# Fileset

[s41598-025-12345-4.pdf](https://mdr.nims.go.jp/filesets/14ac4d57-e1d5-4bd9-ba08-80566437a5fa/download)

## Creator

[Masahiro Goto](https://orcid.org/0000-0002-1003-2781), Michiko Sasaki, [Taku Moronaga](https://orcid.org/0000-0002-6915-0627), [Toru Hara](https://orcid.org/0000-0002-9715-6444), [Yibin Xu](https://orcid.org/0000-0001-8600-8748)

## Rights

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

## Other metadata

[Room-temperature FeSi2-doped Cu2Se thermoelectric films with enhanced figure of merit](https://mdr.nims.go.jp/datasets/d4084a54-0c1a-4987-a3bb-5e635ba7ebea)

## Fulltext

Room-temperature FeSi2-doped Cu2Se thermoelectric films with enhanced figure of meritMasahiro Goto1, Michiko Sasaki1, Taku Moronaga2, Toru Hara3 & Yibin Xu4Thermoelectric (TE) materials offer a promising pathway toward achieving carbon neutrality by converting waste heat into electricity. The enhancement of their figure-of-merit (zT) depends on optimizing the composition of materials and nanostructures, reducing the thermal conductivity, and increasing the power factor. Cu2Se, a superionic material, achieves a zT of 0.4 at 300 K by facilitating Cu ion movement within its face-centered cubic lattice, effectively suppressing thermal conductivity. Herein, we present a novel TE material developed by doping CuxSe crystals of different compositions with FeSi2. We report a remarkable zT of 0.69 at 298 K for Cu2Se-based materials and reveal the presence of the CuO and Cu2O tiny crystals on the material surface, uniform dispersion of Si within the film, and formation of distinctive amorphous FeO. Our strategy holds great potential for notably advancing waste heat recovery in sustainable TE materials.Keywords  Thermoelectric, Cu2Se, Sputter, Thin film, Composite, FeSi2Developing thermoelectric (TE) materials capable of efficiently converting waste heat into electricity is crucial for achieving energy goals aligned with carbon neutrality1,2. Since approximately 90% of the thermal energy is dissipated at temperatures below 600 K, TE materials with enhanced performance near room temperature are in high demand, especially as autonomous energy sources for wearable sensing devices3. The conversion efficiency from thermal to electric energy is characterized by the dimensionless figure-of-merit, zT = S2σ T/κ, where S is the Seebeck coefficient, σ is the electrical conductivity, S2σ is the power factor (PF), T is the temperature, and κ is the thermal conductivity. Improving the TE conversion efficiency requires innovative strategies to reduce thermal conductivity and enhance the power factor through the optimization of the material type, composition, interfaces, and nanostructures. Despite considerable progress, the zT values for commercial TE materials, such as Bi2Te3 (n-type) and Bi0.5Sb1.5Te3 (p-type), remain limited to ~ 0.7 at room temperature (300 K), corresponding to ~ 7% conversion efficiency, which is considerably smaller than ~ 20% efficiencies reported for commercial photovoltaic modules. In particular, there are few p-type thermoelectric materials that exhibit high zT values, and the development of novel p-type thermoelectric materials are expected. This large difference highlights the pressing need to improve the conversion efficiency of sustainable, cost-effective TE materials, which is crucial for their widespread adoption. Achieving this goal aligns with the primary objective of installing TE element modules over large areas for extended periods, satisfying performance and sustainability requirements.Furthermore, thermal conductivity, Seebeck coefficient, and electrical conductivity are the main material properties that dictate the improvement of dimensionless zT. However, their intercorrelated effects hinder the enhancement of the TE performance. The use of phonon-glass electron-crystal (PGEC) materials are among the most popular solutions to address these challenges because they exhibit glass-like behavior for phonons, reducing thermal conductivity. Moreover, they exhibit crystalline behavior for electrons, thereby enhancing electrical conductivity, which can considerably improve zT4.Recent research efforts have focused on optimizing TE materials, including the optimization of the BiN5 thermoelectric properties, prediction of high zT values for ternary transition-metal nitride halide monolayers, such as ZrNI or HfNI6, through data science methods, and investigation of the 2D GeTe/arsenene van der 1Thermal Energy Materials Group, Research Center for Materials Nanoarchitectonics (MANA), National Institute for Materials Science (NIMS), 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan. 2Electron Microscopy Unit, Materials Fabrication and Analysis Platform, Research Network and Facility Services Division, National Institute for Materials Science (NIMS), 