# Fileset

[e-J SSNT 24_90(2026)_Takezawa_ARPES.pdf](https://mdr.nims.go.jp/filesets/2c8e4e65-060a-420a-ac22-cc471ad7c4f3/download)

## Creator

Shingo Takezawa, Daichi Ishikawa, Soichi Takase, Toshiro Osawa, Riku Goto, Arpita Varadwaj, Kentaro Fuku, Masaki Imamura, Yukimi Tanimoto, Hitoshi Sato, Kiyohisa Tanaka, [Yuya Sakuraba](https://orcid.org/0000-0003-4618-9550), [Satoru Ichinokura](https://orcid.org/0000-0002-7968-4016), [Koichiro Yaji](https://orcid.org/0000-0002-0721-1316), Masato Kotsugi, [Naoka Nagamura](https://orcid.org/0000-0002-7697-8983)

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Electronic Structure of Metastable bcc-Cu Ultra-thin Films Grown on FeCo(100) Probed by Angle-resolved Photoemission Spectroscopy](https://mdr.nims.go.jp/datasets/32c3971d-02e1-417b-a2d8-752c0118f234)

## Fulltext

Electronic Structure of Metastable bcc-Cu Ultra-thin Films Grown on FeCo(100) Probed by Angle-resolved Photoemission Spectroscopy  e-Journal of Surface Science and Nanotechnology 24, 90–97 (2026) Electronic Structure of Metastable bcc-Cu Ultra-thin Films Grown on FeCo(100) Probed by Angle-resolved Photoemission Spectroscopy Shingo Takezawa,a, b Daichi Ishikawa,a, b Soichi Takase,a, b Toshiro Osawa,a, b Riku Goto,a, b Arpita Varadwaj,b Kentaro Fuku,c Masaki Imamura,d Yukimi Tanimoto,e Hitoshi Sato,f Kiyohisa Tanaka,g Yuya Sakuraba,b Satoru Ichinokura,b Koichiro Yaji,b, h Masato Kotsugi,a Naoka Nagamura a, b, i, † a Department of Materials Science and Technology, Tokyo University of Science, Katsushika-ku, Tokyo 125-8585, Japan b National Institute for Materials Science, Tsukuba, Ibaraki 305-0047, Japan c Department of Chemistry, Graduate School of Science, Nagoya University, Nagoya, Aichi 464-8602, Japan d Synchrotron Light Application Center, Saga University, Honjo, Saga 840-8502, Japan e Graduate School of Advanced Science and Engineering, Hiroshima University, Higashi-Hiroshima, Hiroshima 739-8526, Japan f Research Institute for Synchrotron Radiation Science, Hiroshima University, Higashi-Hiroshima, Hiroshima 739-0046, Japan g UVSOR Facility, Institute for Molecular Science, Okazaki, Aichi 444-8585, Japan h Unprecedented-scale Data Analytics Center, Tohoku University, Sendai, Miyagi 980-8578, Japan i Research Institute of Electrical Communication, Tohoku University, Sendai, Miyagi 980-8577, Japan † Corresponding author: NAGAMURA.Naoka@nims.go.jp Received: 11 March, 2026; Accepted: 14 July, 2026; J-STAGE Advance Publication: 5 September, 2026; Published: 5 September, 2026 We report an experimental investigation of the electronic band structure of metastable body-centered cubic (bcc)-Cu ultra-thin films epitaxially grown on FeCo(100) substrates. High-quality bcc-Cu films were fabricated by molecular beam epitaxy, ena- bling in-situ synchrotron angle-resolved photoemission spec- troscopy measurements with sufficient energy and momentum resolution. The Fermi surface morphology of bcc-Cu was dis- tinguished from that of conventional face-centered cubic-Cu and showed characteristic features consistent with the bcc structure. Overall, the experimentally observed band dispersions exhibit good qualitative agreement with density functional theory (DFT) calculations. However, the Cu-derived bands are systematically shifted toward higher binding energies compared to bulk-based DFT predictions. These deviations are likely caused by a combination of intrinsic limitations of the DFT approximation and interface-induced effects such as charge transfer at the Cu/FeCo heterointerface. The present results highlight the importance of experimentally determining the electronic structure of metastable ultra-thin films and demonstrate that interface effects must be considered when discussing Fermi surface matching and transport properties in Cu/FeCo heterostructures. Keywords Angle-resolved photoemission; Molecular beam epitaxy; Copper; Metastable states I. INTRODUCTION The crystallographic growth of metallic thin films has long been a central topic in surface science \[1\]. As the film thickness is reduced to the nanometer scale, thin films in- creasingly exhibit low-dimensional characteristics, and their structural, electronic, and transport properties become strongly influenced by the underlying substrates. In such ultra-thin regimes, film growth is governed by a complex interplay of epitaxial strain, interfacial bonding, quantum- size effects, and growth parameters. Consequently, the result- ing electronic structure often deviates from predictions based Regular Paper e-J. Surf. Sci. Nanotechnol. 