# Fileset

[AppliedSurfaceScienceAdvances-Enhancing thermoelectric properties of ScN films through twin domains.pdf](https://mdr.nims.go.jp/filesets/5d301f46-2886-419b-b2ae-3b0b5e73be6e/download)

## Creator

J. More-Chevalier, U.D. Wdowik, J. Martan, T. Baba, S. Cichoň, P. Levinský, D. Legut, E. de Prado, P. Hruška, J. Pokorný, J. Bulíř, C. Beltrami, [T. Mori](https://orcid.org/0000-0003-2682-1846), M. Novotný, I. Gregora, L. Fekete, L. Volfová, J. Lančok

## Rights

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

## Other metadata

[Enhancing thermoelectric properties of ScN films through twin domains](https://mdr.nims.go.jp/datasets/7c74d19e-7e99-4374-ba3e-8c7ae6a1fb6f)

## Fulltext

Enhancing thermoelectric properties of ScN films through twin domainsApplied Surface Science Advances 25 (2025) 100674 A2Contents lists available at ScienceDirectApplied Surface Science Advancesjournal homepage: www.sciencedirect.com/journal/applied-surface-science-advancesFull length articleEnhancing thermoelectric properties of ScN films through twin domainsJ. More-Chevalier a ,∗, U.D. Wdowik b ,∗, J. Martan c , T. Baba d,e, S. Cichoň a , P. Levinský a ,D. Legut b,f , E. de Prado a , P. Hruška a,f , J. Pokorný a , J. Bulíř a , C. Beltrami c,T. Mori d,e , M. Novotný a , I. Gregora a , L. Fekete a , L. Volfová a , J. Lančok aa Institute of Physics of the Czech Academy of Sciences, Na Slovance 2, 18221 Praha 8, Czech Republicb IT4Innovations, VSB - Technical University of Ostrava, 17. listopadu 2172/15, CZ 708 00 Ostrava-Poruba, Czech Republicc New Technologies Research Centre (NTC), University of West Bohemia, Univerzitni 8, 301 00 Plzen, Czech Republicd International Center for Materials Nanoarchitectonics (WPI-MANA), National Institute of Materials Science (NIMS), Namiki 1-1,Tsukuba, Ibaraki 305-0044, Japane Graduate School of Pure and Applied Science, University of Tsukuba, Tennodai 1-1, Tsukuba, Ibaraki 305-8671, Japanf Department of Condensed Matter Physics, Faculty of Mathematics and Physics, Charles University, Ke Karlovu 3, 121 16 Prague 2, Czech RepublicA R T I C L E I N F ODataset link: https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/Keywords:ThermoelectricityScandium nitrideThin filmsSeebeck coefficientA B S T R A C TTailoring thermoelectric properties of ScN-based materials is of vital importance for their application,particularly at high operating temperatures. Here, we report on the thermoelectric properties of the ScNlayers deposited on MgO (001) substrates by the DC reactive magnetron sputtering. The microstructure ofthe produced thin films is examined by X-ray diffraction and atomic force microscopy, while their chemicalcomposition and contamination by defects are determined by X-ray photoelectron spectroscopy. The effect oftemperature on the phonon properties of ScN layers, having implications for their thermoelectric properties, isexplored by Raman spectroscopy. The results of our experiments are confronted with those following from thefirst-principles studies. We find that the ScN/MgO(001) layers with twin-domain structure reveal enhancedthermoelectric properties at elevated temperature as compared to those measured for almost defect- anddomain-free layers, namely, enlarged Seebeck coefficient by about 30% and over two and a half times increasedfigure of merit at 800 K. Therefore, structural twin domains in thin ScN film appear to be a simple and ratherstable solution for the improvement of its thermoelectric properties at elevated temperatures.1. IntroductionIn the last decades, the search for sustainable clean energy alongwith the miniaturization of sensors and electronic circuits has stim-ulated intense research on thermoelectricity and thermoelectric de-vices [1–9]. Such devices are usually employed to convert heat intoelectrical energy as well as for energy harvesting from the environmentand/or harvesting of industrial waste heat [10–12]. Recently, thinfilms of the transition-metal nitrides (ScN, CrN, ZrN, HfN) [13] aswell as metal/semiconductor multilayers and superlattices based onepitaxial ZrN/ScN and HfN/ScN [14–18] have drawn much experi-mental and theoretical interest as they hold considerable promise forthermoelectric applications.Among transition-metal nitrides with potential application in ther-moelectricity, ScN has recently gained considerable attention for itssuperior refractory properties enabling its operation at high tempera-tures. Scandium nitride is a rocksalt-structured, narrow-bandgap n-typesemiconductor with an indirect 𝛤 − 𝑋 energy gap of ∼0.9 eV and a∗ Corresponding authors.E-mail addresses: morechevalier@fzu.cz (J. More-Chevalier), urszula.danuta.wdowik@vsb.cz (U.D. Wdowik).direct 𝛤 −𝛤 gap of about 2.2 eV. ScN, belonging to group III (B) nitridesemiconductors, can overcome some limitations of group III (A) nitridesemiconductors for a variety of applications [19]. This compound isextremely sensitive to defects, especially oxygen impurities [19–23]that can be present in a large amount if ScN is not synthesized in pureultra-high vacuum conditions. A high concentration of charge carriers(1018 − 1022 cm−3) and low electrical resistivity (∼300 μΩcm) of ScNlead to its high thermoelectric power factor of 2.5–3.3 W m−1 K−1 [13,19,24–26]. On the other hand, its high thermal conductivity rangingbetween 10 and 12 W m−1 K−1 at room temperature in thick (∼450 nm)films [27], results in its limited figure of merit (0.2–0.3) [24,26,28].Such a high thermal conductivity requires reduction to enable theapplication of ScN as a thermoelectric material. Thus, ScN has beena subject of doping by various elements, e.g., Cr, Nb, Mg or Li [29–32]. This strategy was successful only to some extent, as yieldingsmaller thermal conductivity of doped ScN films (∼2.2 W m−1 K−1 forSc1−𝑥Nb𝑥N with 𝑥 = 0.1), led also to deterioration of its electricalhttps://doi.org/10.1016/j.apsadv.2024.100674Received 18 October 2024; Received in revised form 18 November 2024; Acceptedvailable online 24 December 2024 666-5239/© 2024 The Author(s). Published by Elsevier B.V. This is an open access a 2 December 2024rticle under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ). https://www.sciencedirect.com/journal/applied-surface-science-advanceshttps://www.sciencedirect.com/journal/applied-surface-science-advanceshttps://orcid.org/0000-0001-5867-7020https://orcid.org/0000-0001-9443-3621https://orcid.org/0000-0002-5832-4425https://orcid.org/0000-0002-3652-1583https://orcid.org/0000-0002-8501-1256https://orcid.org/0000-0001-9185-9934https://orcid.org/0000-0003-1133-3161https://orcid.org/0000-0002-6471-6804https://orcid.org/0000-0002-2614-1667https://orcid.org/0000-0002-1302-4365https://orcid.org/0000-0003-2682-1846https://orcid.org/0000-0002-9994-1024https://orcid.org/0000-0003-1616-4125https://orcid.org/0000-0002-8548-2613https://orcid.org/0000-0002-7178-4872https://orcid.org/0000-0002-4897-140Xhttps://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/mailto:morechevalier@fzu.czmailto:urszula.danuta.wdowik@vsb.czhttps://doi.org/10.1016/j.apsadv.2024.100674https://doi.org/10.1016/j.apsadv.2024.100674http://crossmark.crossref.org/dialog/?doi=10.1016/j.apsadv.2024.100674&domain=pdfhttp://creativecommons.org/licenses/by/4.0/J. More-Chevalier et al.f−ccfpttSmtndbtotlustc0(AopswetdtwTScwbtlolar(tarp(bodrctrP(tsJt(Applied Surface Science Advances 25 (2025) 100674 resistivity and Seebeck coefficient. Recently, the Ar-implanted ScNilms have been shown to exhibit improved Seebeck coefficient (−30 to85 μV K−1 at 600 K) due to introduced deep acceptor levels reducingoncentration of free carriers [33] as well as strongly lowered thermalonductivity (3 W m−1 K−1 at 300 K vs 12 W m−1 K−1 in defect-ree ScN) because of a higher level of scattering by defects, i.e., lowerhonon mean free path. Several experimental studies performed duringhe last years on diversely modified ScN showed a large variation in itshermoelectric properties. A better understanding of the properties ofcN in the form of film for its potential application as a thermoelectricaterial calls for additional investigations that should be carried out forhe almost defect-free ScN films, i.e., containing neither added dopantsor implanted elements. Therefore, in the present work the ScN layerseposited on MgO(001) substrates with a thickness of about 1 μm haveeen prepared. We managed to obtain two kinds of ScN layers, namely,he layer without defects, which is here denoted as ScN, and the secondne revealing a twin-domain structure denoted as ScN-T. We examinehe microstructure, vibrational and thermoelectric properties of theseayers not only at room temperature but also at elevated temperaturesp to 800 K. Our experimental research is supported by ab initioimulations based on the density functional theory (DFT) employedo explore the thermoelectric properties of an ideal ScN crystal andhanges in its phonon dynamics as a function of temperature via abinitio molecular dynamics (AIMD).2. Methodology2.1. ExperimentalThe ScN layers were deposited on double-side polished (10 × 10 ×.5 mm3) MgO (001) substrates in an ultra-high-vacuum (UHV) system∼10−8 Pa) by DC reactive magnetron sputtering using a mixture ofr (99.9999% pure) and N2 (99.9999% pure) discharge with a ratiof 60/40, respectively. The MgO substrates (001 oriented, 99.95%urity, from Alineason Materials and Technology) were cleaned usingeveral steps, first with acetone in an ultrasonic bath, then repeatedith ethanol and blown dry with an N2-gun. The sample holder waslectrically heated first to be degassed at 573 K for 30 min and theno reach the deposition temperature of 1123 K. Before starting theeposition process, the substrate was kept for 1 h at the depositionemperature for the surface reconstruction process. The temperaturesere measured by two thermocouples attached to the sample holder.he working pressure of 2 Pa was used to sputter ScN from threec targets (NEYCO: Sc 99.9%) of one-inch diameter with measuredontamination of Ta at 0.1%. A sputtering power density of 2.8 W cm−2as used to deposit films with thicknesses of 800 ± 50 nm. The distanceetween the targets and the sample holder was fixed at 5 cm. Thehicknesses were measured by a profilometer. Uninterrupted depositioned to the formation of ScN layers. Conversely, periodic interruption ofne cathode every hour during the deposition process produced ScN-Tayers.The X-ray diffraction measurements were performed on three differ-ent diffractometers. We measured high-resolution symmetric reciprocalspace maps (RSMs) on a Bruker D8 DISCOVER diffractometer with four-crystal Bartels monochromator and Cu K𝛼1 radiation, high-esolution 2D-RSMs on a SmartLab SE Multipurpose diffractometerRigaku Corp., Tokyo, Japan) using Cu𝛼1 radiation and a 2-bounce Ge(220) monochromator while the 𝜃 − 2𝜃 scans and pole figures weremeasured on a PANalytical X’Pert PRO powder diffractometer withoutany monochromator.The