Crystal Structure, Vibrational, Spectroscopic and Thermochemical Properties of Double Sulfate Crystalline Hydrate [CsEu(H 2 O) 3 (SO 4 ) 2 ] · H 2 O and Its Thermal Dehydration Product CsEu(SO 4 ) 2

: Crystalline hydrate of double

Among the REE elements, europium compounds are of particular interest, since Eu 3+ ions provide efficient photoluminescence in the red spectral range highly needed for creating white LEDs with similar to daylight emission characteristics [37][38][39][40][41][42][43][44][45][46]. In recent years, a large number of studies related to the synthesis and properties of crystal phosphors doped with Eu 3+ ions have appeared. However, in such systems, the doping level is usually very low and, often, the distribution of Eu 3+ ions in the corresponding crystallographic positions is not obvious. For this reason, in complex compounds, it is difficult to clearly determine the relation between the coordination and spectroscopic parameters of Eu 3+ ions in the host lattice. In such a situation, the compounds with a stoichiometric content of europium ions have attracted the increasing attention of researchers [37,38,[47][48][49][50][51][52][53]. Self-activated phosphors are characterized by an almost complete absence of structural defects, and the precise determination of the crystal structure makes it possible to evaluate the relations between the Eu 3+ ion coordination in the lattice and spectroscopic characteristics of the compound. Simple europium stoichiometric compounds with tetrahedral MO 4 units, where M = Mo, W and S, were thoroughly studied and their applicability as highly efficient polyfunctional materials was shown [54][55][56][57][58][59][60][61][62][63][64][65][66][67]. At the same time, the properties of complex compounds of monovalent cations and rare-earth elements with tetrahedral anions are presented quite sporadically in the literature [37,[48][49][50][51]53,[68][69][70][71][72]. The structures and some properties of several double molybdates and tungstates with general composition AEu(MO 4 ) 2 (A = Li, Na, K, Cs, Rb, Ag + ; M = Mo, W) were investigated in the past and the examples of such contributions can be found elsewhere [72][73][74][75][76]. Contrary to that, the characterizations of the complex sulfate compounds of europium and monovalent cations are very scarce in the literature [53,68,69,71,[77][78][79][80][81]

Methods and Materials
In the synthesis, the solutions of CsNO 3 , Eu(NO 3 ) 3 and H 2 SO 4 were used as starting materials. For the solution preparation, twice distilled deionized water was used. The volumes of liquids were measured using glass pipettes and cylinders with the accuracy of 0.1 mL. Solid reagents were weighed on an analytical balance with the accuracy of 0.1 mg.
An europium nitrate solution was prepared using Eu 2 O 3 (99.995%, TDM-96 Ltd., Russia). To remove carbonate and europium hydroxide impurities occasionally formed during their storage, the commodity oxide was calcined at 900 • C for 12 h, after which it was cooled to room temperature in a desiccator over silica gel. Subsequently, the calcined europium oxide, mass: 17.5963 g was transferred into a flask, and 21.6 mL of concentrated nitric acid (C(HNO 3 ) = 14.6 mol/L, ρ = 1.3956 g/cm 3 , ultrapure, Vekton Ltd., Nizhegorodsky, Russia) was added in small portions to the europium oxide. The mixture was carefully stirred until the oxide was completely dissolved according to equation: After the dissolution, the solution volume in the flask was adjusted to the mark with deionized water and mixed well for homogeneity. A sulfuric acid solution with the molar concentration of 2 mol/L was prepared by diluting concentrated sulfuric acid. To make this, 50 mL of water was poured into a 100.00 mL volumetric flask, then 11.17 mL of concentrated sulfuric acid (C(H 2 SO 4 ) = 17.9 mol/L, ρ = 1.8349 g/cm 3 , ultrapure, Vekton Ltd., Nizhegorodsky, Russia) was carefully poured in small portions, avoiding a strong heating of the solution. After this, the solution was naturally cooled to room temperature and the volume was adjusted to the mark with deionized water.
[CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O was obtained by a slow crystallization of the solution containing stoichiometric amounts of ions. For this, in a glass beaker, 10 mL of the CsNO 3 (C(Cs + ) = 1 mol/L) solution, 10 mL of the Eu(NO 3 ) 3 (C(Eu 3+ ) = 1 mol/L) solution and 10 mL of the H 2 SO 4 (C(SO 4 2− ) = 2 mol/L) solution were mixed. The mixed solution was inserted into a desiccator over silica gel at 25 • C. In 12 h, the crystals precipitated from the mother liquor fell out in the reaction mixture. They were extracted, washed with ice water, pressed between filter paper sheets and dried in an empty desiccator to a constant weight. Anhydrous sulfate CsEu(SO 4 Figure 1a. As it is seen, the crystals were transparent and they were well faceted, and that was a robust indicator of their high structural quality, as it was earlier observed for different materials [79,[87][88][89]. The crystals were partly twinned due to the existence of several crystallization centers. The SEM pattern of CsEu(SO 4 ) 2 particles is given in Figure 1b. The product mainly contained loose aggregates. Such type of the particle micromorphology is commonly formed in powder compounds fabricated by the high-temperature decomposition process due to gas release effects [90,91].
The optical microscopy images of the crystals were fixed with an MS-2 microscope (State Optical Institute, Saint Petersburg, Russia) in reflected unpolarized light. The SEM patterns were exhibited using a JEOL JSM-6510LV scanning electron microscope. To avoid the surface charging effects, the powder samples were deposited on a conductive substrate (carbon tape) and covered with a nanometer gold layer (99.9%).
The single crystal X-ray diffraction data from [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O were recorded by a SMART APEXII diffractometer (Mo K α , λ = 0.7106 Å) at T = 102(2) K. The orientation matrixes and cell parameters were calculated and refined by 45,124 reflections. The main information about the crystal data, data collection and refinement are reported in Table 1. The program APEXII (Bruker) was used to integrate the reflex intensities. Space group P2 1 /c was obtained by the analysis of extinction rules and intensity statistics obtained from all reflections. The multiscan absorption correction of reflection intensities was performed by the APEXII software (Bruker, 2003-2008. Then, the intensities of equivalent reflections were averaged. The structure was solved by the direct methods using package SHELXS and refined in the anisotropic approach for non-hydrogen atoms using the SHELXL program [92]. All hydrogen atoms of H 2 O molecules were found via Fourier difference maps and, further, they were refined in a constrained mode. The structure test for the presence of other missing elements of symmetry and possible voids was produced using the PLATON program [93]. The DIAMOND program was used for the crystal structure plotting [94]. The optical microscopy images of the crystals were fixed with an MS-2 microscope (State Optical Institute, Saint Petersburg, Russia) in reflected unpolarized light. The SEM patterns were exhibited using a JEOL JSM-6510LV scanning electron microscope. To avoid the surface charging effects, the powder samples were deposited on a conductive substrate (carbon tape) and covered with a nanometer gold layer (99.9%).   2 for Rietveld analysis were collected at room temperature with a Bruker D8 ADVANCE powder diffractometer (Cu-Kα radiation) and linear VANTEC detector. The step size of 2θ was 0.02 • , and the counting time was 5 s per step. The Rietveld refinement was performed by using TOPAS 4.2 [95]. The structural parameters of [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O determined by the single crystal analysis were used as a basis in powder pattern Rietveld refinement. For the CsEu(SO 4 ) 2 sample, all peaks were indexed by monoclinic cell (C2/c) with the parameters close to those of RbEu(SO 4 ) 2 [96]. Therefore, the crystal structure of RbEu(SO 4 ) 2 was taken as a starting model for Rietveld refinement, and, in the structure, the Rb ion was replaced by the Cs ion. The refinement was stable and gave low R-factors. The crystallographic data were deposited in Cambridge Crystallographic Data Centre (CSD# 2102324-2102325). The data can be downloaded from the site (www.ccdc.cam.ac.uk/data_request/cif) (accessed on 10 August 2021).
