Single Crystals of EuScCuSe3: Synthesis, Experimental and DFT Investigations

EuScCuSe3 was synthesized from the elements for the first time by the method of cesium-iodide flux. The crystal belongs to the orthorhombic system (Cmcm) with the unit cell parameters a = 3.9883(3) Å, b = 13.2776(9) Å, c = 10.1728(7) Å, V = 538.70(7) Å3. Density functional (DFT) methods were used to study the crystal structure stability of EuScCuSe3 in the experimentally obtained Cmcm and the previously proposed Pnma space groups. It was shown that analysis of elastic properties as Raman and infrared spectroscopy are powerless for this particular task. The instability of EuScCuSe3 in space group Pnma space group is shown on the basis of phonon dispersion curve simulation. The EuScCuSe3 can be assigned to indirect wide-band gap semiconductors. It exhibits the properties of a soft ferromagnet at temperatures below 2 K.

The chemistry of trivalent scandium compounds is of particular interest in terms of crystal structure and properties [8]. Trivalent scandium differs from other trivalent cations of the first transition series due to the presence of a closed outer electron shell with an argon configuration. Scandium chalcogenides are p-type [2,4,[9][10][11] and n-type [8] semiconductors. Doping with heavy metals results in n-type conduction with low resistivity [9] in compounds that exhibit metallic properties [12].
Scandium chalcogenides can be obtained in different ways: -In the form of single crystals by the methods of reactive flux or halide flux [15], then heated in a resistant furnace. A temperature of 1120 K was reached within 30 h and kept for 96 h. Afterward, it was cooled to 570 K at a rate of 4 K h −1 and then to room temperature within 3 h. The reaction proceeded according to the equation: Eu + Cu + Sc + 3 Se → EuScCuSe 3 . The reaction product was purified from flux residues with demineralized water. The synthesized samples were dark red, needle-shaped single crystals of EuScCuSe 3 ( Figure 1). The stoichiometric ratio of the elements of europium (76.39 mg), scandium (22.60 mg), copper (31.94 mg), and selenium (119.07 mg) in the presence of CsI (800 mg) was loaded into silica ampoules. These ampoules were evacuated to a pressure of 2 × 10 −3 mbar, sealed, and then heated in a resistant furnace. A temperature of 1120 K was reached within 30 h and kept for 96 h. Afterward, it was cooled to 570 K at a rate of 4 K h −1 and then to room temperature within 3 h. The reaction proceeded according to the equation: Eu + Cu + Sc + 3 Se ⟶ EuScCuSe3. The reaction product was purified from flux residues with demineralized water. The synthesized samples were dark red, needle-shaped single crystals of EuScCuSe3 ( Figure 1).

Figure 1.
A single crystal of EuScCuSe3 placed in a glass capillary.

Methods
A selected single crystal of EuScCuSe3 0.05 × 0.05 × 0.45 mm 3 in size was sealed into a thin-walled glass capillary ( Figure 1) for X-ray diffraction experiments. The capillary was subsequently mounted on a Bruker-Nonius κ-CCD single-crystal diffractometer (Bruker, Billerica, MA, USA) equipped with a Mo-Kα radiation source, a graphite monochromator, and a CCD detector. The unit cell of this compound belongs to the orthorhombic crystal system. The space group was determined from a statistical analysis of the intensities of all reflections. The DENZO program [26] was used to process the collected intensity data. The HABITUS program [27] was used to numerically correct the absorption. The crystal structure was solved and refined by means of the SHELX-2013 software package [28,29]. CSD 2,239,558 contains supplementary crystallographic data. These data can be obtained free of charge via https://www.ccdc.cam.ac.uk/structures (accessed on 11 February 2023) or from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44)1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk. Crystal structures were visualized in the program package VESTA 3.5.7 [30].
The temperature dependence of the EuScCuSe3 magnetization was measured using a helium-cooled magnetic property measurement system (MPMS3, Quantum Design, San Diego, CA, USA) in the temperature range from 2 to 300 K in zero-field cooling (ZFC) modus and heating in an external magnetic field (FC). The field value was 500 kOe (39.8 mA m −1 ). The field-dependent magnetic moments were measured at room temperature (300 K) and at 2 K.
The ab-initio calculations of the EuScCuSe3 were carried out in the framework of density functional theory (DFT) using the PBE0 exchange-correlation functional [18], which