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan. 3Microstructure Analysis Group, Research Center for Structural Materials, National Institute for Materials Science (NIMS), 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan. 4Data-Driven Inorganic Materials Group, Center for Basic Research on Materials, National Institute for Materials Science (NIMS), 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan. email: goto.masahiro@nims.go.jp; sasaki.michiko@nims.go.jpOPENScientific Reports |        (2025) 15:27278 1| https://doi.org/10.1038/s41598-025-12345-4www.nature.com/scientificreportshttp://www.nature.com/scientificreportshttp://crossmark.crossref.org/dialog/?doi=10.1038/s41598-025-12345-4&domain=pdf&date_stamp=2025-7-24Waals heterostructure7. In addition, studies have explored cobalt-containing sintered silicon–germanium alloys8 and p-type Ta0.42Nb0.3V0.15Ti0.13FeSb thermoelectric materials and performed composition optimizations in the quaternary phase space of half-Heusler compounds9. Further enhancements in the spin-driven thermopower and zT value without compromising electrical conductivity have been achieved using temperature-driven spin crossover, as observed in 5% Cr-doped MnTe10. Additionally, an increase in the room-temperature thermoelectricity with tensile strain in SrTiO3-based superlattices, such as [(SrTiO3)m/(SrTi0.8Nb0.2O3)n]t, on DyScO3(110) substrates has been achieved11. Furthermore, TE property enhancement has been observed in Bi2Te3 through combinatorial gradient annealing technique12. A significant breakthrough is the improvement in TE performance through quadruple-band synglisis, which promotes the convergence of the energies and momenta of four valence bands13. In particular, selenium (Se)-alloyed SnS materials using quadruple-band synglisis demonstrate a high zT of ~ 1.0 at 300 K13, making them promising for advancement of room-temperature TE materials.Cu2Se is a well-known and attractive superionic conductive PGEC material in which Cu ions exhibit liquid-like behavior within a stable Se crystal lattice. Cu2Se undergoes a phase transition from a monoclinic α-phase to a cubic β-phase at ~ 400 K and is expected to improve zT through the PGEC phenomenon. Although the stable α-phase is suitable for the fabrication of TE devices, the high-temperature β-phase improves the Cu2Se ionic conductivity but complicates the device fabrication14. In 2012, Liu et al.15 and Yu et al.16 reported zT values of 0.2 and 0.4, respectively, for bulk Cu2Se at 300  K. In 201617, Perez-Taborda et al. demonstrated a high zT of 0.45 at room temperature by precisely controlling the overstoichometric Cu2.075Se composition in a Cu2Se film. Interestingly, some investigations have reported remarkable zT values exceeding 400 within specific temperature ranges, particularly during structural phase transitions18. However, these achievements remain largely impractical for widespread applications as they have not yet surpass the performance of Bi2Te3 (n-type) and Bi0.5Sb1.5Te3 (p-type). Therefore, various doping techniques have been explored to enhance the TE efficiency of Cu2Se, incorporating substances such as Ga19, Fe20,21, Ni20, Mn20, In20, Zn20, Sm20, Te22,23, K24, Cl25, TiO226, SiC27, W28, Au29, Ti30, Ag31, BiCuSeO + graphene32, Sb33, Sn34, BN35, In(super lattice)36 and Sc, Y, and La (first-principles investigations)37. Among these dopants, K doping yielded the highest zT of 0.56 at room temperature24. However, further improvements driven by simpler fabrication processes, safe and less toxic materials, and higher performance remain strongly desired.Herein, to discover p-type thermoelectric materials with high zT at room temperature, we propose a novel method to improve the zT of Cu2Se-based materials by incorporating small amounts of FeSi2. Our approach involves incorporating multiple CuxSe crystal structures with different compositions to enhance the Seebeck coefficient while maintaining low thermal conductivity through increased interface density. The proposed method defines a strategy for low-cost, high-performance TE that can replace Bi2Te3 at room temperature, offering a pathway for the development of highly efficient devices.Results and discussionWe prepared Cu2SexFeySizOt (CSSFO) films using a combinatorial sputter coating system (COSCOS; see Methods) and modified the coating parameters by adjusting the substrate temperature from 298 K to 773 K and the radiofrequency (RF) power from 60 to 120 W (Supplementary Fig. S1)38,39. Energy-dispersive X-ray (EDX) spectroscopy was used to analyze the elemental composition of the CSSFO films under various fabrication conditions (Table  1, Supplementary Fig. S2). For CSSFO samples prepared at 773  K, the Cu-to-Se ratio was approximately 2.4, and the elemental composition remained stable as the RF power increased from 60 to 120  W. Conversely, the films produced