Structural, Electronic and Vibrational Properties of YAl3(BO3)4

The crystal structure of YAl3(BO3)4 is obtained by Rietveld refinement analysis in the present study. The dynamical properties are studied both theoretically and experimentally. The experimental Raman and Infrared spectra are interpreted using the results of ab initio calculations within density functional theory. The phonon band gap in the Infrared spectrum is observed in both trigonal and hypothetical monoclinic structures of YAl3(BO3)4. The electronic band structure is studied theoretically, and the value of the band gap is obtained. It was found that the YAl3(BO3)4 is an indirect band gap dielectric material.

The Eu 3+ ions doped into the YAl3(BO3)4 lattice pursuing a red phosphor with good colorimetric features for display panels applications [25].
The Eu 3+ ions doped into the YAl 3 (BO 3 ) 4 lattice pursuing a red phosphor with good colorimetric features for display panels applications [25].
The emission with output power within the range of 0.8-1 W is obtained at different wavelengths: 1602, 1550, 1543 and 1520 nm [28]. The spectroscopic properties of Er,Yb:YAl 3 (BO 3 ) 4 crystals have been investigated at both ambient and high temperature (300-800K) conditions [29]. It has been shown that the high-performance eye-safe 1.55 µm microchip laser can be fabricated by the tightly pressurization of two sapphire crystals with high thermal conductivity and the Er:Yb:YAl 3 (BO 3 ) 4 laser crystal between them [30,31]. The narrow lines attributed to the Mn 4+ ions (682, 684 and 686 nm) are observed in the luminescence spectra of YAB:Mn under 514.5 nm excitation [32].
It was previously established that the rare-earth borates represent three polymorphic modifications: the huntite structure (space group R32) and two monoclinic structures with C2/c and C2 space groups [33]. However, the weak bands of possible monoclinic (C2/c) polytype of ReM 3 -borates (Re is Nd, Gd and Y; M is Al, Ga, Cr, and Fe) have been found in the Infrared spectrum of samples with huntite structure [34]. Co-existence of trigonal and monoclinic phases can produce, for example, the effect of local stresses or decreasing of the nonlinear properties. The variation from non-centrosymmetric (R32) to centrosymmetric (C2/c) structure will affect to nonlinear optical and magnetoelectric properties.
The structural characterization of YAl 3 (BO3) 4 host lattice is mainly related to X-ray diffraction [25,[35][36][37]. The main purpose of this study is to study vibrational spectra of YAB and reveal or exclude a possible portion existence of the monoclinic (C2/c) phase in trigonal (R32) YAl 3 (BO 3 ) 4 lattice. The results of this work can be used in part to study vibrational properties of a set of ReM 3 (BO 3 ) 4 family members. The investigation of electronic, structural and vibrational properties of YAB is based on performing ab initio calculations in the framework of density functional theory calculations and a group of experimental techniques: Infrared, energy-dispersive X-ray and Raman spectroscopy, and X-ray diffraction analysis.

Synthesis
Single crystals of YAl 3 (BO 3 ) 4 [38]. The saturation temperature of this solution-melt was determined as T sat = 980 • C. The concentration (n) dependence of the saturation temperature had a slope dT sat /dn =15 • C/wt %.
The solution-melt of 150 g total weight was prepared in a cylindrical platinum crucible (D = 50 mm, h = 60 mm) by sequential melting of oxides (Bi 2 O 3 + MoO 3 ), B 2 O 3 , Y 2 O 3 , (Li 2 CO 3 + MoO 3 ) at T = 1000−1100 • C. The saturation temperature was defined with accurate to within ± 2 • C with the use of spontaneous probe crystals grown previously.
Group method was used to growth crystals. Four seeds with size~1 mm 3 were attached to the rod crystal holder. The initial supercooling was corresponded to the middle of the metastability zone and footed up to 10 • C. After this, the temperature of the solution-melt was reduced according to the program with an increasing rate of 1-3 • C/day. The rate of crystal growth did not exceed 0.5 mm/24 h. The rod crystal holder was rotated reversibly with a 1-min period. After the growth was finished, the rod crystal holder had been raised above the solution-melt and the furnace was cooled to room temperature with the rate of 100 • C/h. The YAl 3 (BO 3 ) 4 crystals with dimensions of 6-10 mm and a total mass of 10 g were obtained in the crystallization temperature interval of 17 • C.

