# Fileset

[Supplementary.pdf](https://mdr.nims.go.jp/filesets/c2928e80-ab8f-42ee-bac8-6ae03a582bcb/download)

## Creator

[Tomoya Nakatani](https://orcid.org/0000-0001-9590-216X), [Rohit Dahule](https://orcid.org/0000-0002-2768-8808), [Nattamon Suwannaharn](https://orcid.org/0000-0003-1285-599X), [Mevhibe Dilara Pütkül](https://orcid.org/0009-0007-0268-1478), [Yusuke Ikeda](https://orcid.org/0009-0001-4394-711X), [Taisuke T. Sasaki](https://orcid.org/0000-0002-5952-7638), [Ryoji Sahara](https://orcid.org/0000-0003-0788-2985)

## Rights

This article may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing. This article appeared in Tomoya Nakatani, Rohit Dahule, Nattamon Suwannaharn, Mevhibe Dilara Pütkül, Yusuke Ikeda, Taisuke T. Sasaki, Ryoji Sahara; Orientation-controlled epitaxy of CoSn anisotropic conductive thin films on sapphire substrates toward interconnect applications. J. Appl. Phys. 21 October 2026; 140 (15): 155101 and may be found at https://doi.org/10.1063/5.0352862.[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[Orientation-controlled epitaxy of CoSn anisotropic conductive thin films on sapphire substrates toward interconnect applications](https://mdr.nims.go.jp/datasets/ef978e03-03e9-47bb-8e99-eb91db04fdd0)

## Fulltext

Supplementary MaterialOrientation-controlled epitaxy of CoSn anisotropic conductive thin films on  sapphire substrates toward interconnect applicationsTomoya Nakatani,1\* Rohit Dahule,2 Nattamon Suwannaharn,2 Mevhibe Dilara Pütkül,1  Yusuke Ikeda,1 Taisuke T. Sasaki,2 and Ryoji Sahara,21 Research Center for Magnetic and Spintronic Materials, National Institute for  Materials Science, 1-2-1, Sengen, Tsukuba, Ibaraki 305-0047, Japan.2 Research Center for Structural Materials, National Institute for Materials Science, 1-2- 1, Sengen, Tsukuba, Ibaraki 305-0047, Japan.\*Corresponding author’s email address: [nakatani.tomoya@nims.go.jp](mailto:nakatani.tomoya@nims.go.jp)1. 1\. Sheet resistance evaluation of buffer layerThe CoSn epitaxial films used in this study were grown on sapphire substrates via  metallic buffer layers: (from the substrate side) CoFe (2 nm)/Co (2 nm)/Ru (2 nm) for  CoSn and Ti (2 nm)/Ru (4 nm) for CoSn. Therefore, the contribution of  these buffer layers to the total sheet resistance (Rs) must be considered when evaluating  the resistivity of CoSn films. Following our previous study, the buffer layer’s Rs was  estimated by measuring the Rs of samples consisting of buffer layer/AgSn (tAgSn) capped  with SiN (3 nm). AgSn is an fcc Ag90Sn10 (at. %) alloy with a resistivity of \~30 μΩ cm.  The details of this method are described in the Supplementary Material of our previous  publication.1左小かっこ 101 バー 0 右小かっこ (101̅0)左小かっこ 112 バー 0 右小かっこ (112̅0)Figures S1(a) and S1(b) show 1/Rs vs. tAgSn plots for CoFe (2 nm)/Co (2 nm)/Ru (2  nm)/AgSn (tAgSn) on MgO(110) and sapphire substrates buffered with MgO (5  nm), respectively. In these cases, the results were almost identical for the current  directions parallel to the c axis and a axis of Ru. The inverse of the extrapolated intercept  at tAgSn = 0 corresponds to the Rs of the buffer layers, yielding  Ω/sq.