# Fileset

[Structural analysis and transport properties of  010 -tilt grain boundaries in Fe Se Te .pdf](https://mdr.nims.go.jp/filesets/a1a456f7-aff8-4766-940a-2e4e31a29012/download)

## Creator

Kazumasa Iida, Yoshihiro Yamauchi, Takafumi Hatano, Kai Walter, Bernhard Holzapfel, Jens Hänisch, Zimeng Guob, Hongye Gao, Haoshan Shi, Shinnosuke Tokuta, Satoshi Hata, Akiyasu Yamamoto, Hiroshi Ikuta

## Rights

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

## Other metadata

[Structural analysis and transport properties of [010]-tilt grain boundaries in Fe(Se,Te)](https://mdr.nims.go.jp/datasets/1614feaf-a226-4d90-a5a1-77179e423a38)

## Fulltext

Microsoft Word - AMO_TSTA_A_2384829.docxFull Terms & Conditions of access and use can be found athttps://www.tandfonline.com/action/journalInformation?journalCode=tsta20Science and Technology of Advanced MaterialsISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/tsta20Structural analysis and transport properties of[010]-tilt grain boundaries in Fe(Se,Te)Kazumasa Iida, Yoshihiro Yamauchi, Takafumi Hatano, Kai Walter, BernhardHolzapfel, Jens Hänisch, Zimeng Guo, Hongye Gao, Haoshan Shi, ShinnosukeTokuta, Satoshi Hata, Akiyasu Yamamoto & Hiroshi IkutaTo cite this article: Kazumasa Iida, Yoshihiro Yamauchi, Takafumi Hatano, Kai Walter, BernhardHolzapfel, Jens Hänisch, Zimeng Guo, Hongye Gao, Haoshan Shi, Shinnosuke Tokuta, SatoshiHata, Akiyasu Yamamoto & Hiroshi Ikuta (08 Aug 2024): Structural analysis and transportproperties of [010]-tilt grain boundaries in Fe(Se,Te), Science and Technology of AdvancedMaterials, DOI: 10.1080/14686996.2024.2384829To link to this article:  https://doi.org/10.1080/14686996.2024.2384829© 2024 The Author(s). Published by NationalInstitute for Materials Science in partnershipwith Taylor & Francis Group.View supplementary material Accepted author version posted online: 08Aug 2024.Submit your article to this journal View related articles View Crossmark datahttps://www.tandfonline.com/action/journalInformation?journalCode=tsta20https://www.tandfonline.com/journals/tsta20?src=pdfhttps://www.tandfonline.com/action/showCitFormats?doi=10.1080/14686996.2024.2384829https://doi.org/10.1080/14686996.2024.2384829https://www.tandfonline.com/doi/suppl/10.1080/14686996.2024.2384829https://www.tandfonline.com/doi/suppl/10.1080/14686996.2024.2384829https://www.tandfonline.com/action/authorSubmission?journalCode=tsta20&show=instructions&src=pdfhttps://www.tandfonline.com/action/authorSubmission?journalCode=tsta20&show=instructions&src=pdfhttps://www.tandfonline.com/doi/mlt/10.1080/14686996.2024.2384829?src=pdfhttps://www.tandfonline.com/doi/mlt/10.1080/14686996.2024.2384829?src=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1080/14686996.2024.2384829&domain=pdf&date_stamp=08 Aug 2024http://crossmark.crossref.org/dialog/?doi=10.1080/14686996.2024.2384829&domain=pdf&date_stamp=08 Aug 2024ACCEPTED MANUSCRIPT  1  Structural analysis and transport properties of [010]-tilt grain boundaries in Fe(Se,Te) Kazumasa Iida1,8, Yoshihiro Yamauchi2, Takafumi Hatano2,8, Kai Walter3, Bernhard Holzapfel3, Jens Hänisch3, Zimeng Guo4,8, Hongye Gao5, Haoshan Shi6, Shinnosuke Tokuta7,8, Satoshi Hata4,5,6,8, Akiyasu Yamamoto7,8 and Hiroshi Ikuta2,9  1College of Industrial Technology, Nihon University, 1-2-1 Izumi-cho, Narashino, Chiba 275-8575, Japan; 2Department of Materials Physics, Nagoya University, Furo-cho, Nagoya 464-8603, Japan; 3Institute for Technical Physics, Karlsruhe Institute of Technology, Hermann-von-Helmholtz-Platz 1, Eggenstein-Leopoldshafen, 76344, Germany; 4Department of Advanced Materials Science and Engineering, Kyushu University, Kasuga, Fukuoka 816-8580, Japan; 5The Ultramicroscopy Research Center, Kyushu University, Motooka, Fukuoka 819-0395, Japan; 6Interdisciplinary Graduate School of Engineering Sciences, Kyushu University, Kasuga, Fukuoka 816-8580, Japan; 7Department of Applied Physics, Tokyo University of Agriculture and Technology, Koganei, Tokyo 184-8588, Japan; 8JST CREST, Kawaguchi, Saitama 332-0012, Japan; 9 Research Center for Crystalline Materials Engineering, Nagoya University, Furo-cho, Nagoya 464-8603, Japan   ARTICLE HISTORY Compiled July 1, 2024  ABSTRACT Understanding the nature of grain boundaries is a prerequisite for fabricating high-performance superconducting bulks and wires. For iron-based superconductors [e.g. Ba(Fe,Co) 2 As 2 , Fe(Se,Te), and NdFeAs(O,F)], the dependence of the critical current density cJ  on misorientation angle ( GBθ ) has been explored on [001]-tilt grain boundaries, but no data for other types of orientations have been reported. Here, we report on the structural and transport properties of Fe(Se,Te) grown on CeO 2 -buffered symmetric [010]-tilt roof-type SrTiO 3  bicrystal substrates by pulsed laser deposition. X-ray diffraction and transmission electron microscopy revealed that GBθ  of Fe(Se,Te) was smaller whereas GBθ  of CeO 2  was larger than that of the substrate. The difference in GBθ  between the CeO 2  buffer layer and the substrate is getting larger with increasing GBθ . For GB 24θ ≥   of the substrates, GBθ  of Fe(Se,Te) was zero, whereas GBθ  of CeO 2  was continuously increasing. The inclined growth of CeO 2  can be explained by the geometrical coherency model.   CONTACT Kazumasa Iida. Email: iida.kazumasa@nihon-u.ac.jp https://crossmark.crossref.org/dialog/?doi=10.1080/14686996.2024.2384829&domain=pdfACCEPTED MANUSCRIPT  2  The c -axis growth of Fe(Se,Te) for GB 24θ ≥   of the substrates is due to the domain matching epitaxy on (221) planes of CeO 2 . Electrical transport measurements confirmed no reduction of inter-grain cJ  for GB 9θ ≤  , indicative of strong coupling between the grains.  