Experimental
The Infrared (IR) absorption spectrum was recorded with a Fourier-transform spectrometer VERTEX 70 V (Bruker, Billerica, MA, USA) in the spectral range from 400 to 1600 cm −1 with spectra resolution 4 cm −1 . The spectrum was taken from a tablet sample shaped as about 0.4 mm thick tablet of 13 mm in diameter and a weight of 0.15 g. The tablet was prepared as follows: 0.00338 g of YAl 3 (BO 3 ) 4 was thoroughly ground with 0.20 g of KBr. The Globar was used for light source, and it was equipped with a KBr wide beamsplitter and RT-DLaTGS as a detector (Bruker, Billerica, MA, USA).
The morphology of the sample was characterized with a Tabletop Microscope TM3000 (Hitachi, Tokyo, Japan) equipped with an EDX X-Flash 430 (Bruker, Billerica, MA, USA) with an acceleration voltage of 15 kV. Its chemical composition (mapping) was investigated with a detection time of 100 s. To avoid a surface charge-up as well as to improve an imaging quality of the SEM (scanning electron microscope) micrographs a thin platinum film was deposited with a sputter coater K575XD (Emitech, Houston, TX, USA) with 3 cycles. The average coating time was of the order of 1 min using a plasma current of 10 mA.
The X-Ray powder diffraction data of YAl 3 (BO 3 ) 4 was collected at room temperature with a Bruker D8 ADVANCE powder diffractometer (Cu-Kα radiation, 40 kV, 40 Ma, Bruker, Billerica, MA, USA) and linear VANTEC detector. The step size of 2θ was 0.016 • , and the counting time was 1 s per step. The intensities from single crystal YAl 3 (BO 3 ) 4 of 0.2 × 0.1 × 0.1 mm dimensions were collected at 296 K using the SMART APEXII X-ray four-circle single crystal diffractometer (Bruker) equipped with a CCD-detector, graphite monochromator and Mo Kα radiation source. The cell parameters were refined by 1525 reflections. The X-ray data from crystal were measured with the exposure time of 10 s on each frame. Crystal rotated along ω-axis by 0.5 • at the fixed ϕ angle and the ω value was increased from 0 • to 182 • . Totally the 364 frames were measured at each fixed ϕ equal to 0 • , 120 • and 240 • . After that, the program APEXII from Bruker integrated the intensities of reflections. Space group R32 was defined by the analysis of extinction rules and intensity statistics obtained from all reflections. Multiscan absorption correction of reflection intensities was performed by APEXII software (Bruker, Billerica, MA, USA). Then, the intensities of equivalent reflections were averaged.
The Raman spectra study of the single crystal sample was carried out at room temperature in a back-scattering geometry. The laser irradiation of solid-state laser (532.1 nm, Spectra-Physics Millennia) was used for the Raman experiment after passing a monochromator to suppress parasitic laser lines. A triple-grating spectrometer TriVista 777 (Princeton Instruments, Acton, USA) was used for the Raman scattering registration in a frequency range from 18 to 1700 cm −1 with spectral resolution ∼1 cm −1 . For the wavelength calibration of the spectrometer, a neon-discharge lamp was used.

Calculation Details
Density functional (DFT) calculations were performed using the plane-wave pseudopotential method as implemented in the CASTEP code [39]. The structural parameters of YAl 3 (BO 3 ) 4 were fully optimized using the local density approximation (LDA) provided by the Perdew and Zunger [40] parameterization of the numerical results of Ceperley and Alder (CA-PZ) [41]. The calculations were performed using norm conserving pseudopotentials with 2s 2 2p 1 electrons for B, 2s 2 2p 4 electrons for O, 3s 2 3p 1 electrons for Al, and 4d 1 5s 2 electrons for Y atom treated as a valence ones. The tolerance in a self-consistent field (SCF) procedure was set to be 5.0 × 10 −8 eV/atom and total energy was corrected for a finite basis set. The convergence tolerance for geometry optimization was selected with the differences in maximal force and stress tensor within 0.0001 eVÅ −1 and 0.01 GPa correspondingly. The energy cutoff of 900 eV was used with 4 × 4 × 4 sampling of the Brillouin zone (BZ) using the Monkhorst-Pack scheme [42]. The phonon spectra at the Γ-point of the BZ was calculated within density functional perturbation theory and finite displacement method [43,44] based on the crystal system type. The dispersion of phonon branches along high symmetry directions of the BZ was calculated using a linear response formalism [45].

