V. R.
Reddy
, and
A.
Gupta
,
J. Magn. Magn. Mater.
497
,
165934
(
2020
).
17.
A. K.
Bera
,
A. S.
Dev
,
M.
Kuila
,
M.
Ranjan
,
P.
Pandit
,
M.
Schwartzkopf
,
S. V.
Roth
,
V. R.
Reddy
, and
D.
Kumar
,
Appl. Surf. Sci.
581
,
152377
(
2022
).
18.
M. O.
Liedke
,
M.
Körner
,
K.
Lenz
,
F.
Grossmann
,
S.
Facsko
, and
J.
Fassbender
,
Appl. Phys. Lett.
100
,
242405
(
2012
).
19.
R.
Moroni
,
D.
Sekiba
,
F.
Buatier de Mongeot
,
G.
Gonella
,
C.
Boragno
,
L.
Mattera
, and
U.
Valbusa
,
Phys. Rev. Lett.
91
,
167207
(
2003
).
20.
F.
Bisio
,
A.
Toma
,
R.
Moroni
,
R.
Pasero
,
F.
Buatier De Mongeot
,
C.
Boragno
,
M.
Canepa
,
U.
Valbusa
, and
L.
Mattera
,
Phys. Rev. B
75
,
054407
(
2007
).
21.
M. A.
Arranz
,
J. M.
Colino
, and
F. J.
Palomares
,
J. Appl. Phys.
115
,
183906
(
2014
).
22.
F.
Bisio
,
R.
Moroni
,
F.
Buatier de Mongeot
,
M.
Canepa
, and
L.
Mattera
,
Phys. Rev. Lett.
96
,
057204
(
2006
).
23.
E.
Schlömann
,
J. Appl. Phys.
41
,
1617
(
1970
).
24.
R.
Moroni
,
F.
Bisio
,
F.
Buatier De Mongeot
,
C.
Boragno
, and
L.
Mattera
,
Phys. Rev. B
76
,
214423
(
2007
).
25.
A. K.
Bera
and
D.
Kumar
,
ACS Appl. Electron. Mater.
2
,
3686
(
2020
).
26.
C. A. F.
Vaz
,
S. J.
Steinmuller
, and
J. A. C.
Bland
,
Phys. Rev. B
75
,
132402
(
2007
).
27.
J. M.
Colino
,
M. A.
Arranz
,
A. J.
Barbero
,
A.
Bollero
, and
J.
Camarero
,
J. Phys. D: Appl. Phys.
49
,
135002
(
2016
).
28.
A.
Lopez-Cazalilla
,
D.
Chowdhury
,
A.
Ilinov
,
S.
Mondal
,
P.
Barman
,
S. R.
Bhattacharyya
,
D.
Ghose
,
F.
Djurabekova
,
K.
Nordlund
, and
S.
Norris
,
J. Appl. Phys.
123
,
235108
(
2018
).
29.
R. M.
Bradley
and
T.
Sharath
,
Phys. Rev. E
103
,
022804
(
2021
).
30.
B.
Davidovitch
,
M. J.
Aziz
, and
M. P.
Brenner
,
Phys. Rev. B
76
,
205420
(
2007
).
31.
U.
Welp
,
V. K.
Vlasko-Vlasov
,
X.
Liu
,
J. K.
Furdyna
, and
T.
Wojtowicz
,
Phys. Rev. Lett.
90
,
4
(
2003
).
32.
A. K.
Bera
,
A. S.
Dev
, and
D.
Kumar
,
Appl. Phys. Lett.
122
,
022405
(
2023
).
33.
J. W.
Koo
,
S.
Mitani
,
T. T.
Sasaki
,
H.
Sukegawa
,
Z. C.
Wen
,
T.
Ohkubo
,
T.
Niizeki
,
K.
Inomata
, and
K.
Hono
,
Appl. Phys. Lett.
103
,
192401
(
2013
).
34.
S.
Yuasa
,
T.
Nagahama
,
A.
Fukushima
,
Y.
Suzuki
, and
K.