KEYWORDS Fe(Se,Te); roof-type grain boundary; domain matching epitaxy   1.  Introduction Grain boundaries (GBs) are interfaces between crystalline grains at which the crystallographic orientation abruptly changes. Microscopically, the overlap of the wave functions is perturbed by GBs, leading to a change in the electronic structure. The electronic structure is also affected by local strain and dislocations in and around the GB. Hence, physical properties across GBs are expectedly altered, and understanding the nature of GBs is therefore an important step for further improvement of the functionalities of materials. Polycrystalline samples contain many types of GBs, which complicates the investigations of specific GBs. To understand the nature of such a specific GB, they have to be fabricated artificially. For instance, the attempt at realizing artificial GBs in silicon ingots has been reported recently [1]. For high-temperature superconductors (HTS, e.g. YBa 2 Cu 3 O 7 δ− , YBCO) as well as iron-based superconductors (IBSs), thin films containing a well-defined single GB have been fabricated, since the critical current would be too large to evaluate by electrical transport measurements on bulk samples. In this case, superconducting thin films have been grown biaxially on bicrystal substrates, which consist of two single crystals having a, usually common symmetric, rotation along [001] that are joined by a solid-state reaction [2]. After growth, the electrical transport properties across the GB are investigated as a function of misorientation angle. Such experiments are recognized as a powerful method for understanding the GB properties of HTS, for reviews see [3, 4]. For cuprates, not only GBs with in-plane misorientation ([001]-tilt GB) but also with out-of-plane misorientation ([010]-tilt GB) as well as [100]- and [001]-twist GBs have been realized [5, 6]. The inter-grain cJ  across [001]-tilt GBs was shown to decrease exponentially above a GBθ  around 3  ∼ 5   [2, 3, 5]. This angle is defined as the critical angle cθ . Similar to the [001]-tilt GBs, the inter-grain cJ  reduced significantly at the [100]-twist type GBs. On the other hand, the inter-grain cJ  of [001]-twist GBs for Bi 2 Sr 2CaCu 2 O 8 δ+  was unaltered regardless of misorientation angle [7]. For YBCO, the inter-grain cJ  of [010]-tilt GBs was almost constant even for GBθ =8   [5, 6], indicating that cθ  can depend on the type of GB. For Ba(Fe,Co) 2 As 2  [8], Fe(Se,Te) [9, 10], and NdFeAs(O,F) [11, 12], only [001]-tilt GBs have been investigated so far. The common feature of those IBSs is that cθ  is around 9  , which is 2-3 times larger than for YBCO of the same type of GB. Additionally, the inter-grain cJ  stayed constant in the range 15  ≤ GBθ ≤ 45  , whereas for YBCO it decreases further exponentially with GBθ . These prominent features of GBs in IBSs may originate from their s ±  wave symmetry. However, no data for other types of orientations have been reported. Hence, it is interesting how cJ  is affected by [010]-tilt as well as twist GBs. To address this issue, we have fabricated Fe(Se,Te) thin films on symmetric [010]-tilt ACCEPTED MANUSCRIPT  3  roof-type SrTiO 3  bicrystals with GBθ  up to 30   and investigated the structural and transport properties. We have selected Fe(Se,Te), since it has the simplest crystal structure among IBSs. Hence, it is considered easy to extract the factors governing the superconducting properties. However, growing Fe(Se,Te) thin films with good superconducting properties is not easy due to the excess Fe, which localizes conducting carriers, leading to a lower cJ  [13]. In fact, as-grown films under our growth conditions contain excess Fe. In this paper, we firstly optimize the post-annealing conditions for Fe(Se,Te) to remove excess Fe. Then Fe(Se,Te) bicrystal films are fabricated by employing the optimized post-annealing condition, followed by structural and electrical transport characterizations.  2.  Experiment CeO 2  was grown on SrTiO 3 (001) (K&R Creation Co., Ltd.) in p O 2 =1 Pa at 600  C by pulsed laser deposition (PLD), where a commercially available CeO 2  sintered target (Toshima Manufacturing Co., Ltd.) was ablated by a KrF excimer laser (wavelength λ=248  nm) with 1 Hz. An energy density of ∼ 1.2 J/cm 2  at the target surface was employed. A total pulse number of 1320 yielded a 30 nm-thick CeO 2  film confirmed by X-ray reflectivity measurements (Supplementary fig. S1). After deposition, the CeO 2 -buffered SrTiO 3  substrates were transferred to the UHV chamber (base pressure ∼ 1×107−  Pa) for deposition of Fe(Se,Te) without exposing them to air. The Fe(Se,Te) target with nominal composition Fe:Se:Te=1:0.5:0.5 was prepared by spark plasma sintering [14]. The precursor powders were mechanically alloyed prior to the sintering [15]. The nominal FeSe0.5 Te 0.5  films were also grown on CeO 2 -buffered [010]-tilt roof-type SrTiO 3  bicrystal substrates (8  ≤ STOGBθ ≤ 30  , Furuuchi Chemical Co.) at 300  C and with 5 Hz laser repetition rate. The energy density of laser was the same as for the CeO 2  deposition. A pulse number of 7500 yielded a 135–155 nm-thick FeSe 0.5 Te 0.5  layer, which is an optimum thickness for achieving a high cT  [16, 17]. Post annealing has been conducted by referring to [18, 19]. The samples were again transferred to the CeO 2  deposition chamber after growth of FeSe 0.5 Te 0.5  followed by annealing at 100  C ≤ annealT ≤ 350  C in a fixed p O 2  of 1 Pa. The dwell time at the maximum annealT  for each experimental run was fixed at 10 min. Structural properties of the films were characterized by X-ray diffraction (XRD) using Cu Kα  radiation and transmission electron microscopy (TEM). The [001] directions of both FeSe 0.5 Te 0.5  and CeO 2  are expected to be away from the substrate normal by STOGB / 2θ , when FeSe 0.5 Te 0.5  is grown on CeO 2 -buffered symmetric [010]-tilt SrTiO 3  bicrystal substrates having STOGBθ . Hence, the growth angles (i.e. offset angle) for FeSe 0.5 Te0.5  and CeO 2  were determined by ω -scans, where the angle 2θ  was fixed as the 002 reflections of each layer. TEM was performed on a cross-sectional foil sample covering the grain boundary. The foil sample was made by focused ion beam (FIB) in a scanning electron microscope (SEM) called Helios Hydra CX (Thermo Fisher Sci., USA). The scanning TEM (STEM) observations were carried out for high-resolution microstructural analyses by a TEM called Titan Cubed G2 (Thermo Fisher Sci., USA). In order to ACCEPTED MANUSCRIPT  4  accurately assess the grain boundary angle in each layer, the automated crystal orientation mapping (ACOM) technique in a TEM called ARM-200F (JEOL Ltd., Japan) was performed by using ASTAR device (NanoMEGAS, Belgium) with a spatial resolution at 4 nm and an acceleration voltage of 200 kV. Details of the ACOM in TEM are described in refs. [20, 21]. After structural characterization, micro-bridges for electrical transport measurements were formed by laser cutting and conducted using a 4-probe method. The bridges of 100 m width had a length of 2 mm for inter-grain measurements, and 1 mm for intra-grain measurements, respectively. The superconducting transition temperature (c,90T ) was defined as a 10% drop of the normal state resistance nR , at which the resistance deviated from the linear fit to the normal state in the vicinity of the superconducting transition. cJ  was determined by an electrical field criterion of 1 µV/cm.  3.  Results and discussion   3.1.  Removal of Excess Fe The as-grown FeSe 0.5 Te 0.5  films on CeO 2 -buffered ordinary SrTiO 3 (001) substrates contained excess Fe, inferred from a resistance upturn before the superconducting transition [fig. 1(a)]. This is due to the charge carrier localization by excess Fe in Fe(Se,Te) [13]. Figure 1(a) shows the normalized resistance curves of the FeSe 0.5 Te 0.5  thin films after post-annealing. The resistance upturn was gradually suppressed with increasing annealT. At 200  C ≤ annealT ≤ 220  C, the upturn disappeared. Simultaneously, the superconducting transition temperature c,90T  increased with annealT  and reached a maximum c,90T  around 15 K at annealT =200  C [fig. 1(b)]. Further increasing annealT  reduced c,90T . For anneal > 300T  C, superconductivity disappeared completely. Additionally, the resistance curve for the film annealed at 300  C showed semiconducting behavior. Figure 1(c) shows the XRD patterns of FeSe 0.5 Te 0.5  annealed at various temperatures. In the XRD 2θ -ω  scans, no appreciable differences between the as-grown film and the film annealed at 200  C were observed. On the other hand, significant shifts of the 00 l  reflections toward higher 2θ  values were observed for the film annealed at 300  C, indicative of a decrease in c -axis length. This is mainly due to the loss of Te, since severe annealing conditions may terminate the Fe-Te bonds leading to a loss in Te [22] and the c -axis length is decreasing with decreasing Te content in FeSe 0.5 Te 0.5  single crystal [23]. When the film was annealed at 350  C, further shifting of the 00 l  peaks together with peaks originating from impurities was recognized. In fact, the c -axis length significantly reduced at annealT ≥ 300  C [fig. 1(d)], whereas the c -axis length of the superconducting films was located between 6.0 Å  and 6.1 Å . From those results, the optimum post-annealing temperature was determined as 200  C. The post-annealing conditions in this study differed from the ones reported by Zhang et  al . [18] with respect to p O 2 , annealing temperature and dwell time, which were there 100 mbar (∼ 13.3 Pa), 90  C and 1∼ 2 h. The annealing temperature of 200  C in our case is almost the double of Zhang’s study, whereas our dwell time is shorter. In our study, the resistance upturn was suppressed even at annealT =100  C. Hence, it may be possible to remove more Fe with further increasing the holding time. The post-annealing reported by Zhang et  al . not only led to removal of excess Fe but also to a significant ACCEPTED MANUSCRIPT  5  enhancement of critical currents, although cT  was slightly reduced. Post-annealing at low temperatures may indeed be used to tune the properties of superconducting films further, such as critical current properties of REBCO films [24]. Nevertheless, in the following, the FeSe 0.5 Te 0.5  films on CeO 2 -buffered [010]-tilt SrTiO 3  bicrystal substrates were post-annealed at 200  C for 10 minutes in 1 Pa of oxygen.  3.2.  Structural Analyses Figure 2(a) exhibits the XRD 2θ -ω  patterns of FeSe 0.5 Te 0.5  grown on CeO 2 -buffered [010]-tilt SrTiO 3  bicrystal substrates with various misorientation angles. The film for STOGBθ= 0   was grown on an ordinary SrTiO 3 (001) substrate. The angle FSTGBθ  shown in the panel indicates the measured offset angle of FeSe 0.5 Te 0.5  multiplied by two [i.e. the actual misorientation angle of FeSe 0.5 Te 0.5 ], and the angle in parenthesis is the misorientation angle of the SrTiO 3  bicrystals ( STOGBθ ). For STOGBθ =0  , the 00 l  reflections of FeSe 0.5 Te 0.5  and CeO 2  together with SrTiO 3  were observed. Additionally, the 101 reflection of the φ  scan showed a fourfold symmetry [Supplementary fig. S2(a)], which proves the phase-pure and epitaxial growth of FeSe 0.5 Te 0.5 . On the other hand, almost only the 00 l  reflections of FeSe 0.5 Te 0.5  were observed for STOGB > 0θ  , indicating