composition of the films was investigated using X-ray photo-electron spectroscopy (XPS) NanoESCA Omicron instrument. A monochromatic Al anode K𝛼 radiation was used as an X-ray source. Theanalyzed spot size was 100 × 300 μm2 and several positions on thesample were probed. The depth profiling process was described inour previous works [34,35]. Spectra are presented as measured, which2 means that no charge correction was performed. Under these condi-ions, the binding energy of the 𝑁 1s peak measured on the surfacemounts to 396.0 eV and remains in close agreement with the valueeported in other studies [35].The Raman spectra were recorded in a backscattering and parallel-olarization (VV) configuration using a Renishaw Raman spectrometerSystem 1000) equipped with Bragg filters. The spectrometer was cali-rated using the silicon F1 g peak at 520.2 cm−1. An Ar-laser with anexcitation wavelength of 514 nm, power ∼10 mW, and spot size of 5 μm(× 50, NA = 0.55, Olympus microscope objective) was used. The spectrafrom ScN and ScN-T were scanned at 10 s per frame, × 5 accumulations,in a temperature chamber Linkam HFS600E-PB4 stage at temperaturesranging from 300 to 800 K. The measured phonon modes covered therange of frequencies of 15–1800 cm−1.An atomic force microscope (AFM Dimension ICON, Bruker) wasused to investigate the surface morphology and roughness. The mea-surements were performed under ambient conditions and images wereobtained in the Peak Force Tapping mode using ScanAsystAir tips withscan areas of 4.4 × 4.4 μm2.Electrical resistivity and Seebeck coefficient were measured simulta-neously using a custom-made instrument which employs the four-probemethod. Platinum electrodes were used as current probes and Pt-PtRh (type S) thermocouples served both as voltage probes and tomeasure the sample temperature and its gradient. Measurements wereperformed in a nitrogen atmosphere on samples typically 10 mm longand 3 mm wide.We used two techniques to measure thermal effusivity. In the firstne, thermal effusivity in various depths under the sample surface isetermined by a measurement system based on pulsed photothermaladiometry described in detail elsewhere [36,37]. This technique usesa pulsed laser to heat the material surface and a fast infrared detectorto observe temperature decrease after the laser pulse impact. Fromthe temperature decrease with time in the nanosecond time range,the thermal effusivity in depth of hundreds of nanometers is evalu-ated. The three-layer 1D model with surface absorption of laser lightwithout thermal interface resistances between layers is employed, asdescribed in Ref. [38]. Semitransparency correction is applied accord-ing to Ref. [37]. In the second technique, thermal conductivities in theross-plane direction of the ScN films are measured with a picosecondhermo-reflectance apparatus (PicoTR, NETZSCH-Gerätebau GmbH) atoom temperature [39,40]. Before the measurement, a 100 nm-thickt layer is deposited on the films by sputtering. Front-heat front-detectFF) configuration is applied to the samples and thermal effusivities ofhe films are determined. To calculate the thermal conductivity, thepecific heat capacity and density of the films were assumed to be 846 kg−1K−1 and 4400 kg m−3, respectively.2.2. TheoreticalCalculations were performed within the framework of density func-ional perturbation theory (DFPT) implemented in the quantum espressoQE) package [41,42]. The Perdew–Burke–Ernzerhof (PBE) generalizedgradient approximation (GGA) exchange–correlation functional [43]and the projector augmented wave (PAW) pseudopotentials for thedescription of the electron–ion interaction were employed. The PAWpseudopotential with 3 and 5 electrons in valence for Sc and 𝑁 atoms,respectively, were adopted from the QE database. A kinetic energy cut-off of 51 Ry was used for the plane-wave basis set. The DFPT was usedto determine band structures, phonon dispersions, and the electron–phonon (e-ph) coupling matrix elements on a coarse k and q-pointgrids of 12 × 12 × 12 and 4 × 4 × 4, respectively. Subsequently,the quantities required to evaluate the e-ph self-energy ∑𝑒−ph𝑛,𝐤 wereinterpolated on significantly finer 483-k and 163-q grids via maximallylocalized Wannier functions formalism implemented in the electron–phonon Wannier (EPW) package [44–46]. Assuming that the majorscattering mechanism is due to the (e-ph) interaction, the scatteringJ. More-Chevalier et al.smil𝑘ecTCotd1dfp2TRwstdotghPibespa(sic1sscccSpi2iXSmtgApplied Surface Science Advances 25 (2025) 100674 rates can be expressed in terms of an effective transport phonon fre-quency distribution 𝛼2t r𝐤(𝜔)𝐹 (𝜔) [47], which was calculated using theEPW code. The e-e scattering has been neglected as being usuallyignificantly smaller in comparison with that arising from the e-phechanism [48,49]. Hence, in the present consideration the calculatedresistivity 𝜌(𝑇 ), which is determined in the scattering time approx-mation is related solely to the scattering of electrons by phonons.Also, the computed electronic contribution to thermal conductivity 𝑘𝑒 isapproximated by 𝑘𝑒−ph and determined via the Wiedemann–Franz lawfrom the electrical conductivity 𝜎(𝑇 ), which is an inverse of 𝜌(𝑇 ). Theattice thermal conductivity 𝑘𝐿 contains contributions from acoustic𝐿𝑎 and optical 𝑘𝐿𝑜 phonons, i.e., 𝑘𝐿 = 𝑘𝐿𝑎 + 𝑘𝐿𝑜. The 𝑘𝐿𝑎 wasstimated using the semi-empirical Slack model [50,51], while the 𝑘𝐿𝑜was evaluated according to the approach proposed by Cahill [52,53].Detailed formulas are given in the Appendix A. To calculate the Seebeckcoefficient the Boltztrap2 code has been applied [54].To study the temperature evolution of the ScN Raman spectra, theAIMD simulations within the isothermal–isobaric (NpT) and canonical(NVT) ensembles were performed using the Vienna Ab Initio Simula-tion Package (VASP) [55,56]. Here, also the PAW and PBE exchange–orrelation potential were adopted to describe electron–ion interaction.he energy cutoff of 520 eV for the plane wave expansion was applied.alculations were carried out at the 𝛤 -point for the system consistingf 216 atoms in a cubic supercell. The system was heated from 300o 800 K with a temperature step of 100 K in the NpT ensembleuring 20 ps and subsequently equilibrated at each temperature for0 ps. The Langevin thermostat was used to control the temperatureuring the NpT simulations. The shape constraints for the volumeluctuations experienced under Parinello–Rahman dynamics were im-osed. Production runs were carried out within the NVT scheme for0 ps. The temperature was controlled by the Nosé–Hoover thermostat.o compare the results of our AIMD simulations with the measuredaman spectra, the power spectra of the autocorrelation function 𝐺𝐤(𝜔)ere projected onto the 𝛤 -point of the cubic 𝐹 𝑚3̄𝑚 structure of ScN.Details of the formalism and applied procedure are provided and widelydiscussed in the pertinent literature [57,58].3. Results and discussion3.1. X-ray diffractionSymmetric (black curve) and skew-symmetric (red and blue curves)𝜃 − 2𝜃 scans of the ScN-T films deposited at 1123 K on MgO (001)ubstrates are shown in Fig. 1. Peaks marked with (*) and (**) cor-respond to the Cu K𝛽 reflections from the MgO substrate and ScN-Tlayer, respectively. The symmetric and skew-symmetric 𝜃 − 2𝜃 scans ofhe ScN films are presented in Fig. S1 (supplementary information). Noifference between diffraction patterns measured for both samples isbserved. The ScN films show a unique reflection at 𝜒 = 0◦, being dueo the 002 orientation which corresponds to the 002 grains epitaxiallyrown cube-on-cube with respect to the MgO substrates. Similar resultsave been reported for the ScN films deposited on MgO by differentVD techniques [19,21,24,59–61].Fig. 2 shows the pole figures of the 002, 022, and 111 orientationsof the ScN-T and ScN layers. The center of the 002 pole figure displayedin Fig. 2(a) corresponds to the main 002 orientation peak in ScN-T. Liken our previous work dealing with the ScN films [23], the eight spotsat 𝜒 ≈ 48◦ and four at 𝜒 ≈ 70◦ come from four twin domains grown onthe 111 twinning planes of the 002-oriented grains. Each twin domainproduces three poles (the poles generated by one of those domains haveeen marked with red circles). Beyond the four intense spots, which arexpected for the epitaxial films due to the 001 orientation in a cubictructure, the four extra poles arising from the twins are observed in theole figures of the 022 and 111 orientations that are shown in Fig. 2(b)nd Fig. 2(c), respectively. s3 The reciprocal space maps (RSM) and the 2D reciprocal space map2D-RSM) of the ScN-T films are depicted in Fig. 3. For comparison, theRSM and 2D-RSM of the ScN layer are given in Fig. S2 (Supplementaryinformation). The FWHM values of the 002 peak are equal to 𝐾2𝜃(002)= 0.364◦ and 𝐾𝜔(002) = 0.596◦. The 2D-RSM shows the different peakpositions as well as the peaks from twins in the 111 planes.3.2. Atomic force microscopyThe images of the ScN-T film surface obtained from the atomic forcemicroscopy confirm the film crystallization that we have observed usingX-ray diffraction. Additionally, they reveal two kinds of morphologiesmixed in the entire ScN-T surface, as shown in Fig. 4. The first surface,which is shown in Fig. 4(a) is composed of grains with square shapes,while the second one, depicted in Fig. 4(b), consists of grains with 3D-pyramidal shapes. In comparison to the ScN-T film surface, the almostdefect-free ScN film surface presents only square shape morphologiesequivalent to the one shown in Fig. 4(a).The flat surface made of square-shaped grains corresponds to theurface for the ScN films oriented 001, whereas the surface presentingsosceles-pyramidal-shape grains reflects 3 surfaces from the family 001oming from the disoriented crystals through the twinning effect in the11 planes. The grains with an isosceles-pyramidal shape are rotated by90◦ relative to each other, confirming the results from XRD pole figuremeasurements visible in Fig. 2(a). The measured values of root meanquare (RMS) surface roughness amount to 1.4 nm and 93.1 nm for theurface with square shapes and the surface exhibiting 3D-pyramidal-shaped grains, respectively. For comparison, the AFM image of the ScNfilms is given in supplementary information, Fig. S3. We also note thatthe formation of a mound structure is a characteristic and very commonfeature of the ScN as well as other nitride films [21,25,30,59,62,63].It is connected with the effect of adatom mobility on a film surfaceascribed to the presence of defects, which limits the down-step motionof atoms due to the Ehrlich–Schwoebel barrier and favors the uphillmigration on terraces [30,64–66]. In the present work, intentionalinterruptions of one of the cathodes during the ScN-T layer depositionmodify the energy landscape for adatoms. This disruption preventsthe complete overcoming of the Ehrlich–Schwoebel barrier, leadingto the formation of twin domains. As evidenced by X-ray diffraction,the absence of equilateral 3D pyramidal grains characteristic of 111-oriented grains on the mound structures of ScN-T layers confirms thelack of such grains within the layer.3.3. XPS spectraMeasured XPS spectra of ScN films are displayed in Fig. 5. Thepresence of all expected elements, O, Sc, N and C, on the surface isonfirmed. Based on high-resolution spectra, it is clear that the surfaceomposition corresponds to Sc nitride, Sc oxide and airborne carbonontamination. After the Ar+ depth profiling sputtering, a highly purecN material is observed. Carbon contamination disappears and theroportion of oxides is significantly lower; the oxygen concentrationn the bulk of the material is below 2.0% atomic. Further, there is a Scp peak broadening and Ar ion implantation as a result of sputtering-nduced damage. In summary, in agreement with our previous studies,PS results confirm surface composition typical for an air-exposedcN and a high quality of produced ScN films [35]. As regards XPSeasurements, there are no apparent differences between the ScN andhe ScN-T films.3.4. Raman spectroscopyFig. 