左小かっこ 101 バー 0 右小かっこ (101̅0)大文字 R 下付き s じょう buffer イコール 217 下付き 左小かっこ マイナス 36 右小かっこ じょう 左小かっこ プラス 52 右小かっこ 𝑅s buffer = 217−36 +521[mailto:nakatani.tomoya@nims.go.jp](mailto:nakatani.tomoya@nims.go.jp)on the MgO(110) substrate and   Ω/sq. on the sapphire   substrate. The slope of the linear fit corresponds to the resistivity of the AgSn film, as  shown in Figs. S1(a) and S1(b).大文字 R 下付き s じょう buffer イコール 258 下付き 左小かっこ マイナス 48 右小かっこ じょう 左小かっこ プラス 76 右小かっこ 𝑅s buffer = 258−48 +76左小かっこ 101 バー 0 右小かっこ (101̅0)Figures S1(c) and S1(d) show the average resistivity of the whole film including the  buffer layers,   by open symbols and the resistivity of  the CoSn film,   by closed symbols.   is  the sheet resistance of the entire film, and  is the total thickness of the buffer layers  (6 nm). For large tCoSn, the difference between  and  becomes negligible,  particularly along the low-resistivity c axis. In contrast, a difference of approximately 10  μΩ cm was observed for tCoSn = 10 nm.ロー じょう ave イコール 大文字 R 下付き s じょう whole 内積 左小かっこ t 下付き CoSn プラス t 下付き buffer 右小かっこ 𝜌ave = 𝑅s whole ∙ (𝑡CoSn + 𝑡buffer)ロー じょう CoSn イコール 下付き文字 てい , t 下付き CoSn 割る 左小かっこ 1 割る 大文字 R 、 下付き s じょう whole マイナス 下付き文字 てい , 1 割る 大文字 R 、 下付き s じょう buffer 右小かっこ 𝜌CoSn = 𝑡CoSn/(1/𝑅s whole − 1/𝑅s buffer)大文字 R 下付き s じょう total 𝑅s totalt 下付き buffer 𝑡bufferロー じょう ave 𝜌aveロー じょう CoSn 𝜌CoSnNext, the Rs of Ti (2 nm)/Ru (4 nm) buffer layers on a sapphire substrate was  evaluated by the same method, as shown in Fig. S2. In these samples, Rs depended on the  current direction. The buffer layers’ Rs was   Ω/sq. and   Ω/sq. along the  c axis and \[ of Ru. Along the direction 45° from the c axis, Rs was intermediate  between these values:   Ω/sq, as shown in Fig. S2(c). The observed orientation  dependence of Rs is consistent with the resistivity anisotropy of Ru predicted by first- principles calculations.2 Figure S2(d) shows the average resistivity of the whole films  (open symbols) and the resistivity of the CoSn films (closed symbols).左小かっこ 11 バー 02 右小かっこ (11̅02)135 下付き 左小かっこ マイナス 15 右小かっこ じょう 左小かっこ プラス 20 右小かっこ 135−15 +20169 下付き 左小かっこ マイナス 45 右小かっこ じょう 左小かっこ プラス 98 右小かっこ 169−45 +98左大かっこ 11 バー 00 右大かっこ 11̅00\]149 下付き 左小かっこ マイナス 7 右小かっこ じょう 左小かっこ プラス 8 右小かっこ 149−7 +820 10 20 30 40 50 0.000\.050\.100\.150\.201 /Rs  (1 /W)tAgSn (nm)Buffer Rs = 217+52 -36  W/sq.rAgSn = 24.2±0.2 mW cm0 10 20 30 40 0.000\.050\.100\.150\.201 /Rs  (1 /W)tAgSn (nm)Buffer Rs = 258+76 -48  W/sq.rAgSn = 23.3±0.2 mW cm(a) on MgO(110) substrate (b) on sapphire substrate(c) on MgO(110) substrate (d) on sapphire substrate10 20 30 40 50 020406080100120140160r  ( m Wc m)tCoSn (nm)∥basal plane∥c axisOpen symbols: average resistivity including bufferClosed symbols: CoSn only resistivity10 20 30 40 50 020406080100120140160 r  ( m Wc m)tCoSn (nm)∥basal plane∥c axisOpen symbols: average resistivity including bufferClosed symbols: CoSn only resistivityFIG. S1. 