Methods
A selected single crystal of EuScCuSe 3 0.05 × 0.05 × 0.45 mm 3 in size was sealed into a thin-walled glass capillary ( Figure 1) for X-ray diffraction experiments. The capillary was subsequently mounted on a Bruker-Nonius κ-CCD single-crystal diffractometer (Bruker, Billerica, MA, USA) equipped with a Mo-K α radiation source, a graphite monochromator, and a CCD detector. The unit cell of this compound belongs to the orthorhombic crystal system. The space group was determined from a statistical analysis of the intensities of all reflections. The DENZO program [26] was used to process the collected intensity data. The HABITUS program [27] was used to numerically correct the absorption. The crystal structure was solved and refined by means of the SHELX-2013 software package [28,29]. CSD 2,239,558 contains supplementary crystallographic data. These data can be obtained free of charge via https://www.ccdc.cam.ac.uk/structures (accessed on 11 February 2023) or from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44)1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk. Crystal structures were visualized in the program package VESTA 3.5.7 [30].
The temperature dependence of the EuScCuSe 3 magnetization was measured using a helium-cooled magnetic property measurement system (MPMS3, Quantum Design, San Diego, CA, USA) in the temperature range from 2 to 300 K in zero-field cooling (ZFC) modus and heating in an external magnetic field (FC). The field value was 500 kOe (39.8 mA m −1 ). The field-dependent magnetic moments were measured at room temperature (300 K) and at 2 K.
The ab-initio calculations of the EuScCuSe 3 were carried out in the framework of density functional theory (DFT) using the PBE0 exchange-correlation functional [18], which takes into account both local and nonlocal Hartree-Fock exchanges. The calculations were performed in the CRYSTAL17 program designed to simulate periodic structures [31,32]. For Eu 2+ , the ECP53MWB quasi-relativistic pseudopotential was used to describe the inner shells of this lanthanoid cation. Thus, the inner shells, including 4, were replaced by a pseudopotential. To describe the outer shells (5s 2 5p 6 ) involved in chemical bonds, a valence basis set of TZVP type was used. The pseudopotential and the valence basis set are available on the site [4].
For scandium, copper, and selenium, the full-electron basis sets were used. The basis sets are available on the CRYSTAL program site as «Sc_864-11G*_harrison_2006», «Cu_86-4111(41D)G_doll_2000» and «Se_976-311d51G_towler_1995» [32]. Gaussian primitives with orbital exponent values less than 0.1 were removed from the basis sets since these calculations are periodic. The exponent in the outer orbital of selenium was set to 0.14. The accuracy of calculating the self-consistent field was set to 10 −9 a.u. The accuracy of the calculation of the two-electron integrals was set to at least 10 −8 . Integration over the Brillouin zone was carried out according to the Monkhorst-Pack scheme with a grid of kpoints equal to 8 × 8 × 8. The sequence of calculations was as follows. The optimization of the crystal structure was carried out first. After that, the phonon spectrum was calculated at the Г point, or the elastic constants were calculated for the crystal structure corresponding to the minimum energy. The octahedral [ScSe 6 ] 9− units in the EuScCuSe 3 structure are interconnected to each other through the (Se1) 2− ions along the z axis, as shown in Figure 2a, and though the (Se2) 2− anions along a axis (see Figure 2b). The [CuSe 4 ] 7− tetrahedra are linked via common (Se1) 2− anions along a axis. The [ScSe 6 ] 9− and [CuSe 4 ] 7− units have common Se1 and Se2 vertices. The nearest neighbors around Eu 2+ cations form trigonal prisms [EuSe 6 ] 10− ( Figure S1). The four Eu-Se1 bond lengths are equal to 3.0605 Å, while the remaining two Eu-Se2 bonds are 3.1711 Å long (Table S2).

Density Functional Theory Calculations
As has been mentioned above, a crystal structure prediction was previously made for EuScCuSe3, and the space group Pnma was supposed [3]. Due to the fact that the sample experimentally synthesized in our work was solved in space group Cmcm, we did a comprehensive investigation of the EuScCuSe3 crystal structure stability in both space groups, Pnma and Cmcm.
At the first step of density functional theory calculations, crystal structures of EuSc-CuSe3 in Pnma and Cmcm space groups were totally optimized, and the obtained lattice parameters are presented in Table 3. The simulated structural data get close to the experiments in both cases. It should be noted that the energy per formula unit is almost the same for both structure types and differs only in the fifth decimal place: −9634.327521165 at. un (Pnma), −9634.327541515 at. un (Cmcm).