6 shows the Raman spectra of ScN and ScN-T measured atthe VV polarization and temperatures ranging from 298 to 800 K. Ineneral, the spectra are very similar to those determined in our previoustudy [23] as well as consistent with the reported Raman spectra ofJ. More-Chevalier et al.Fig. 1. Symmetric (black curve) and skew-symmetric (red and blue curves) scans in ScN-T films deposited on MgO (001) substrates. Peaks are labeled by their phase and (hkl)indices. Peaks marked with (*) and (**) correspond to the Cu K𝛽 reflections from the MgO substrate and ScN-T layer, respectively. (For interpretation of the references to colorin this figure legend, the reader is referred to the web version of this article.)Fig. 2. Pole figures of the 002, 022, and 111 orientations of the ScN-T (a, b, and c) and ScN (d, e, and f) films deposited at 1123 K. The red circles in (a) show the peak positionsof one of the twins observed in the figure.Fig. 3. (a) The 002 reciprocal space map (RSM) and (b) the 2D reciprocal space map (2D-RMS) of ScN-T films. The peaks in the 2D-RMS map denoted in red correspond to thepeaks from the twins in the 111 plane.Applied Surface Science Advances 25 (2025) 100674 4 J. More-Chevalier et al. Applied Surface Science Advances 25 (2025) 100674 Fig. 4. The AFM images from the ScN-T surfaces. (a) Surface formed by square shapes, (b) surface formed by triangle 3D-pyramidal shapes. The arrows denote rotated grains.The RMS values of 1.4 nm and 93.1 nm are measured for surfaces shown in (a) and (b), respectively.Fig. 5. XPS spectra of ScN film. (a) Survey. Low-binding energy peak labeling is omitted for clarity. (b) O 1s. Surface spectrum consists of two peak components. Bulk spectrumis composed of a single low-intensity peak component. (c) Sc 2p and 𝑁 1s. Surface Sc 2p is typical with a doublet attributed to Sc–N bonds and a second doublet attributed toSc–O bonds. Surface 𝑁 1s consists of a single peak component attributed to N–Sc bonds. Bulk Sc 2p is characteristic of a large FWHM due to sputtering-induced damage. Bulk 𝑁1s consists of a single peak component attributed to N–Sc bonds. (d) C 1s. Several peak components form surface C 1s. Bulk spectrum does not show the presence of carbon.epitaxial ScN layers [67–69]. The main Raman spectrum measuredat each temperature covers the energy range extending from 300 to800 cm−1, while an additional shallow peak observed above 800 cm−1represents the second-order Raman effect.The Raman spectra of both ScN and ScN-T consist of two, weaklyseparated broad bands extending between 300 and 550 cm−1 and from550 to 800 cm−1. The lower band is mainly composed of the transverseoptical (TO) phonons, whereas the longitudinal optical (LO) modesgather in the upper band. In fact, defect-free ScN should exhibit noRaman-active modes due to symmetry reasons as each atom occupies asite of inversion symmetry. On the other hand, any perturbation andlowering of a local symmetry of the system, which results from thepresence of point or extended defects, can lead to the appearance of5 additional phonon modes, including those Raman or infrared-active.Indeed, the experimental spectra of vacancy-containing, Al- and Ti- orMg-doped ScN with low content of defects confirm the existence ofthe Raman modes in this system [23,67–69]. Additional confirmationcomes from our AIMD simulations, as illustrated in Fig. 7, whichprovides temperature behavior of the autocorrelation function powerspectra projected onto the Brillouin zone center of the 𝐹 𝑚3̄𝑚 symmetryScN crystal.Although the calculated spectra are obtained for defect-free ScN,the local symmetry of our cubic ScN supercell is no longer of 𝐹 𝑚3̄𝑚since the simulations were performed without symmetry constraintsimposed on the atomic positions. Therefore, one observes quite broadlower and upper bands, alike in the measured Raman spectra of ScNJ. More-Chevalier et al. Applied Surface Science Advances 25 (2025) 100674 Fig. 6. Temperature evolution of the measured Raman spectra for (a) ScN and (b) ScN-T films deposited on MgO (001) substrate. The spectra are measured at the VV scatteringgeometry. Baseline corrections are applied. The vertical lines denote the positions of the main fitted peaks for the room temperature spectra. ScN: (1) 404.6 cm−1, (2) 479.9 cm−1,(3) 641.8 cm−1, and (4) 678.8 cm−1. ScN-T: (1) 394.4 cm−1, (2) 462.8 cm−1, (3) 639.3 cm−1, and (4) 676.1 cm−1.Fig. 7. Calculated wave vector projected power spectra of the autocorrelation function𝐺𝐤(𝜔) onto the 𝛤 -point of the 𝐹 𝑚3̄𝑚 structure of ScN at temperatures ranging from 300to 800 K. Positions of peaks at 300 K: 𝜔1 = 448.4, 𝜔2 = 457.6, 𝜔3 = 619.8, 𝜔4 = 652.1cm−1.and ScN-T. These broad bands indicate substantial distribution of theRaman phonons. Actually, both lower and upper spectral bands featurea double-peak structure, which is clearly visible below 500 K in thesimulated spectra and it smears out at higher temperatures. The double-peak structure remains, however, less pronounced in the experimentalspectra. Nevertheless, the measured spectra of ScN and ScN-T couldbe fit with four Gaussian profiles, as shown in Fig. S4 and Fig. S5(Supplementary information). For each temperature, the positions ofparticular peaks resulting from fits are shown in Fig. 8 as well as theyare collected in Table S1 and Table S2 (Supplementary information).The peaks from the lower band in ScN, denoted as (1) and (2),that arise mostly from the TO phonons, shift their positions downward,i.e., to lower frequencies with increasing temperature T, whereas aslight upward shift is observed for the higher frequency peak (4)belonging to the upper band. Despite some deviation in the vicinity of500 K, the peak (3) from the upper band practically does not change itsposition (within the experimental error) with increased T. On the otherhand, the behavior of peaks’ positions as a function of temperature inScN-T remains not so clear as in ScN. Only peak (4) tends to increaseits position upward, but there is no general trend in the temperaturebehavior of the remaining peaks, except for a small jump in frequencyat 500 K. The observed fluctuations in positions of peaks in ScN-T mayresult from local microstresses generated by twin domains, while thedifference in the respective peaks’ shifts between ScN and ScN-T is6 likely due to much higher concentration of structural defects in ScN-T, which are responsible for the formation of twin domains, as alreadydiscussed in the previous sections.The AIMD simulations also produce shifting of peaks with heatingthe ScN system, as illustrated in Fig. 9. We observe an almost lineardecrease in frequencies of the 𝜔1, 𝜔2, 𝜔3, and 𝜔4 peaks, albeit withdifferent rates.The origin of a shift in peak position with temperature, i.e., thechange of a phonon frequency, can be manifold. Mostly these arethe anharmonic interactions that affect the lattice vibrations. Theyare responsible not only for the modification of a phonon frequencybut also for influencing its lifetime and intensity. Here we limit ourconsideration to a qualitative description of the effect of tempera-ture on the changes in frequencies of the Raman-active phonons. Inprinciple, the frequency shift of the normal modes with temperatureat constant pressure arises from thermal expansion of the lattice 𝛥𝐸and pure temperature contribution (phonon–phonon coupling) 𝛥𝐴, thelatter being due to cubic 𝛥(3) and quartic 𝛥(4) anharmonicities (three-and four-phonon couplings). Therefore, the temperature dependence offrequency can be written as [70–72]: 𝜔(𝑇 ) = 𝜔0 + 𝛥𝐸 + 𝛥𝐴, where 𝜔0is the harmonic frequency. In most solids 𝛥𝐸 < 0, i.e., lattice dilationresults in mode softening (red shift). The 𝛥𝐸 can be evaluated from thefollowing expression:𝛥𝐸 = 𝜔0[exp(−3𝛾 ∫𝑇0𝛼(𝑇 )𝑑 𝑇)− 1](1)𝛥𝐴 = 𝛥(3) + 𝛥(4),with 𝜔0, 𝛾𝑖, and 𝛼(𝑇 ) denoting respectively the harmonic frequency,Grüneisen parameter for the optical Raman mode, and the coefficientof linear thermal expansion. The frequency shifts 𝛥(3) and 𝛥(4) arise fromthe phonon–phonon interactions due to the lowest-order cubic andquartic terms in the interaction potential. The multi-phonon processesassociated with the 𝛥(3) usually give rise to a negative frequency shift(𝛥(3) < 0), whereas the shift 𝛥(4) resulting from the quartic anharmonic-ity can be either positive or negative [73]. Hence, the overall frequencyshift may be either positive or negative, depending on the relativemagnitudes of the anharmonic terms in the interatomic potentials andthe term following from thermal expansion (volume change).The simulated Raman spectra that are shown in Fig. 7 contain both𝛥𝐸 and 𝛥𝐴 terms. The contributions 𝛥(3) and 𝛥(4) appear, however,to be dominated by the thermal expansion term 𝛥𝐸 , which drivescontinuous red shift of the optical phonons in ScN, as depicted inFig. 