that the offset angle of FeSe 0.5 Te 0.5  differs from those of CeO 2  and SrTiO 3 . In fact, the respective misorientation angles of FeSe 0.5 Te 0.5  and CeO 2  are different from each other and from those of the SrTiO 3  bicrystals [fig. 2(b)]. As can be seen, the actual misorientation angle of CeO 2  ( CeO2GBθ ) is getting larger than STOGBθ , whereas FSTGBθ  is always smaller than STOGBθ . A similar effect was observed in FeSe 0.5 Te 0.5  thin films on vicinal CaF 2  substrates deposited at 260  C [25]. For STOGB 24θ ≥  , FSTGBθ  was zero, indicating the absence of a GB in FeSe 0.5 Te 0.5 . These observations can be explained by the geometrical coherency model [26–28], according to which CeO STO2GB GB GB1=θ θ θ− Δ  and CeOFST 2GB GB GB2=θ θ θ− Δ  can be calculated by   STOSTO CeO1GB1 GB2STO= tan tan2 2d ddθ θ−− Δ    (1)  CeO2CeO FST1GB2 GB2CeO2= tan tan2 2d ddθ θ− −Δ     (2)  where STOd , CeO2d , and FSTd  are the out-of-plane, monolayer step height of SrTiO 3  (3.91 Å ), CeO 2  (5.41 Å ), and FeSe 0.5 Te 0.5  (5.96 Å ), respectively. The direction of the tilt of [001] CeO 2  from [001] SrTiO 3  is away from the substrate normal, because CeO STO2>d d . Similarly, the direction of the tilt of [001] FeSe 0.5 Te 0.5  from [001] CeO 2  is away from the substrate normal. For STOGBθ =30  , GB1 / 2θΔ  is calculated to 5.9  , which is close to the measured angle from the STEM image shown in fig. 2(c). The grain boundary angles of ACCEPTED MANUSCRIPT  6  CeO 2  ( CeO2GBθ ) lie on the calculated lines (dotted blue line) [fig. 2(b)], indicating that the geometrical coherency model is valid. However, this model seems not to be valid for FeSe0.5 Te 0.5 /CeO 2 , since the experimental data did not lie on the dashed red line calculated from the model. The vicinal angles of FeSe 0.5 Te 0.5  grown on off-cut CaF 2  substrates at 260  C deviated similarly from the calculation (supplementary fig. S3). This may be due to the low growth temperature, leading to a low surface mobility of atoms [28]. In fact, vicinal angles of FeSe 0.5 Te 0.5  grown at a higher temperature of 400  C were almost identical to those of the CaF 2  substrates (fig. S3). Possibly, film surfaces and CaF 2  at low temperatures do not have well defined terraces needed for the geometry coherency mechanism. Finally, for a proper analysis, the lattice parameters at growth temperature should be considered, which we omitted here for our estimates. Figure 3(a) shows the cross-sectional view of FeSe 0.5 Te 0.5  grown on the CeO 2 -buffered SrTiO 3  bicrystal with STOGBθ =30  . The respective layer thicknesses of FeSe 0.5 Te0.5  and CeO 2  were 135 nm and 30 nm. The film contained planar defects with a thickness of ∼ 1.5 nm along the ab -plane, fig. 3(b). Atomic-resolution images of SrTiO 3  and CeO 2  buffer layer around the GB confirmed that the respective GB angles are STOGBθ =30   and CeO2GBθ =42.4   [figs. 3(d) and 3(e)]. Those values are consistent with the ones evaluated by XRD measurements. Figure 3(c) confirms the presence of a GB in the CeO 2  buffer layer, whereas no visible GB was present in the FeSe 0.5 Te 0.5  layer as stated above. Additionally, the FeSe 0.5 Te 0.5  layer grew biaxially textured as shown in fig. 3(f). The in-plane texture was also confirmed by the φ  scan of the 101 reflection [Supplementary fig. S2(b)]. According to the geometric considerations based on the TEM observation, the epitaxial relation (001)[100]FeSe 0.5 Te 0.5 (114)[22 1 ]CeO 2  is realized as domain growth [29]. In fact, a domain wall structure was observed in the FeSe 0.5 Te 0.5  film along [010], i.e. across the GB for STOGBθ =30  , and their average width was 32 ± 12 nm [figs. 4(a) and 4(b)]. Note that such a structure has not been observed in the FeSe 0.5 Te 0.5  film grown on CeO 2 -buffered single-crystal SrTiO 3  substrate (Supplementary fig. S4). The relation (001)[100]FeSe 0.5 Te 0.5 (114)[22 1 ]CeO 2  also holds for STOGBθ =24  . Due to the extinction rule, the diffraction peak arising from the 114 reflection of CeO 2  could not be observed in XRD pattern. The domain growth is expressed by the following index, , ,, ,m n oh k lC , where ( )h k l× ×  lattice of the CeO 2  buffer layer and ( )m n o× ×  lattice of the Fe(Se,Te), refer to [29]. The respective indices are 2,0,01,1,0C  for along the GB and 4,0,02,2,1C  for across the GB. However, the most probable index for the latter is 9,0,04,4,2C , since the domain misfit ( dε ) expressed by eq. (3) is smaller, as shown in table 1. Additionally, the domain width FST9 a× =34 nm ( FSTa : in-plane lattice parameter of FeSe 0.5 Te 0.5 ) corresponds well to the average domain width of 32 nm observed in ACOM, and the opposite mismatch compared to the FeSe 0.5 Te 0.5(100) direction may slightly lower the total energy.  ACCEPTED MANUSCRIPT  7   2 2 2 2 2 2FST CeO2d 2 2 2 2 2 2FST CeO2= 2m n o a h k l am n o a h k l a+ + − + ++ + + + +ε  (3)  The dε  of 2,0,01,1,0C  is smaller than that of 9,0,04,4,2C , which is reflected in the full width at half maximum values ( ωΔ ) of the 00 l  rocking curves [Supplementary fig. S2(h)∼ (k)]. As can be seen, the ωΔ  for the [1 1 0] (along the GB, denoted as “L” in fig. S2) is smaller than that for the [22 1 ] (across the GB, denoted as “T” in fig. S2). For cubic lattices, the Σ  value of symmetrical GB is expressed by the sum of the squares of the Miller indices [30]. In our experimental results, a 9[110] /{221}Σ  GB with an ideal GB angle of 38.9   has formed in CeO 2  on both 24   and 30   substrates with sufficiently close real GB angles of 35.4   and 42.2  , respectively. Unlike other GBs (e.g. 11[110] /{332}Σ ), the 9[110] /{221}Σ  GB is, together with the twin 3[110] /{111}Σ  (not observed