at 60  W and 298  K exhibited a Cu-to-Se ratio of approximately 1.9 with a marginal increase in Fe and O concentrations, indicating that the substrate temperature and RF power considerably influence the elemental composition and properties of the films.Figure 1 shows the dependence of the CSSFO films’ TE properties on the RF power and substrate temperature. Cu2Se crystals undergo a transition from the monoclinic α-phase to the face-centered cubic β-phase below 400 K. The results show that the CSSFO samples fabricated at 773 K exhibit a similar phase transition, although the phase transition temperature (PTT) shifts to 350 K and increases to 370 K with increasing RF power during coating. Repeating measurements three times up to 573 K confirmed that the material is thermally stable (Supplementary Fig. S3). However, the CSSFO films prepared at 60 W and 298 K exhibited no observable phase change in the Seebeck coefficient. electrical conductivity and PF measurements indicated a phase transition during the second measurement cycle, with the PTT occurring at ~ 400 K, which is consistent with that of conventional Cu2Se. The electrical conductivity shown in Fig. 1a exhibits a peak value of ~ 4 MS/m for films produced at 60 W and 298 K. The sample prepared at 60 W and 773 K exhibited a lower electrical conductivity of ~ 1.5 MS/m, whereas the films produced at 80–120 W and 773 K demonstrated electrical conductivity values of 2.0–2.7 MS/m, with Sample Thickness (µm)Composition (at%)Cu/Se ratioO Si Fe Cu Se60 W, 298 K 9.37 8.8 0.2 1.4 58.8 30.8 1.960 W, 773 K 13.10 4.9 0.4 0.5 67.9 26.3 2.680 W, 773 K 11.75 3.3 0.2 0.5 67.5 28.5 2.4100 W, 773 K 13.00 3.5 0.3 0.4 67.0 28.8 2.3120 W, 773 K 13.25 5.1 0.3 0.3 65.7 28.6 2.3Table 1.  Elementary concentration of Cu2SexSiyFezOt films prepared under different synthesis conditions. Scientific Reports |        (2025) 15:27278 2| https://doi.org/10.1038/s41598-025-12345-4www.nature.com/scientificreports/http://www.nature.com/scientificreportsa maximum near the PTT. The electrical conductivity value for the 120-W sample at room temperature was twice as that of previously reported values15, exhibiting excellent stability with temperature changes up to 573 K. The Seebeck coefficient shown in Fig. 1b varies from 20 to 38 µV/K in the α-phase (298–350 K) and from 20 to 53 µV/K in the β-phase for the 773 K samples. In contrast, the Seebeck coefficient for the 60 W and 298 K samples ranged from 10 to 30 µV/K, indicating no evidence of phase changes. The PF variation in the 298–573 K range is shown in Fig. 1c, ranging from 0.5 to 5.0 mW/mK2. The samples prepared on the substrates at 773 K exhibit a consistent inflection point at PTT, which was absent in those prepared on the substrates at 298 K. At room temperature, the PF for CSSFO films reached 2.6 mW/mK2, which is approximately 2.4 times greater than the previously reported value of 1.1 mW/mK217. This result represents the highest PF value reported for the Cu2Se-based TEs, highlighting the potential of the CSSFO films for improved TE efficiency. Interestingly, among the 773 K samples, the Seebeck coefficient increases with RF power from 60 to 120 W, which suggests a decrease in carrier density. However, the electrical conductivity also increases across this range, remaining nearly constant between 80 and 120 W at 298 K. This simultaneous decrease in carrier density and increase in the electrical conductivity implies a substantial enhancement in carrier mobility. This effect is likely attributed to the promotion of microcrystallization and the influence of interfacial contributions—factors that will be discussed in detail later. Overall, these effects are strongly correlated with the dramatic improvement in PF.Figure 2 shows the thermal conductivity and zT of the CSSFO films measured at room temperature. The thermal conductivity values range from 1.13 to 1.25 W/mK, which is slightly higher than that reported in the previous study (1.0 W/mK)15. The zT values exhibit considerable dependence on the RF power and substrate temperature. At 60 W, zT was measured at 0.11 and 0.35 for the substrate temperatures at 298 K and 773 K, respectively. For the samples on 773 K substrates, zT increased from 0.35 to 0.69 as the RF power increased from 60 to 120 W. Figure 3 compares the zT values of Cu2Se-based materials in the literature depending on different types of doping. Liu et al.15, Yu et al.16, and Prez-Taborda et al.17 reported zT values of 0.2–0.4 at room temperature for Cu2Se without doping, which was recently reported at 0.45 by Ang et al. in 202331. Although most doping attempts to exceed these zT values were unsuccessful, doping with K (zT = 0.56) and W (zT = 0.49) proved to be successful. In this study, we achieved a notable improvement in zT, reaching 0.69, approaching that of Bi2Te3 at room temperature, which is