9. The observed softening of 𝜔1, 𝜔2, 𝜔3, and 𝜔4 modes followsthe temperature behavior of 𝛥𝐸 predicted for particular peaks, asindicated by the straight dashed lines in Fig. 9. The respective 𝛥𝐸shifts were obtained according to Eq. (1), with calculated GrüneisenJ. More-Chevalier et al. Applied Surface Science Advances 25 (2025) 100674 Fig. 8. Positions of the measured Raman peaks in (a) ScN and (b) ScN-T as a function of temperature.Fig. 9. Temperature evolution of peaks’ frequencies in ScN. (a) Positions of peaks 𝜔1 and 𝜔2 belonging to the lower band. (b) Positions of peaks 𝜔3 and 𝜔4 belonging to theupper band. Dashed lines correspond to 𝛥𝐸 term evaluated according to Eq. (1).parameters for the optical Raman modes and by applying the calculatedlinear thermal expansion coefficient of 7.92 × 10−6 K−1, determined as𝛼(𝑇 ) = 13 𝛽(𝑇 ), where 𝛽(𝑇 ) stands for the volumetric thermal expansioncoefficient 𝛽(𝑇 ) = 1𝑉𝑑 𝑉 (𝑇 )𝑑 𝑇 derived from the NpT simulations. We notethe close correspondence between the calculated 𝛼(𝑇 ) and the reportedexperimental value of 6.61 − 7.98 × 10−6 K−1 [74,75]. Alike the shiftsresulting from the AIMD simulations, the experimental peaks (1) and(2) from the lower frequency band in ScN seem to exhibit similar trendsin their position shifts with temperature. Quite the opposite effect is,however, noticed for the peaks (3) and (4) from the upper-frequencyband. The former maintains its position almost on the constant level,while the latter one experiences a slight, but visible increase in itsfrequency (blue shift) with the increasing temperature. Thus, it islikely that the intrinsic anharmonicity due to the quartic term couldbe indeed strong for the highest phonon frequency band in ScN asit overcomes the sum of 𝛥(3) and 𝛥𝐸 terms. Similarly, the peak (4)in ScN-T reveals a blue shift with temperature, indicating that, alsoin the twinned-ScN, the quartic anharmonicity plays a significant rolein the phonon–phonon interaction, especially among modes with thehighest frequencies. It also seems that the interplay between the quarticand cubic contributions in 𝛥𝐴 and the thermal expansion term 𝛥𝐸in ScN-T takes place, and hence the shifts of peaks (1)-(3) do not7 show straightforward behavior with increasing temperature. In fact,the effect of hardening (blue shift) of the Raman-active phonons asa function of temperature has been less commonly reported [76–78]than the softening (red shift) of such modes [79–84]. The observedblue shift has, however, been assigned to some other effects ratherthan directly to the anharmonicity of lattice vibrations [76,77]. Inboth ScN and ScN-T, the higher terms in anharmonicity, i.e., at leasta positive quartic term seems to dominate at elevated temperatures forthe highest-frequency phonons, suppressing contributions from thermalexpansion and the lowest-order cubic term in the interatomic potential.The temperature evolution of the Raman peak intensities in bothScN and ScN-T follows a typical trend, i.e., a decrease in the intensitiesof peaks with increased temperature. The Raman scattering intensityis proportional to the population difference between the ground andexcited vibrational states, the latter being governed by the Boltzmanndistribution function. The population difference decreases as temper-ature increases and this leads to the reduction of the Raman peaks’intensities. An additional factor lowering intensities of the Raman peakscan be correlated with the increase of the absorption light in ScN, whichreduces the number of excited phonon modes [79,80]. Such observationhas already been made for V2O5, WO3, and MoO3. In these oxidesthe closer the absorption band edge of material to the laser frequencyJ. More-Chevalier et al. Applied Surface Science Advances 25 (2025) 100674 Fig. 10. (a) Temperature dependence of the Seebeck coefficient in ScN and ScN-T layers. (b) Temperature dependence of resistivity in ScN and ScN-T layers. Solid lines representthe calculated Seebeck coefficient and resistivity of ScN.and the weaker the metal–oxygen bonds, the more meaningful theeffect of temperature on the intensity of the Raman spectra [80]. Itsband gap (∼2.25 eV) lies quite close to the laser incident energy of514 nm (2.41 eV), which is used in the present Raman spectroscopyexperiments, and this facilitates a resonance effect contributing to theRaman intensity of each phonon mode [79]. This effect decreases uponincreased temperature due to a larger deviation of the transition energyfrom the photon energy of the laser light, which results in a lowernumber of excitonic transitions. This finally leads to reduced intensityof the scattered radiation.3.5. Thermoelectric propertiesThe Seebeck coefficient and resistivity for ScN and ScN-T layersdetermined in the present experiments along with the results of calcu-lations carried out for ScN are shown in Fig. 10. One notices that bothcomputed temperature dependencies of the Seebeck coefficient andresistivity in ScN remain in close agreement with the measured data.The Seebeck coefficient in ScN varies between −30 and −64 μV K−1in ScN, while it is considerably lower in the ScN-T layer, spanning therange from −50 to −82 μV K−1 over the temperatures between 300 and800 K. The resistivity in ScN and ScN-T layers increases as a functionof increasing temperature, which indicates that both systems behave assemimetals. The resistivity ranges from 0.21 to 0.39 mΩ cm in ScN andfrom 0.45 to 0.89 mΩ cm in ScN-T. Hence, the resistivity in the twinnedScN layer is nearly twice as big as that in the almost defect-free ScNlayer.The thermal conductivity measured at room temperature amountsto 10.06 and 2.85 W m−1 K−1 for ScN and ScN-T, respectively. Again,we find an exceptionally good agreement between the experimental andcalculated (10.48 W m−1 K−1) thermal conductivity in ScN at roomtemperature. The calculated thermal conductivity decreases by about40% in the temperature range of 300 - 800 K, as shown in Fig. 11.The present theoretical study allows us to analyze in more detail theelectronic (𝑘𝑒) and lattice/phonon (𝑘𝐿) contributions to the overallthermal conductivity (𝑘), especially their temperature dependence. Theresulting evolution of 𝑘𝑒 and 𝑘𝐿 with temperature is depicted in theinset of Fig. 11. We find the phonon term to be about two timeshigher than the electronic term at room temperature. The 𝑘𝐿 decreases,while the 𝑘𝑒 increases with temperature. The 𝑘𝐿 is the major termin the thermal conductivity up to 500 K, but at higher temperatures,the electronic contribution 𝑘𝑒 becomes a dominating term in the heattransfer of ScN. It is also interesting to note that 𝑘𝐿 is governed bythe acoustic phonons, as the contribution from the optical modes isnearly three orders of magnitude smaller in the entire temperaturerange covered by our theoretical calculations.We should also mention that the values of Seebeck coefficient, resis-tivity, and thermal conductivity of ScN determined in our experimental8 Fig. 11. Calculated temperature dependence of thermal conductivity in ScN. Inset:electronic and lattice contributions to thermal conductivity in ScN.and theoretical investigations generally correspond to those reported byother studies [19,26,31,33]. The Seebeck coefficient of ScN ranges be-tween −20 and −40 μV K−1 at room temperature and between −40 and−160 μV K−1 at 800 K. The room temperature resistivity of ScN variesbetween 0.1 and 1.8 mΩ cm, while the room temperature thermalconductivity of ScN lies in the range 8–12.5 W m−1 K−2. Even thoughstronger Seebeck coefficient at 800 K (−170 μV K−1), lower electricalconductivity (∼500 − 1500 Scm−1) as compared to that reported in thepresent work (∼5000 Scm−1), and higher thermal conductivity (∼15W m−1 K−1) are sometimes encountered in the literature [24,25,28,32].Such a spread in the quantities determining thermoelectric properties ofScN can arise from the contamination of samples by a higher amountof oxygen than in our case (∼1 − 2%), nitrogen vacancies as well ashigh-level doping of the scandium sublattice with various elements. Allthose defects modify the thermoelectric properties of the ScN layers.In particular, oxygen plays an important role as an electron donor, andhence the O-impurities, especially when in high concentration, improveelectrical conductivity, and reduce the Seebeck coefficient and thermalconductivity of the ScN films by lowering the 𝑘𝐿 term.Finally, we discuss thermoelectric efficiency to convert heat toelectricity, which is governed by the dimensionless figure of merit𝑍 𝑇 = 𝑆2𝜎 𝑇 ∕𝑘, where 𝑆, 𝜎, 𝑘, and 𝑇 stand for the Seebeck coefficient,electrical conductivity, thermal conductivity, and temperature, respec-tively. In fact, the experimentally determined temperature evolution ofZT, which is shown in Fig. 12, is its lower limit, as the value of 𝑘 hasbeen measured only at room temperature.J. More-Chevalier et al. Applied Surface Science Advances 25 (2025) 100674 Fig. 12. Lower limit of figure of merit (ZT) as a function of temperature for ScN andScN-T layers on Mg (001) substrates. Solid (dashed) lines denote calculated ZT (lowerlimit of ZT) for ScN.Theoretical temperature dependence of ZT, which is also includedin Fig. 12 does not have such limitation, however, to compare withthe present experiments the calculated lower limit of ZT is provided aswell. Despite some agreement between the calculated and experimentallower limit of ZT in the vicinity of room temperature, the theoreticallower limit of ZT underestimates experimental ZT at elevated temper-atures. The experimental lower limit of ZT is approximated better bythe calculated real ZT up to about 600 K. At still higher temperatures,the lower limit of ZT obtained with the constant value of experimentalthermal conductivity measured at room temperature is not sufficientand such approximation suffers from underestimation of real ZT. Ofcourse, the higher the temperature the bigger the discrepancy. Never-theless, the determined values of the lower limit of ZT (0.01–0.03) liein the range usually reported for ScN films [19,24,28,32,33,85]. TheScN-T layer reveals much higher ZT than the ScN layer, which arisespredominantly from the presence of twin domains. They lead to a de-crease in thermal conductivity, presumably by suppressing the acousticphonon propagation, and by lowering the electronic conductivity andenhancing the Seebeck coefficient due to twin domain boundaries.4. Summary and conclusionsWe have undertaken an examination of the microstructural features,dynamical and thermoelectric properties of the ScN/MgO(001) layersproduced by the DC reactive magnetron sputtering. In particular, theeffect of twin domains on the aforementioned properties of ScN filmshas been explored. Twin domains were observed in pole figures grow-ing on the 111 twinning planes of the 002-oriented grains. A substantialdifference in thermoelectric properties between almost defect-free ScNand that with twin domains is found. Twin domains in the ScN layergive rise to: (i) an increased figure of merit (until at 800 K), (ii) anincreased Seebeck coefficient (from −64 to −82 μV K−1 at 800 K), and(iii) reduced thermal conductivity (from 10.06 to 2.85 W m−1 K−1 at300 K). Thus, a clear improvement in global thermoelectric propertiesof the ScN films with twin domains is perceived. In both defect-freeScN and that containing twin domains, the quartic anharmonicity ofthe interatomic potentials is present and reflected by the temperaturebehavior of their high-frequency optical phonon modes, as revealedby the Raman spectroscopy studies. The results of present researchperformed on thin ScN epitaxial layers suggest that structural domainscan be a rather simple and stable solution enhancing thermoelectricproperties of materials in the form of films.9 CRediT authorship contribution statementJ. More-Chevalier: Writing – review & editing, Writing – origi-nal draft, Visualization, Validation, Supervision, Software, Resources,Methodology, Investigation, Formal