24, 90–97 (2026) \| DOI: 10.1380/ejssnt.2026-014 90[mailto:NAGAMURA.Naoka@nims.go.jp](mailto:NAGAMURA.Naoka@nims.go.jp)[https://doi.org/10.1380/ejssnt.2026-014](https://doi.org/10.1380/ejssnt.2026-014)on idealized bulk models obtained from density functional theory (DFT) calculations. Copper (Cu) is one of the most ubiquitous metallic ele- ments and is widely used in electronic devices as an inter- connect material. In particular, FeCo/Cu/FeCo multilayer heterostructures have attracted considerable attention because they exhibit extremely large magnetoresistance (MR) ratios when employed as current-in-plane giant magnetoresistance (CIP-GMR) devices \[2–4\]. Previous studies have attributed this enhanced MR effect to strong spin-dependent scattering at the Cu/FeCo interfaces. The underlying mechanism has been interpreted in terms of Fermi surface matching, or electron band matching, in which the Fermi surface of the nonmagnetic Cu layer closely matches that of the majority- spin band of the ferromagnetic FeCo layer, while showing a significant mismatch with the minority-spin band \[2, 5\]. As a result, the Cu/FeCo interface is expected to act as an efficient spin filter, preferentially transmitting majority-spin electrons. It has been reported that ultra-thin Cu layers grown on body-centered cubic (bcc)-FeCo(100) substrates by sputter- ing do not adopt the bulk-stable face-centered cubic (fcc) structure, but instead form a metastable bcc phase templated by the underlying FeCo lattice \[2–4, 6, 7\]. These findings highlight that the physical properties of Cu thin films in such heterostructures are strongly influenced by the substrate and may differ substantially from those of bulk Cu. Nevertheless, most theoretical discussions about electronic states of bcc- Cu, including Fermi surface matching at the bcc-Cu/FeCo interface, have been based on calculations assuming ideal bulk bcc-Cu \[2, 5, 8\], neglecting the effects of reduced dimensionality, epitaxial strain, and interfacial interaction inherent to ultra-thin films. Therefore, an experimental deter- mination of the electronic structure of bcc-Cu in realistic Cu/ FeCo heterostructures is essential for a quantitative evalua- tion of Fermi surface and electron band matching in actual devices. Despite its importance, experimental investigations of the electronic structure of bcc-Cu remain scarce. Although DFT calculations have extensively explored the band structure of bulk bcc-Cu, previous angle-resolved photoemission spec- troscopy (ARPES) studies on bcc-Cu films grown on Ag(100) \[9\] were limited by insufficient energy and momen- tum resolution. Moreover, the bcc-Cu films often contained substantial fractions of fcc domains and structural defects \[9, 10\] making it difficult to unambiguously identify the intrinsic electronic states of bcc-Cu. Because both Cu and FeCo surfaces are highly susceptible to oxidation, reliable ARPES measurements further require the preparation of large-area, structurally uniform films and in-situ spectroscopic character- ization. In our previous work, we demonstrated that high-quality, large-area bcc-Cu ultra-thin films can be fabricated on Fe- Co(100) substrates by molecular beam epitaxy (MBE), with minimal structural disorder and negligible fcc-Cu contami- nation \[11\]. Building on this growth technique, the present study reports high-resolution in-situ synchrotron ARPES measurements of the electronic structure of bcc-Cu epitax- ially grown on FeCo(100). By combining multi-photon-en- ergy ARPES with first-principles calculations, we experi- mentally determine the band dispersions and Fermi surface morphology of bcc-Cu in this heterostructure and quantita- tively assess their deviations from bulk DFT predictions. These results provide essential insight into the electronic structure of metastable bcc-Cu ultra-thin films and an exper- imental foundation for evaluating Fermi surface matching at Cu/FeCo interfaces in spintronic devices. II. EXPERIMENTAL FeCo and Cu thin films were grown on MgO(100) sub- strates by MBE in an ultra-high vacuum (UHV) system with a base pressure of \~9 × 10−8 Pa. MgO(100) substrates were annealed at 800K for 30min prior to deposition to obtain clean and well-ordered surfaces. The evaporation sources were a high-purity Fe0.5Co0.5 rod (TOSHIMA Manufactur- ing, 2 × 2 × 20mm, 4N) and a Cu wire (Nilaco, 0.25mm diameter, 5N). The surface crystallinity and structural phase of the FeCo layers and Cu overlayers were examined by in- situ low-energy electron diffraction (LEED). FeCo layers were first deposited on MgO(100) substrates and served as the structural and electronic template for the subsequent growth of Cu overlayers. FeCo films were grown using an electron-beam (EB) evaporator with the emission current fixed at 20mA, while maintaining the substrate tem- perature at 600K. The FeCo film thickness was confirmed ex-situ by X-ray reflectivity (XRR) measurements using a laboratory-based X-ray diffractometer (Rigaku SmartLab), performed after the ARPES experiments, and a thickness of at least 20 nm was verified for all samples prepared at each synchrotron facility. The thickness was chosen to ensure bulk-like electronic properties of the FeCo layer. After the growth of FeCo, Cu overlayers were deposited at room temperature using an EB evaporator with the emission current fixed at 30mA. In our previous studies, Cu films grown on FeCo(100) were shown to exhibit a bcc structure only in the ultra-thin regime below approximately 1 nm, whereas thicker films relaxed toward the bulk-stable fcc phase \[11\]. Consistent with this thickness-dependent struc- tural transition, the bcc nature of the present Cu overlayers was verified by in-situ LEED I-V analysis over an incident electron-energy range of 150–400 eV. Accordingly, the nom- inal thickness of the Cu layers in the present study was below 1 nm. The presence of Cu in the overlayer was further con- firmed by laboratory-based in-situ X-ray photoelectron spec- troscopy (XPS) measurements performed on reference sam- ples prepared under identical growth conditions. The XPS measurements were carried out using a VG Scienta SES-100 hemispherical analyzer with an Al Kα radiation source (hν = 1486.6 eV). ARPES measurements were carried out at three synchro- tron radiation facilities: the BL-7 beamline of HiSOR, the BL13 beamline of SAGA-LS \[12\], and the BL7U beamline of UVSOR \[13, 14\]. The photon-energy ranges used were 30–400, 45–180, and 12–38 eV at HiSOR, SAGA-LS, and Regular Paper e-J. Surf. Sci. Nanotechnol. 