The Fourier-transformed infrared spectroscopy (FTIR) measurements were carried out with the use of a Fourier Transform Infrared Spectrometer FSM 1201 (Infraspek Ltd., Saint Petersburg, Russia). The sample for the investigation was prepared as a tablet with the addition of annealed KBr. The Raman spectra were recorded using an i-Raman Plus spectrometer (B&W Tek, Lubeck, Germany) at a laser excitation wavelength of 785 nm. The diffuse reflectance spectra were measured on a UV-2600 spectrophotometer (Shimadzu, Kyoto, Japan) equipped by the ISR-2600Plus attachment with an integrating sphere. The optical bandgap was estimated on the base of the measurements of diffuse reflectance spectra.
The calculation of electronic bandgap structures of [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O and CsEu(SO 4 ) 2 was performed by using the DFT (density functional theory) method as implemented in the CASTEP code [97]. On-the-fly generated norm-conserving potentials were used and 5s5p6s, 4f 5s5p6s, 3s3p, 2s2p and 1s electrons were treated as the valence ones for Cs, Eu, S, O and H, respectively. The self-consistent field tolerance was set to 2.0 10 −7 eV/atom. The energy cutoff was chosen as 1143 eV for both compounds and superimposed by the 3 × 1 × 2 k-point grid, in the case of [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O, and by the 4 × 4 × 2 k-point grid in the case of CsEu(SO 4 ) 2 . The local density approximation based on the Perdew and Zunger [98] parameterization of the numerical results of Ceperley and Alder [99] was used. The Hubbard U energy term U f = 6 eV for the Eu 4f orbital was applied. The thermal analysis was carried out in the argon flow with the use of a Simultaneous Thermal Analysis (STA) equipment 499 F5 Jupiter NETZSCH (NETZSCH Holding, Selb, Germany). The powder samples were inserted into alumina crucibles. The heating rate was 3 K/min. For the enthalpy determination, the equipment was initially calibrated with the use of standard metal substances, such as In, Sn, Bi, Zn, Al, Ag, Au, Ni. The heat effect peaks were determined with the package Proteus 6 2012.
The luminescence spectra under room temperature were registered on a HORIBA Jobin Yvon T64000 triple spectrometer with the spectral resolution 2.1 cm −1 using the excitation from the GaN laser at 410 nm and the power of 5 mW on the sample. The microscope based on Olympus BX-41 with the Olympus LMPlanFl 50 × objective lens f = 10.2 mm with numerical aperture N.A. = 0.5 was used. The unfocused laser radiation illuminated the small sample powder quantity tangentially. The angle between incident laser light and the registered luminescence was about 60 degrees.

Crystal, Vibrational and Electronic Structure
According to the single crystal and powder diffraction analysis ( Figure 2a, Tables 1 and S1- Figure 3a). The EuO 6 (H 2 O) 3 polyhedron was joined with two SO 4 2− tetrahedra by nodes and edges, respectively, forming, in total, a 2D net. The tridentate bridge-chelate µ2 coordination of the anion towards Eu atoms was observed. One H 2 O molecule was not coordinated to any metal and it should be considered as an isolated one. It was interesting to consider the stability of this type of structure in reference to the metal ion substitution. The collection of the known compounds [A(Ln,Ac)(H 2 O) 3 (SO 4 ) 2 ]·H 2 O is presented in Table S6 (see Supplementary Materials) and the dependence of unit cell volume V A on the ion radius IR of the Ln or Ac element is shown in Figure 4 [69,[80][81][82][83][84][85][86]. It was evident that only such big-sized cations as A = NH 4 , Tl, Rb and Cs provided a stable monoclinic structure. In this crystal family, the upper limit of V A = 1142. 11         According to the results of the powder diffraction analysis (Figure 1b, Tables 1 and S7 and S8), CsEu(SO4)2 crystallizes in the monoclinic space group C2/c. As it is seen in Figure  3b, the structure was of layered type. There were a half of Eu, a half of Cs ions and one SO4 group in the asymmetric part of the unit cell. The Cs + ion in CsEu(SO4)2 was coordinated by 14 O − ions forming a hexagonal dipyramid. In the CsEu(SO4)2 structure, the Cs + ion was coordinated by six Eu ions and eight SO4 tetrahedra. Each Eu 3+ ion was coordinated by six sulfate groups SO4 2− via oxygen atoms. Two sulfate groups were chelately coordinated, while the rest were monodentate, resulting in the formation of a two-capped trigonal prism, and the coordination number of europium was equal to eight (Figure 3b). The tetradentate bridge-chelate μ3 coordination mode of the anion towards Eu 3+ was observed for CsEu(SO4)2. The structure of CsEu(SO4)2 was isostructural to that of RbEu(SO4)2 [96].