analysis, Data curation, Concep-tualization. U.D. Wdowik: Writing – review & editing, Writing –original draft, Visualization, Validation, Software, Methodology, Inves-tigation, Formal analysis, Data curation, Conceptualization. J. Martan:Writing – original draft, Visualization, Validation, Investigation, For-mal analysis, Data curation. T. Baba: Investigation, Formal analysis,Data curation. S. Cichoň: Writing – review & editing, Writing – orig-inal draft, Visualization, Validation, Investigation, Formal analysis,Data curation. P. Levinský: Visualization, Validation, Investigation,Formal analysis, Data curation. D. Legut: Writing – review & edit-ing, Validation, Resources, Project administration, Funding acquisition,Conceptualization. E. de Prado: Writing – review & editing, Writing –original draft, Visualization, Validation, Investigation, Formal analysis,Data curation. P. Hruška: Validation, Formal analysis, Data curation. J.Pokorný: Validation, Formal analysis, Data curation. J. Bulíř: Valida-tion, Formal analysis, Data curation. C. Beltrami: Visualization, Formalanalysis, Data curation. T. Mori: Writing – original draft, Visualiza-tion, Formal analysis, Data curation. M. Novotný: Writing – originaldraft, Validation, Resources, Project administration, Funding acquisi-tion, Formal analysis, Data curation. I. Gregora: Formal analysis, Datacuration. L. Fekete: Formal analysis, Data curation. L. Volfová: Formalanalysis, Data curation. J. Lančok: Visualization, Resources, Projectadministration, Funding acquisition, Formal analysis, Data curation.Declaration of competing interestThe authors declare that they have no known competing finan-cial interests or personal relationships that could have appeared toinfluence the work reported in this paper.AcknowledgmentsThe authors acknowledge the Czech Science Foundation (GAČR)project No. 23-07228S. This work was also supported by the Min-istry of Education, Youth and Sports of the Czech Republic throughthe SENDISO - CZ.02.01.01/00/22_008/0004596 and the e-INFRA CZ(ID:90254). The Interdisciplinary Centre of Mathematical and Com-putational Modeling (ICM), Warsaw University, Poland is acknowl-edged for providing computer facilities to perform part of the presentcalculations.Appendix A. Formulas to obtain resistivity and thermal conduc-tivity of ScNThe scattering rates 1𝜏(𝐤,𝑇 ) are evaluated according to Ref. [47]1𝜏(𝐤, 𝑇 ) = 4𝜋 𝛽 ∫∞0ℏ𝜔 𝛼2t r𝐤(𝜔)𝐹 (𝜔)(exp(𝛽 ℏ𝜔) − 1) (1 − exp(−𝛽 ℏ𝜔))𝑑 𝜔, (A.1)where 𝛽 = 1𝑘𝐵𝑇and 𝛼2t r𝐤(𝜔)𝐹 (𝜔) stands for the effective transportphonon frequency distribution computed by the EPW package. In thescattering time approximation, the resistivity 𝜌(𝑇 ) can be obtained as:𝜌(𝑇 ) = 𝑚𝑛𝑒2⟨1𝜏(𝐤, 𝑇 )⟩, (A.2)where 𝑚 is the electron mass, 𝑒 denotes its charge, and 𝑛 is the numberof free charge carriers per unit volume. The bracket ⟨⋯⟩ indicates anaverage over the Fermi surface. The 𝜌(𝑇 ) is used to calculate electricalconductivity 𝜎(𝑇 ) = 𝜌(𝑇 )−1 and subsequently the electronic contribu-tion to the thermal conductivity 𝑘𝑒 from the Wiedemann–Franz law𝑘𝑒 = 𝑘𝑒−ph = 𝐿0 𝜎 𝑇 , with 𝐿0 =𝜋3(𝑘𝐵𝑒)2= 2.44 × 10−8 W Ω K−2 denotingthe Lorentz number.J. More-Chevalier et al.ps𝛩Titrbitoeb𝜏orel -Applied Surface Science Advances 25 (2025) 100674 The contribution to the lattice thermal conductivity from acousticphonons 𝑘𝐿𝑎 is calculated according to the Slack model and it takes onthe following form [50,51]:𝑘𝐿𝑎(𝑇 ) = 𝑘𝐿𝑎(𝛩𝑎)𝛩𝑎𝑇(A.3)𝑘𝐿𝑎(𝛩𝑎) = 𝐴(𝑘𝐵𝛩𝑎ℏ)2 𝑘𝐵𝑀 𝑉 13ℏ𝛾2𝑎(A.4)𝐴 =0.849 × 3 3√420𝜋3(1 − 0.514𝛾−1𝑎 + 0.228𝛾−2𝑎 ), (A.5)where 𝛾𝑎, 𝑀 , and 𝑉 are the overall Grüneisen parameter for acoustichonon modes, the average atomic mass, and the unit cell volume, re-pectively. The 𝛩𝑎 stands for the so-called acoustic Debye temperature,which is related to the Debye temperature 𝛩𝐷 via the relationship:𝑎 = 𝑛−13 𝛩𝐷, with 𝑛 denoting the number of atoms per unit cell.he 𝛩𝑎 is obtained by considering only the acoustic phonon modes,.e., under the assumption that the optical phonons do not contributeo heat transport [86]. The 𝛩𝐷 is derived from the second moment ofthe calculated phonon spectrum⟨𝜇2⟩ [87]:𝛩𝐷 = ℏ𝑘𝐵√53⟨𝜇2⟩(A.6)⟨𝜇2⟩ =∫ 𝜔2𝑔(𝜔)𝑑 𝜔∫ 𝑔(𝜔)𝑑 𝜔 , (A.7)where 𝑔(𝜔) is the computed density of phonon states over the acousticange. The factor 𝛩𝑎𝑇 is introduced to comply with the observed 𝑇 −1ehavior of the thermal conductivity at temperatures above 𝛩𝐷. The 𝛾𝑎s the weighted average of the (𝐤, 𝑗) mode-specific Grüneisen parame-ers 𝛾(𝐤, 𝑗), where the weighting factors 𝐶𝑉 (𝐤, 𝑗) are the contributionsf individual acoustic (𝐤, 𝑗) modes to the heat capacity. The followingxpressions apply:𝛾𝑎 =∑𝐤,𝑗 𝛾(𝐤, 𝑗)𝐶𝑉 (𝐤, 𝑗)∑𝐤,𝑗 𝐶𝑉 (𝐤, 𝑗)(A.8)𝛾(𝐤, 𝑗) = − 𝜕(ln𝜔(𝐤, 𝑗))𝜕 ln𝑉= − 𝑉𝜔(𝐤, 𝑗)𝜕 𝜔(𝐤, 𝑗)𝜕 𝑉 . (A.9)The contribution to the lattice thermal conductivity from opticalphonons 𝑘𝐿𝑜 is obtained according to the Cahill model [52,53], which isased on the Einstein random walk with the lifetime of each oscillator= 𝜋𝜔 , i.e., one half the period of vibration. The 𝑘𝐿𝑜 resulting from therandom walk between such localized excitations can be expressed as:𝑘𝐿𝑜 =(𝜋6)13 𝑁23∑𝑖𝑣𝑖(𝑇𝛩𝑖)2∫𝛩𝑖∕𝑇0𝑥3𝑒𝑥(𝑒𝑥 − 1)2 𝑑 𝑥 (A.10)𝛩𝑖 = 𝑣𝑖(ℏ𝑘𝐵)(6𝜋2𝑁)13 , (A.11)where the sum runs over the three sound modes (two transverse andne longitudinal) with speeds of sound 𝑣𝑖. The 𝑁 and 𝛩𝑖 denote,espectively, the number density of atoms and the cutoff frequency forach polarization (expressed in degrees Kelvin).Appendix B. Supplementary dataSupplementary material related to this article can be found onlineat https://doi.org/10.1016/j.apsadv.2024.100674.Data availabilityThe data as well as the figures are available through the followingink https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScNdata/.10 References[1] T. Hendricks, T. Caillat, T. Mori, Advanced thermoelectric generationtechnologies 2022, Energies 15 (2022) 7307.[2] M. Mukherjee, A. Srivastava, A.K. Singh, Recent advances in designingthermoelectric materials, J. Mater. Chem. C 10 (2022) 12524–12555.[3] J.Z. X.-L. Shi, Z.-G. Chen, Advanced thermoelectric design: From materials andstructures to devices, Chem. Rev. 120 (2020) 7399–7515.[4] F. Tohidi, S.G. Holagh, A. Chitsaz, Thermoelectric generators: A comprehensivereview of characteristics and applications, App. Therm. Eng. 201 (2022) 117793.[5] J. Wei, L. Yang, Z. Ma, P. Song, M. Zhang, J. Ma, F. Yang, X. Wang, Review ofcurrent high-ZT thermoelectric materials, J. Mater. Sci. 55 (2020) 12642–12704.[6] L. Jin, Y. Hao, A.K. Tareen, K. Khan, S. Wageh, O.A. Al-Hartomy, A.G. Al-Sehemi,H. Zhang, Y. Zhang, Tellurium/polymers for flexible thermoelectrics: status andchallenges, J. Mater. Chem. A 11 (2023) 3771–3788.[7] F. Hossein-Babaei, S. Masoumi, A. Noori, Linking thermoelectric generation inpolycrystalline semiconductors to grain boundary effects sets a platform for novelseebeck effect-based sensors, J. Mater. Chem. A 6 (2018) 10370–10378.[8] W. Wu, G.-K. Ren, X. Chen, Y. Liu, Z. Zhou, J. Song, Y. Shi, J.-M. Jiang, Y.-H.Lin, Interfacial advances yielding high efficiencies for thermoelectric devices, J.Mater. Chem. A 9 (2021) 3209–3230.[9] T. Degousée, V. Untilova, V. Vijayakumar, X. Xu, Y. Sun, M. Palma, M.Brinkmann, L. Biniek, O. Fenwick, High thermal conductivity states and en-hanced figure of merit in aligned polymer thermoelectric materials, J. Mater.Chem. A 9 (2021) 16065–16075.[10] H.J. Goldsmid, Introduction to Thermoelectricity, Springer, Berlin, Heidelberg,2016.[11] G.J. Snyder, E.S. Toberer, Complex thermoelectric materials, Nature Mater. 7(2008) 105–114.[12] M.N. Hasan, H. Wahid, N. Nayan, M.S.M. Ali, Inorganic thermoelectric materials:A review, Int. J. Energy Res. 44 (2020) 6170–6222.[13] P. Eklund, S. Kerdsongpanya, B. Alling, Transition-metal-nitride-based thin filmsas novel energy harvesting materials, J. Mater. Chem. C 4 (2016) 3905–3914.[14] M. Garbrecht, I. McCarroll, L. Yang, V. Bhatia, B. Biswas, D. Rao, J.M. Cair-ney, B. Saha, Thermally stable epitaxial ZrN/carrier-compensated Sc0.99Mg0.01Nmetal/semiconductor multilayers for thermionic energy conversion, J. Mater. Sci.55 (2020) 1592–1602.[15] B. Saha, T.D. Sands, U.V. Waghmare, First-principles analysis of ZrN/ScNmetal/semiconductor superlattices for thermoelectric energy conversion, J. Appl.Phys. 109 (2011) 083717.[16] V. Rawat, Y.K. Koh, D.G. Cahill, T.D. Sands, Thermal conductivity of(Zr,W)N/ScN metal/semiconductor multilayers and superlattices, J. Appl. Phys.105 (2009) 024909.[17] M. Zebarjadi, Z. Bian, R. Singh, A. Shakouri, R. Wortman, V. Rawat, T. Sands,Thermoelectric transport in a ZrN/ScN superlattice, J. Elect. Mater. 38 (2009)960–963.[18] R. Kumar, L. Yang, I. McCarroll, S.M. Shivaprasad, J.M. Cairney, M. Garbrecht,B. Saha, Atomistic structure and three-dimensional spatial distribution of oxideclusters along voids in nitride metal/semiconductor superlattices, Phys. Rev.Mater. 5 (2021) 084601.[19] B. Biswas, B. Saha, Development of semiconducting ScN, Phys. Rev. Mater. 3(2019) 020301.[20] R. Deng, B.D. Ozsdolay, P.Y. Zheng, S.V. Khare, D. Gall, Optical and transportmeasurement and first-principles determination of the ScN band gap, Phys. Rev.B 91 (2015) 045104.[21] A. le Febvrier, N. Tureson, N. Stilkerich, G. Greczynski, P. Eklund, Effect ofimpurities on morphology, growth mode, and thermoelectric properties of (111)and (001) epitaxial-like ScN films, J. Phys. D: Appl. Phys. 52 (3) (2018) 035302.[22] J. More-Chevalier, S. Cichoň, J. Bulíř, M. Poupon, P. Hubík, L. Fekete, J. Lančok,Electrical and optical properties of scandium nitride nanolayers on MgO(100)substrate, AIP Adv. 9 (1) (2019) 015317.[23] J. More-Chevalier, U.D. Wdowik, D. Legut, S. Cichoň, E. de Prado, I. Gregora, J.Bulíř, M. Novotný, L. Fekete, J. Lančok, Effect of oxygen defects on microstruc-ture, optical and vibrational properties of ScN films deposited on MgO substratefrom experiment and first principles, Appl. Surf. Sci. 615 (2023) 156203.[24] P.V. Burmistrova, J. Maassen, T. Favaloro, B. Saha, S. Salamat, Y. Rui Koh, M.S.Lundstrom, A. Shakouri, T.D. Sands, Thermoelectric properties of epitaxial ScNfilms deposited by reactive magnetron sputtering onto MgO(001) substrates, J.Appl. Phys. 113 (2013) 153704.[25] P.V. Burmistrova, D.N. Zakharov, T. Favaloro, A. Mohammed, E.A. Stach,A. Shakouri, T.D. Sands, Effect of deposition pressure on the microstructureand thermoelectric properties of epitaxial ScN(001) thin films sputtered ontoMgO(001) substrates, J. Mater. Res. 30 (5) (2015) 626–634.[26] S. Kerdsongpanya, N. Van Nong, N. Pryds, A. Žukauskaitė, J. Jensen, J. Birch,J. Lu, L. Hultman, G. Wingqvist, P. Eklund, Anomalously high thermoelectricpower factor in epitaxial ScN thin films, Appl. Phys. Lett. 99 (2011) 232113.[27] B. Saha, J.A. Perez-Taborda, J.