24, 90–97 (2026) \| DOI: 10.1380/ejssnt.2026-014 91[https://doi.org/10.1380/ejssnt.2026-014](https://doi.org/10.1380/ejssnt.2026-014)UVSOR, respectively. The total energy resolution was set to approximately 130meV (hν = 140 eV), 75meV (hν = 140 eV), and 38meV (hν = 38 eV), at HiSOR, SA- GA-LS, and UVSOR, respectively. The angular resolution was about 0.1° at all endstations. Linearly polarized light with horizontal polarization was used in all ARPES measure- ments. During data acquisition, the pressure in the measure- ment chambers was maintained below 1 × 10−8 Pa, and the sample temperature was kept below 20K. All beamlines were equipped with UHV preparation chambers and LEED optics directly connected to the ARPES measurement cham- bers, allowing the as-grown samples to be transferred and measured in-situ without exposure to air. The binding-energy scale was calibrated using the Fermi edge of each sample and cross-checked with reference measurements on a gold sur- face. First-principles calculations of the electronic band struc- ture of bcc-Cu and bcc-FeCo were performed within DFT using the Vienna Ab initio Simulation Package (VASP) \[15\]. The exchange-correlation functional was treated within the semi-local generalized gradient approximation (GGA-PBE), employing a 20 × 20 × 20 Monkhorst-Pack k-point mesh. Spin polarization was included, and both the lattice parame- ters and magnetic moments were allowed to relax. The bcc- FeCo alloy was modeled using the virtual crystal approxi- mation (VCA) \[16\], where Fe sites were replaced by an effective potential corresponding to 50% Co substitution, yielding an Fe0.5Co0.5 potential. Such interpolation ap- proaches have been widely adopted and shown to provide reliable descriptions of electronic structures in substitutional alloys \[17–21\]. Additional slab calculations were also carried out to estimate the work functions of bcc-Cu(100) and Fe- Co(100), and to visualize the charge-density redistribution at the bcc-Cu(100)/FeCo(100) interface. Details of the slab models, structural relaxation, vacuum thickness, dipole cor- rection, and charge-density-difference analysis are described in Supplementary Material. III. RESULTS AND DISCUSSIONS Figure 1(a, b) shows the LEED patterns of clean Fe- Co(100) and Cu/FeCo(100) recorded at the same sample– screen distance. Under this identical geometry, the spacing between diffraction spots directly reflects the in-plane atomic periodicity of the surface. The in-plane atomic spacing of fcc Cu (2.5Å) is smaller than that of bcc FeCo and metastable bcc Cu (2.9Å), which is confirmed in Figure S1 in Supple- mental Material. No additional spots corresponding to an fcc lattice are observed in Figure 1(b), indicating the absence of the fcc-Cu phase under the present growth conditions. The bcc character of the LEED patterns is further supported by the incident-electron-energy dependence of the 01 diffraction spot in the LEED I-V measurements \[11\]. Figure 1(c, d) shows XPS spectra acquired from a sample prepared in a laboratory-based system following essentially the same deposition process as that for the ARPES measure- ments, confirming the presence of Cu at the sample surface. As the surface energy of Cu is lower than that of both Fe(100) and Co(100) \[22–24\], Cu is expected to wet the FeCo(100) surface, suggesting that the substrate is uniformly covered by the Cu overlayer. Moreover, previous transmis- sion electron microscopy studies have reported that no sig- nificant atomic interdiffusion occurs at the Cu/FeCo interface \[4, 11\]. These observations indicate that a bcc-Cu(100) thin film is formed on a sufficiently relaxed FeCo(100) surface under the present growth conditions. Figure 2 presents ARPES intensity maps measured along the Γ–N direction at SAGA-LS, together with the corre- sponding second-derivative plots and the DFT-calculated band structures. The raw ARPES intensity maps shown in Figure 2(a, b) exhibit intensity variations depending on emis- sion angle, mainly due to instrumental response and matrix- element effects. Since such intensity modulations make it difficult to directly evaluate weak dispersive features from the raw maps alone, second-derivative maps are also shown in Figure 2(c, d) to enhance the visibility of the band dis- persions. For the clean FeCo(100) substrate \[Figure 2(a)\], a parabolic band (dashed line) is observed at binding energies EB = 2.0–3.0 eV. This feature is also clearly identified in the corresponding second-derivative map \[Figure 2(c)\]. After Cu deposition, an additional parabolic band appears at EB = 4.0– 5.0 eV, as shown in Figure 2(b) (solid line), which is absent in the spectrum of the clean FeCo(100) surface. This addi- tional band is more clearly resolved in the second-derivative Figure 1: Representative LEED patterns of (a) a clean FeCo(100) surface and (b) a Cu/FeCo(100) surface recorded under the same sample–screen geometry at UHV preparation chambers in synchro- tron radiation facilities. (c, d) XPS spectrum of Cu/FeCo(100): a wide-scan spectrum (c) and an enlarged view of the Cu core-level region (d) on reference samples taken at a UHV preparation cham- ber in laboratory-based system, confirming the presence of Cu at the sample surface. Regular Paper e-J. Surf. Sci. Nanotechnol. 