here), the most stable structure [31]. Due to the difference between the ideal and real GB angle in CeO 2  (3.5   for STOGBθ=24   and 3.3   for STOGBθ =30  , respectively), the (114) planes on either side are tilted by half of this difference, and a GB angle of ∼ 3.5   should be expected in the FeSe 0.5 Te 0.5  films, which however is not observed in fig. 4(c). From figs. 4(c) and 4(d), the respective misorientation angles between domains were within 4   and 3   for in-plane and out-of-plane. The artificial GB (or rather the two sides of the bicrystal) may be still recognized as a more macroscopic shift of the base line (average) misorientation (with respect to a common starting point) of ∼ 0.7   out-of-plane and ∼ 0.9   in-plane, which, however, is well within the range of domain-to-domain misorientations. Two more effects may explain that. First, the geometry coherency growth may happen again, now on (114) instead of (001), and since the c -axis of FeSe 0.5 Te 0.5  is shorter than the single-layer distance in (114) direction in CeO 2 , the c -axis will tend towards the substrate normal, although just about negligible ∼ 0.1  . Secondly, since FeSe 0.5 Te 0.5  single crystals typically grow in ab-oriented platelets, the surface energy of (001) may be concluded to be by far the lowest one. Hence, the system tends to adjust (001) parallel to the surface. For the low-angle GBs, a similar combination of special GB in CeO 2  (33[110] /{441}Σ  or 51[110] /{551}Σ  may be candidates), geometry coherency on the relevant, vicinal planes [(118) or (1110 ) for the abovementioned GBs], and surface energy reduction may explain the FST GB angles being lower than expected.  3.3.  Transport Properties The temperature dependence of the resistivity ρ  of the inter- and intra-grain bridges is shown in fig. 5. Although FeSe 0.5 Te 0.5  GBs were absent for the STOGBθ =24   and 30   of films, the data for inter-grain were acquired from the bridges located on the GB of CeO 2  and SrTiO 3 . For STOGB 24θ ≥  , the normal state resistivity of the intra-grain bridges was somewhat higher than that of the inter-grain bridges. The semi-logarithmic plots of fig. 5 (g)-(k) proved the resistivity dropped to the detection limit of the voltmeter below the transition. Additionally, the transition temperature of the intra-grain and the inter-grain bridges was almost the same for all samples. ACCEPTED MANUSCRIPT  8  Figure 6(a) shows the inter- and intra-grain cJ  as a function of STOGBθ  at 4.2 K. For FSTGBθ =0  , the micro-bridge was fabricated from the film grown on the ordinary SrTiO 3  substrate. All bridges showed a cJ  of 8×10 4  A/cm 2  except for the bridge with FSTGBθ =0   (cJ =1.6×10 5  A/cm 2 ). The reason for higher cJ  is that the only cJ  component is the ab -plane. On the other hand, for STOGB > 0θ  , the inter- and intra-grain measurements contained two components of cJ : along the c -axis and the ab -plane. However, inter-grain cJ  for the films having a STOGB = 24θ   and 30   was ∼ 8×10 4  A/cm 2  [fig. 6(b)] although those films had a FSTGBθ  close to 0  , fig. 2(b). These results infer that the domain wall structure gave a negative impact on cJ . Additionally, the maximum in-plane misorientation was ∼3   [fig. 4(d)], which also reduces cJ  although the in-plane misorientation angles are less than the critical angle. In fact, the inter-grain cJ  of the [001]-tilt Fe(Se,Te) GB having a misorientation angle of 3   was reduced around 20% relative to the intra-grain cJ  [9]. The ratio of inter-grain to intra-grain cJ  as a function of FSTGBθ  is shown in fig. 6(c). The data for the [001]-tilt GB are also shown for comparison [9, 10]. The ratio was almost 1 up to FSTGB 9.5θ ∼ , which is similar to the [001]-tilt GBs. Hence, absence of weak-link behavior also for [010]-tilt GBs up to ∼ 9.5   is confirmed in FeSe 0.5 Te 0.5 .  4.  Conclusion FeSe 0.5 Te 0.5  thin films have been grown on CeO 2 -buffered symmetric [010]-tilt roof-type SrTiO 3  bicrystal substrates by pulsed laser deposition. Excess Fe was successfully removed by post-annealing at 200  C for 10 min in p O 2 =1 Pa. The misorientation angle of the CeO 2  buffer layers and FeSe 0.5 Te 0.5  were different from those of the SrTiO 3  bicrystal substrates. The inclined growth of CeO 2  can be explained by the geometrical coherency model. For the nominal STOGBθ = 24   and 30   of the [010]-tilt SrTiO 3  bicrystal substrates, domain wall boundaries rather than grain boundaries were formed in FeSe 0.5 Te0.5  due to the epitaxial relation (001)[100]FeSe 0.5 Te 0.5 (114)[22 1 ]CeO 2 . The inter-grain cJ  of the [010]-tilt GB did not decay below 9.5   of the misorientation angle. The current results offer implications for mitigating the weak-link issue in HTS, since CeO 2  has been used as common buffer layers for HTS.  Disclosure statement The authors declare that they have no known competing financial interests or personal relationships that could influence the work reported in this paper.  Funding This work was supported by JST CREST Grant Number JPMJCR18J4. This work was also partly supported by the Advanced Characterization Platform of the Nanotechnology Platform Japan sponsored by the Ministry of Education Culture, Sports, Science and Technology (MEXT), Japan.  ACCEPTED MANUSCRIPT  9  References  [1] Fukuda Y, Kutsukake K, Kojima T, Usami N. Effects of grain boundary structure and shape of the solid-liquid interface on the growth direction of the grain boundaries in multicrystalline silicon. Cryst. Eng. Comm. 2022; 24: 1948–1954. doi:10.1039/D1CE01573G  [2] Dimos D, Chaudhari P, Mannhart J, LeGoues F. K. Orientation Dependence of Grain-Boundary Critical Currents in YBa 2 Cu 3 O 7 δ−  Bicrystals. Phys. Rev. Lett. 1988; 61: 219–222. doi:10.1103/PhysRevLett.61.219  [3] Hilgenkamp H, Mannhart J. Grain boundaries in high- cT  superconductors. Rev. Mod. Phys. 2002; 74: 485–549. doi: 10.1103/RevModPhys.74.485  [4] Iida K, Hänisch J, Yamamoto A. Grain boundary characteristics of Fe-based superconductors. Supercond. Sci. Technol. 