-H. Bahk, Y.R. Koh, A. Shakouri, M. Martin-Gonzalez, T.D. Sands, Temperature-dependent thermal and thermoelectricproperties of n-type and p-type Sc1-xMgxN, Phys. Rev. B 97 (2018) 085301.https://doi.org/10.1016/j.apsadv.2024.100674https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/https://asep.lib.cas.cz/arl-cav/cs/detail-cav_un_epca-0602125-ScN-data/http://refhub.elsevier.com/S2666-5239(24)00102-8/sb1http://refhub.elsevier.com/S2666-5239(24)00102-8/sb1http://refhub.elsevier.com/S2666-5239(24)00102-8/sb1http://refhub.elsevier.com/S2666-5239(24)00102-8/sb2http://refhub.elsevier.com/S2666-5239(24)00102-8/sb2http://refhub.elsevier.com/S2666-5239(24)00102-8/sb2http://refhub.elsevier.com/S2666-5239(24)00102-8/sb3http://refhub.elsevier.com/S2666-5239(24)00102-8/sb3http://refhub.elsevier.com/S2666-5239(24)00102-8/sb3http://refhub.elsevier.com/S2666-5239(24)00102-8/sb4http://refhub.elsevier.com/S2666-5239(24)00102-8/sb4http://refhub.elsevier.com/S2666-5239(24)00102-8/sb4http://refhub.elsevier.com/S2666-5239(24)00102-8/sb5http://refhub.elsevier.com/S2666-5239(24)00102-8/sb5http://refhub.elsevier.com/S2666-5239(24)00102-8/sb5http://refhub.elsevier.com/S2666-5239(24)00102-8/sb6http://refhub.elsevier.com/S2666-5239(24)00102-8/sb6http://refhub.elsevier.com/S2666-5239(24)00102-8/sb6http://refhub.elsevier.com/S2666-5239(24)00102-8/sb6http://refhub.elsevier.com/S2666-5239(24)00102-8/sb6http://refhub.elsevier.com/S2666-5239(24)00102-8/sb7http://refhub.elsevier.com/S2666-5239(24)00102-8/sb7http://refhub.elsevier.com/S2666-5239(24)00102-8/sb7http://refhub.elsevier.com/S2666-5239(24)00102-8/sb7http://refhub.elsevier.com/S2666-5239(24)00102-8/sb7http://refhub.elsevier.com/S2666-5239(24)00102-8/sb8http://refhub.elsevier.com/S2666-5239(24)00102-8/sb8http://refhub.elsevier.com/S2666-5239(24)00102-8/sb8http://refhub.elsevier.com/S2666-5239(24)00102-8/sb8http://refhub.elsevier.com/S2666-5239(24)00102-8/sb8http://refhub.elsevier.com/S2666-5239(24)00102-8/sb9http://refhub.elsevier.com/S2666-5239(24)00102-8/sb9http://refhub.elsevier.com/S2666-5239(24)00102-8/sb9http://refhub.elsevier.com/S2666-5239(24)00102-8/sb9http://refhub.elsevier.com/S2666-5239(24)00102-8/sb9http://refhub.elsevier.com/S2666-5239(24)00102-8/sb9http://refhub.elsevier.com/S2666-5239(24)00102-8/sb9http://refhub.elsevier.com/S2666-5239(24)00102-8/sb10http://refhub.elsevier.com/S2666-5239(24)00102-8/sb10http://refhub.elsevier.com/S2666-5239(24)00102-8/sb10http://refhub.elsevier.com/S2666-5239(24)00102-8/sb11http://refhub.elsevier.com/S2666-5239(24)00102-8/sb11http://refhub.elsevier.com/S2666-5239(24)00102-8/sb11http://refhub.elsevier.com/S2666-5239(24)00102-8/sb12http://refhub.elsevier.com/S2666-5239(24)00102-8/sb12http://refhub.elsevier.com/S2666-5239(24)00102-8/sb12http://refhub.elsevier.com/S2666-5239(24)00102-8/sb13http://refhub.elsevier.com/S2666-5239(24)00102-8/sb13http://refhub.elsevier.com/S2666-5239(24)00102-8/sb13http://refhub.elsevier.com/S2666-5239(24)00102-8/sb14http://refhub.elsevier.com/S2666-5239(24)00102-8/sb14http://refhub.elsevier.com/S2666-5239(24)00102-8/sb14http://refhub.elsevier.com/S2666-5239(24)00102-8/sb14http://refhub.elsevier.com/S2666-5239(24)00102-8/sb14http://refhub.elsevier.com/S2666-5239(24)00102-8/sb14http://refhub.elsevier.com/S2666-5239(24)00102-8/sb14http://refhub.elsevier.com/S2666-5239(24)00102-8/sb15http://refhub.elsevier.com/S2666-5239(24)00102-8/sb15http://refhub.elsevier.com/S2666-5239(24)00102-8/sb15http://refhub.elsevier.com/S2666-5239(24)00102-8/sb15http://refhub.elsevier.com/S2666-5239(24)00102-8/sb15http://refhub.elsevier.com/S2666-5239(24)00102-8/sb16http://refhub.elsevier.com/S2666-5239(24)00102-8/sb16http://refhub.elsevier.com/S2666-5239(24)00102-8/sb16http://refhub.elsevier.com/S2666-5239(24)00102-8/sb16http://refhub.elsevier.com/S2666-5239(24)00102-8/sb16http://refhub.elsevier.com/S2666-5239(24)00102-8/sb17http://refhub.elsevier.com/S2666-5239(24)00102-8/sb17http://refhub.elsevier.com/S2666-5239(24)00102-8/sb17http://refhub.elsevier.com/S2666-5239(24)00102-8/sb17http://refhub.elsevier.com/S2666-5239(24)00102-8/sb17http://refhub.elsevier.com/S2666-5239(24)00102-8/sb18http://refhub.elsevier.com/S2666-5239(24)00102-8/sb18http://refhub.elsevier.com/S2666-5239(24)00102-8/sb18http://refhub.elsevier.com/S2666-5239(24)00102-8/sb18http://refhub.elsevier.com/S2666-5239(24)00102-8/sb18http://refhub.elsevier.com/S2666-5239(24)00102-8/sb18http://refhub.elsevier.com/S2666-5239(24)00102-8/sb18http://refhub.elsevier.com/S2666-5239(24)00102-8/sb19http://refhub.elsevier.com/S2666-5239(24)00102-8/sb19http://refhub.elsevier.com/S2666-5239(24)00102-8/sb19http://refhub.elsevier.com/S2666-5239(24)00102-8/sb20http://refhub.elsevier.com/S2666-5239(24)00102-8/sb20http://refhub.elsevier.com/S2666-5239(24)00102-8/sb20http://refhub.elsevier.com/S2666-5239(24)00102-8/sb20http://refhub.elsevier.com/S2666-5239(24)00102-8/sb20http://refhub.elsevier.com/S2666-5239(24)00102-8/sb21http://refhub.elsevier.com/S2666-5239(24)00102-8/sb21http://refhub.elsevier.com/S2666-5239(24)00102-8/sb21http://refhub.elsevier.com/S2666-5239(24)00102-8/sb21http://refhub.elsevier.com/S2666-5239(24)00102-8/sb21http://refhub.elsevier.com/S2666-5239(24)00102-8/sb22http://refhub.elsevier.com/S2666-5239(24)00102-8/sb22http://refhub.elsevier.com/S2666-5239(24)00102-8/sb22http://refhub.elsevier.com/S2666-5239(24)00102-8/sb22http://refhub.elsevier.com/S2666-5239(24)00102-8/sb22http://refhub.elsevier.com/S2666-5239(24)00102-8/sb23http://refhub.elsevier.com/S2666-5239(24)00102-8/sb23http://refhub.elsevier.com/S2666-5239(24)00102-8/sb23http://refhub.elsevier.com/S2666-5239(24)00102-8/sb23http://refhub.elsevier.com/S2666-5239(24)00102-8/sb23http://refhub.elsevier.com/S2666-5239(24)00102-8/sb23http://refhub.elsevier.com/S2666-5239(24)00102-8/sb23http://refhub.elsevier.com/S2666-5239(24)00102-8/sb24http://refhub.elsevier.com/S2666-5239(24)00102-8/sb24http://refhub.elsevier.com/S2666-5239(24)00102-8/sb24http://refhub.elsevier.com/S2666-5239(24)00102-8/sb24http://refhub.elsevier.com/S2666-5239(24)00102-8/sb24http://refhub.elsevier.com/S2666-5239(24)00102-8/sb24http://refhub.elsevier.com/S2666-5239(24)00102-8/sb24http://refhub.elsevier.com/S2666-5239(24)00102-8/sb25http://refhub.elsevier.com/S2666-5239(24)00102-8/sb25http://refhub.elsevier.com/S2666-5239(24)00102-8/sb25http://refhub.elsevier.com/S2666-5239(24)00102-8/sb25http://refhub.elsevier.com/S2666-5239(24)00102-8/sb25http://refhub.elsevier.com/S2666-5239(24)00102-8/sb25http://refhub.elsevier.com/S2666-5239(24)00102-8/sb25http://refhub.elsevier.com/S2666-5239(24)00102-8/sb26http://refhub.elsevier.com/S2666-5239(24)00102-8/sb26http://refhub.elsevier.com/S2666-5239(24)00102-8/sb26http://refhub.elsevier.com/S2666-5239(24)00102-8/sb26http://refhub.elsevier.com/S2666-5239(24)00102-8/sb26http://refhub.elsevier.com/S2666-5239(24)00102-8/sb27http://refhub.elsevier.com/S2666-5239(24)00102-8/sb27http://refhub.elsevier.com/S2666-5239(24)00102-8/sb27http://refhub.elsevier.com/S2666-5239(24)00102-8/sb27http://refhub.elsevier.com/S2666-5239(24)00102-8/sb27J. More-Chevalier et al. Applied Surface Science Advances 25 (2025) 100674 [28] D. Rao, B. Biswas, E. Flores, A. Chatterjee, M. Garbrecht, Y.R. Koh, V. Bhatia,A.I.K. Pillai, P.E. Hopkins, M. Martin-Gonzalez, B. Saha, High mobility andhigh thermoelectric power factor in epitaxial ScN thin films deposited withplasma-assisted molecular beam epitaxy, Appl. Phys. Lett. 116 (2020) 152103.[29] S. Kerdsongpanya, B. Sun, F. Eriksson, J. Jensen, J. Lu, Y.K. Koh, N.V. Nong,B. Balke, B. Alling, P. Eklund, Experimental and theoretical investigation ofCr1-xScxN solid solutions for thermoelectrics, J. Appl. Phys. 120 (21) (2016)215103.[30] N. Tureson, N. Van Nong, D. Fournier, N. Singh, S. Acharya, S. Schmidt, L.Belliard, A. Soni, A. le Febvrier, P. Eklund, Reduction of the thermal conductivityof the thermoelectric material ScN by Nb alloying, J. Appl. Phys. 122 (2) (2017)025116.[31] N. Tureson, M. Marteau, T. Cabioch, N. Van Nong, J. Jensen, J. Lu, G.Greczynski, D. Fournier, N. Singh, A. Soni, L. Belliard, P. Eklund, A. le Febvrier,Effect of ion-implantation-induced defects and Mg dopants on the thermoelectricproperties of ScN, Phys. Rev. B 98 (2018) 205307.[32] D. Rao, O. Chowdhury, A.I.K. Pillai, G.K. Pradhan, S. Sahoo, J.P. Feser, M.Garbrecht, B. Saha, Multifunctional irradiation-induced defects for enhancingthermoelectric properties of scandium nitride thin films, ACS Appl. Energy Mater.5 (2022) 6847–6854.[33] R. Burcea, J.-F. Barbot, P.-O. Renault, D. Eyidi, T. Girardeau, M. Marteau, F.Giovannelli, A. Zenji, J.-M. Rampnoux, S. Dilhaire, P. Eklund, A. le Febvrier, In-fluence of generated defects by Ar implantation on the thermoelectric propertiesof ScN, ACS Appl. Energy Mater. 5 (2022) 11025–11033.[34] J. More-Chevalier, S. Cichoň, L. Horák, J. Bulíř, P. Hubík, Z. Gedeonová, L.Fekete, M. Poupon, J. Lančok, Correlation between crystallization and oxidationprocess of ScN films exposed to air, Appl. Surf. Sci. 515 (2020) 145968.[35] S. Cichoň, J. More-Chevalier, U.D. Wdowik, E. de Prado, J. Bulíř, M. Novotný,L. Fekete, J. Duchoň, D. Legut, J. Lančok, Environmental stability and ageingof ScN thin films from XPS Ar+ depth profiling, Appl. Surf. Sci. 674 (2024)160867.[36] J. Martan, O. Hervé, V. Lang, Two-detector measurement system of pulsephotothermal radiometry for the investigation of the thermal properties of thinfilms, J. Appl. Phys. 102 (2007) 064903.[37] J. Martan, Optical layer development for thin films thermal conductivity mea-surement by pulsed photothermal radiometry, Rev. Sci. Instrum. 86 (2015)014902.[38] D.L. Balageas, J.C. Krapez, P. Cielo, Pulsed photothermal modeling of layeredmaterials, J. Appl. Phys. 59 (1986) 348–357.[39] T. Baba, T. Baba, K. Ishikawa, T. Mori, Determination of thermal diffusivity ofthin films by applying Fourier expansion analysis to thermo-reflectance signalafter periodic pulse heating, J. Appl. Phys. 130 (2021) 225107.[40] T. Baba, T. Baba, T. Mori, Fourier transform thermoreflectance method underfront-heat front-detect configuration, Int. J. Thermophys. 45 (2024) 61.[41] P. Giannozzi, et al., QUANTUM ESPRESSO: a modular and open-source softwareproject for quantum simulations of materials, J. Phys.: Condens. Matter. 21(2009) 395502.[42] P. Giannozzi, et al., Advanced capabilities for materials modelling with quantumESPRESSO, J. Phys.: Condens. Matter. 29 (2017) 465901.[43] J.P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient approximation madesimple, Phys. Rev. Lett. 78 (1997) 1396.[44] J. Noffsinger, F. Giustino, B.D. Malone, C.-H. Park, S.G. Louie, M.L. Cohen,EPW: Electron–phonon coupling, transport and superconducting properties usingmaximally localized wannier functions, Comput. Phys. Commun. 181 (2010)2140.[45] S. Poncé, E.R. Margine, C. Verdi, F. Giustino, EPW: Electron–phonon coupling,transport and superconducting properties using maximally localized wannierfunctions, Comput. Phys. Comm. 209 (2016) 116.[46] F. Giustino, M.L. Cohen, S.G. Louie, Electron-phonon interaction using wannierfunctions, Phys. Rev. B 76 (2007) 165108.[47] B. Hayman, J.P. Carbotte, Anisotropic transport scattering times in the alkalimetals, J. Phys. F: Met. Phys. 2 (1972) 915.