beneficial for TE applications due to its reduced cost and lower toxicity. The thermal conductivity of the Cu2Se-based materials remains relatively stable across a broad temperature spectrum. Liu et al.15 observed that thermal conductivity at 573 K is comparable to that at room temperature; thus, using our measured thermal conductivity value of 1.13 W/mK at room temperature, zT for the CSSFO samples at 573 K is estimated to be 2.45, which is remarkably high. The stoichiometric composition of Cu2Se often yields a maximum zT of ~ 0.417. Notably, the improvement in zT values achieved here strongly depends on the limitation of the RF power source of the sputter coating apparatus. Therefore, further research utilizing enhanced equipment and optimized fabrication configurations holds remarkable potential for achieving even higher zT values for CSSFO materials, offering low-cost, high-performance alternatives to Bi2Te3.Figure  4 shows the scanning electron microscopy (SEM) and high-resolution transmission electron microscopy (HRTEM) micrographs of the CSSFO sample prepared at 120 W and 773 K (micrographs of the remaining samples are shown in Supplementary Fig. S4). Several fissures are visible on the thin film’s surface in Fig. 4a. A close look at these fissures reveals clearly visible approximately 1–3-µm-sized crystal grains, as shown in the inset. When the surface is magnified, a tetrahedral-like structure is observed. A detailed observation using HRTEM reveals that the structure corresponds to a Cu2O single crystal, as shown in Fig.  4b. High-resolution micrographs in Fig. 4c reveal that these micron-sized grains consist of a large number of fine crystals and interfaces, whereas the atomic-level observations in Fig. 4d show their polycrystalline orientations. X-ray diffraction (XRD) analysis was also carried out (Fig. 5 and Supplementary Fig. S5). The XRD and TE property data were analyzed using principal component analysis (PCA). The Lasso and random forest algorithms were used for analysis, yielding R2 values of 0.90 and 0.60, respectively. Considering it higher accuracy, Lasso was selected for further analysis. The heatmap of the Lasso analysis results is shown in Supplementary Fig. S9. The zT contribution rates of various components were 90% for PC1 and 9% for PC2, respectively, accounting for nearly the entire variance. To improve the zT value, it is essential to reduce PC1 and PC2, which corresponds to weakening the 002 peak intensity of Cu2Se at ~ 13°, while strengthening the 211 and 311 peak intensities Fig. 1.  Variation of the thermoelectric (TE) properties of Cu2SexSiyFezOt films with the radiofrequency (RF) power and substrate temperature. Temperature dependences of electrical conductivity (a), Seebeck coefficient (b), and power factor (c). Scientific Reports |        (2025) 15:27278 3| https://doi.org/10.1038/s41598-025-12345-4www.nature.com/scientificreports/http://www.nature.com/scientificreportsobserved at ~ 25.5° and 220 peak intensity at ~ 44°. In addition, increasing the peak intensities of Cu1.82Se and Cu1.8Se is beneficial. This suggests that the sufficient growths of the Cu2Se, Cu1.82Se, and Cu1.8Se, tiny crystals are crucial for improving the TE performance of the material.The XRD spectrum of the CSSFO sample at 120 W and a 773 K substrate shows several phases, including Cu2Se, Cu1.82Se, Cu1.8Se, Cu2O, and CuO. However, the CSSFO sample prepared at 298 K mainly comprised Cu2Se crystals, while the CSSFO samples at 773 K were a mixture of Cu1.82Se, Cu1.8Se, and Cu2Se tiny crystals. Additionally, as the RF power increased, the peak intensities of the Cu1.82Se and Cu1.8Se crystals increased, whereas the Cu2Se peak decreased. No XRD peaks corresponding to Fe and Si were observed due to their metastable structures. Figure 6 shows the EDX elemental maps for the CSSFO sample at 120 W and 773 K. Si was uniformly distributed across the film, whereas Fe was detected in granular form within an amorphous iron oxide structure. Thus, the integration of a small quantity of FeSi2 during the sputter coating process led to the formation of various complex crystal structures within the CuxSe matrix. These structural differences improved the Seebeck coefficient by creating multiple surfaces, which led to a decrease in thermal conductivity. Furthermore, the oxidation across the entire film led to the formation of Cu2O and CuO on the surface, whereas FeO was evenly distributed within the film structure.First-principles calculations are an effective method for analyzing the thermoelectric properties; however, the calculations of nonstoichiometric Cu2 − xSe materials are difficult due to the existence of 3d electrons in Cu. Domashevskaya et al.40 investigated the band structure and electron density of Cu2 − xSe using X-ray photoelectron spectroscopy (XPS) and X-ray emission spectroscopy(XES). They