The vibrational spectra of [CsEu(H2O)3(SO4)2]·H2O and CsEu(SO4)2 are shown in Figure 5a,b, respectively. The normal vibrational modes of free (SO4) 2− ions had the wavenumbers of 450, 611, 983 and 1105 cm −1 for ν2, ν4, ν1 and ν3 vibrations, respectively [100]. The correlation for the internal vibrational modes of free sulfate ion and its site symmetry in the lattice and crystal symmetry for the investigated compounds are given in Table 2. Both sulfates had the same factor group symmetry and (SO4) 2− units occupied the identical symmetry sites. According to the structure refinement results, CsEu(SO4)2 was characterized by only one crystallographically independent SO4 tetrahedron, while [CsEu(H2O)3(SO4)2]·H2O had two independent SO4 units in its structure. Thus, the number of bands in the Raman and Infrared spectra in the regions of (SO4) 2− vibrations should have been twice as big in [CsEu(H2O)3(SO4)2]·H2O than in CsEu(SO4)2. This relation is clearly seen in Figure 6, where one strong band was found in the region of  According to the results of the powder diffraction analysis (Figure 1b, Tables 1, S7 and S8), CsEu(SO 4 ) 2 crystallizes in the monoclinic space group C2/c. As it is seen in Figure 3b, the structure was of layered type. There were a half of Eu, a half of Cs ions and one SO 4 group in the asymmetric part of the unit cell. The Cs + ion in CsEu(SO 4 ) 2 was coordinated by 14 O − ions forming a hexagonal dipyramid. In the CsEu(SO 4 ) 2 structure, the Cs + ion was coordinated by six Eu ions and eight SO 4 tetrahedra. Each Eu 3+ ion was coordinated by six sulfate groups SO 4 2− via oxygen atoms. Two sulfate groups were chelately coordinated, while the rest were monodentate, resulting in the formation of a two-capped trigonal prism, and the coordination number of europium was equal to eight (Figure 3b). The tetradentate bridge-chelate µ3 coordination mode of the anion towards Eu 3+ was observed for CsEu(SO 4 ) 2 . The structure of CsEu(SO 4 ) 2 was isostructural to that of RbEu(SO 4 ) 2 [96].
The vibrational spectra of [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O and CsEu(SO 4 ) 2 are shown in Figure 5a,b, respectively. The normal vibrational modes of free (SO 4 ) 2− ions had the wavenumbers of 450, 611, 983 and 1105 cm −1 for ν 2 , ν 4 , ν 1 and ν 3 vibrations, respectively [100]. The correlation for the internal vibrational modes of free sulfate ion and its site symmetry in the lattice and crystal symmetry for the investigated compounds are given in Table 2. Both sulfates had the same factor group symmetry and (SO 4 ) 2− units occupied the identical symmetry sites. According to the structure refinement results, CsEu(SO 4 ) 2 was characterized by only one crystallographically independent SO 4 tetrahedron, while  Table S9.  Table S9.       Figure 7. The nature of electronic transition was determined by the exponent factor n = 2 or 1/2 for direct or indirect electronic transitions, respectively. As it is seen in Figure 7a was dependent on photon energy hν, are shown in Figure 7. The nature of electronic transition was determined by the exponent factor n = 2 or 1/2 for direct or indirect electronic transitions, respectively. As it is seen in Figure 7a Figure 9. The bandgap calculated value was determined as the difference between the valence band top (VBT) and the conduction band bottom (CBB). As europium is a lanthanide, the band structure was presented as spin up and spin down components. The VBT and CBB points of [CsEu(H2O)3(SO4)2]·H2O and CsEu(SO4)2 were located in the center of the Brillouin zone (see Figure 9) and, thus, we can say that both compounds were direct band gap materials. The calculated bandgap value for [CsEu(H2O)3(SO4)2]·H2O was 5.09 eV, while, for CsEu(SO4)2, Eg = 3.30 eV. Thus, the transformation from the hydrate to the anhydrous compound reduced the bandgap value in the pair of sulfates.  