[48] M. Bernardi, D. Vigil-Fowler, J. Lischner, J.B. Neaton, S.G. Louie, Ab initio studyof hot carriers in the first picosecond after sunlight absorption in silicon, Phys.Rev. Lett. 112 (2014) 257402.[49] J. Liu, Y. Zhao, C. Lian, Z. Dai, J.-T. Sun, S. Meng, Ab initio study on anisotropicthermoelectric transport in ternary pnictide KZnP, J. Phys.: Condens. Matter. 2(2019) 024001.[50] G.A. Slack, Nonmetallic crystals with high thermal conductivity, J. Phys. Chem.Solids 34 (1973) 321.[51] G.A. Slack, The thermal conductivity of nonmetallic crystals, Solid State Phys.34 (1979) 1.[52] D.G. Cahill, R.O. Pohl, Heat flow and lattice vibrations in glasses, Solid StateCommun. 70 (1989) 927.[53] D.G. Cahill, S.K. Watson, R.O. Pohl, Lower limit to the thermal conductivity ofdisordered crystals, Phys. Rev. B 46 (1992) 6131.[54] G.K.H. Madsen, J. Carrete, M.J. Verstraete, BoltzTraP2, a program for interpolat-ing band structures and calculating semi-classical transport coefficients, Comput.Phys. Comm. 231 (2018) 140–145.11 [55] G. Kresse, J. Hafner, Ab initio molecular dynamics for liquid metals, Phys. Rev.B 47 (1993) 558.[56] G. Kresse, J. Furthmüller, Efficient iterative schemes for ab initio total-energycalculations using a plane-wave basis set, Comput. Mater. Sci. 6 (1996) 15.[57] D.B. Zhang, T. Sun, R.M. Wentzcovitch, Phonon quasiparticles and anharmonicfree energy in complex systems, Phys. Rev. Lett. 112 (2014) 058501.[58] T. Sun, D.B. Zhang, R.M. Wentzcovitch, Dynamic stabilization of cubic CaSiO3perovskite at high temperatures and pressures from ab initio molecular dynamics,Phys. Rev. B 89 (2014) 094109.[59] D. Gall, I. Petrov, N. Hellgren, L. Hultman, J.E. Sundgren, J.E. Greene, Growthof poly- and single-crystal ScN on MgO(001): Role of low-energy N2+ irradiationin determining texture, microstructure evolution, and mechanical properties, J.Appl. Phys. 84 (1998) 6034–6041.[60] R.T. Haasch, T.Y. Lee, D. Gall, J.E. Greene, I. Petrov, Epitaxial ScN(001) grownand analyzed in situ by XPS and UPS. I. analysis of As-deposited layers, Surf.Sci. Spec. 7 (2000) 169–177.[61] R.T. Haasch, T.Y. Lee, D. Gall, J.E. Greene, I. Petrov, Epitaxial ScN(001) grownand analyzed in situ by XPS and ups. II. analysis of Ar+ sputter etched layers,Surf. Sci. Spec. 7 (2000) 178–184.[62] D. Gall, I. Petrov, P. Desjardins, J.E. Greene, Microstructural evolution and Pois-son ratio of epitaxial ScN grown on TiN(001)/MgO(001) by ultrahigh vacuumreactive magnetron sputter deposition, J. Appl. Phys. 86 (1999) 5524–5529.[63] B.W. Karr, I. Petrov, D.G. Cahill, J.E. Greene, Morphology of epitaxial TiN(001)grown by magnetron sputtering, Appl. Phys. Lett. 70 (1997) 1703–1705.[64] F.F. Leal, S.C. Ferreira, S.O. Ferreira, Modelling of epitaxial film growth withan Ehrlich–Schwoebel barrier dependent on the step height, J. Phys.: Condens.Matter. 23 (2011) 292201.[65] S.C. Wang, G. Ehrlich, Adatom motion to lattice steps: A direct view, Phys. Rev.Lett. 70 (1993) 41–44.[66] R. Dheemahi, B. Bidesh, A. Shashidhara, B. Vijay, P.A.I. Kamalasanan, G.Magnus, S. Bivas, Effects of adatom mobility and Ehrlich–Schwoebel barrier onheteroepitaxial growth of scandium nitride (ScN) thin films, Appl. Phys. Lett.117 (2020) 212101.[67] G. Travaglini, F. Marabelli, R. Monnier, E. Kaldis, P. Wachter, Electronic structureof ScN, Phys. Rev. B 34 (1986) 3876.[68] D. Gall, M. Stoehr, J.E. Greene, Vibrational modes in epitaxial Ti1-xScxN(001)layers: An ab initio calculations and Raman spectroscopy study, Phys. Rev. B 64(2001) 174302.[69] R. Deng, K. Jiang, D. Gall, Optical phonon modes in Al1–xScxN, J. Appl. Phys.115 (2014) 013506.[70] M. Balkanski, R.F. Wallis, E. Haro, Anharmonic effects in light scattering due tooptical phonons in silicon, Phys. Rev. B 28 (1982) 1928.[71] J. Menéndez, M. Cardona, Temperature dependence of the first-order Ramanscattering by phonons in Si, Ge, and alpha-Sn: anharmonic effects, Phys. Rev. B29 (1982) 2051.[72] J. González, E. Moya, J.C. Chervin, Anharmonic effects in light scattering dueto optical phonons in CuGaS2, Phys. Rev. B 54 (1996) 4707.[73] A.A. Maradudin, A.E. Fein, Scattering of neutron by an anharmonic crystal, Phys.Rev. B 128 (1962) 2589.[74] L. Ciprian, S. Mihalic, C. Lüttich, F. Hörich, E. Wade, B. Christian, A. Dadgar,O. Ambacher, Thermal expansion coefficient of ScN(111) thin films grown onSi(111) determined by X-ray diffraction, Appl. Phys. Lett. 124 (2024) 052203.[75] S. Tahri, A. Qteish, I.I. Al-Qasir, N. Meskin, Vibrational and thermal properties ofScN and YN: quasi-harmonic approximation calculations and anharmonic effects,J. Phys.: Condens. Matter. 24 (2011) 035401.[76] I. Antoniazzi, N. Zawadzka, M. Grzeszczyk, T. Woźniak, J.I. nez, Z. Muhammad,W. Zhao, M.R. Molas, A. Babiński, The effect of temperature and excitationenergy on Raman scattering in bulk HfS2, J. Phys.: Condens. Matter. 35 (2023)305401.[77] J. Peng, S. Najmaei, M. Dubey, P.W. Chung, A reversible structural transitionat 300 K to a low-symmetry polytype of hafnium disulfide atomic layer, Mater.Today Commun. 26 (2021) 101722.[78] S. Mann, V.K. Jindal, Blue and red shifted temperature dependence of implicitphonon shifts in graphene, Mater. Res. Express 4 (2017) 075038.[79] H. Zobeiri, S. Xu, Y. Yue, Q. Zhang, Y. Xie, X. Wang, Effect of temperature onRaman intensity of nm-thick WS2: combined effects of resonance Raman, opticalproperties, and interface optical interference, Nanoscale 12 (2020) 6064–6078.[80] S. Xie, E. Iglesia, A.T. Bell, Effects of temperature on the Raman spectra anddispersed oxides, J. Phys. Chem. B 105 (2001) 5144–5152.[81] Z. Zhao, J. Elwood, M.A. Carpenter, Phonon anharmonicity of PdO studied byRaman spectrometry, J. Phys. Chem. C 119 (2015) 23094–23102.[82] H. Malekpour, A.A. Balandin, Raman-based technique for measuring thermalconductivity of graphene and related materials, J. Raman Spectrosc. 49 (1)(2018) 106–120.[83] A. Özden, E. Zuñiga-Puelles, J. Kortus, R. Gumeniuk, C. Himcinschi, Thermalconductivity and phonon anharmonicity of chemical vapor transport grown andmineral–FeS2 single crystals: An optothermal Raman study, J. Raman Spectrosc.54 (1) (2023) 84–92.[84] Q. Xie, L. Xu, C. Hu, L. Chen, J. Zheng, W. Wang, H. Yin, G. Cheng, X. Ai, Phononanharmonicity of thermoelectric material HfTe5 studied by Raman spectroscopy,J. Raman Spectrosc. 52 (5) (2021) 988–994.http://refhub.elsevier.com/S2666-5239(24)00102-8/sb28http://refhub.elsevier.com/S2666-5239(24)00102-8/sb28http://refhub.elsevier.com/S2666-5239(24)00102-8/sb28http://refhub.elsevier.com/S2666-5239(24)00102-8/sb28http://refhub.elsevier.com/S2666-5239(24)00102-8/sb28http://refhub.elsevier.com/S2666-5239(24)00102-8/sb28http://refhub.elsevier.com/S2666-5239(24)00102-8/sb28http://refhub.elsevier.com/S2666-5239(24)00102-8/sb29http://refhub.elsevier.com/S2666-5239(24)00102-8/sb29http://refhub.elsevier.com/S2666-5239(24)00102-8/sb29http://refhub.elsevier.com/S2666-5239(24)00102-8/sb29http://refhub.elsevier.com/S2666-5239(24)00102-8/sb29http://refhub.elsevier.com/S2666-5239(24)00102-8/sb29http://refhub.elsevier.com/S2666-5239(24)00102-8/sb29http://refhub.elsevier.com/S2666-5239(24)00102-8/sb30http://refhub.elsevier.com/S2666-5239(24)00102-8/sb30http://refhub.elsevier.com/S2666-5239(24)00102-8/sb30http://refhub.elsevier.com/S2666-5239(24)00102-8/sb30http://refhub.elsevier.com/S2666-5239(24)00102-8/sb30http://refhub.elsevier.com/S2666-5239(24)00102-8/sb30http://refhub.elsevier.com/S2666-5239(24)00102-8/sb30http://refhub.elsevier.com/S2666-5239(24)00102-8/sb31http://refhub.elsevier.com/S2666-5239(24)00102-8/sb31http://refhub.elsevier.com/S2666-5239(24)00102-8/sb31http://refhub.elsevier.com/S2666-5239(24)00102-8/sb31http://refhub.elsevier.com/S2666-5239(24)00102-8/sb31http://refhub.elsevier.com/S2666-5239(24)00102-8/sb31http://refhub.elsevier.com/S2666-5239(24)00102-8/sb31http://refhub.elsevier.com/S2666-5239(24)00102-8/sb32http://refhub.elsevier.com/S2666-5239(24)00102-8/sb32http://refhub.elsevier.com/S2666-5239(24)00102-8/sb32http://refhub.elsevier.com/S2666-5239(24)00102-8/sb32http://refhub.elsevier.com/S2666-5239(24)00102-8/sb32http://refhub.elsevier.com/S2666-5239(24)00102-8/sb32http://refhub.elsevier.com/S2666-5239(24)00102-8/sb32http://refhub.elsevier.com/S2666-5239(24)00102-8/sb33http://refhub.elsevier.com/S2666-5239(24)00102-8/sb33http://refhub.elsevier.com/S2666-5239(24)00102-8/sb33http://refhub.elsevier.com/S2666-5239(24)00102-8/sb33http://refhub.elsevier.com/S2666-5239(24)00102-8/sb33http://refhub.elsevier.com/S2666-5239(24)00102-8/sb33http://refhub.elsevier.com/S2666-5239(24)00102-8/sb33http://refhub.elsevier.com/S2666-5239(24)00102-8/sb34http://refhub.elsevier.com/S2666-5239(24)00102-8/sb34http://refhub.elsevier.com/S2666-5239(24)00102-8/sb34http://refhub.elsevier.com/S2666-5239(24)00102-8/sb34http://refhub.elsevier.com/S2666-5239(24)00102-8/sb34http://refhub.elsevier.com/S2666-5239(24)00102-8/sb35http://refhub.elsevier.com/S2666-5239(24)00102-8/sb35http://refhub.elsevier.com/S2666-5239(24)00102-8/sb35http://refhub.elsevier.com/S2666-5239(24)00102-8/sb35http://refhub.elsevier.com/S2666-5239(24)00102-8/sb35http://refhub.elsevier.com/S2666-5239(24)00102-8/sb35http://refhub.elsevier.com/S2666-5239(24)00102-8/sb35http://refhub.elsevier.com/S2666-5239(24)00102-8/sb36http://refhub.elsevier.com/S2666-5239(24)00102-8/sb36http://refhub.elsevier.com/S2666-5239(24)00102-8/sb36http://refhub.elsevier.com/S2666-5239(24)00102-8/sb36http://refhub.elsevier.com/S2666-5239(24)00102-8/sb36http://refhub.elsevier.com/S2666-5239(24)00102-8/sb37http://refhub.elsevier.com/S2666-5239(24)00102-8/sb37http://refhub.elsevier.com/S2666-5239(24)00102-8/sb37http://refhub.elsevier.com/S2666-5239(24)00102-8/sb37http://refhub.elsevier.com/S2666-5239(24)00102-8/sb37http://refhub.elsevier.com/S2666-5239(24)00102-8/sb38http://refhub.elsevier.com/S2666-5239(24)00102-8/sb38http://refhub.elsevier.com/S2666-5239(24)00102-8/sb38http://refhub.elsevier.com/S2666-5239(24)00102-8/sb39http://refhub.elsevier.com/S2666-5239(24)00102-8/sb39http://refhub.elsevier.com/S2666-5239(24)00102-8/sb39http://refhub.elsevier.com/S2666-5239(24)00102-8/sb39http://refhub.elsevier.com/S2666-5239(24)00102-8/sb39http://refhub.elsevier.com/S2666-5239(24)00102-8/sb40http://refhub.elsevier.com/S2666-5239(24)00102-8/sb40http://refhub.elsevier.com/S2666-5239(24)00102-8/sb40http://refhub.elsevier.com/S2666-5239(24)00102-8/sb41http://refhub.elsevier.com/S2666-5239(24)00102-8/sb41http://refhub.elsevier.com/S2666-5239(24)00102-8/sb41http://refhub.elsevier.com/S2666-5239(24)00102-8/sb41http://refhub.elsevier.com/S2666-5239(24)00102-8/sb41http://refhub.elsevier.com/S2666-5239(24)00102-8/sb42http://refhub.elsevier.com/S2666-5239(24)00102-8/sb42http://refhub.elsevier.com/S2666-5239(24)00102-8/sb42http://refhub.elsevier.com/S2666-5239(24)00102-8/sb43http://refhub.elsevier.com/S2666-5239(24)00102-8/sb43http://refhub.elsevier.com/S2666-5239(24)00102-8/sb43http://refhub.elsevier.com/S2666-5239(24)00102