showed that the magnitude of p-band splitting depended on the change in the value of x in Cu2 − xSe and that the value of the band gap varied with the x value. This result indicates that present materials are a complex mixture of nanocrystals with three band gaps. TE efficiency in SnS crystals arises due to an increased convergence of the energy and momentum of four valance bands, termed as quadruple-band synglisis13. Furthermore, TE properties can be improved by actively introducing vacancies and activating multiple band synthesis. Our samples are composite materials with different band gaps of Cu2Se, Cu1.82Se, and Cu1.8Se. While the interaction among these three band gaps may contribute to an enhancement in zT, isolating the underlying mechanisms proved difficult due to Fig. 2.  Figure of merit (zT) and thermal conductivity (κ) of Cu2SexSiyFezOt thin films as a function of RF power and substrate temperature at room temperature (298 K). Scientific Reports |        (2025) 15:27278 4| https://doi.org/10.1038/s41598-025-12345-4www.nature.com/scientificreports/http://www.nature.com/scientificreportsthe concurrent effects of interfacial phenomena, interlayer stress, and electronic interactions between different phases. Importantly, the coexistence of multiple thermoelectric tiny crystals and the introduction of numerous interfaces contributed to the observed improvement in zT. These results demonstrate the importance of FeSi2 doping in facilitating diverse crystal formations and optimizing TE properties through nanostructuring and interface engineering.MethodsCSSFO samples were synthesized using COSCOS, which is a homemade sputter coating system, as shown in Supplementary Fig. S1. The samples were prepared on floating potential quartz substrates measuring 22 × 4 × 1.5 Fig. 3.  Comparison of zT values at room temperature with previous studies and this work. Scientific Reports |        (2025) 15:27278 5| https://doi.org/10.1038/s41598-025-12345-4www.nature.com/scientificreports/http://www.nature.com/scientificreportsmm3, supplied by Hiraoka Special Glass Mfg. Co. (Japan). Sputter coating was conducted with ultra-high purity argon gas (99.999%) and a Cu2Se sputter target with a diameter and thickness of 50 and 4 mm, respectively. The operational argon gas pressure was maintained at 0.4 Pa. A 3-mm thick FeSi2 sputter-coated shutter was used for FeSi2 doping, with 5% of the coated area overlapping the target during the coating process to ensure precise doping. The introduction of FeSi2 into the plasma facilitated the co-sputtering of Fe and Si atoms, allowing trace quantities to be incorporated into the films. The overlapped area was kept constant but could be modified to control the doping levels. The distance between the target and substrate was fixed at 55 mm. Pre-sputtering was performed for 15 min to ensure target stability, and a quartz crystal thickness monitor along with cross-sectional scanning transmission electron microscopy (STEM) measurement (9.37–13.25 μm) were used to ensure uniform thickness. During sputtering, the RF power was varied between 60 and 120 W, and the substrate temperature was adjusted from 298 K to 773 K, respectively. The surface morphology and elemental composition were examined using SEM (JEOL JSM-7900 F, FEI Electron Optics Thermo Scientific Scios2 HiVac, and Hitachi High-Tech Fig. 4.  Scanning electron microscopy (SEM) and high-resolution transmission electron microscopy (HRTEM) images of Cu2Se0.87 Si0.01Fe0.01O0.16 thin films. SEM images of the Cu2Se0.87Si0.01Fe0.01O0.16 films with radiofrequency (RF) power of 120 W and substrate temperature of 773 K (a). The thin film exhibits numerous cracks and micron-sized crystal grains grown on its surface. (inset). A tetrahedron-like structure is observed when zooming in on the crack areas. Detailed observation of the tetrahedron-like structure using HRTEM revealed that it was a single crystal of Cu2O (b). Cross-sectional HRTEM image of the film (c). Multiple small crystals exist within the film in different orientations (d). The sample surface of panels (c) and (d) covered by a carbon protective film. Scientific Reports |        (2025) 15:27278 6| https://doi.org/10.1038/s41598-025-12345-4www.nature.com/scientificreports/http://www.nature.com/scientificreportsS-3700) and EDX (JEOL JED-2300 Analysis Station Plus), respectively. The Seebeck coefficient and electrical conductivity were measured using a Netzsch SBA 458 Nemesis® (Spain) system with the 4-point method and He gas for thermal insulation. Thermal conductivity (κ) was estimated using the formula κ = TD × Cp × d, where TD denotes the thermal diffusivity, Cp denotes the heat capacity, and d denotes the density. Thermal diffusivity was measured using ai-Phase type-1u at 298 K. The Cp and d values of 0.354 J/gK and 6.3 g/cm3, respectively, were taken from the literature41. STEM and electron diffraction were performed using an atomic resolution electron microscope (JEOL JEM-ARM300F GRAND ARM, Japan). The crystal structure of the samples was determined using XRD (Rigaku SmartLab, Japan). The XRD peaks were assigned based on the material database system (AtomWork, NIMS), and PCA analysis was performed by the WAVEBASE system (Toyota Motor Corporation, Japan).Fig. 5.  