Density Functional Theory Calculations
As has been mentioned above, a crystal structure prediction was previously made for EuScCuSe 3 , and the space group Pnma was supposed [3]. Due to the fact that the sample experimentally synthesized in our work was solved in space group Cmcm, we did a comprehensive investigation of the EuScCuSe 3 crystal structure stability in both space groups, Pnma and Cmcm.
At the first step of density functional theory calculations, crystal structures of EuScCuSe 3 in Pnma and Cmcm space groups were totally optimized, and the obtained lattice parameters are presented in Table 3. The simulated structural data get close to the experiments in both cases. It should be noted that the energy per formula unit is almost the same for both structure types and differs only in the fifth decimal place: −9634.327521165 at. un (Pnma), −9634.327541515 at. un (Cmcm).  (9) 10.1728 (7) 538.70 (7) 6.132 The next mandatory part of the crystal structure stability investigation is the simulation of elastic properties [33]. Calculations of the elastic constants were performed using the built-in functionality of CRYSTAL17 code. The obtained data for EuScCuSe 3 in Cmcm and Pnma structures are presented in Table 4. The necessary and sufficient Born criteria [34] for the orthorhombic crystal-system stability are C 11 > 0, C 11 C 22 > C 12 , C 11 C 22 C 33 + 2C 12 C 13 C 23 − C 11 C 23 2 − C 22 C 13 2 − C 33 C 12 2 > 0, C 44 > 0, C 55 > 0, C 66 > 0. All the above conditions are satisfied both for the real Cmcm and predicted Pnma-structure of EuScCuSe3 previously. As any data on elastic properties of EuScCuSe 3 are absent at this time in databases or articles, we present calculation of the bulk modulus, Young's modulus, and shear modulus in the Voigt, Reuss, and Hill approximations ( Table 5). The dependence of Young's modulus on the crystal directions demonstrates a significant anisotropy of the elastic properties in both the Cmcm and the Pnma structure ( Figure S2). The calculated values of the shear modulus and bulk modulus make it possible to estimate the Vickers hardness for EuScCuSe 3 (Table 5). To estimate the Vickers hardness, the empirical formula (3.3.1) from work [1] was used.
This formula well describes the hardness of a row of compounds with an ionic and covalent type of chemical bond (about 40 compounds were considered in work [1]). In Formula (1), G and B are the shear modulus, and bulk modulus by Hill is estimated. The experimental values of hardness are absent from research papers. According to calculations, the elastic constants and hardness of EuScCuSe 3 differ significantly for the Cmcm and Pnma structures ( Table 4).
As vibrational spectroscopy is a powerful tool for the determination of crystal structure details, simulation of Raman and infrared spectra for the experimentally obtained data in this work (Cmcm structure) and possibly earlier predicted Pnma structure [1] were done. The results for the infrared-active modes, Raman modes, and "silent" modes at the Г point are given in Tables S4 and S5 of the SM. The degree of participation of each ion in a particular mode is estimated from the analysis of displacement vectors obtained from these ab-initio calculations. The ions that are shifted significantly in the mode are listed in the column "participants" (Tables S4 and S5). The values of ion displacements for vibrational modes are shown in Figure S3.
The number of formula units in the Pnma structure is equal to 4 (Z = 4), and this value is the same for the Cmcm structure, see Table 1. However, the primitive cell of the Cmcm structure contains only two formula units ( Figure S4). Thus, the number of vibrational modes should be larger in the Pnma case. The Raman-active modes for Pnma and Cmcm structures should be listed as 12 A g + 6 B 1g +12 B 2g + 6 B 3g and 5 A g + 4 B 1g + B 2g + 5 B 3g , correspondingly [35]. The result of Raman and infrared spectra simulations for both structures are presented in Figure 3. Despite the fact that the number of vibrational modes is different for the structures in Cmcm and Pnma, the simulated Raman and infrared spectra are quite similar. Thus, we suppose that the definition of the correct space group (Cmcm or Pnma) using experimental vibrational spectroscopy is almost impossible in this case. data in this work (Cmcm structure) and possibly earlier predicted Pnma structure [1] were done. The results for the infrared-active modes, Raman modes, and "silent" modes at the Г point are given in Tables S4 and S5 of the SM. The degree of participation of each ion in a particular mode is estimated from the analysis of displacement vectors obtained from these ab-initio calculations. The ions that are shifted significantly in the mode are listed in the column "participants" (Tables S4 and S5). The values of ion displacements for vibrational modes are shown in Figure S3.
The number of formula units in the Pnma structure is equal to 4 (Z = 4), and this value is the same for the Cmcm structure, see Table 1. However, the primitive cell of the Cmcm structure contains only two formula units ( Figure S4). Thus, the number of vibrational modes should be larger in the Pnma case. The Raman-active modes for Pnma and Cmcm structures should be listed as 12 Ag + 6 B1g +12 B2g + 6 B3g and 5 Ag + 4 B1g + B2g + 5 B3g, correspondingly [35]. The result of Raman and infrared spectra simulations for both structures are presented in Figure 3. Despite the fact that the number of vibrational modes is different for the structures in Cmcm and Pnma, the simulated Raman and infrared spectra are quite similar. Thus, we suppose that the definition of the correct space group (Cmcm or Pnma) using experimental vibrational spectroscopy is almost impossible in this case.  