Results and Discussion
The main information about crystal data, data collection and refinement are reported in Table 1. The structure was solved by the direct methods using package SHELXS and refined in the anisotropic approach for non-boron atoms using SHELXL program [46]. The structural tests for the presence of missing symmetry elements and possible voids were produced using the PLATON program [47]. The main crystal data are shown in Table 1. The coordinates of atoms are reported in Table S1 and main bond lengths are shown in Table S2 of supplementary materials. Table 1. Main parameters of processing and refinement of the YAl 3 (BO 3 ) 4 single crystal.

Molecular weight
405.09

Refinement Results
The weighed refinement of 0.072 (5) Almost all peaks of the powder X-ray diffraction pattern, besides impurity SiO 2 peaks, were indexed by trigonal cell (R32) with parameters close to the previously published YAl 3 (BO 3 ) 4 [10] and identical to parameters of investigated single crystal (see Table 1). The SiO 2 impurity was appeared after grinding YAl 3 (BO 3 ) 4 in the agate mortar, while the initial YAl 3 (BO 3 ) 4 bulk material was pure. The structure obtained from single crystal examination was taken as a starting model for multiphase Rietveld refinement method [48] which was performed using TOPAS 4.2 [49] software package. Refinement was stable and gave low R-factors (Table 2, Figure 2). Coordinates of atoms and main bond lengths are presented in Tables S3 and S4 of supplementary materials, respectively. The crystallographic data are deposited in Cambridge Crystallographic Data Centre (CCDC #1960228). The data can be downloaded from the site (www.ccdc.cam.ac.uk/data_request/cif). Table 2. Main parameters of processing and refinement of the YAl 3 (BO 4 ) 3 powder.  [50,51] and points are shown in Figure 3a. The results of the calculation of the yttrium aluminum borate band structure are presented in Figure 3b.