24, 90–97 (2026) \| DOI: 10.1380/ejssnt.2026-014 92[https://doi.org/10.1380/ejssnt.2026-014](https://doi.org/10.1380/ejssnt.2026-014)map \[Figure 2(d)\]. As this band exhibits clear dispersive behavior and appears only after Cu deposition, it is reason- able to assign it to bcc-Cu, although the presence of a minor amorphous Cu component cannot be completely excluded. The parabolic band originating from the FeCo substrate (dashed line), which is also visible in Figure 2(a), remains observable even after Cu deposition because the thickness of the Cu overlayer is sufficiently small compared with the probing depth of the present ARPES measurement (hν = 140 eV) \[25\]. In addition, the photoemission intensity in the binding-energy range of EB = 0.0–2.0 eV is reduced after Cu deposition. This behavior can be attributed to the differ- ence in the electronic configurations of FeCo and Cu. In FeCo, partially occupied 3d states extend up to the vicinity of the Fermi level, resulting in high-photoemission intensity at EB \< 2.0 eV. In contrast, for Cu, the 3d states are fully occupied in both the bcc and fcc phases located at higher binding energies by stabilization of the 3d electrons. As a result, the photoemission intensity near EF is significantly reduced after Cu deposition. A similar reduction of near EF intensity is also observed for thicker fcc-Cu films (see Fig- ure S1 in Supplementary Material). Therefore, the dispersive band observed at higher binding energies than the bcc-FeCo- derived band is attributed to bcc-Cu. Figure 3 shows Fermi surface mappings of Cu/FeCo(100) obtained by ARPES experiments at HiSOR BL-7, together with the corresponding DFT-calculated Fermi surfaces. From the periodicity of the k⊥–k// mapping in Figure 3(c), the inner potential V0 was experimentally estimated to be V0 = 31 eV, and a photon energy of hν = 140 eV corresponds to the Γ point. The Fermi surface mapping in Figure 3(c) was meas- ured in an experimental geometry where the in-plane mo- mentum axis was aligned parallel to the bulk MgO\[110\] direction. Previous works \[2, 11\] established the epitaxial relationship MgO\[110\] \|\| bcc\[100\] \|\| fcc-Cu\[110\] in Cu/FeCo/ MgO(100) heterostructures. Accordingly, under the assump- tion of a bcc-Cu lattice, the present geometry probes the Γ– H–N plane of the bcc Brillouin zone \[Figure 3(a)\]. In con- trast, if fcc-Cu is assumed, the measurement corresponds to the Γ–K–X plane of the fcc Brillouin zone \[Figure 3(b)\]. Figure 2: (a, b) ARPES intensity maps of (a) FeCo(100) and (b) bcc-Cu/FeCo(100) as a function of k// and EB measured along the Γ–N line with hν = 140 eV, respectively. (c, d) The corresponding second derivative plots of (c) FeCo(100) and (d) bcc-Cu/FeCo(100). (e, f ) The DFT- calculated band structure of (e) FeCo and (f ) bcc-Cu. Regular Paper e-J. Surf. Sci. Nanotechnol. 24, 90–97 (2026) \| DOI: 10.1380/ejssnt.2026-014 93[https://doi.org/10.1380/ejssnt.2026-014](https://doi.org/10.1380/ejssnt.2026-014)Figure 3(d–g) present bulk-band DFT calculations corre- sponding to these Brillouin-zone cuts for bcc-FeCo \[Fig- ure 3(d, e)\], bcc-Cu \[Figure 3(f )\], and fcc-Cu \[Figure 3(g)\] at Fermi level. Because the photon-energy range used here (30–400 eV) yield an inelastic mean free path of approxi- mately 1 nm \[25\], comparable to the Cu film thickness, the detected photoelectrons originate not only from the Cu over- layer but also from the underlying FeCo layer. Thus, the measured ARPES intensity in Figure 3(c) should not be interpreted as a fully Cu-derived Fermi surface, but rather as a superposition of spectral weight from the Cu/FeCo heterostructure. Despite this overlap, the experimentally observed Fermi- surface morphology is clearly inconsistent with the neckless Fermi surface expected for conventional fcc-Cu \[Fig- ure 3(g)\]. In particular, no significant spectral intensity is observed between the Γ and N points in the central Brillouin zone, whereas such intensity would be expected for the fcc- Cu Fermi surface in the corresponding Γ–K–X plane. The presence of a neck along the \[100\] direction in bcc-Cu is also consistent with earlier positron-annihilation measurements \[26\]. In addition, weak spectral weight observed near the Γ point of the first Brillouin zone can be attributed to the minority-spin component of the FeCo substrate. The Fermi surface observed in Figure 3(c) appears broadened. The ori- gin of this broadening will be discussed in a later section. Thus, although the finite momentum resolution and the coexistence of Cu- and FeCo-derived photoemission signals prevent a fully quantitative determination of the detailed Fermi-surface morphology, the