2020; 33: 043001. doi:10.1088/1361-6668/ab73ef  [5] Held R, Schneider C. W, J. Mannhart J, Allard L. F, More K. L, Goyal A. Low-angle grain boundaries in YBa 2 Cu 3 O 7 δ−  with high critical current densities. Phys. Rev. B 2009; 79: 014515. doi:10.1103/PhysRevB.79.014515  [6] Goyal A, Field D. P, Held R, Mannhart J. Grain boundary networks in high-performance, heteroepitaxial, YBCO films on polycrystalline, cube-textured metals. Philos. Mag. Lett. 2011; 91: 246–255. doi:10.1080/09500839.2010.548345  [7] Zhu Y, Li Q, Tsay Y. N, Suenaga M, Gu G. D, Koshizuka N. Structural origin of misorientation-independent superconducting behavior at [001] twist boundaries in Bi 2Sr 2 CaCu 2 O 8 δ+ . Phys. Rev. B 1998; 57: 8601–8608. doi:10.1103/PhysRevB.57.8601  [8] Katase T, Ishimaru Y, Tsukamoto A, Hiramatsu H, Kamiya T, Tanabe K, Hosono H. Advantageous grain boundaries in iron-pnictide superconductors. Nat. Commun. 2011; 2: 409. doi:10.1038/ncomms1419  [9] Si W, Zhang C, Shi. X, Ozaki. T, Jaroszynski. J, Li Q. Grain boundary junctions of FeSe 0.5 Te 0.5  thin films on SrTiO 3  bi-crystal substrates. Appl. Phys. Lett. 2015; 106: 032602. doi: 10.1063/1.4906429  [10] Sarnelli. E, Nappi C, Camerlingo C, Enrico E, Bellingeri E, Kawale S, Braccini V, Leveratto A, Ferdeghini C. Properties of Fe(Se, Te) Bicrystal Grain Boundary Junctions, SQUIDs, and Nanostrips. IEEE Trans. Appl. Supercond. 2017; 27, 7400104. doi: 10.1109/TASC.2016.2636248  [11] Omura T, Matsumoto T, Hatano T, Iida K, Ikuta H. FABRICATION OF GRAIN BOUNDARY JUNCTIONS USING NdFeAs(O,F) SUPERCONDUCTING THIN FILMS. J. Phys. Conf. Ser. 2018; 1054: 012024. doi: 10.1088/1742-6596/1054/1/012024  [12] Iida K, Omura T, Matsumoto T, Hatano T, Ikuta H. Grain boundary ACCEPTED MANUSCRIPT  1  characteristics of oxypnictide NdFeAs(O,F) superconductors. Supercond. Sci. Technol. 2019; 32: 074003. doi: 10.1088/1361-6668/ab1660  [13] Liu T. J, Ke X, Qian B, Hu J, Fobes D, Vehstedt E. K, Pham H, Yang J. H, Fang M. H, Spinu L, Schiffer P, Liu Y, Mao Z. Q. Charge-carrier localization induced by excess Fe in the superconductor Fe1 y+ Te 1 x− Se x . Phys. Rev. B 2009; 80: 174509. doi: 10.1103/PhysRevB.80.174509  [14] Tokuta S, Hasegawa Y, Shimada Y, Yamamoto A. Enhanced critical current density in K-doped Ba122 polycrystalline bulk superconductors via fast densification. iScience 2022; 25: 103992. doi: 10.1016/ j.isci.2022.103992  [15] Tokuta S, Yamamoto A. Enhanced upper critical field in Co-doped Ba122 superconductors by lattice defect tuning. APL Mater. 2019; 7: 111107. doi: 10.1063/1.5098057  [16] Si W, Han S, Shi X, Ehrlich S, Jaroszynski J, Goyal A, Li Q. High current superconductivity in FeSe 0.5 Te 0.5 -coated conductors at 30 tesla. Nat. Commun. 2013; 4: 1347. doi: 10.1038/ncomms2337  [17] Ozaki T, Wu L, Zhang C, Jaroszynski J, Si W, Zhou J, Zhu Y, Li Q. A route for a strong increase of critical current in nanostrained iron-based superconductors. Nat. Commun. 2016; 7: 13036. doi: 10.1038/ncomms13036  [18] Zhang C, Si W, Li Q. Doubling the critical current density in superconducting FeSe 0.5 Te 0.5  thin films by low temperature oxygen annealing. Appl. Phys. Lett. 2016; 109: 202601. doi: 10.1063/1.4967879  [19] Sun Y, Shi Z, Tamegai T. Review of annealing effects and superconductivity in Fe1 y+ Te 1 x− Se x  superconductors. Supercond. Sci. Technol. 2019; 32: 103001. doi: 10.1088/1361-6668/ab30c2  [20] Rauch E. F, Portillo J, Nicolopoulos S, Bultreys D, Rouvimov S, Moeck P. Automated nanocrystal orientation and phase mapping in the transmission electron microscope on the basis of precession electron diffraction. Zeitschrift fur Krist. 2010; 225: 103–109. doi: 10.1524/zkri.2010.1205  [21] Guo Z, Gao H, Kondo K, Hatano T, Iida K, Hänisch J, Ikuta H, Hata S. Nanoscale Texture and Microstructure in a NdFeAs(O,F)/IBAD-MgO Superconducting Thin Film with Superior Critical Current Properties. ACS Appl. Electron. Mater. 2021; 3: 3158–3166. doi: 10.1021/acsaelm.1c00364  [22] Zhang N, Liu C, Zhao J-L, Lei T, Wang J-O, Qian H-J, Wu R, Yan L, Guo H-Z, Ibrahim K. Observation of selective surface element substitution in FeTe 0.5 Se 0.5  superconductor thin film exposed to ambient air by synchrotron radiation spectroscopy. Chin. Phys. B 2016; 25: 097402. doi: 10.1088/1674-1056/25/9/097402  [23] Mukasa K, Matsuura K, Qiu M, Saito M, Sugiura Y, Ishida K, Otani M, Onishi Y, Mizukami Y, Hashimoto K, Gouchi J, Kumai R, Uwatoko Y, Shibauchi T. ACCEPTED MANUSCRIPT  1  High-pressure phase diagrams of FeSe1 x− Te x : correlation between suppressed nematicity and enhanced superconductivity. Nat. Commun. (2021);12: 381. doi: 10.1038/s41467-020-20621-2  [24] Popov R, Ackermann K, Rijckaert H, Hänisch J, Driessche I. Van, Holzapfel B. Effect of oxygenation process on flux pinning in pristine and BaHfO 3  nanocomposite GdBa 2 Cu 3 O 7  superconducting thin films. J. Phys.: Conf. Ser. 2020; 1559: 012038. doi: 10.1088/1742-6596/1559/1/012038  [25] Bryja H, Hühne R, Iida K, Molata S, Sala A, Putti M, Schultz L, Nielsch K, Hänisch J. Deposition and properties of Fe(Se,Te) thin films on vicinal CaF 2  substrates. Supercond. Sci. Technol. 