X-ray diffraction (XRD) spectra of Cu2Se0.87 Si0.01Fe0.01O0.16 thin film at 120 W and 773 K. Peaks corresponding to Cu2Se, Cu1.82Se, Cu1.8Se, Cu2O, and CuO crystals are observed. Peaks marked with a star are due to the substrate. Scientific Reports |        (2025) 15:27278 7| https://doi.org/10.1038/s41598-025-12345-4www.nature.com/scientificreports/http://www.nature.com/scientificreportsData availabilityThe authors declare that the data supporting the findings of this study are available within the paper, its supple-mentary information file.Received: 21 March 2025; Accepted: 16 July 2025References  1.  Snyder, G. J. & Toberer, E. S. Complex thermoelectric materials. Nat. Mater. 7, 105–114. https://doi.org/10.1038/nmat2090 (2008).  2.  Mao, J., Chen, G. & Ren, Z. F. Thermoelectric cooling materials. Nat. Mater. 20, 454–461. ​h​t​t​p​s​:​/​/​d​o​i​.​o​r​g​/​1​0​.​1​0​3​8​/​s​4​1​5​6​3​-​0​2​0​-​0​0​8​5​2​-​w​​​​ (2021).  3.  Raj, A. & Steingart, D. Review-power sources for the internet of things. J. Electrochem. Soc. 165, B3130–B3136. ​h​t​t​p​s​:​/​/​d​o​i​.​o​r​g​/​1​0​.​1​1​4​9​/​2​.​0​1​8​1​8​0​8​j​e​s​​​​ (2018).  4.  Wei, J. T. et al. Review of current high-ZT thermoelectric materials. J. Mater. Sci. 55, 12642–12704. ​h​t​t​p​s​:​/​/​d​o​i​.​o​r​g​/​1​0​.​1​0​0​7​/​s​1​0​8​5​3​-​0​2​0​-​0​4​9​4​9​-​0​​​​ (2020).  5.  Remesal, E. R. et al. Enhancing the thermoelectric figure of merit of BiN polymorphism, pressure, and nanostructuring. J. Mater. Chem. A. 13, 220–229. https://doi.org/10.1039/d4ta05891g (2024).  6.  Anbarasan, R., Kim, D. & Park, J. H. Ternary transition-metal nitride halide monolayers MNI (M = Zr, Hf) with low thermal conductivity and high thermoelectric figure of merit. Comp. Mater. Sci. 247, 113508. ​h​t​t​p​s​:​/​/​d​o​i​.​o​r​g​/​1​0​.​1​0​1​6​/​j​.​c​o​m​m​a​t​s​c​i​.​2​0​2​4​.​1​1​3​5​0​8​​​​ (2025).  7.  Balaji, A. S., Ramasamy, A., Sivasankar, K. J., Mohanraj, H. R. & Thiruvadigal, J. DFT + NEGF insights on boosting the thermoelectric figure of merit of 2D GeTe/arsenene vdW heterostructure device: Interface engineering. Surf. Interfaces 56, 105534. https://doi.org/10.1016/j.surfin.2024.105534  (2025).  8.  Grevtsov, N. et al. Thermoelectric materials based on cobalt-containing sintered silicon-germanium alloys. Mater. Res. Bull. 184, 113258. https://doi.org/10.1016/j.materresbull.2024.113258 (2025).  9.  Naderloo, R. H. et al. Performance advancements in P-type TaFeSb-based thermoelectric materials through composition and composite optimizations. Energy Environ. Sci. 18, 738–749. https://doi.org/10.1039/d4ee04819a (2025). 10.  Polash, M. M. H., Stone, M., Chi, S. X. & Vashaee, D. Designing spin-crossover systems to enhance thermopower and thermoelectric figure-of-merit in paramagnetic materials. Energy Environ. Mater. 8, e12822. https://doi.org/10.1002/eem2.12822 (2025). 11.  Zhu, Y. et al. Boosting room-temperature thermoelectricity in SrTiO-based superlattices. J. Mater. Chem. C. 13, 2279–2285. https://doi.org/10.1039/d4tc04200j (2025). 12.  Sasaki, M., Ju, S. H., Xu, Y. B., Shiomi, J. & Goto, M. Identifying optimal strain in bismuth telluride thermoelectric film by combinatorial gradient thermal annealing and machine learning. ACS Comb. Sci. 22, 782–790. ​h​t​t​p​s​:​/​/​d​o​i​.​o​r​g​/​1​0​.​1​0​2​1​/​a​c​s​c​o​m​b​s​c​i​.​0​c​0​0​1​1​2​​​​ (2020).Fig. 6.  Elemental maps of CSSFO films prepared at 120 W and 773 K using energy-dispersive X-ray spectroscopy. Scientific Reports |        (2025) 15:27278 8| https://doi.org/10.1038/s41598-025-12345-4www.nature.com/scientificreports/https://doi.org/10.1038/nmat2090https://doi.org/10.1038/s41563-020-00852-whttps://doi.org/10.1038/s41563-020-00852-whttps://doi.org/10.1149/2.0181808jeshttps://doi.org/10.1149/2.0181808jeshttps://doi.org/10.1007/s10853-020-04949-0https://doi.org/10.1007/s10853-020-04949-0https://doi.org/10.1039/d4ta05891ghttps://doi.org/10.1016/j.commatsci.2024.113508https://doi.org/10.1016/j.commatsci.2024.113508https://doi.org/10.1016/j.surfin.2024.105534https://doi.org/10.1016/j.materresbull.2024.113258https://doi.org/10.1039/d4ee04819ahttps://doi.org/10.1002/eem2.12822https://doi.org/10.1039/d4tc04200jhttps://doi.org/10.1021/acscombsci.0c00112https://doi.org/10.1021/acscombsci.0c00112http://www.nature.com/scientificreports 13.  Liu, S. et al. Quadruple-band synglisis enables high thermoelectric efficiency in earth-abundant Tin sulfide crystals. Science 387, 202–208. https://doi.org/10.1126/science.ado1133 (2025). 