YAl 3 (BO 4 ) 3 Powder
The value of the band gap is defined as the difference between the conduction band minimum (CBM) and the valence band maximum (VBM). It is found that the VBM is well localized in the vicinity of the T-point and the CBM is located between the P2 and Γ points. The band gap value for indirect electronic transitions is E i g = 5.161 eV. The lowest energy direct transition is found in the vicinity of P2 point of the BZ (the point in the P2→Г direction), also the direct transition with approximately the same energy is obtained in L-point of the BZ (see Figure 3b). The value of the direct bandgap is equal to E d g = 5.308 eV. The obtained value of band gap is underestimated compared with the experiment value of 5.7 eV [52] which can be explained as a systematic DFT problem due to wellknown band gap underestimation problem [53].
The hybrid functional HSE06 [54] method was developed to improve the accuracy of the band structure calculations. The value of bandgap E i g =7.2 eV was calculated using the hybrid functional method. The obtained value is significantly overestimate the experimental value reported in [52]. There is no experimental absorption spectrum presented in the paper [52] only the theoretical one, therefore no evidence that the value of the band gap is correctly extracted (for example extrapolated with Kubelka-Munk equation [55]) from experimental data. We suggest the new experiments on the determination of the band gap would clarify more accurate value. Next, obtained structural parameters were taken as initial for the ab initio geometry optimization included the unit cell parameters and atomic positions. The optimized structure is consistent with experimental data as shown in Table S5 Figure 3b.
The value of the band gap is defined as the difference between the conduction band minimum (CBM) and the valence band maximum (VBM). It is found that the VBM is well localized in the vicinity of the T-point and the CBM is located between the P 2 and Γ points. The band gap value for indirect electronic transitions is E i g = 5.161 eV. The lowest energy direct transition is found in the vicinity of P 2 point of the BZ (the point in the P 2 →Γ direction), also the direct transition with approximately the same energy is obtained in L-point of the BZ (see Figure 3b). The value of the direct bandgap is equal to E d g = 5.308 eV. The obtained value of band gap is underestimated compared with the experiment value of 5.7 eV [52] which can be explained as a systematic DFT problem due to well-known band gap underestimation problem [53].
The hybrid functional HSE06 [54] method was developed to improve the accuracy of the band structure calculations. The value of bandgap E i g =7.2 eV was calculated using the hybrid functional method. The obtained value is significantly overestimate the experimental value reported in [52]. There is no experimental absorption spectrum presented in the paper [52] only the theoretical one, therefore no evidence that the value of the band gap is correctly extracted (for example extrapolated with Kubelka-Munk equation [55]) from experimental data. We suggest the new experiments on the determination of the band gap would clarify more accurate value. According to the Y. Wang et. al. [56] the VBM is at the M point and the CBM is at A point of the BZ (crystallographic hexagonal unitcell); calculated band structure of YAB is presented in work of M.G. Brik et. al. [57] but the nature of electronic transitions is not discussed; according to the work of R. He [58] the YAB is a material with a direct band gap (Γ-point) but it is noteworthy that the band structure was calculated along two paths in BZ only, therefore the bandstructure in [58] is not complete. The direct transition in Г-point was also obtained in Ali H. Reshak's work [59], but a significantly lower value of the cutoff energy was used in the calculations, therefore the basis set is not complete. No experimental investigation of the band structure was found, hence if the YAB is a direct transition crystal or not is an open question. The hexagonal unitcell is three times bigger than the primitive rhombohedral one, hence the volume of the Brillouin zone is three times lower. We perform calculations using rhombohedral unit cell and along all with known high-symmetry directions, therefore the results obtained in recent research more straightforward. According to the Y. Wang et. al. [56] the VBM is at the M point and the CBM is at A point of the BZ (crystallographic hexagonal unitcell); calculated band structure of YAB is presented in work of M.G. Brik et. al. [57] but the nature of electronic transitions is not discussed; according to the work of R. He [58] the YAB is a material with a direct band gap (Γ-point) but it is noteworthy that the band structure was calculated along two paths in BZ only, therefore the bandstructure in [58] is not complete. The direct transition in Γ-point was also obtained in Ali H. Reshak's work [59], but a significantly lower value of the cutoff energy was used in the calculations, therefore the basis set is not complete. No experimental investigation of the band structure was found, hence if the YAB is a direct transition crystal or not is an open question. The hexagonal unitcell is three times bigger than the primitive rhombohedral one, hence the volume of the Brillouin zone is three times lower. We perform calculations using rhombohedral unit cell and along all with known high-symmetry directions, therefore the results obtained in recent research more straightforward.