1/Rs vs. tAgSn plots of CoFe (2 nm)/Co (2 nm)/Ru (2 nm)/AgSn (tAgSn) films  deposited on (a) MgO(110) and (b) sapphire  substrates. The 1/Rs values  extrapolated to tAgSn = 0 correspond to Rs of the CoFe (2 nm)/Co (2 nm)/Ru (2 nm) buffer  layers. Average resistivity of the whole CoFe/Co/Ru/CoSn films (open symbols) and the  resistivity of only the CoSn films (closed symbols) on (c) MgO(110) and (d)  sapphire substrates.左小かっこ 101 バー 0 右小かっこ (101̅0)左小かっこ 101 バー 0 右小かっこ (101̅0)Alt Text: Sheet resistance analysis of CoFe/Co/Ru buffer layers and its effect on the  extracted resistivity of CoSn films. Buffer-layer contributions are significant for thin  CoSn films and become less important with increasing CoSn thickness.30 10 20 30 40 0.000\.050\.100\.151 /Rs  (1 /W)tAgSn (nm)Buffer Rs = 169+98 -45  W/sq.AgSn resistivity = 27.2±0.6 mW cm0 10 20 30 40 0.000\.050\.100\.151 /Rs  (1 /W)tAgSn (nm)Buffer Rs = 135+20 -15  W/sq.AgSn resistivity = 26.8±0.2 mW cm0 10 20 30 40 0.000\.050\.100\.151 /Rs  (1 /W)tAgSn (nm)Buffer Rs = 149+8 -7  W/sq.AgSn resistivity = 27.2±0.1 mW cm(a) Current c axis (b) Current(c) Current at 45° from c axis (d)10 20 30 40 50 0204060140160180200220240r  (m Wc m)tCoSn (nm)∥c axis∥basal planeOpen symbols: average resistivity including bufferClosed symbols: CoSn only resistivityFIG. S2. 1/Rs vs. tAgSn plots of Ti (2 nm)/Ru (4 nm)/AgSn (tAgSn) films deposited on a  sapphire substrate for current directions parallel to (a) Ru\[0001\], (b) Ru\[,  and (c) 45° from Ru\[0001\]. (d) Average resistivity of the whole Ti/Ru/CoSn films (open  symbols) and the resistivity of only the CoSn films (closed symbols).左小かっこ 11 バー 02 右小かっこ (11̅02)左大かっこ 11 バー 00 右大かっこ 11̅00\]Alt Text: Sheet resistance analysis of Ti/Ru buffer layers for different current directions.  The buffer layers exhibit resistivity anisotropy, and correction of their contribution  enables evaluation of the intrinsic resistivity of CoSn films.41. 2\. Chemical states in CoSn filmsTo examine the oxygen incorporation in the CoSn films, X-ray photoelectron  spectroscopy (XPS) was measured for a 50-nm-thick CoSn  films using  QuanteraSXM (Ulvac Phi). Figure S3 shows the depth-dependent XPS spectra from Co  2p, Sn 3d, and O 1s. Co-O, Sn-O, and O peak were observed only in the near-surface  region of the film, indicating the formation of a surface oxide layer, as confirmed by the  STEM observation (Fig. 4). After removal of the surface oxide layer by Ar sputtering  etching, no detectable oxygen signal was detected in the film interior.左小かっこ 101 バー 0 右小かっこ (101̅0)Sn 3d542 540 538 536 534 532 530 528 526 524010203040506070In ten s it yDe pt h  / nmBinding energy / eVO 1sCo 2p500 498 496 494 492 490 488 486 484 482 480010203040506070In ten s it yDe pt h  / nmBinding energy / eV810 800 790 780 770010203040506070In ten s it yDe pt h  / nmBinding energy / eVCo 2p3/2Sn 3d5/2Sn 3d3/2Sn-O 3d5/2Sn-O 3d3/2Co 2p3/2Co-O 2p3/2Co2O3 2p3/2FIG. S3. Depth-dependent XPS spectra of the CoSn 50-nm-thick film for Co 2p, Sn 3d,  and O 1s core-levels. “Co-O” can be either CoO and Co2O3, and “Sn-O” can be either  SnO and SnO2.Alt Text: Depth-dependent XPS spectra of a CoSn film. Oxygen-related signals are  observed only near the surface, indicating formation of a surface oxide layer without  detectable oxygen incorporation in the film interior.1. 