Figure 9. The bandgap calculated value was determined as the difference between the valence band top (VBT) and the conduction band bottom (CBB). As europium is a lanthanide, the band structure was presented as spin up and spin down components. The VBT and CBB points of [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O and CsEu(SO 4 ) 2 were located in the center of the Brillouin zone (see Figure 9) and, thus, we can say that both compounds were direct band gap materials. The calculated bandgap value for [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O was 5.09 eV, while, for CsEu(SO 4 ) 2 , E g = 3.30 eV. Thus, the transformation from the hydrate to the anhydrous compound reduced the bandgap value in the pair of sulfates. The partial density of states (PDOS) for [CsEu(H2O)3(SO4)2]·H2O and CsEu(SO4)2 are shown in Figure 10 and the contribution of each type of atoms can be considered. It can be stated that the valence band top in both compounds was governed by the p electrons of oxygen, while the conduction band bottom was formed by the d electrons of Eu 3+ ions. The small peak related to the f-electron state of Eu 3+ ions appeared near the Fermi level in both cases.  Figure 10 and the contribution of each type of atoms can be considered. It can be stated that the valence band top in both compounds was governed by the p electrons of oxygen, while the conduction band bottom was formed by the d electrons of Eu 3+ ions. The small peak related to the f-electron state of Eu 3+ ions appeared near the Fermi level in both cases.

Thermochemical Properties
Since anhydrous sulfate CsEu(SO 4 ) 2 was formed as a result of the [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O dehydration, a full-scale study of thermochemical properties can be performed based only on the thermal analysis data shown for [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O in a wide temperature range ( Figure 11, Table 3). The [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O dehydration proceeded in three stages and led to the formation of anhydrous sulfate CsEu(SO 4 ) 2 . In the first stage, three water molecules were pinched off (effect A). The remaining water molecule was firmly bound in the structure and the dehydration process occurred in two stages, which corresponded to the formations of a hemihydrate (effect B) and anhydrous salt (effect C), respectively. Anhydrous sulfate CsEu(SO 4 ) 2 was stable up to 800 • C, and, at higher temperatures, a two-stage decomposition was observed. At the first stage (effect D), the decomposition into simple sulfates and decomposition of europium (III) sulfate occurred with the formation of europium oxysulfate Eu 2 O 2 SO 4 . At the second stage (effect E), the europium oxysulfate decomposition took place. Thus, the final thermal destruction product at~1200 • C was a mixture of cesium sulfate and europium oxide. This destruction mechanism resembled that of AgEu(SO 4 ) 2 [53], but, in the case of AgEu(SO 4 ) 2 , the decomposition effects of the complex sulfate and those of europium sulfate were differentiated.