The total and partial density of states of the YAl 3 (BO 3 ) 4 structure are plotted in Figure 4 and Figure S1 of supplementary materials. As a result of the figure analysis, one can find that the valence band maximum is formed mostly by p-electrons of oxygen atoms while the conduction band minimum is constructed mostly by d-electrons of yttrium, p-electrons of boron and p-electrons of oxygen atoms. It clearly seen, that the contribution of the aluminum ions to the total DOS significantly less than other ions. Therefore, one can assume the Al and Y atoms of YAl 3 (BO 3 ) 4 crystal are found to be in (III) valence state. The total and partial density of states of the YAl3(BO3)4 structure are plotted in Figure 4 and Figure S1 of supplementary materials. As a result of the figure analysis, one can find that the valence band maximum is formed mostly by p-electrons of oxygen atoms while the conduction band minimum is constructed mostly by d-electrons of yttrium, p-electrons of boron and p-electrons of oxygen atoms. It clearly seen, that the contribution of the aluminum ions to the total DOS significantly less than other ions. Therefore, one can assume the Al and Y atoms of YAl3(BO3)4 crystal are found to be in (III) valence state. The absorption coefficient calculated by LDA using a scissor operator (the difference between the theoretical and experimental [52] band gap values) equal to 0.539 eV is plotted in Figure 5. From the partial density of states analysis (Figure 4 and S1), it follows that the first peak the spectrum is associated with electronic transitions mainly from the 2p orbitals of the O atom to the 4d orbitals of Y atoms.  The absorption coefficient calculated by LDA using a scissor operator (the difference between the theoretical and experimental [52] band gap values) equal to 0.539 eV is plotted in Figure 5. From the partial density of states analysis (Figure 4 and Figure S1), it follows that the first peak the spectrum is associated with electronic transitions mainly from the 2p orbitals of the O atom to the 4d orbitals of Y atoms. The total and partial density of states of the YAl3(BO3)4 structure are plotted in Figure 4 and Figure S1 of supplementary materials. As a result of the figure analysis, one can find that the valence band maximum is formed mostly by p-electrons of oxygen atoms while the conduction band minimum is constructed mostly by d-electrons of yttrium, p-electrons of boron and p-electrons of oxygen atoms. It clearly seen, that the contribution of the aluminum ions to the total DOS significantly less than other ions. Therefore, one can assume the Al and Y atoms of YAl3(BO3)4 crystal are found to be in (III) valence state. The absorption coefficient calculated by LDA using a scissor operator (the difference between the theoretical and experimental [52] band gap values) equal to 0.539 eV is plotted in Figure 5. From the partial density of states analysis (Figure 4 and S1), it follows that the first peak the spectrum is associated with electronic transitions mainly from the 2p orbitals of the O atom to the 4d orbitals of Y atoms.  The nonpolarized Raman and Infrared spectra are shown in Figures 6 and 7 correspondingly. Polarized Raman spectra are plotted in Figures S2 and S3 and simulated Raman spectra for specific Raman tensor components are shown in Figures S2-S4 of supplementary materials. The mechanical representation for the YAl 3 (BO 3 ) 4 at Brillouin zone center is Γ vibr = 7A 1 + 13A 2 + 20E [60] where Raman active modes are Γ Raman = 7A 1 + 19E, and infrared active modes are Γ Infrared = 12A 2 + 19E. The acoustic modes are Γ Acoustic = A 2 + E. The A and E letters correspond to nondegenerate and doubly degenerate vibrations correspondingly. The E modes are polar and active as in Raman as in IR spectra. The nonpolarized Raman and Infrared spectra are shown in Figures 6 and 7 correspondingly. Polarized Raman spectra are plotted in Figures S2 and S3 and simulated Raman spectra for specific Raman tensor components are shown in Figures S2-S4 of supplementary materials. The mechanical representation for the YAl3(BO3)4 at Brillouin zone center is Γvibr = 7A1 + 13A2 + 20E [60] where Raman active modes are ΓRaman = 7A1 + 19E, and infrared active modes are ΓInfrared = 12A2 + 19E. The acoustic modes are ΓAcoustic = A2 + E. The A and E letters correspond to nondegenerate and doubly degenerate vibrations correspondingly. The E modes are polar and active as in Raman as in IR spectra.   The nonpolarized Raman and Infrared spectra are shown in Figures 6 and 7 correspondingly. Polarized Raman spectra are plotted in Figures S2 and S3 and simulated Raman spectra for specific Raman tensor components are shown in Figures S2-S4 of supplementary materials. The mechanical representation for the YAl3(BO3)4 at Brillouin zone center is Γvibr = 7A1 + 13A2 + 20E [60] where Raman active modes are ΓRaman = 7A1 + 19E, and infrared active modes are ΓInfrared = 12A2 + 19E. The acoustic modes are ΓAcoustic = A2 + E. The A and E letters correspond to nondegenerate and doubly degenerate vibrations correspondingly. The E modes are polar and active as in Raman as in IR spectra.   The symmetry of [BO 3 ] 3− ions and type of vibrations was described by Nakamoto [61]. It was found that the point group of [BO 3 ] 3− is D 3h and the decomposition of vibrational spectra by irreducible representations is as follows: A 1 ' + 2A 2 " + 3E' + A 2 ' + E". The mode ν 1 (A 1 ') is a symmetric stretching vibration, ν 2 (A 2 ") is off-plane deformational vibration, ν 3 and ν 4 (E') is in-plane deformational vibration. Normal modes of vibrations of [BO 3 ] 3− ions discussed above are presented in Figure S5 of supplementary materials. Finally, there are three translational vibrations, one (along high symmetry axis) with symmetry A 2 " and two E", and three rotational vibrations A 2 ' and E". The boron atoms in the host YAl 3 (BO 3 ) 4 unit cell is found to take two Wyckoff positions, namely 3b (site symmetry D 3 ) and 9e (site symmetry C 2 ). The correlation diagram of internal vibrations between the free [BO 3 ] 3− ions with D 3h symmetry, its site symmetries (D 3 and C 2 ) and factor group symmetry D 3 of host unit cell is shown in Table 3. The calculated phonon frequencies of the YAl 3 (BO 3 ) 4 are given in Table S6. Table 3. Correlation diagram of internal vibrations of the BO 3 3− in the YAB.