Ando
,
Nat. Mater.
3
,
868
(
2004
).
35.
A. K.
Bera
and
D.
Kumar
,
Metall. Mater. Trans. A
49
,
5205
(
2018
).
36.
S.
Olivier
,
A.
Saúl
, and
G.
Tréglia
,
Appl. Surf. Sci.
212–213
,
866
(
2003
).
37.
C. W.
Burrows
,
S. A.
Hatfield
,
F.
Bastiman
, and
G. R.
Bell
,
J. Phys.: Condens. Matter
26
,
395006
(
2014
).
38.
B. V.
Rao
,
D.
Gruznev
,
T.
Tambo
, and
C.
Tatsuyama
,
Appl. Surf. Sci.
175–176
,
187
(
2001
).
40.
R.
Cuerno
and
A.-L.
Barabási
,
Phys. Rev. Lett.
74
,
4746
(
1995
).
41.
M.
Nastasi
,
J.
Mayer
, and
J. K.
Hirvonen
,
Ion-Solid Interactions: Fundamentals and Applications
(
Cambridge University Press
,
1996
).
42.
A.
Metya
and
D.
Ghose
,
Appl. Phys. Lett.
103
,
161602
(
2013
).
43.
A.
Lopez-Cazalilla
,
A.
Ilinov
,
K.
Nordlund
, and
F.
Djurabekova
,
J. Phys.: Condens. Matter
31
,
075302
(
2019
).
44.
U.
Valbusa
,
C.
Boragno
, and
F. B.
De Mongeot
,
J. Phys.: Condens. Matter
14
,
8153
(
2002
).
45.
H. D.
Shih
,
F.
Jona
,
D. W.
Jepsen
, and
P. M.
Marcus
,
Phys. Rev. Lett.
46
,
731
(
1981
).
46.
M.
Sicot
,
S.
Andrieu
,
C.
Tiusan
,
F.
Montaigne
, and
F.
Bertran
,
J. Appl. Phys.
99
,
08D301
(
2006
).
47.
A. T.
Hindmarch
,
D. A.
Arena
,
K. J.
Dempsey
,
M.
Henini
, and
C. H.
Marrows
,
Phys. Rev. B
81
,
100407(R)
(
2010
).
48.
M.
Dumm
,
M.
Zölfl
,
R.
Moosbühler
,
M.
Brockmann
,
T.
Schmidt
, and
G.
Bayreuther
,
J. Appl. Phys.
87
,
5457
(
2000
).
49.
H. W.
Zhao
,
Y. Z.
Wu
,
C.
Won
, and
Z. Q.
Qiu
,
J. Appl. Phys.
95
,
7300
(
2004
).
50.
W.
Weber
,
C. H.
Back
,
A.
Bischof
,
C.
Würsch
, and
R.
Allenspach
,
Phys. Rev. Lett.
76
,
1940
(
1996
).
51.
D.
Sekiba
,
R.
Moroni
,
G.
Gonella
,
F.
Buatier de Mongeot
,
C.
Boragno
,
L.
Mattera
, and
U.
Valbusa
,
Appl. Phys. Lett.
84
,
762
(
2004
).
52.
M. S.
Jamal
,
P.
Gupta
,
I.
Sergeev
,
O.
Leupold
, and
D.
Kumar
,
Phys. Rev. B
107
,
075416
(
2023
).
53.
M. S.
Jamal
,
Y.
Kumar
,
M.
Gupta
,
P.
Gupta
,
I.
Sergeev
,
H. C.
Wille
, and
D.
Kumar
,
Hyperfine Interact.
242
,
5
(
2021
).
54.
R.
Rüffer
and
A. I.
Chumakov
,
Hyperfine Interact.
97–98
,
589
(
1996
).
55.
56.
A. G.
Khanderao
,
I.
Sergueev
,
H. C.
Wille
, and
D.
Kumar
,
Appl. Phys. Lett.
116
,
101603
(
2020
).
57.
M. A.
Andreeva
,
Hyperfine Interact.
185
,
17
(
2008
).