3\. DFT calculations of CoSn surface energy2. a. Structural relaxationFigure S4 shows the unrelaxed and relaxed slab geometries of each plane. Upon  cleavage, structural relaxation partially compensates for the loss of coordination at the  surface. The (0001), (0002), , , and  surfaces undergo relatively  modest atomic rearrangements, with relaxation energies ranging from –0.009 to –0.001左小かっこ 101 バー 0 右小かっこ (101̅0)左小かっこ 202 バー 0 右小かっこ (202̅0)左小かっこ 112 バー 0 右小かっこ (112̅0)5eV Å–2. For the (0001) and (0002) surfaces, relaxation is dominated by vertical  displacements of the outermost atoms (0.03–0.25 Å) along the surface normal, while in- plane displacements remain negligible. The  ,  , and  surfaces  display similar behavior, characterized by vertical displacements of 0.03–0.19 Å along  with a small lateral shift of 0.03–0.09 Å. In contrast, the  surface exhibits a  stronger, predominantly vertical relaxation, in which the top-layer Co atom shifts inward  by 0.27 Å toward the subsurface Sn layer, while the corresponding surface Sn atom moves  outward toward the vacuum by 0.25 Å. This results in the largest relaxation energy  contribution (–0.017 eV Å–2) among the studied surfaces. Atomic displacements decrease  rapidly below the top layer, confirming that both structural and electronic effects are  highly localized at the surface. Nevertheless, the cleavage contribution remains the  dominant factor determining the total surface energy, and the relative surface stability  follows the sequence (0001) \<  \< (0002) \<  \<  \< .左小かっこ 101 バー 0 右小かっこ (101̅0)左小かっこ 202 バー 0 右小かっこ (202̅0)左小かっこ 112 バー 0 右小かっこ (112̅0)左小かっこ 224 バー 0 右小かっこ (224̅0)左小かっこ 202 バー 0 右小かっこ (202̅0)左小かっこ 101 バー 0 右小かっこ (101̅0)左小かっこ 224 バー 0 右小かっこ (224̅0)左小かっこ 112 バー 0 右小かっこ (112̅0)b. Bader charge analysisThe electronic response to surface relaxation was further examined using Bader charge  analysis,3 and the resulting charge variations are summarized in Table S1. Here, ΔQ is  defined as the difference between relaxed and unrelaxed slabs (ΔQ = Qrelaxed − Qunrelaxed).  ΔQlayer and ΔQsurface denote the average charge changes in the outermost layer and the  exposed surface atoms, respectively. The magnitude of charge redistribution within the  outermost layer remains small for all investigated orientations (\|ΔQ\| ≤ 0.036 e), indicating  that relaxation induces only localized electronic adjustments. The (0001), (0002),   ,  , and (11 0) surfaces show minimal changes, whereas the   surface exhibits the largest charge redistribution at the exposed surface atoms (+0.247 e),  consistent with its larger relaxation energy and more pronounced atomic displacements.  The correlation between relaxation energy and charge redistribution suggests a coupling  between structural relaxation and electronic reconstruction at the surface. The   surface shows larger relaxation and enhanced charge redistribution, indicating a less  favorable atomic coordination at the surface and a stronger tendency toward surface  roughening.4 Among the two experimentally observed orientations, the CoSn  surface is substantially more stable than the CoSn surface. This difference