-8/sb44http://refhub.elsevier.com/S2666-5239(24)00102-8/sb44http://refhub.elsevier.com/S2666-5239(24)00102-8/sb44http://refhub.elsevier.com/S2666-5239(24)00102-8/sb44http://refhub.elsevier.com/S2666-5239(24)00102-8/sb44http://refhub.elsevier.com/S2666-5239(24)00102-8/sb44http://refhub.elsevier.com/S2666-5239(24)00102-8/sb44http://refhub.elsevier.com/S2666-5239(24)00102-8/sb45http://refhub.elsevier.com/S2666-5239(24)00102-8/sb45http://refhub.elsevier.com/S2666-5239(24)00102-8/sb45http://refhub.elsevier.com/S2666-5239(24)00102-8/sb45http://refhub.elsevier.com/S2666-5239(24)00102-8/sb45http://refhub.elsevier.com/S2666-5239(24)00102-8/sb46http://refhub.elsevier.com/S2666-5239(24)00102-8/sb46http://refhub.elsevier.com/S2666-5239(24)00102-8/sb46http://refhub.elsevier.com/S2666-5239(24)00102-8/sb47http://refhub.elsevier.com/S2666-5239(24)00102-8/sb47http://refhub.elsevier.com/S2666-5239(24)00102-8/sb47http://refhub.elsevier.com/S2666-5239(24)00102-8/sb48http://refhub.elsevier.com/S2666-5239(24)00102-8/sb48http://refhub.elsevier.com/S2666-5239(24)00102-8/sb48http://refhub.elsevier.com/S2666-5239(24)00102-8/sb48http://refhub.elsevier.com/S2666-5239(24)00102-8/sb48http://refhub.elsevier.com/S2666-5239(24)00102-8/sb49http://refhub.elsevier.com/S2666-5239(24)00102-8/sb49http://refhub.elsevier.com/S2666-5239(24)00102-8/sb49http://refhub.elsevier.com/S2666-5239(24)00102-8/sb49http://refhub.elsevier.com/S2666-5239(24)00102-8/sb49http://refhub.elsevier.com/S2666-5239(24)00102-8/sb50http://refhub.elsevier.com/S2666-5239(24)00102-8/sb50http://refhub.elsevier.com/S2666-5239(24)00102-8/sb50http://refhub.elsevier.com/S2666-5239(24)00102-8/sb51http://refhub.elsevier.com/S2666-5239(24)00102-8/sb51http://refhub.elsevier.com/S2666-5239(24)00102-8/sb51http://refhub.elsevier.com/S2666-5239(24)00102-8/sb52http://refhub.elsevier.com/S2666-5239(24)00102-8/sb52http://refhub.elsevier.com/S2666-5239(24)00102-8/sb52http://refhub.elsevier.com/S2666-5239(24)00102-8/sb53http://refhub.elsevier.com/S2666-5239(24)00102-8/sb53http://refhub.elsevier.com/S2666-5239(24)00102-8/sb53http://refhub.elsevier.com/S2666-5239(24)00102-8/sb54http://refhub.elsevier.com/S2666-5239(24)00102-8/sb54http://refhub.elsevier.com/S2666-5239(24)00102-8/sb54http://refhub.elsevier.com/S2666-5239(24)00102-8/sb54http://refhub.elsevier.com/S2666-5239(24)00102-8/sb54http://refhub.elsevier.com/S2666-5239(24)00102-8/sb55http://refhub.elsevier.com/S2666-5239(24)00102-8/sb55http://refhub.elsevier.com/S2666-5239(24)00102-8/sb55http://refhub.elsevier.com/S2666-5239(24)00102-8/sb56http://refhub.elsevier.com/S2666-5239(24)00102-8/sb56http://refhub.elsevier.com/S2666-5239(24)00102-8/sb56http://refhub.elsevier.com/S2666-5239(24)00102-8/sb57http://refhub.elsevier.com/S2666-5239(24)00102-8/sb57http://refhub.elsevier.com/S2666-5239(24)00102-8/sb57http://refhub.elsevier.com/S2666-5239(24)00102-8/sb58http://refhub.elsevier.com/S2666-5239(24)00102-8/sb58http://refhub.elsevier.com/S2666-5239(24)00102-8/sb58http://refhub.elsevier.com/S2666-5239(24)00102-8/sb58http://refhub.elsevier.com/S2666-5239(24)00102-8/sb58http://refhub.elsevier.com/S2666-5239(24)00102-8/sb59http://refhub.elsevier.com/S2666-5239(24)00102-8/sb59http://refhub.elsevier.com/S2666-5239(24)00102-8/sb59http://refhub.elsevier.com/S2666-5239(24)00102-8/sb59http://refhub.elsevier.com/S2666-5239(24)00102-8/sb59http://refhub.elsevier.com/S2666-5239(24)00102-8/sb59http://refhub.elsevier.com/S2666-5239(24)00102-8/sb59http://refhub.elsevier.com/S2666-5239(24)00102-8/sb60http://refhub.elsevier.com/S2666-5239(24)00102-8/sb60http://refhub.elsevier.com/S2666-5239(24)00102-8/sb60http://refhub.elsevier.com/S2666-5239(24)00102-8/sb60http://refhub.elsevier.com/S2666-5239(24)00102-8/sb60http://refhub.elsevier.com/S2666-5239(24)00102-8/sb61http://refhub.elsevier.com/S2666-5239(24)00102-8/sb61http://refhub.elsevier.com/S2666-5239(24)00102-8/sb61http://refhub.elsevier.com/S2666-5239(24)00102-8/sb61http://refhub.elsevier.com/S2666-5239(24)00102-8/sb61http://refhub.elsevier.com/S2666-5239(24)00102-8/sb62http://refhub.elsevier.com/S2666-5239(24)00102-8/sb62http://refhub.elsevier.com/S2666-5239(24)00102-8/sb62http://refhub.elsevier.com/S2666-5239(24)00102-8/sb62http://refhub.elsevier.com/S2666-5239(24)00102-8/sb62http://refhub.elsevier.com/S2666-5239(24)00102-8/sb63http://refhub.elsevier.com/S2666-5239(24)00102-8/sb63http://refhub.elsevier.com/S2666-5239(24)00102-8/sb63http://refhub.elsevier.com/S2666-5239(24)00102-8/sb64http://refhub.elsevier.com/S2666-5239(24)00102-8/sb64http://refhub.elsevier.com/S2666-5239(24)00102-8/sb64http://refhub.elsevier.com/S2666-5239(24)00102-8/sb64http://refhub.elsevier.com/S2666-5239(24)00102-8/sb64http://refhub.elsevier.com/S2666-5239(24)00102-8/sb65http://refhub.elsevier.com/S2666-5239(24)00102-8/sb65http://refhub.elsevier.com/S2666-5239(24)00102-8/sb65http://refhub.elsevier.com/S2666-5239(24)00102-8/sb66http://refhub.elsevier.com/S2666-5239(24)00102-8/sb66http://refhub.elsevier.com/S2666-5239(24)00102-8/sb66http://refhub.elsevier.com/S2666-5239(24)00102-8/sb66http://refhub.elsevier.com/S2666-5239(24)00102-8/sb66http://refhub.elsevier.com/S2666-5239(24)00102-8/sb66http://refhub.elsevier.com/S2666-5239(24)00102-8/sb66http://refhub.elsevier.com/S2666-5239(24)00102-8/sb67http://refhub.elsevier.com/S2666-5239(24)00102-8/sb67http://refhub.elsevier.com/S2666-5239(24)00102-8/sb67http://refhub.elsevier.com/S2666-5239(24)00102-8/sb68http://refhub.elsevier.com/S2666-5239(24)00102-8/sb68http://refhub.elsevier.com/S2666-5239(24)00102-8/sb68http://refhub.elsevier.com/S2666-5239(24)00102-8/sb68http://refhub.elsevier.com/S2666-5239(24)00102-8/sb68http://refhub.elsevier.com/S2666-5239(24)00102-8/sb69http://refhub.elsevier.com/S2666-5239(24)00102-8/sb69http://refhub.elsevier.com/S2666-5239(24)00102-8/sb69http://refhub.elsevier.com/S2666-5239(24)00102-8/sb70http://refhub.elsevier.com/S2666-5239(24)00102-8/sb70http://refhub.elsevier.com/S2666-5239(24)00102-8/sb70http://refhub.elsevier.com/S2666-5239(24)00102-8/sb71http://refhub.elsevier.com/S2666-5239(24)00102-8/sb71http://refhub.elsevier.com/S2666-5239(24)00102-8/sb71http://refhub.elsevier.com/S2666-5239(24)00102-8/sb71http://refhub.elsevier.com/S2666-5239(24)00102-8/sb71http://refhub.elsevier.com/S2666-5239(24)00102-8/sb72http://refhub.elsevier.com/S2666-5239(24)00102-8/sb72http://refhub.elsevier.com/S2666-5239(24)00102-8/sb72http://refhub.elsevier.com/S2666-5239(24)00102-8/sb73http://refhub.elsevier.com/S2666-5239(24)00102-8/sb73http://refhub.elsevier.com/S2666-5239(24)00102-8/sb73http://refhub.elsevier.com/S2666-5239(24)00102-8/sb74http://refhub.elsevier.com/S2666-5239(24)00102-8/sb74http://refhub.elsevier.com/S2666-5239(24)00102-8/sb74http://refhub.elsevier.com/S2666-5239(24)00102-8/sb74http://refhub.elsevier.com/S2666-5239(24)00102-8/sb74http://refhub.elsevier.com/S2666-5239(24)00102-8/sb75http://refhub.elsevier.com/S2666-5239(24)00102-8/sb75http://refhub.elsevier.com/S2666-5239(24)00102-8/sb75http://refhub.elsevier.com/S2666-5239(24)00102-8/sb75http://refhub.elsevier.com/S2666-5239(24)00102-8/sb75http://refhub.elsevier.com/S2666-5239(24)00102-8/sb76http://refhub.elsevier.com/S2666-5239(24)00102-8/sb76http://refhub.elsevier.com/S2666-5239(24)00102-8/sb76http://refhub.elsevier.com/S2666-5239(24)00102-8/sb76http://refhub.elsevier.com/S2666-5239(24)00102-8/sb76http://refhub.elsevier.com/S2666-5239(24)00102-8/sb76http://refhub.elsevier.com/S2666-5239(24)00102-8/sb76http://refhub.elsevier.com/S2666-5239(24)00102-8/sb77http://refhub.elsevier.com/S2666-5239(24)00102-8/sb77http://refhub.elsevier.com/S2666-5239(24)00102-8/sb77http://refhub.elsevier.com/S2666-5239(24)00102-8/sb77http://refhub.elsevier.com/S2666-5239(24)00102-8/sb77http://refhub.elsevier.com/S2666-5239(24)00102-8/sb78http://refhub.elsevier.com/S2666-5239(24)00102-8/sb78http://refhub.elsevier.com/S2666-5239(24)00102-8/sb78http://refhub.elsevier.com/S2666-5239(24)00102-8/sb79http://refhub.elsevier.com/S2666-5239(24)00102-8/sb79http://refhub.elsevier.com/S2666-5239(24)00102-8/sb79http://refhub.elsevier.com/S2666-5239(24)00102-8/sb79http://refhub.elsevier.com/S2666-5239(24)00102-8/sb79http://refhub.elsevier.com/S2666-5239(24)00102-8/sb80http://refhub.elsevier.com/S2666-5239(24)00102-8/sb80http://refhub.elsevier.com/S2666-5239(24)00102-8/sb80http://refhub.elsevier.com/S2666-5239(24)00102-8/sb81http://refhub.elsevier.com/S2666-5239(24)00102-8/sb81http://refhub.elsevier.com/S2666-5239(24)00102-8/sb81http://refhub.elsevier.com/S2666-5239(24)00102-8/sb82http://refhub.elsevier.com/S2666-5239(24)00102-8/sb82http://refhub.elsevier.com/S2666-5239(24)00102-8/sb82http://refhub.elsevier.com/S2666-5239(24)00102-8/sb82http://refhub.elsevier.com/S2666-5239(24)00102-8/sb82http://refhub.elsevier.com/S2666-5239(24)00102-8/sb83http://refhub.elsevier.com/S2666-5239(24)00102-8/sb83http://refhub.elsevier.com/S2666-5239(24)00102-8/sb83http://refhub.elsevier.com/S2666-5239(24)00102-8/sb83http://refhub.elsevier.com/S2666-5239(24)00102-8/sb83http://refhub.elsevier.com/S2666-5239(24)00102-8/sb83http://refhub.elsevier.com/S2666-5239(24)00102-8/sb83http://refhub.elsevier.com/S2666-5239(24)00102-8/sb84http://refhub.elsevier.com/S2666-5239(24)00102-8/sb84http://refhub.elsevier.com/S2666-5239(24)00102-8/sb84http://refhub.elsevier.com/S2666-5239(24)00102-8/sb84http://refhub.elsevier.com/S2666-5239(24)00102-8/sb84J. More-Chevalier et al. Applied Surface Science Advances 25 (2025) 100674 [85] B. Saha, T.D. Sands, U.V. Waghmare, Thermoelectric properties of HfN/ScNmetal/semiconductor superlattices: a first-principles study, J. Phys.: Condens.Matter. 24 (41) (2012) 415303.12 [86] D.T. Morelli, G.A. Slack, High lattice thermal conductivity solids, in: HighThermal Conductivity Materials, Springer, New York, 2006, p. 37.[87] G. Grimvall, Thermophysical Properties of Materials, Elsevier, 1999.http://refhub.elsevier.com/S2666-5239(24)00102-8/sb85http://refhub.elsevier.com/S2666-5239(24)00102-8/sb85http://refhub.elsevier.com/S2666-5239(24)00102-8/sb85http://refhub.elsevier.com/S2666-5239(24)00102-8/sb85http://refhub.elsevier.com/S2666-5239(24)00102-8/sb85http://refhub.elsevier.com/S2666-5239(24)00102-8/sb86http://refhub.elsevier.com/S2666-5239(24)00102-8/sb86http://refhub.elsevier.com/S2666-5239(24)00102-8/sb86http://refhub.elsevier.com/S2666-5239(24)00102-8/sb87 Enhancing thermoelectric properties of ScN films through twin domains Introduction Methodology Experimental Theoretical Results and discussion X-ray diffraction  Atomic force microscopy XPS spectra Raman spectroscopy Thermoelectric properties Summary and conclusions CRediT authorship contribution statement Declaration of competing interest Acknowledgments Formulas to obtain resistivity and thermal conductivity of ScN Appendix A. Formulas to obtain resistivity and thermal conductivity of ScN Supplementary data Appendix B. Supplementary data Supplementary data Appendix B. Supplementary data Data availability Appendix . Data availability References