The only possible indicator for the Pnma structure is the low-lying weak band in the Raman spectrum (Figure 3a) which is associated with very strong movements of all ions except for Cu + ( Figure S3). However, the wavenumber value of this vibrational mode is the lowest in both structures. In this regard, the calculation of phonon dispersion curves was done for the Pnma structure, and the results of the simulation in Γ-X direction are shown in Figure 4a. The key factor of the dynamical stability of crystal lattice is the absence of imaginary (unstable) phonon modes and this approach works in for the case of experimentally observed crystal structures [36] as for crystal structure stability prediction [37,38]. According to the obtained data (Figure 4a), we can say that the crystal structure of EuScCuSe3 in the previously supposed space group Pnma should be unstable. This fact, among other things, is consistent with the experimentally obtained space group Cmcm obtained for the real EuScCuSe3 in this work. At the same time, simulated phonon dispersion for the Cmcm structure do not contain unstable phonon modes over all of the highsymmetric Brillouin zone points (Cmcm). The only possible indicator for the Pnma structure is the low-lying weak band in the Raman spectrum (Figure 3a) which is associated with very strong movements of all ions except for Cu + ( Figure S3). However, the wavenumber value of this vibrational mode is the lowest in both structures. In this regard, the calculation of phonon dispersion curves was done for the Pnma structure, and the results of the simulation in Γ-X direction are shown in Figure 4a. The key factor of the dynamical stability of crystal lattice is the absence of imaginary (unstable) phonon modes and this approach works in for the case of experimentally observed crystal structures [36] as for crystal structure stability prediction [37,38]. According to the obtained data (Figure 4a), we can say that the crystal structure of EuScCuSe 3 in the previously supposed space group Pnma should be unstable. This fact, among other things, is consistent with the experimentally obtained space group Cmcm obtained for the real EuScCuSe 3 in this work. At the same time, simulated phonon dispersion for the Cmcm structure do not contain unstable phonon modes over all of the high-symmetric Brillouin zone points (Cmcm).
The band structure and the density of states for EuScCuSe 3 calculated using hybrid PBE0 functional are shown in Figure 5. The path in the Brillouin zone is plotted through the most highly symmetric points. For the space group Cmcm, the path is made along Г-Y-T-Z-S-R-Г. The coordinates of the points are (0,0,0,), ( 1 / 2 , 1 / 2 ,0), ( 1 / 2 , 1 / 2 , 1 / 2 ), (0,0, 1 / 2 ), (0, 1 / 2 ,0), (0, 1 / 2 , 1 / 2 ), (0,0,0) respectively. The Bilbao crystallographic server was used [35]. Since for europium pseudopotential that replaced their core shells, the 4f inclusive was used, the band structure does not include 4f states. For the Eu 2+ cations, only outer shells (5s 2 5p 6 ) were taken into account by means of valence basis sets [39]. The projected DOS onto the whole set of atomic orbitals of Eu, Sc, Cu, and Se atoms was calculated near the band gap. According to these calculations, the DOS of copper and selenium are located near the top of the valence band. The DOS of scandium and europium are located near the bottom of the conduction band. The band gap value is defined as the difference in energy between the top of the valence band and the bottom of the conduction band. Calculations predict for EuScCuSe 3 the indirect electronic transition with a band gap value of 3.27 eV. It should be noted, that in the case of the dynamically unstable Pnma structure, the band gap value is the same, but the calculated electronic transition is direct ( Figure S5).  The band structure and the density of states for EuScCuSe3 calculated using hybrid PBE0 functional are shown in Figure 5. The path in the Brillouin zone is plotted through the most highly symmetric points. For the space group Cmcm, the path is made along Г-Y-T-Z-S-R-Г. The coordinates of the points are (0,0,0,), ( 1 /2, 1 /2,0), ( 1 /2, 1 /2, 1 /2), (0,0, 1 /2), (0, 1 /2,0), (0, 1 /2, 1 /2), (0,0,0) respectively. The Bilbao crystallographic server was used [35]. Since for europium pseudopotential that replaced their core shells, the 4f inclusive was used, the band structure does not include 4f states. For the Eu 2+ cations, only outer shells (5s 2 5p 6 ) were taken into account by means of valence basis sets [39]. The projected DOS onto the whole set of atomic orbitals of Eu, Sc, Cu, and Se atoms was calculated near the band gap. According to these calculations, the DOS of copper and selenium are located near the top of the valence band. The DOS of scandium and europium are located near the bottom of the conduction band. The band gap value is defined as the difference in energy between the top of the valence band and the bottom of the conduction band. Calculations predict for EuScCuSe3 the indirect electronic transition with a band gap value of 3.27 eV. It should be noted, that in the case of the dynamically unstable Pnma structure, the band gap value is the same, but the calculated electronic transition is direct ( Figure  S5).   The band structure and the density of states for EuScCuSe3 calculated using hybrid PBE0 functional are shown in Figure 5. The path in the Brillouin zone is plotted through the most highly symmetric points. For the space group Cmcm, the path is made along Г-Y-T-Z-S-R-Г. The coordinates of the points are (0,0,0,), ( 1 /2, 1 /2,0), ( 1 /2, 1 /2, 1 /2), (0,0, 1 /2), (0, 1 /2,0), (0, 1 /2, 1 /2), (0,0,0) respectively. The Bilbao crystallographic server was used [35]. Since for europium pseudopotential that replaced their core shells, the 4f inclusive was used, the band structure does not include 4f states. For the Eu 2+ cations, only outer shells (5s 2 5p 6 ) were taken into account by means of valence basis sets [39]. The projected DOS onto the whole set of atomic orbitals of Eu, Sc, Cu, and Se atoms was calculated near the band gap. According to these calculations, the DOS of copper and selenium are located near the top of the valence band. The DOS of scandium and europium are located near the bottom of the conduction band. The band gap value is defined as the difference in energy between the top of the valence band and the bottom of the conduction band. Calculations predict for EuScCuSe3 the indirect electronic transition with a band gap value of 3.27 eV. It should be noted, that in the case of the dynamically unstable Pnma structure, the band gap value is the same, but the calculated electronic transition is direct ( Figure  S5).