in左小かっこ 101 バー 0 右小かっこ (101̅0)左小かっこ 202 バー 0 右小かっこ (202̅0)2 バー 2̅左小かっこ 224 バー 0 右小かっこ (224̅0)左小かっこ 224 バー 0 右小かっこ (224̅0)左小かっこ 101 バー 0 右小かっこ (101̅0)左小かっこ 112 バー 0 右小かっこ (112̅0)6surface stability is consistent with the experimentally observed morphology evolution.FIG. S4. Unrelaxed and relaxed slab geometries for the CoSn (0001), (0002), (100),  (20 0), (11 0), and (22 0) surfaces. Co and Sn atoms are shown in blue and gray,  respectively. Arrows denote atomic displacements induced by structural relaxation.1 バー 1̅2 バー 2̅2 バー 2̅4 バー 4̅Alt Text: Unrelaxed and relaxed atomic structures of representative CoSn surfaces.  Surface relaxation is generally localized near the surface, with the largest structural  rearrangement occurring for the (220) surface.4 バー 4̅7Table S1. Bader charge redistribution in the outermost layer and exposed atoms of CoSn  surfaces.Alt Text: Bader charge redistribution associated with relaxation of CoSn surfaces. Most  surfaces exhibit only small charge redistribution, whereas the (22 0 surface shows a  substantially larger electronic reconstruction.4 バー 4̅|||---|||||||||||||SurfaceNumber of atoms  in layer (surface)ΔQlayer (e)ΔQsurface (e)(0001)6 (4)\-0.035\-0.030(0002)6 (2)0\.026\-0.011(100)太字 1 バー 𝟏̅6 (2)\-0.0360\.007(200)太字 2 バー 𝟐̅6 (4)\-0.033\-0.011(110)太字 2 バー 𝟐̅12 (4)\-0.022\-0.032(220)太字 4 バー 𝟒̅12 (2)\-0.0210\.247References1 T. Nakatani, N. Suwannaharn, and T.T. Sasaki, “Fabrication and characterization of single- crystalline CoSn(101¯0) kagome metal thin films for interconnect applications: Structure and  anisotropic electrical resistivity,” Appl. Phys. Lett. 127(20), 203302 (2025).2 E. Milosevic, S. Kerdsongpanya, A. Zangiabadi, K. Barmak, K.R. Coffey, and D. Gall,  “Resistivity size effect in epitaxial Ru(0001) layers,” J. Appl. Phys. 124(16), 165105 (2018).3 G. Henkelman, A. Arnaldsson, and H. Jónsson, “A fast and robust algorithm for Bader  decomposition of charge density,” Comput. Mater. Sci. 36(3), 354–360 (2006).4 M. Methfessel, D. Hennig, and M. Scheffler, “Trends of the surface relaxations, surface  energies, and work functions of the 4d transition metals,” Phys. Rev. B 46(8), 4816 (1992).8&#32; ( 10  1 - 0 )&#32; ( 11  2 - 0 )&#32; ( 10  1 - 0 )&#32;   R s buffer =  217  − 36  + 52&#32;   R s buffer =  258  − 48  + 76&#32; ( 10  1 - 0 )&#32;   𝜌 ave =   R s whole ∙ (   t CoSn +   t buffer )&#32;   𝜌 CoSn =     t CoSn / ( 1 / R s whole −   1 / R s buffer )&#32;   R s total&#32;   t buffer&#32;   𝜌 ave&#32;   𝜌 CoSn&#32; ( 1  1 - 02 )&#32;  135  − 15  + 20&#32;  169  − 45  + 98&#32;  \[ 1  1 - 00 \]&#32;  149  − 7  + 8&#32; ( 10  1 - 0 )&#32; ( 10  1 - 0 )&#32; ( 1  1 - 02 )&#32;  \[ 1  1 - 00 \]&#32; ( 10  1 - 0 )&#32; ( 10  1 - 0 )&#32; ( 20  2 - 0 )&#32; ( 11  2 - 0 )&#32; ( 10  1 - 0 )&#32; ( 20  2 - 0 )&#32; ( 11  2 - 0 )&#32; ( 22  4 - 0 )&#32; ( 20  2 - 0 )&#32; ( 10  1 - 0 )&#32; ( 22  4 - 0 )&#32; ( 11  2 - 0 )&#32; ( 10  1 - 0 )&#32; ( 20  2 - 0 )&#32;  2 -&#32; ( 22  4 - 0 )&#32; ( 22  4 - 0 )&#32; ( 10  1 - 0 )&#32; ( 11  2 - 0 )&#32;  1 -&#32;  2 -&#32;  2 -&#32;  4 -&#32;  4 -&#32;  4 -&#32;   𝟏 -&#32;   𝟐 -&#32;   𝟐 -&#32;   𝟒 -