14.  Brown, D. R., Day, T., Caillat, T. & Snyder, G. J. Chemical stability of (Ag,Cu)2Se: A historical overview. J. Electron. Mater. 42, 2014–2019. https://doi.org/10.1007/s11664-013-2506-2 (2013). 15.  Liu, H. L. et al. Copper ion liquid-like thermoelectrics. Nat. Mater. 11, 422–425. https://doi.org/10.1038/Nmat3273 (2012). 16.  Yu, B. et al. Thermoelectric properties of copper Selenide with ordered selenium layer and disordered copper layer. Nano Energy. 1, 472–478. https://doi.org/10.1016/j.nanoen.2012.02.010 (2012). 17.  Perez-Taborda, J. A. et al. Pulsed hybrid reactive Magnetron sputtering for high Cu2Se thermoelectric films. Adv. Mater. Technol. -US. 2, 1700012. https://doi.org/10.1002/admt.201700012 (2017). 18.  Byeon, D. et al. Discovery of colossal Seebeck effect in metallic Cu2Se. Nat. Commun. 10, 72. ​h​t​t​p​s​:​/​/​d​o​i​.​o​r​g​/​1​0​.​1​0​3​8​/​s​4​1​4​6​7​-​0​1​8​-​0​7​8​7​7​-​5​​​​ (2019). 19.  Liu, S. et al. Enhanced thermoelectric properties of β-Cu2Se by incorporating CuGaSe. J. Alloy Compd. 688, 521–526. ​h​t​t​p​s​:​/​/​d​o​i​.​o​r​g​/​1​0​.​1​0​1​6​/​j​.​j​a​l​l​c​o​m​.​2​0​1​6​.​0​7​.​2​1​8​​​​ (2016). 20.  Peng, P. et al. Structure and thermoelectric performance of β-Cu2Se doped with fe, ni, Mn. Zn Or. Sm Intermetallics. 75, 72–78. https://doi.org/10.1016/j.intermet.2016.05.012 (2016). 21.  Lu, R. M., Bailey, T. P., Uher, C. & Poudeu, P. F. P. Ultrafine interwoven dendritic Cu2Se/CuFeSe composites with enhanced thermoelectric performance. ACS Appl. Energy Mater. 3, 9133–9142. https://doi.org/10.1021/acsaem.0c01525 (2020). 22.  Yang, L. et al. Te-doped Cu2Se nanoplates with a high average thermoelectric figure of merit. J. Mater. Chem. A. 4, 9213–9219. https://doi.org/10.1039/c6ta02998a (2016). 23.  Zhu, Y. B., Zhang, B. P. & Liu, Y. Enhancing thermoelectric performance of Cu2Se by doping Te. Phys. Chem. Chem. Phys. 19, 27664–27669. https://doi.org/10.1039/c7cp05149b (2017). 24.  Zhu, Z., Zhang, Y. W., Song, H. Z. & Li, X. J. Enhancement of thermoelectric performance of Cu2Se by K doping. Appl. Phys. A-Mater. 124 https://doi.org/10.1007/s00339-018-2299-5 (2018). 25.  Kim, M. J. et al. Effects of Cl-doping on thermoelectric transport properties of Cu2Se prepared by spark plasma sintering. J. Electron. Mater. 48, 1958–1964. https://doi.org/10.1007/s11664-018-6708-5 (2019). 26.  Kong, F. F. et al. Ultralow thermal conductivity and high thermoelectric performance of Cu2Se/TiO nanocomposite. Appl. Phys. Lett. 115, 203901. https://doi.org/10.1063/1.5126152 (2019). 27.  Lei, J. D. et al. High thermoelectric performance in Cu2Se superionic conductor with enhanced liquid-like behaviour by dispersing SiC. J. Mater. Chem. A. 7, 7006–7014. https://doi.org/10.1039/c8ta12210e (2019). 28.  Bohra, A. K. et al. Stabilizing thermoelectric figure-of-merit of superionic conductor Cu2Se through W nanoinclusions. Phys. Status Solidi-R. 14, 2000102. https://doi.org/10.1002/pssr.202000102 (2020). 29.  Jin, Y. et al. Size-controlled Au-Cu2Se core-shell nanoparticles and their thermoelectric properties. ACS Appl. Mater. Inter. 12, 36589–36599. https://doi.org/10.1021/acsami.0c08149 (2020). 30.  Chen, Y. X., Zhang, Y. Y., Uher, C. & Poudeu, P. F. P. Carrier mobility modulation in Cu2Se composites using coherent Cu2TiSe inclusions leads to enhanced thermoelectric performance. ACS Appl. Mater. Inter. 14, 56817–56826. ​h​t​t​p​s​:​/​/​d​o​i​.​o​r​g​/​1​0​.​1​0​2​1​/​a​c​s​a​m​i​.​2​c​1​7​1​4​6​​​​ (2022). 31.  Ang, A. K. R. et al. Development of Cu2Se/Ag(S,Se)-Based monolithic thermoelectric generators for low-grade waste heat energy harvesting. ACS Appl. Mater. Inter. 15, 46962–46970. https://doi.org/10.1021/acsami.3c09823 (2023). 32.  Zhou, Z. F. et al. Compositing effects for high thermoelectric performance of Cu2Se-based materials. Nat. Commun. 14, 2410. https://doi.org/10.1038/s41467-023-38054-y (2023). 33.  Mac, T. K. et al. Synthesis and thermoelectric characterization of Sb-doped Cu2Se by mechanical alloying and solid-state reaction. J. Electron. Mater. 53, 1026–1034. https://doi.org/10.1007/s11664-023-10808-w (2024). 34.  Wang, Q. et al. Enhanced thermoelectric performance of Cu2Se thin film derived from potential barrier scattering by incorporating SnSe nano-dispersions. Appl. Phys. Lett. 124, 192104. https://doi.org/10.1063/5.0201400 (2024). 35.  Lin, Z. H. et al. Impact of Boron nitride on the thermoelectric properties and service stability of Cu2Se. ACS Appl. Mater. Inter. 17, 1922–1930. https://doi.org/10.1021/acsami.4c16857 (2024). 36.  Chen, Y., Song, G. H., Ben, Z. H., Wu, Y. S. & You, J. H. Improved thermoelectric properties of the β-Cu2Se/CuInSe multilayer films by layer interface scattering. Curr. Appl. Phys. 70, 1–10. https://doi.org/10.1016/j.cap.2024.11.010 (2025). 37.  