Magnetic Properties
The temperature dependence of the specific magnetization was measured in the temperature range from 2 to 300 K ( Figure 6). Based on it, the temperature dependences of the direct and reciprocal values of the molar magnetic susceptibility are calculated.

Magnetic Properties
The temperature dependence of the specific magnetization was measured in the temperature range from 2 to 300 K ( Figure 6). Based on it, the temperature dependences of the direct and reciprocal values of the molar magnetic susceptibility are calculated.
The main contribution to the magnetic properties of EuScCuSe 3 is made by the Eu 2+ cations with unfilled f-shells. There is no significant effect of the crystal field on the magnetic moment since, in the ground state ( 8 S 7/2 ), this cation has a zero-orbital momentum. Its temperature dependence of magnetic susceptibility in the paramagnetic region should be well described by the Curie-Weiss law: χ = χ TIP + C T−θ W considering the temperatureindependent term χ TIP . Approximation of the experimental dependence by this formula gives the following values: χ TIP = 1.04·10 −5 m 3 kmol −1 , C = 0.0977 K m 3 kmol −1 , θ W = 6.0 K. The deviations of the experimental points from the approximating curve in the temperature ranging from 40 to 300 K are no more than 1%, and from 10 to 40 K about 2.5%. A comparison of the characteristics obtained with those calculated for non-interacting Eu 2+ cations is given in Table 6. The main contribution to the magnetic properties of EuScCuSe3 is made by the Eu 2+ cations with unfilled f-shells. There is no significant effect of the crystal field on the magnetic moment since, in the ground state ( 8 S7/2), this cation has a zero-orbital momentum. Its temperature dependence of magnetic susceptibility in the paramagnetic region should be well described by the Curie-Weiss law: considering the temperatureindependent term χTIP. Approximation of the experimental dependence by this formula gives the following values: χTIP = 1.04·10 −5 m 3 kmol -1 , C = 0.0977 K m 3 kmol -1 , θW = 6.0 K. The deviations of the experimental points from the approximating curve in the temperature ranging from 40 to 300 K are no more than 1%, and from 10 to 40 K about 2.5%. A comparison of the characteristics obtained with those calculated for non-interacting Eu 2+ cations is given in Table 6. There is a sharp deviation from the Curie-Weiss law at temperatures below 5 K. This deviation is obviously due to the ferromagnetic transition, although there is no noticeable discrepancy in the data for the FC and ZFC The experimental curve of magnetization at a temperature of 2 K (Figure 7b) has the form characteristic of magnetically soft ferromagnets. The coercive force is less than 2 kA m −1 , and saturation occurs in a field of about 500 kA m −1 . The magnetization in a field of 4 MA m −1 per formula unit is 6.5 μB, which is close to the theoretical value of about 7 μB for a free Eu 2+ cation. Figure 6. Temperature dependences of specific magnetization σ (left axis) and reciprocal molar susceptibility χ −1 (right axis) in the temperature range from 2 K to 300 K (a) and to 12 K (b).
There is a sharp deviation from the Curie-Weiss law at temperatures below 5 K. This deviation is obviously due to the ferromagnetic transition, although there is no noticeable discrepancy in the data for the FC and ZFC The experimental curve of magnetization at a temperature of 2 K (Figure 7b) has the form characteristic of magnetically soft ferromagnets. The coercive force is less than 2 kA m −1 , and saturation occurs in a field of about 500 kA m −1 . The magnetization in a field of 4 MA m −1 per formula unit is 6.5 µ B , which is close to the theoretical value of about 7 µ B for a free Eu 2+ cation.