data in Figure 3 support the formation of a bcc-type Cu electronic structure and show no indication of the neckless Fermi-surface morphology charac- teristic of fcc-Cu. Figure 4(a) shows the ARPES intensity map of the Cu/ FeCo along the Γ-H direction at a photon energy of hν = 140 eV, which was measured at HiSOR BL-7, together with the corresponding energy distribution curves (EDCs) shown in Figure 4(b). The ARPES intensity map of Fig- ure 4(a) was normalized so that the integrated intensity of each spectrum is identical. In the binding-energy range from 1.0 to 4.0 eV, two dispersive bands \[B1 and B2 in Fig- ure 4(a)\] are clearly observed, both exhibiting a monotonic upward dispersion from Γ toward H. Based on the compar- ison of the Γ–N band dispersions in Figure 2, the deeper band (B1) is attributed to Cu-derived states, while the shal- lower one (B2) is assigned to FeCo-derived states. The corresponding DFT band structures for FeCo and bcc- Cu are shown in Figure 4(c, d). Along the Γ–H direction, both systems exhibit Δ5 bands with predominantly d-orbital character: a Δ5 band for Cu and a majority-spin Δ5 band for FeCo. The experimentally observed bands marked as B1 and B2 in Figure 4 are therefore attributed to these Δ5 states. In addition, the bands observed near the Γ point at EB \< 1 eVare assigned to the minority-spin Δ1 and Δ5 band of FeCo, in Figure 3: The three-dimensional first Brillouin zones and the two-dimensional projection of (a) bcc lattice, and (b) fcc lattice, respectively. (c) Photon-energy-dependent ARPES constant-energy surface in the k⊥–k// plane of Cu/FeCo(100) at the Fermi level (hν = 30–400 eV), overlaid by (d–f ), showing the bcc-like neck structure. (d–g) DFT-calculated Fermi surfaces of (d) majority-spin of bcc-FeCo, (e) minority-spin of bcc- FeCo, (f ) bcc-Cu, at the cross section indicated in (a), and (g) fcc-Cu, at the cross section indicated in (b). Regular Paper e-J. Surf. Sci. Nanotechnol. 24, 90–97 (2026) \| DOI: 10.1380/ejssnt.2026-014 94[https://doi.org/10.1380/ejssnt.2026-014](https://doi.org/10.1380/ejssnt.2026-014)good agreement with the DFT calculations. Near the H point, two bands crossing the Fermi level (B3, S1) are observed. The band showing an upward dispersion from Γ to H (B3) is attributed to the Cu-derived Δ1 band with predominantly s–p character. In contrast, the other band (S1) cannot be uniquely assigned to any bulk band predicted by the DFT calculations. Moreover, this feature shows no no- ticeable photon-energy dependence, suggesting that it is not a simple bulk-derived state. Possible origins include surface- related states, as reported for fcc-Cu surfaces, or quantum- well states (QWS) appearing away from the surface normal due to umklapp scattering \[27–29\]. The S1 feature is located close to the Fermi level and exhibits an apparently linear dispersion. Therefore, although its microscopic origin re- mains unclear, it may represent an electronic state with a relatively small effective mass or high group velocity, which could influence carrier transport in bcc-Cu/FeCo heterostruc- tures. A detailed identification of this band, however, re- quires further experimental investigation. One possible approach is systematic thickness-dependent ARPES measurements. Previous studies on QWS in fcc-Cu/ Co(100) demonstrated that Cu thicknesses in the range of approximately 1–10ML are sufficient to clarify the QWS character \[30\]. Considering the lattice constant of bcc-Cu, 10ML is equivalent to a thickness of less than 2 nm. In addition, metastable bcc-Cu layers with thickness up to 2 nm have been realized in FeCo/bcc-Cu/FeCo multilayer structures \[2\]. Therefore, although experimentally demand- ing, monolayer-scale thickness-dependent ARPES measure- ments should be feasible in principle within the thickness range where the metastable bcc-Cu structure is preserved. Such measurements will be important for clarifying whether S1 originates from a QWS. Although the overall qualitative behavior of the experi- mentally observed band dispersions is well reproduced by the DFT calculations, a pronounced quantitative discrepancy is found in the energy positions of the Cu-derived bands. In particular, the Cu-derived Δ5 band is observed at a binding energy deeper by approximately 1.0 eV than predicted by bulk-based DFT calculations. A similar tendency is also found for the Δ1 band, which crosses the Fermi level closer to the H point than expected from the calculations. Overall, the Cu-derived bands are systematically shifted toward high- er binding energies compared to the DFT results. Such band shifts directly modify the Fermi surface mor- phology, indicating that interfacial transport properties can- not be fully understood solely on the basis of Fermi surface matching arguments derived from ideal bulk DFT calcula- tions. Several factors may contribute to the observed discrep- ancy between experiment and theory. First, it may partially originate from the intrinsic limitations of the DFT approx- imation employed. In particular, the GGA-PBE functional is known to suffer from self-interaction errors, which can lead to systematic shifts in band energies \[31, 32\]. Second, inter- face-induced charge redistribution may also play an impor- tant role. When two