2017; 30: 115008. doi: 10.1088/1361-6668/aa8421  [26] Nagai H. Structure of vapor deposited Ga x In 1 x− As crystals. J. Appl. Phys. 1974; 45: 3789–3794. doi: 10.1063/1.1663861  [27] Ayers J. E, Ghandhi S. K, Schowalter L. J. Crystallographic tilting of heteroepitaxial layers. J. Cry. Growth 1991; 113: 430–440. doi: 10.1016/0022-0248(91)90077-I  [28] Budai J. D, Yang W, Tamura N, Chung J-S, Tischler J. Z, Larson B. C, ICE G. E, Park C, Norton D. P. X-ray microdiffraction study of growth modes and crystallographic tilts in oxide films on metal substrates. Nat. Mater. 2003; 2: 487–. doi: 10.1038/nmat916  [29] Kaneko S, Akiyama K, Ito T, Yasui M, Soga M, Hirabayashi Y, Funakubo H, Yoshimoto M. Large Lattice Misfit on Epitaxial Thin Film: Coincidence Site Lattice Expanded on Polar Coordinate System. Jpn. J. Appl. Phys. 2010; 49: 08JE02. doi: 10.1143/JJAP.49.08JE02  [30] Lucid A. K, Plunkett A. C, Watson G. W. Predicting the Structure of Grain Boundaries in Fluorite-Structured Materials. Johnson Matthey Technol. Rev. 2019; 63: 247–254. doi: 10.1595/205651319X15598975874659  [31] Feng B, Sugiyama I, Hojo H, Ohta H, Shibata N, Ikuhara Y. Atomic structures and oxygen dynamics of CeO2 grain boundaries. Sci. Rep. 2016; 6: 20288. doi: 10.1038/srep20288    ACCEPTED MANUSCRIPT  1  Table 1.  The domain indices , ,, ,m n oh k lC  and the corresponding domain mismatch calculated from eq. (3).    , ,, ,m n oh k lC    d (%)ε    2,0,01,1,0C    -0.93  4,0,02,2,1C    -6.82  9,0,04,4,2C    4.96       Figure  1. (a) The resistance curves of the as-grown FeSe 0.5 Te 0.5  (FST) and the FST films annealed at various temperatures normalized to the value at 300 K. Inset shows the magnified image of (a) around the superconducting transition. STO represents the SrTiO 3  substrate. (b) The transition temperature c,90T  as a function of the annealing temperature annealT . The maximum c,90T  around 15 K was observed at annealT =200  C. (c) The XRD patterns of the FeSe 0.5 Te 0.5  thin films annealed at 200  C, 300  C, and 350  C. For comparison, the data for the as-grown film is also shown. Beyond the optimum annealT , the 00 l  diffraction peaks shifted to higher angles. For annealT = 350  C, some diffraction peaks marked as “• “ other than FST and CeO 2  were observed. The peaks marked as “∗ “ originate from SrTiO 3 . (d) The c -axis length as a function of the annealing temperature annealT . The c -axis length of the superconducting films was located between 6.0 Å  and 6.1 Å . ACCEPTED MANUSCRIPT  1      Figure  2. (a) The XRD pattern of the FeSe 0.5 Te 0.5  thin films grown on CeO 2 -buffered [010]-tilt symmetric SrTiO 3  (STO) bicrystal substrates having various grain boundary angles STOGBθ . Here, the angle FSTGBθ  corresponds to twice the offset angle of FeSe 0.5 Te 0.5 . The peaks marked as “∗ “ originate from SrTiO 3 . Because of the different offset angles between FeSe 0.5 Te 0.5  and CeO 2  as well as FeSe 0.5 Te 0.5  and SrTiO 3 , almost only the 00 l  peaks from FeSe 0.5 Te 0.5  were observed. (b) The FSTGBθ  for FeSe 0.5 Te 0.5  (closed symbol) and CeO2GBθ  for CeO 2  (open symbol) as a function of STOGBθ . The dashed red and dotted blue lines are calculations using the geometrical coherency model [26, 27, 28]. (c) The atomic resolution HAADF-STEM image of the interface between CeO 2  and SrTiO 3  having a STOGBθ = 30  . The calculated value of CeO STO2GB GB( ) / 2θ θ−  is 5.9  , which is close to the measured value of ∼ 5.5  .     ACCEPTED MANUSCRIPT  1  Figure  3. Microstructure of the FeSe 0.5 Te 0.5 /CeO 2  sample grown on the 30   [010]-tilt symmetric SrTiO 3  bicrystal substrate. (a) Cross-sectional view near the GB acquired by ADF-STEM. (b) ADF-STEM image taken away from GB. Planar defects shown by a black arrow are visible. (c) Magnified image of (a). The GB is absent in FeSe 0.5 Te 0.5 . Atomic-resolution HAADF-STEM image of the GBs in SrTiO 3  (d) and CeO 2  (e). The GB angle in CeO 2 , CeO2GBθ , is 42.4  , consistent with the value by XRD. (f) Atomic-resolution image of the interface between CeO 2  and FeSe 0.5 Te 0.5 , which was clean and without reaction layer.    Figure  4. (a) Automatic crystal orientation mapping of FeSe 0.5 Te 0.5  grown on CeO 2 -buffered SrTiO 3  with STOGBθ =30   by scanning precession diffraction. (b) Inverse pole figure map. (c) Out-of-plane and (d) in-plane misorientation profiles from the first point along the orange line shown in (b). ACCEPTED MANUSCRIPT  1     Figure  5. The electrical measurement using the intra- and inter-grain bridges is schematized in (a). The resistivity curves of the inter- and intra-grain bridges with various STOGBθ  [(b)∼ (k)]. The open and solid symbols represent the intra- and inter-grain bridges, respectively. STOGBθ =0   [(b)] refers to the ordinary substrate. Semi-logarithmic plot of (b)∼(f) in the vicinity of the transition [(g)∼ (k)].    Figure  6. (a) cJ  of the inter- and intra-grain bridges as a function of STOGBθ  at 4 K. The micro-bridge with STOGBθ =0   was fabricated from the film grown on the ordinary SrTiO 3  substrate. cJ  was almost constant around 8×10 4  A/cm 2  except for the film grown on the ordinary SrTiO 3  substrate. (b) Data of (a) replotted as a function of FSTGBθ . (c) The FSTGBθ  dependence of the normalized cJ  for the Fe(Se,Te) bicrystal films measured at 4 K in comparison to data of [001]-tilt GBs.  