Conclusions
In summary, we report on the new quaternary scandium selenide EuScCuSe3, which was synthesized from a mixture of the elements with CsI as a flux in sealed silica ampoules at elevated temperatures. The structural, vibrational, and elastic-property calculations

Conclusions
In summary, we report on the new quaternary scandium selenide EuScCuSe 3 , which was synthesized from a mixture of the elements with CsI as a flux in sealed silica ampoules at elevated temperatures. The structural, vibrational, and elastic-property calculations have been performed for EuScCuSe 3 in the framework of the density functional theory (DFT) by using the PBE0 hybrid functional and LCAO-MO approach. The calculation results predict the Cmcm structure, which agrees very well with the obtained crystallographic data (a = 3.9883(3), b = 13.2776(9), c = 10.1728(7) Å). The calculation results can be used to interpret the Raman and infrared spectra.
The crystal structure, according to single-crystal data, showed that EuScCuSe 3 belongs to the orthorhombic crystal system with the space group Cmcm. The structure type corresponds to KZrCuS 3, and thus the structure includes trigonal prisms [EuSe 6 ] 10− , octahedra [ScSe 6 ] 9− , and tetrahedra [CuSe 4 ] 7− . The title compound is paramagnetic above 4.5 K and soft ferromagnetic at lower temperatures.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/ma16041555/s1, Figure S1: Crystal structure of EuScCuSe 3 . Projection onto the bc plane (a) and onto the ab plane (b). The trigonal prisms [EuSe 6 ] 10− are colored in turquoise; Figure S2: Dependence of Young's modulus in GPa on the crystallographic directions in EuScCuSe 3 for both possible orthorhombic structures; Figure S3: Displacement of ions at the phonon modes in the crystal structure of EuScCuSe 3 in both possible descriptions (Cmcm and Pnma); Figure S4: Primitive cell of EuScCuSe 3 ; Figure S5: Band structure and electronic density of states of EuScCuSe 3 calculated for the dynamically unstable Pnma structure; Table S1: Anisotropic displacement parameters in Å 2 of EuScCuSe 3 ; Table S2: Main bond lengths in Å of EuScCuSe 3 ; Table S3: Geometric parameters for EuScCuSe 3 ; Table S4: Wavenumbers in cm −1 and types of the phonon modes at the Г-point for EuScCuSe 3 in the Cmcm structure. The intensity of the Raman modes was calculated for λ = 532 nm and T = 298 K; Table S5: Wavenumbers in cm −1 and types of the phonon modes at the Г-point for EuScCuSe 3 in the Pnma structure. The intensity of the Raman modes was calculated for λ = 532 nm and T = 298 K.