materials with different work functions form a heterointerface, charge redistribution occurs near the interface to align their Fermi levels. To examine this effect, additional slab calculations were performed for bcc-Cu(100), FeCo(100), and the bcc-Cu(100)/FeCo(100) interface. The calculated work functions were 4.395 eV for bcc-Cu(100) and 3.869 eV for FeCo(100), indicating a clear work-function difference between the two materials. Figure 5 shows the calculated charge-density-difference distribution at the bcc- Cu(100)/FeCo(100) interface. Enhanced electron density is observed on the bcc-Cu side near the interface, indicating electron transfer from FeCo to bcc-Cu and the occurrence of interfacial electron reconstruction. These results support the presence of interface-induced charge redistribution and local electrostatic potential modification in the bcc-Cu/FeCo het- erostructure. This charge redistribution is consistent with the direction of the experimentally observed shift of the bcc-Cu- derived bands toward higher binding energy in the ARPES measurements. However, the magnitude of the observed shift Figure 4: (a) ARPES intensity map of Cu/FeCo(100) as a function of k// and EB along Γ–H line with hν = 140 eV. (b) Energy-distribution curves from (a). (c) DFT-calculated band structure of FeCo (majority/minority). (d) DFT-calculated band structure of bcc-Cu. Regular Paper e-J. Surf. Sci. Nanotechnol. 24, 90–97 (2026) \| DOI: 10.1380/ejssnt.2026-014 95[https://doi.org/10.1380/ejssnt.2026-014](https://doi.org/10.1380/ejssnt.2026-014)is likely influenced by multiple factors, including reduced screening in the ultrathin Cu layer, epitaxial strain, quantum- size effects, and limitations of the bulk-based DFT calcula- tions. Recent studies about the MR ratio of Cu/FeβCo1−β with varying compositions \[5\] revealed a small but systematic deviation between theoretical predictions and experimental results, suggesting that idealized bulk electronic structures do not fully capture the electronic states relevant to transport in realistic heterostructures. Our observation of interface-in- duced band shifts in bcc-Cu/FeCo suggests that such devia- tions can arise from modifications of the electronic structure at heterointerfaces. These results indicate that, in realistic heterostructures, the electronic bands can be significantly modified from their ideal bulk counterparts, and that such modifications can be directly probed by ARPES. The present observations thus highlight the importance of considering interface-induced band shifts when discussing the electronic structure and transport properties of Cu/FeCo heterojunctions. In the present study, the ability to grow high-quality bcc- Cu ultra-thin films by MBE on FeCo(100) substrates enabled direct ARPES observation of the bcc-Cu band dispersions. This experimental access reveals not only the characteristic Fermi surface morphology of bcc-Cu but also significant deviations from bulk-based DFT predictions, highlighting the importance of experimentally determining the electronic structure of metastable phases realized under epitaxial con- straints. IV. CONCLUSION In summary, the electronic band structure of metastable bcc-Cu has been experimentally determined by combining our MBE growth technique with in-situ ARPES measure- ments. The Fermi surface morphology of bcc-Cu was clearly distinguished from that of conventional fcc-Cu, demonstrat- ing the unique electronic structure of this metastable phase. This achievement provides a long-sought experimental benchmark for bcc-Cu, which has previously been discussed primarily on the basis of theoretical calculations. Overall, the experimentally observed band dispersions show good qualitative agreement with DFT calculations. However, for bcc-Cu grown on FeCo, the Cu-derived elec- tronic bands are systematically shifted toward higher binding energies compared to predictions based on ideal bulk DFT calculations. This discrepancy is likely caused not only by intrinsic limitations of the DFT approximation but also by interface-induced effects such as charge transfer at the Cu/ FeCo interface. These findings indicate that, in ultra-thin heterostructures, environmental effects such as interface formation play a crucial role in determining the electronic structure. Conse- quently, device performance predictions based solely on idealized bulk electronic structures, such as Fermi surface matching arguments, may be insufficient. Direct experimen- tal determination of electronic states is therefore essential. In this context, future studies employing high-throughput AR- PES measurement techniques, are expected to enable more reliable discussions of interface transport properties based on experimentally determined Fermi surfaces. Acknowledgments The authors thank Dr. Shunsuke Tsuda and Dr. Fumio Komori for fruitful discussions. We also thank Ms. Sanae Kuramochi, Ms. Natsuko Kojima and Mr. Fumiya Arai for technical supports. The synchrotron radiation experiments were performed with the approval of UVSOR (Proposal Number: 25IMS6667), SL-Center, Saga Uni- versity (Proposal Numbers: R6-102V, R6-103V) and HiSOR (Pro- posal Number: 24BU002). This work is particularly supported by JST CREST (Grant Numbers: JPMJCR21O1, JPMJCR21O4), JST