Graphical-abstract ACCEPTED MANUSCRIPT      "State The isby em    ement of Nossue of weakmploying Ceovelty". k-links inheeO2-buffer lerent in ironayer contain 1 n-based supning the Σ9erconductor[110]/{221rs, Fe(Se,Te} grain boue), can be avundary. voided  ACCEPTED MANUSCRIPT  1   Supplementary information on “Structural analysis and transport properties of the [010]- tilt grain boundaries in Fe(Se,Te)”  Kazumasa Iida1,8, Yoshihiro Yamauchi2, Takafumi Hatano2,8, Kai Walter3, Bernhard Holzapfel3, Jens Hänisch3, Zimeng Guo4,8, Hongye Gao5, Haoshan Shi6, Shinnosuke Tokuta7,8, Satoshi Hata4,5,6,8, Akiyasu Yamamoto7,8, Hiroshi Ikuta2,9 1 College of Industrial Technology, Nihon University, 1-2-1 Izumi-cho, Narashino, Chiba 275- 8575, Japan; 2 Department of Materials Physics, Nagoya University, Furo-cho, Nagoya 464-8603, Japan; 3 Institute for Technical Physics, Karlsruhe Institute of Technology, Hermann-von-Helmholtz- Platz 1, Eggenstein-Leopoldshafen, 76344, Germany; 4 Department of Advanced Materials Science and Engineering, Kyushu University, Kasuga, Fukuoka 816-8580, Japan; 5 The Ultramicroscopy Research Center, Kyushu University, Motooka, Fukuoka 819-0395, Japan; 6 Interdisciplinary Graduate School of Engineering Sciences, Kyushu University, Kasuga, Fukuoka 816-8580, Japan; 7 Department of Applied Physics, Tokyo University of Agriculture and Technology, Koganei, Tokyo 184-8588, Japan; 8 JST CREST, Kawaguchi, Saitama 332-0012, Japan; 9 Research Center for Crystalline Materials Engineering, Nagoya University, Furo-cho, Nagoya 464-8603, Japan  CONTACT Kazumasa Iida: iida.kazumasa@nihon-u.ac.jp ACCEPTED MANUSCRIPT    1. X  Figuris almX-ray reflecre S1| X-ray most identicactivity measureflectivity mal to the one murement formeasurementmeasured fro1 CeO2 grownt confirmed tom the cross-n on SrTiO3 that the thick-sectional AD3(001)  kness of CeODF-STEM imO2 is 32 nm, wmage [fig. 3(awhich a)]. ACCEPTED MANUSCRIPT  1   2. The 101 reflection of φ scans and the 00l rocking curves of FeSe0.5Te0.5 grown on the CeO2- 𝐒𝐓𝐎 buffered SrTiO3 ordinary substrate and bicrystal substrate with  𝜽𝐆𝐁 =30º. (a) (b) (c)     Figure S2| (a) The φ scans of the 101 reflection of FeSe0.5Te0.5 grown on CeO2-buffered ordinary #$% SrTiO3 substrate and (b) bicrystal substrate with  !" =30°. (c) The schematic illustration of the X-ray scan direction for the 00l rocking curves. (d)-(g) The 00l (l=1, 2, 3 and 4) rocking curves of FeSe0.5Te0.5  grown on CeO2-buffered ordinary SrTiO3  substrate. “T” and “L” denote the transverse and longitudinal directions of the X-ray scans, shown in (c). As expected, no difference #$% in FWHM were observed. On the other hand, for the film grown on bicrystal substrate (𝜃!" the FWHM for “T” (i.e., perpendicular to the GB) are larger than those for “L” [(h)~(k)]. =30°), ACCEPTED MANUSCRIPT  2  !"  !"  ,-.  Table S1 shows the FWHM exhibited in fig. S2(a) and (b). The ∆  of the FeSe0.5Te0.5 film on #$% single crystal SrTiO3  (𝜃!" = 0º) is smaller than of the film on bicrystal substrate. Table S2 summarizes the crystalline quality of the out-of-plane direction of the FeSe0.5Te0.5 films on single crystal and bicrystal substrates. The FWHM of FeSe0.5Te0.5 on ordinary SrTiO3 are almost the same values regardless of the scan directions. On the other hand, for FeSe0.5Te0.5 on bicrystal substrate, FWHM shows a strong directional dependence: FWHM of “T”-direction are always larger than those of “L”-direction.  Table S1| The FWHM (∆𝜙) of FeSe0.5Te0.5 grown on the CeO2-buffered ordinary SrTiO3 substrate and bicrystal SrTiO3 substrate (𝜃#$%=30°).  FeSe0.5Te0.5 on ∆𝜙&'&   (°) ∆𝜙'&&     (°) ∆𝜙&('&   (°) ∆𝜙'&(&   (°) ordinary SrTiO3 2.20 2.24 2.15 2.28 #$% bicrystal SrTiO3  (𝜃!" =30º) 3.82 3.95 3.82 4.07 Table S2| The FWHM (∆𝜔) of FeSe0.5Te0.5 grown on the CeO2-buffered ordinary SrTiO3 substrate and bicrystal SrTiO3 substrate (𝜃#$%=30°).  ordinary SrTiO3 bicrys 2. Vicinal angle evaluated from the geometrical coherency model To evaluate vicinal angles of FeSe0.5Te0.5 grown on off-cut CaF2 substrates, the following equation is employed: /#$  = &tan0& *𝑑12/) − 𝑑/#$ tan𝜃12/)-& (S1) ∆𝜃,-. /#$ /#$ 𝑑12/) 12/) ,-. with  𝜃,-. = ∆𝜃,-. + 𝜃,-. (S2) where  /#$ is the vicinal angle of FeSe0.5Te0.5,  𝑑12/)   is the lattice parameter of CaF2 (5.462 Å), 12/) 𝑑/#$   is the c-axis length of FeSe0.5Te0.5 (5.96 Å) [S1] and  ,-. is the vicinal angle of the CaF2 substrates. Due to  /#$  > 𝑑12/), the direction of the tilt of [001] FeSe0.5Te0.5 from [001] CaF2 is away from the substrate normal. The FeSe0.5Te0.5 thin films were grown on off-cut CaF2 substrates at 260 °C and 400 °C, respectively [S2]. Figure S3 shows the measured vicinal angle of 12/) FeSe0.5Te0.5  by X-ray diffraction as a function of  𝜃,-. . The dashed lines are calculation from equations S1 and S2. As can be seen, the vicinal angles of FeSe0.5Te0.5 grown at 260 °C deviate FeSe0.5Te0.5 on direction ∆𝜔''& (°) ∆𝜔'') (°) ∆𝜔''* (°) ∆𝜔''+ (°) T 1.26 1.15 1.15  1.13   L 1.25 1.16 1.13  1.07  #$% T 2.73 2.27 1.64  1.59  tal SrTiO3 (𝜃!"   =30º) L 1.19 0.83 0.87  0.91  ACCEPTED MANUSCRIPT  2    from the calculation, whereas they lie on the calculated lines for the films grown at 400 °C. These results suggest that the inclined growth mechanism based on the geometrical coherence model is operative at high growth temperature, but not at low growth temperature.   Figure S3| The vicinal angle of FeSe0.5Te0.5 on off-cut CaF2 substrates grown at 260 ºC and 400 12/) ºC as a function of  ,-.      . 4. Cross-sectional TEM image of FeSe0.5Te0.5 on CeO2-buffered ordinary SrTiO3 substrate   Figure S4| The cross-sectional view of FeSe0.5Te0.5 grown on the CeO2-buffered SrTiO3 substrate ()* (𝜃&' = 0°) obtained by ADF-STEM. Reference [S1] Musaka K, Matsuura K, Qin M, Saito M, Sugiura Y, Ishida K, Otani M, Oishi Y, Mizukami Y, Hashimoto K, Gouchi J, Kumai R, Uwatoko Y, Shibauchi T. High-pressure phase diagrams of FeSe1-xTex: correlation between suppressed nematicity and enhanced superconductivity. Nat. Commun. (2021);12: 381. doi: 10.1038/s41467-020-20621-2 [S2] Bryja H, Hühne R, Iida K, Molata S, Sala A, Putti M, Schultz L, Nielsch K, Hänisch J. ACCEPTED MANUSCRIPT  2    Deposition and properties of Fe(Se,Te) thin films on vicinal CaF2  substrates. Supercond. Sci. Technol. 2017; 30: 115008. doi: 10.1088/1361-6668/aa8421