Benaadad, M., Nafidi, A., Melkoud, S., Khan, M. S. & Soubane, D. First-principles investigations of structural, optoelectronic and thermoelectric properties of Cu-based chalcogenides compounds. J. Mater. Sci. 56, 15882–15897. ​h​t​t​p​s​:​/​/​d​o​i​.​o​r​g​/​1​0​.​1​0​0​7​/​s​1​0​8​5​3​-​0​2​1​-​0​6​3​2​5​-​y​​​​ (2021). 38.  Goto, M., Kasahara, A. & Tosa, M. Low frictional property of copper oxide thin films optimised using a combinatorial sputter coating system. Appl. Surf. Sci. 252, 2482–2487. https://doi.org/10.1016/j.apsusc.2005.03.236 (2006). 39.  Goto, M. et al. Control of p-type and n-type thermoelectric properties of bismuth telluride thin films by combinatorial sputter coating technology. Appl. Surf. Sci. 407, 405–411. https://doi.org/10.1016/j.apsusc.2017.02.187 (2017). 40.  Domashevskaya, E. P. et al. XPS and XES emission investigations of d-p resonance in some copper chalcogenides. J. Electron. Spectrosc. 114, 901–908. https://doi.org/10.1016/S0368-2048(00)00406-0 (2001). 41.  Brown, D. R. et al. Relating phase transition heat capacity to thermal conductivity and effusivity in Cu2Se. Phys. Status Solidi-R. 10, 618–621. https://doi.org/10.1002/pssr.201600160 (2016).AcknowledgementsThis research was supported by Japan Science and Technology Agency (JST) CREST under the projects “Ex-ploring Innovative Materials in Unknown Search Space” (JPMJCR21O2), CREST and “Scientific Innovation for Energy Harvesting Technology” (JPMJCR16Q5). A part of this work was also supported by Grant-in-Aid for Scientific Research (C) 24K07349 and 25K08337 from the Japan Society for the Promotion of Science (JSPS). A part of this work was supported by the Electron Microscopy Unit, National Institute for Materials Science (NIMS). We would like to thank Toyota Motor Corporation for allowing us to use the material analysis cloud service “WAVEBASE”.Author contributionsM.G. planned and supervised the study. M.S. and M.G. prepared the samples, characterized them using XRD and X-ray photoelectron spectroscopy, and measured their thermoelectric properties. M.G. and Y.X. performed thermal conductivity measurements. SEM, STEM, and EDX measurements were performed by T.M. and T.H. All authors discussed the results, developed explanations for experiments, and provided feedback on the manu-script. M.G. and M.S. wrote and edited the manuscript.Scientific Reports |        (2025) 15:27278 9| https://doi.org/10.1038/s41598-025-12345-4www.nature.com/scientificreports/https://doi.org/10.1126/science.ado1133https://doi.org/10.1007/s11664-013-2506-2https://doi.org/10.1038/Nmat3273https://doi.org/10.1016/j.nanoen.2012.02.010https://doi.org/10.1002/admt.201700012https://doi.org/10.1038/s41467-018-07877-5https://doi.org/10.1038/s41467-018-07877-5https://doi.org/10.1016/j.jallcom.2016.07.218https://doi.org/10.1016/j.jallcom.2016.07.218https://doi.org/10.1016/j.intermet.2016.05.012https://doi.org/10.1021/acsaem.0c01525https://doi.org/10.1039/c6ta02998ahttps://doi.org/10.1039/c7cp05149bhttps://doi.org/10.1007/s00339-018-2299-5https://doi.org/10.1007/s11664-018-6708-5https://doi.org/10.1063/1.5126152https://doi.org/10.1039/c8ta12210ehttps://doi.org/10.1002/pssr.202000102https://doi.org/10.1021/acsami.0c08149https://doi.org/10.1021/acsami.2c17146https://doi.org/10.1021/acsami.2c17146https://doi.org/10.1021/acsami.3c09823https://doi.org/10.1038/s41467-023-38054-yhttps://doi.org/10.1007/s11664-023-10808-whttps://doi.org/10.1063/5.0201400https://doi.org/10.1021/acsami.4c16857https://doi.org/10.1016/j.cap.2024.11.010https://doi.org/10.1007/s10853-021-06325-yhttps://doi.org/10.1007/s10853-021-06325-yhttps://doi.org/10.1016/j.apsusc.2005.03.236https://doi.org/10.1016/j.apsusc.2017.02.187https://doi.org/10.1016/S0368-2048(00)00406-0https://doi.org/10.1002/pssr.201600160http://www.nature.com/scientificreportsDeclarationsCompeting interestsThe authors declare no competing interests.Additional informationSupplementary Information The online version contains supplementary material available at ​h​t​t​p​s​:​/​/​d​o​i​.​o​r​g​/​1​0​.​1​0​3​8​/​s​4​1​5​9​8​-​0​2​5​-​1​2​3​4​5​-​4​​​​​.​​Correspondence and requests for materials should be addressed to M.G. or M.S.Reprints and permissions information is available at www.nature.com/reprints.Publisher’s note  Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Open Access   This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit ​h​t​t​p​:​/​/​c​r​e​a​t​i​v​e​c​o​m​m​o​n​s​.​o​r​g​/​l​i​c​e​n​s​e​s​/​b​y​-​n​c​-​n​d​/​4​.​0​/​​​​​.​​© The Author(s) 2025 Scientific Reports |        (2025) 15:27278 10| https://doi.org/10.1038/s41598-025-12345-4www.nature.com/scientificreports/https://doi.org/10.1038/s41598-025-12345-4https://doi.org/10.1038/s41598-025-12345-4http://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://www.nature.com/scientificreports Room-temperature FeSi2-doped Cu2Se thermoelectric films with enhanced figure of merit Results and discussion Methods References