PRESTO, Japan (Grant Numbers: JPMJPR17NB, JPMJPR20T7), JSPS/MEXT KAKENHI (Grant Numbers: 25K01628, 24K01243, 24K01352, 21H01638, 21H04696, 25H00743), the Innovative Sci- ence and Technology Initiative for Security, ATLA (Grant Number: JPJ004596), the Cooperative Research Project Program of the Re- search Institute of Electrical Communication, Tohoku University, and JST SPRING (Grant Number: JPMJSP2151). The computation was performed using supercomputing facilities of Research Center for Computational Science, Okazaki, Japan (Project: 26-IMS-C337). Appendix The additional LEED patterns, corresponding ARPES intensity maps, and DFT calculation conditions are available in Supplemen- tary Material at https://doi.org/10.1380/ejssnt.2026-014. Figure 5: Charge-density-difference distribution at the bcc- Cu(100)/FeCo(100) interface calculated by DFT. (a) Top-view two-dimensional slice of the charge-density-difference distribution on a plane parallel to the interface. (b) Side-view three-dimensional charge-density-difference iso-surfaces. The charge-density differ- ence was evaluated as Δρ = ρ(interface) − ρ(bcc-Cu slab) − ρ(FeCo slab), where the isolated slab charge densities were calculated using the same atomic positions and lattice parameters as those in the interface model. In panel (a), the color map represents the charge- density difference on the selected plane. In panel (b), red and blue iso-surfaces indicate charge depletion and charge accumulation, respectively. The iso-surface value in panel (b) is ±0.003 eÅ−3. Regular Paper e-J. Surf. Sci. Nanotechnol. 24, 90–97 (2026) \| DOI: 10.1380/ejssnt.2026-014 96[https://doi.org/10.1380/ejssnt.2026-014](https://doi.org/10.1380/ejssnt.2026-014)[https://doi.org/10.1380/ejssnt.2026-014](https://doi.org/10.1380/ejssnt.2026-014)References \[1\] N. Kaiser, Appl. Opt. 41, 3053 (2002). \[2\] K. B. Fathoni, Y. Sakuraba, T. Sasaki, Y. Miura, J. W. Jung, T. Nakatani, and K. Hono, APL Mater. 7, 111106 (2019). \[3\] D. Taparia, K. B. Fathoni, P. D. Kulkarni, A. Srinvasan, P. Alagarsamy, Y. Sakuraba, T. Nakatani, and K. Hono, J. Magn. Magn. Mater. 538, 168321 (2021). \[4\] K. B. Fathoni, Z. Li, T. Nakatani, Y. Sakuraba, T. Sasaki, and K. Hono, J. Magn. Magn. Mater. 551, 169154 (2022). \[5\] V. K. Kushwaha, R. Toyama, Y. Miura, Y. Iwasaki, and Y. Sakuraba, AIP Adv. 14, 125329 (2024). \[6\] H. Nakayama, B. Xu, S. Iwamoto, K. Yamamoto, R. Iguchi, A. Miura, T. Hirai, Y. Miura, Y. Sakuraba, J. Shiomi, and K. Uchida, Appl. Phys. Lett. 118, 042409 (2021). \[7\] F. Makino, T. Hirai, T. Shiga, H. Suto, H. Fujihisa, K. Oyanagi, S. Kobayashi, T. Sasaki, T. Yagi, K. Uchida, and Y. Sakuraba, Phys. Rev. B 112, 054407 (2025). \[8\] Z. Tang, M. Hasegawa, Y. Nagai, and M. Saito, Phys. Rev. B 65, 195108 (2002). \[9\] H. Li, D. Tian, J. Quinn, Y. S. Li, F. Jona, and P. M. Marcus, Phys. Rev. B 43, 6342 (1991). \[10\] Z. Q. Wang, S. H. Lu, Y. S. Li, F. Jona, and P. M. Marcus, Phys. Rev. B 35, 9322(R) (1987). \[11\] S. Takezawa, M. Riva, F. Dörr, M. Schmid, T. Sasaki, T. Hiroto, M. Kotsugi, K. Yaji, Y. Sakuraba, and N. Nagamura, e-J. Surf. Sci. Nanotechnol. 24, 21 (2026). \[12\] K. Takahashi, M. Imamura, I. Yamamoto, J. Azuma, K. Ogawa, M. Kamada, H. Ohkuma, and S. Yamamoto, J. Phys.: Conf. Ser. 425, 072007 (2013). \[13\] S. Kimura, T. Ito, M. Sakai, E. Nakamura, N. Kondo, T. Horigome, K. Hayashi, M. Hosaka, M. Katoh, T. Goto, T. Ejima, and K. Soda, Rev. Sci. Instrum. 81, 053104 (2010). \[14\] H. Ota, E. Salehi, M. Fujimoto, K. Hayashi, T. Horigome, H. Iwayama, M. Katoh, N. Kondo, S. Makita, F. Matsui, H. Matsuda, T. Mizukawa, A. Minakuchi, E. Nakamura, M. Nagasaka, Y. Okano, T. Ohigashi, M. Sakai, K. Sugita, K. Tanaka, Y. Taira, F. Teshima, J. Yamazaki, T. Yano, H. Yuzawa, and S. Kera, J. Phys.: Conf. Ser. 2380, 012003 (2022). \[15\] G. Kresse and J. Furthmüller, Phys. Rev. B 54, 11169 (1996). \[16\] C. Eckhardt, K. Hummer, and G. Kresse, Phys. Rev. B 89, 165201 (2014). \[17\] W. E. Pickett and D. J. Singh, Phys. Rev. B 53, 1146 (1996). \[18\] D. A. Papaconstantopoulos and W. E. Pickett, Phys. Rev. B 57, 12751 (1998). \[19\] A. M. Saitta, S. de Gironcoli, and S. Baroni, Phys. Rev. Lett. 80, 4939 (1998). \[20\] N. Marzari, S. de Gironcoli, and S. Baroni, Phys. Rev. Lett. 72, 4001 (1994). \[21\] S. Steiner, S. Khmelevskyi, M. Marsmann, and G. Kresse, Phys. Rev. B 93, 224425 (2016). \[22\] B. Heinrich, Z. Celinski, J. F. Cochran, W. B. Muir, J. Rudd, Q. M. Zhong, A. S. Arrott, K. Myrtle, and J. Kirschner, Phys. Rev. Lett. 64, 673 (1990). \[23\] P. LeClair, H. J. M. Swagten, J. T. Kohlhepp, R. J. M. van de Veerdonk, and W. J. M. de Jounge, Phys. Rev. Lett. 84, 2933 (2000). \[24\] D. Kerkmann, D. Pescia, J. W. Krewer, and E. Vescovo, Z. Phys. B 85, 311 (1991). \[25\] M. P. Seah and W. A. Dench, Surf. Interface Anal. 1, 2 (1979). \[26\] Y. Nagai, T. Chiba, Z. Tang, T. Akahane, T. Kanai, M. Hasegawa, M. Takenaka, and E. Kuramoto, Phys. Rev. Lett. 87, 176402 (2001). \[27\] T.