Free ion Symmetry Site Symmetry Factor Group Symmetry Site Symmetry Factor Group Symmetry
According to Table 3 and Table S6 in Table 3. According to the results of calculations, these vibrations should be in the range of 1000-1500 cm −1 . The spectral band at 990 cm -1 corresponds to E(TO) + E(LO) modes and shift in band positions due to TO-LO splitting is insignificant. The spectral range of 1250-1500 cm −1 should consist of remain modes (A 2 + 3E(TO) + 3E(LO)). However, the decomposition of experimental spectra in the range of stretching vibration revealed extra bands that is not in accordance with calculations. The clearly seen extra band around 1100 cm −1 is marked with an asterisk in Figure 7. The typical Infrared spectra of huntite-like ReM 3 (BO 3 ) 4 (Re = Y, rare-earth element, M = Al, Ga, Fe, Cr) compounds with noncentrosymmetric trigonal structure (R32 space group) should contain an empty gap in the range 1050-1200 cm −1 [62][63][64]. However, as discussed earlier [65][66][67], the borates with large rare-earth elements can form not only trigonal but also monoclinic structures depending on the growth conditions. Some extra bands (in comparison with trigonal structure) were observed in the range of 1050-1200 cm −1 and these bands assigned to ν 3 vibrations of BO 3 3− ions [67,68]. The presence of the band at 1100 cm −1 has been attributed to the presence of monoclinically ordered domains incorporated into the trigonal structure [68]. We have carried out first principles calculations of the vibrational spectrum of YAl 3 (BO 3 ) 4 isostructural to published monoclinic structure of β-NdAl 3 (BO 3 ) 4 [70]. The comparison of experimental Infrared spectra in the range of [BO 3 ] 3− stretching vibrations (950-1500 cm −1 ) and calculated wavenumber values are shown in Figure 8. According to the factor group analysis ( Table 4) and results of ab initio calculations, one can conclude that two crystallographically independent BO 3 3− ions should produce four spectral bands in the range of ν 1 vibrations, empty gap between 1050-1250 cm −1 and eight spectral bands related to ν 3 vibrations in the range of 1250-1450 cm −1 . Similar characteristics of Infrared spectra observed only for EuAl 3 (BO 3 ) 4 [69] but not for other ReM 3 (BO 3 ) 4 [34,63,[66][67][68].
In case of Sm 3+ doped YAl 3 (BO 3 ) 4 several bands have been found at 869, 918 and 1064 cm −1 but X-ray diffraction diffractograms do not contain reflexes related to monoclinic phases [71]. We have carried out first principles calculations of the vibrational spectrum of YAl3(BO3)4 isostructural to published monoclinic structure of β-NdAl3(BO3)4 [70]. The comparison of experimental Infrared spectra in the range of [BO3] 3− stretching vibrations (950-1500 cm −1 ) and calculated wavenumber values are shown in Figure 8. According to the factor group analysis ( Table  4) and results of ab initio calculations, one can conclude that two crystallographically independent BO3 3− ions should produce four spectral bands in the range of ν1 vibrations, empty gap between 1050-1250 cm −1 and eight spectral bands related to ν3 vibrations in the range of 1250-1450 cm −1 . Similar characteristics of Infrared spectra observed only for EuAl3(BO3)4 [69] but not for other ReM3(BO3)4 [34,63,[66][67][68]. In case of Sm 3+ doped YAl3(BO3)4 several bands have been found at 869, 918 and 1064 cm −1 but X-ray diffraction diffractograms do not contain reflexes related to monoclinic phases [71].