Thermochemical Properties
Since anhydrous sulfate CsEu(SO4)2 was formed as a result of the [CsEu(H2O)3(SO4)2]·H2O dehydration, a full-scale study of thermochemical properties can be performed based only on the thermal analysis data shown for [CsEu(H2O)3(SO4)2]·H2O in a wide temperature range ( Figure 11, Table 3). The [CsEu(H2O)3(SO4)2]·H2O dehydration proceeded in three stages and led to the formation of anhydrous sulfate CsEu(SO4)2. higher temperatures, a two-stage decomposition was observed. At the first stage (effect D), the decomposition into simple sulfates and decomposition of europium (III) sulfate occurred with the formation of europium oxysulfate Eu2O2SO4. At the second stage (effect E), the europium oxysulfate decomposition took place. Thus, the final thermal destruction product at ~1200 °C was a mixture of cesium sulfate and europium oxide. This destruction mechanism resembled that of AgEu(SO4)2 [53], but, in the case of AgEu(SO4)2, the decomposition effects of the complex sulfate and those of europium sulfate were differentiated.  The most interesting feature of the [CsEu(H2O)3(SO4)2]·H2O dehydration process was the unusual water molecules evaporation order, which seemed to be impossible on the base of the crystal structure, where three water molecules were coordinated to the  The most interesting feature of the [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O dehydration process was the unusual water molecules evaporation order, which seemed to be impossible on the base of the crystal structure, where three water molecules were coordinated to the europium atom, and one water molecule was in the void of the crystal structure. Commonly, in solids under heating, water molecules in voids are lost first and, then, coordinated water molecules are evaporated. However, in [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O, the order was opposite. To explain this phenomenon, it is necessary to consider in detail the coordination of water molecules in the structure and the system of hydrogen bonds shown in Figure 12. It was obvious that the detachment of an uncoordinated water molecule would cause the destabilization of the molecules bound to the O10 and O15 atoms, and it determined the pinching off of these three molecules in one stage. At the same time, the water molecule bound to the O9 atom was very tightly coordinated by the europium polyhedron and two sulfate tetrahedra, and this fact determined its increased stability. It could be intriguing europium atom, and one water molecule was in the void of the crystal structure. Com-monly, in solids under heating, water molecules in voids are lost first and, then, coordinated water molecules are evaporated. However, in [CsEu(H2O)3(SO4)2]·H2O, the order was opposite. To explain this phenomenon, it is necessary to consider in detail the coordination of water molecules in the structure and the system of hydrogen bonds shown in Figure 12. It was obvious that the detachment of an uncoordinated water molecule would cause the destabilization of the molecules bound to the O10 and O15 atoms, and it determined the pinching off of these three molecules in one stage. At the same time, the water molecule bound to the O9 atom was very tightly coordinated by the europium polyhedron and two sulfate tetrahedra, and this fact determined its increased stability. It could be intriguing to compare the thermal dehydration processes in [CsEu(H2O)3(SO4)2]·H2O and other isostructural compounds listed in Table S6. However, to our best knowledge, the results of the thermochemical analysis are available only for [Tl(Ln,Ac)(H2O)3(SO4)2]·H2O [69]. In [Tl(Ln,Ac)(H2O)3(SO4)2]·H2O, three water molecules were evaporated first, and, at the second stage, the fourth water molecule was lost. Thus, the dehydration routes in [Tl(Ln,Ac)(H2O)3(SO4)2]·H2O and [CsEu(H2O)3(SO4)2]·H2O were different. Unfortunately, a detailed analysis of the different behavior of these crystals was impossible because only the cell parameters were reported for [Tl(Ln,Ac)(H2O)3(SO4)2]·H2O [69], and their crystal structures remain unknown.

Luminescence Properties
The exciting radiation at 410 nm used for luminescent measurements fell into the resonance with the transition from the ground state 7 F 0 to the 5 D 3 state of the Eu 3+ ion. The luminescence from 5 D 3 , 5 D 2 and 5 D 1 states was negligible, as compared to that from the 5 D 0 state. The spectra of luminescence from the 5 D 0 state are presented in Figure 13 for both cesium europium sulfate and cesium europium sulfate hydrate. Both crystals belong to the monoclinic symmetry class but to different space groups (C2/c and P2 1 /c, correspondingly), and the luminescent spectra of the Eu 3+ ion drastically differed. The local symmetry of the Eu 3+ ion in cesium europium sulfate was C 2, while in cesium europium sulfate hydrate it was C 1 . This difference seemed to be of minor importance; however, from the spectra, additional features of the local environment could be deduced. The amplitudes of luminescent bands at the magnetic dipole 5 D 0 → 7 F 1 transition and at