-C. Chiang, Surf. Sci. Rep. 39, 181 (2000). \[28\] J. M. An, D. Raczkowski, Y. Z. Wu, C. Y. Won, L. W. Wang, A. Canning, M. A. Van Hove, E. Rotenberg, and Z. Q. Qiu, Phys. Rev. B 68, 045419 (2003). \[29\] P. Heimann, J. Hermanson, H. Miosga, and H. Neddermeyer, Phys. Rev. B 20, 3059 (1979). \[30\] C. Carbone, E. Vescovo, O. Rader, W. Gudat, and W. Eberhardt, Phys. Rev. Lett. 71, 2805 (1993). \[31\] J. L. Bao, L. Galiardi, and D. G. Truhlar, J. Phys. Chem. Lett. 9, 2353 (2018). \[32\] W. J. Kim, S. Xing, G. Kremer, M. Sicot, B. Kierren, D. Malterre, G. Contini, J. Rault, P. Le Fèvre, F. Bertran, D. Rocca, Y. Fagot-Revurat, and S. Lebègue, J. Phys. Chem. C 123, 26309 (2019). All articles published on e-J. Surf. Sci. Nanotechnol. are licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0). You are free to copy and redistribute articles in any medium or format and also free to remix, transform, and build upon articles for any purpose (including a commercial use) as long as you give appropriate credit to the original source and provide a link to the Creative Commons (CC) license. If you modify the material, you must indicate changes in a proper way. Copyright: ©2026 The author(s) Published by The Japan Society of Vacuum and Surface Science Regular Paper e-J. Surf. Sci. Nanotechnol. 24, 90–97 (2026) \| DOI: 10.1380/ejssnt.2026-014 97[https://doi.org/10.1364/AO.41.003053](https://doi.org/10.1364/AO.41.003053)[https://doi.org/10.1063/1.5119370](https://doi.org/10.1063/1.5119370)[https://doi.org/10.1016/j.jmmm.2021.168321](https://doi.org/10.1016/j.jmmm.2021.168321)[https://doi.org/10.1016/j.jmmm.2021.168321](https://doi.org/10.1016/j.jmmm.2021.168321)[https://doi.org/10.1016/j.jmmm.2022.169154](https://doi.org/10.1016/j.jmmm.2022.169154)[https://doi.org/10.1063/5.0216909](https://doi.org/10.1063/5.0216909)[https://doi.org/10.1063/5.0032531](https://doi.org/10.1063/5.0032531)[https://doi.org/10.1103/pd57-gcd5](https://doi.org/10.1103/pd57-gcd5)[https://doi.org/10.1103/pd57-gcd5](https://doi.org/10.1103/pd57-gcd5)[https://doi.org/10.1103/PhysRevB.65.195108](https://doi.org/10.1103/PhysRevB.65.195108)[https://doi.org/10.1103/PhysRevB.65.195108](https://doi.org/10.1103/PhysRevB.65.195108)[https://doi.org/10.1103/PhysRevB.43.6342](https://doi.org/10.1103/PhysRevB.43.6342)[https://doi.org/10.1103/PhysRevB.35.9322](https://doi.org/10.1103/PhysRevB.35.9322)[https://doi.org/10.1380/ejssnt.2026-004](https://doi.org/10.1380/ejssnt.2026-004)[https://doi.org/10.1380/ejssnt.2026-004](https://doi.org/10.1380/ejssnt.2026-004)[https://doi.org/10.1088/1742-6596/425/7/072007](https://doi.org/10.1088/1742-6596/425/7/072007)[https://doi.org/10.1088/1742-6596/425/7/072007](https://doi.org/10.1088/1742-6596/425/7/072007)[https://doi.org/10.1063/1.3425778](https://doi.org/10.1063/1.3425778)[https://doi.org/10.1088/1742-6596/2380/1/012003](https://doi.org/10.1088/1742-6596/2380/1/012003)[https://doi.org/10.1088/1742-6596/2380/1/012003](https://doi.org/10.1088/1742-6596/2380/1/012003)[https://doi.org/10.1103/PhysRevB.54.11169](https://doi.org/10.1103/PhysRevB.54.11169)[https://doi.org/10.1103/PhysRevB.89.165201](https://doi.org/10.1103/PhysRevB.89.165201)[https://doi.org/10.1103/PhysRevB.89.165201](https://doi.org/10.1103/PhysRevB.89.165201)[https://doi.org/10.1103/PhysRevB.53.1146](https://doi.org/10.1103/PhysRevB.53.1146)[https://doi.org/10.1103/PhysRevB.57.12751](https://doi.org/10.1103/PhysRevB.57.12751)[https://doi.org/10.1103/PhysRevB.57.12751](https://doi.org/10.1103/PhysRevB.57.12751)[https://doi.org/10.1103/PhysRevLett.80.4939](https://doi.org/10.1103/PhysRevLett.80.4939)[https://doi.org/10.1103/PhysRevLett.80.4939](https://doi.org/10.1103/PhysRevLett.80.4939)[https://doi.org/10.1103/PhysRevLett.72.4001](https://doi.org/10.1103/PhysRevLett.72.4001)[https://doi.org/10.1103/PhysRevLett.72.4001](https://doi.org/10.1103/PhysRevLett.72.4001)[https://doi.org/10.1103/PhysRevB.93.224425](https://doi.org/10.1103/PhysRevB.93.224425)[https://doi.org/10.1103/PhysRevLett.64.673](https://doi.org/10.1103/PhysRevLett.64.673)[https://doi.org/10.1103/PhysRevLett.64.673](https://doi.org/10.1103/PhysRevLett.64.673)[https://doi.org/10.1103/PhysRevLett.84.2933](https://doi.org/10.1103/PhysRevLett.84.2933)[https://doi.org/10.1103/PhysRevLett.84.2933](https://doi.org/10.1103/PhysRevLett.84.2933)[https://doi.org/10.1007/BF01313234](https://doi.org/10.1007/BF01313234)[https://doi.org/10.1007/BF01313234](https://doi.org/10.1007/BF01313234)[https://doi.org/10.1002/sia.740010103](https://doi.org/10.1002/sia.740010103)[https://doi.org/10.1103/PhysRevLett.87.176402](https://doi.org/10.1103/PhysRevLett.87.176402)[https://doi.org/10.1103/PhysRevLett.87.176402](https://doi.org/10.1103/PhysRevLett.87.176402)[https://doi.org/10.1016/S0167-5729(00)00006-6](<https://doi.org/10.1016/S0167-5729(00)00006-6>)[https://doi.org/10.1103/PhysRevB.68.045419](https://doi.org/10.1103/PhysRevB.68.045419)[https://doi.org/10.1103/PhysRevB.68.045419](https://doi.org/10.1103/PhysRevB.68.045419)[https://doi.org/10.1103/PhysRevB.20.3059](https://doi.org/10.1103/PhysRevB.20.3059)[https://doi.org/10.1103/PhysRevLett.71.2805](https://doi.org/10.1103/PhysRevLett.71.2805)[https://doi.org/10.1021/acs.jpclett.8b00242](https://doi.org/10.1021/acs.jpclett.8b00242)[https://doi.org/10.1021/acs.jpclett.8b00242](https://doi.org/10.1021/acs.jpclett.8b00242)[https://doi.org/10.1021/acs.jpcc.9b07057](https://doi.org/10.1021/acs.jpcc.9b07057)[https://doi.org/10.1021/acs.jpcc.9b07057](https://doi.org/10.1021/acs.jpcc.9b07057)[https://doi.org/10.1380/ejssnt.2026-014](https://doi.org/10.1380/ejssnt.2026-014)