Free ion Symmetry Site Symmetry Factor Group Symmetry
D3h The energy-dispersive X-ray (EDX) microanalysis was used to study the elemental composition of YAl3(BO3)4 crystals (Figure 9a). The component spectrum (Figure 9b) contains peaks of boron, oxygen, aluminum, yttrium, carbon and copper. The last one is related to the copper substrate. It is well known that EDS, in contrast to Auger spectroscopy, is a more accurate method for heavy elements (atomic number > 33). That is why the carbon quantity in the spectrum is overestimated. Moreover, there are a lot of carbon contaminants in any SEM chambers, that affect carbon quantity in spectra. In addition, a carbon conductive double-coated tape was used to mount the sample to operate in SEM. We cannot exclude from the discussion the molybdenum because molybdenum oxides are part of the synthesis components, however, molybdenum is not found.   3 3− in case of hypothetical monoclinic structure of the YAB.

Free ion Symmetry Site Symmetry Factor Group Symmetry
The energy-dispersive X-ray (EDX) microanalysis was used to study the elemental composition of YAl 3 (BO 3 ) 4 crystals (Figure 9a). The component spectrum (Figure 9b) contains peaks of boron, oxygen, aluminum, yttrium, carbon and copper. The last one is related to the copper substrate. It is well known that EDS, in contrast to Auger spectroscopy, is a more accurate method for heavy elements (atomic number > 33). That is why the carbon quantity in the spectrum is overestimated. Moreover, there are a lot of carbon contaminants in any SEM chambers, that affect carbon quantity in spectra. In addition, a carbon conductive double-coated tape was used to mount the sample to operate in SEM. We cannot exclude from the discussion the molybdenum because molybdenum oxides are part of the synthesis components, however, molybdenum is not found. On the other hand, the frequency of Si-O-Si stretching vibration in SiO2 (1100 cm -1 ) [72] perfectly matches the frequency of extra band in Infrared spectra of YAl3(BO3)4 obtained in Infrared spectrum ( Figure 7). Therefore, the nature of the spectra band is an open question.

Conclusions
As a result of this work, we can conclude that the investigated sample of YAl3(BO3)4 belongs to a group of borates with huntite structure. The calculated band structure shows YAl3(BO3)4 to be indirect band gap dielectric with E i g = 5.161 eV. The value of the direct bandgap is equal to E d g = 5.308 eV, which is close to the value of indirect transition. It was clearly shown that the structural analysis of YAl3(BO3)4 should be done on a framework of several methods, for example, a combination of computational, diffraction and spectroscopic methods. It was obtained that the excess bands in the range of 1050-1200 cm −1 of the Infrared spectrum do not correspond to the possible monoclinic phase of YAl3(BO3)4 suggested by Dobretsova et al. [68].
Based on current research results the future activities can be aimed to obtain the vibrational spectra of monoclinic domains in ReAl3(BO3)4, where Re = Y or rare-earth elements (with the exception of EuAl3(BO3)4 [69]), or vibrational spectra of ReAl3(BO3)4 with totally monoclinic structure.

Conclusions
As a result of this work, we can conclude that the investigated sample of YAl 3 (BO 3 ) 4 belongs to a group of borates with huntite structure. The calculated band structure shows YAl 3 (BO 3 ) 4 to be indirect band gap dielectric with E i g = 5.161 eV. The value of the direct bandgap is equal to E d g = 5.308 eV, which is close to the value of indirect transition. It was clearly shown that the structural analysis of YAl 3 (BO 3 ) 4 should be done on a framework of several methods, for example, a combination of computational, diffraction and spectroscopic methods. It was obtained that the excess bands in the range of 1050-1200 cm −1 of the Infrared spectrum do not correspond to the possible monoclinic phase of YAl 3 (BO 3 ) 4 suggested by Dobretsova et al. [68].
Based on current research results the future activities can be aimed to obtain the vibrational spectra of monoclinic domains in ReAl 3 (BO 3 ) 4 , where Re = Y or rare-earth elements (with the exception of EuAl 3 (BO 3 ) 4 [69]), or vibrational spectra of ReAl 3 (BO 3 ) 4 with totally monoclinic structure.