the crystal-field-induced 5 D 0 → 7 F 2 transition were almost equal, and that indicated a relatively low deviation from the inversion symmetry at the Eu 3+ ion site in cesium europium sulfate (Figure 14a). Alternatively, the crystal-field-induced 5 D 0 → 7 F 2 transition confidently dominated in cesium europium sulfate hydrate, indicating a much larger violation of the inversion symmetry at the Eu site in this hydrate crystal (Figure 14b). Using the Judd-Ofelt analysis (see, e.g., paper by Kolesnikov et al. [103]), the radiative lifetime of Eu ion in cesium europium sulfate hydrate was 2.27 times smaller than in cesium europium sulfate due to a larger violation of inversion symmetry specified above. At the same time, the ultranarrow line amplitude of at the 5 D 0 → 7 F 0 transition in cesium europium sulfate was of the same order of magnitude as the amplitude of magnetic dipole transition that evidenced a relatively stronger extent of the mirror symmetry violation at the Eu site in cesium europium sulfate, with respect to that in cesium europium sulfate hydrate.   The extent of the chemical shift of the ultranarrow Eu line induced by the presence of H2O molecules in the vicinity of the Eu site in cesium europium sulfate hydrate, with respect to cesium europium sulfate, is illustrated in more detail in Figure 15. The ultranarrow line position in cesium europium sulfate was at 578.8 nm, while in cesium europium sulfate hydrate it shifted to 579.3 nm. The extent of the chemical shift of the ultranarrow Eu line induced by the presence of H 2 O molecules in the vicinity of the Eu site in cesium europium sulfate hydrate, with respect to cesium europium sulfate, is illustrated in more detail in Figure 15. The ultranarrow line position in cesium europium sulfate was at 578.8 nm, while in cesium europium sulfate hydrate it shifted to 579.3 nm.

Conclusions
Thus, two new double sulfates [CsEu(H2O)3(SO4)2]·H2O and CsEu(SO4)2 were obtained and systematically investigated. The method of simple crystallization of their aqueous solution made it possible to obtain high quality single crystals of [CsEu(H2O)3(SO4)2]·H2O. The thermal dehydration provided the powder of anhydrous double sulfate CsEu(SO4)2 with a high stoichiometry, which is unattainable in a solidphase reaction between simple sulfates. Both sulfates crystallized in a monoclinic system, but in different space groups, and it led to a significant difference in their vibrational, optical and luminescent properties. The band gap decreased on the transition from [CsEu(H2O)3(SO4)2]·H2O to CsEu(SO4)2. The thermochemical behavior of crystalline hydrate, which seemed illogical at first sight, was well explained by a detailed examination of the coordination of water molecules in the structure. A decisive aspect was found by the consideration of a system of hydrogen bonds, leading to an increased stability of one water molecule in the structure. The noticeable difference of the luminescence spectra between cesium europium sulfate and cesium europium sulfate hydrate was found and explained by the variation of the extent of local symmetry violation at the crystallographic sites occupied by Eu 3+ ions, namely, the inversion symmetry and mirror symmetry. The chemical shift of the 5 D0 energy level in cesium europium sulfate hydrate, with respect to cesium europium sulfate, was associated with the presence of H2O molecules in the vicinity of the Eu 3+ ion.

Conclusions
Thus, two new double sulfates [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O and CsEu(SO 4 ) 2 were obtained and systematically investigated. The method of simple crystallization of their aqueous solution made it possible to obtain high quality single crystals of [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O. The thermal dehydration provided the powder of anhydrous double sulfate CsEu(SO 4 ) 2 with a high stoichiometry, which is unattainable in a solid-phase reaction between simple sulfates. Both sulfates crystallized in a monoclinic system, but in different space groups, and it led to a significant difference in their vibrational, optical and luminescent properties. The band gap decreased on the transition from [CsEu(H 2 O) 3 (SO 4 ) 2 ]·H 2 O to CsEu(SO 4 ) 2 . The thermochemical behavior of crystalline hydrate, which seemed illogical at first sight, was well explained by a detailed examination of the coordination of water molecules in the structure. A decisive aspect was found by the consideration of a system of hydrogen bonds, leading to an increased stability of one water molecule in the structure. The noticeable difference of the luminescence spectra between cesium europium sulfate and cesium europium sulfate hydrate was found and explained by the variation of the extent of local symmetry violation at the crystallographic sites occupied by Eu 3+ ions, namely, the inversion symmetry and mirror symmetry. The chemical shift of the 5 D 0 energy level in cesium europium sulfate hydrate, with respect to cesium europium sulfate, was associated with the presence of H 2 O molecules in the vicinity of the Eu 3+ ion.