The rare earth borates RE2B8O15 (RE = La, Pr, Nd) were synthesized in a Walker-type multianvil apparatus under conditions of 5.5 GPa and 1100 °C. Starting from the corresponding rare earth oxides and boron oxide, the syntheses yielded crystalline products of all new compounds that allowed crystal structure analyses based on single-crystal X-ray diffraction data for La2B8O15 and Nd2B8O15. The compound Pr2B8O15 could be characterized via X-ray powder diffractometry. The results show that the new compounds crystallize isotypically to Ce2B8O15 in the monoclinic space group P2/c. The infrared spectra of RE2B8O15 (RE = La, Pr, Nd) have also been studied.
Borates are known for their ability to form various crystal structures based on the two main borate groups [BO3]3− and [BO4]5− and their connections via common corners or edges. Ongoing research in the field of rare earth (RE) borates led to interesting discoveries of new ortho-, meta-, and polyborates exhibiting new crystal structure types. One of the structurally most remarkable example was the synthesis of Dy4B6O15, which is the first borate exhibiting the structural motif of edge-sharing BO4 tetrahedra [1–3]. From the compositional aspect, there is a large variety of rare earth borates, e.g. orthoborates with the general composition REBO3, which crystallize in various structure types depending on the synthesis conditions and the size of the rare earth cations . The same is known for the metaborates (REB3O6), which form two different modifications under normal-pressure conditions, namely α-RE(BO2)3 and β-RE(BO2)3, as well as two more structural variants accessible via high-pressure/high-temperature conditions: γ-RE(BO2)3 and δ-RE(BO2)3 . Furthermore, several borates with various compositions and crystal structures like RE26(BO3)8O27  and RE2B4O9 [6–9] are known. In contrast, only three compositions are known in the field of boron-rich ternary rare earth borates with a ratio RE2O3:B2O3 larger than 1:3. The monoclinic “ultra-oxoborate” La4B14O27 (RE2O3:B2O3 = 1:3.5) was synthesized in sealed platinum ampoules at 710 °C with an excess of CsCl as fluxing agent by Nikelski et al. in 2007 , and the isotypic compound Ce4B14O27 was obtained by Hinteregger et al. applying high-pressure/high-temperature conditions in 2013 . With REB5O9 (RE = La–Nd, Sm–Tm) [12, 13], Li et al. were able to synthesize the most boron-rich ternary rare earth borate so far (ratio of RE to B = 1:5) by annealing of the hydrated rare earth borates H3REB6O12 (RE = Sm–Lu), RE[B8O11(OH)5] (RE = La–Nd), and RE[B9O13(OH)4]·H2O (RE = Pr–Eu). The compounds REB5O9 (RE = La–Nd, Sm–Tm) crystallize in two different structure types depending on the size of the rare earth cations. Tetragonal α-REB5O9 is formed with RE = Pr, Nd, Sm–Tm and monoclinic β-REB5O9 with the rare earth cations RE = La, Ce. Both structures consist of BO3 as well as BO4 groups that are interconnected via common corners. In 2013, we were able to synthesize a new boron-rich rare earth borate with the composition Ce2B8O15  (ratio of RE to B = 1:4) by applying high-pressure/high-temperature conditions to a mixture of ceria and boron oxide. It is exclusively built up from BO4 tetrahedra that are strongly interconnected via common corners forming a complex three-dimensional anionic framework that hosts the rare earth cations in channels formed by nine-membered rings. Up to now, this borate contains the highest amount of boron in the family of high-pressure rare earth borates synthesized so far. In the context of our research, we were now able to synthesize three new polyborates RE2B8O15 (RE = La, Pr, Nd) crystallizing isotypically to Ce2B8O15. In the following, we report the single-crystal X-ray structure determinations of La2B8O15 and Nd2B8O15 as well as the identification of Pr2B8O15via X-ray powder diffractometry. Furthermore, all three compounds were characterized by infrared spectroscopy.
2 Experimental section
For the syntheses of the compounds RE2B8O15 (RE = La, Pr, Nd), mixtures of the corresponding rare earth oxides (La2O3, Sigma Aldrich, 99.9 %; Pr6O11, Strem Chemicals, 99.9 %; Nd2O3, Strem Chemicals, 99.9 %) with boron oxide (B2O3, Strem Chemicals, 99.9 %) were used as the starting materials according to the following equations.
A slight stoichiometric excess of boron oxide was found to be crucial to avoid the formation of undesirable side products like δ-RE(BO2)3 [4, 15], γ-RE(BO2)3 (RE = La, Nd)  or λ-PrBO3  in all syntheses. The educt mixtures were finely ground under inert gas atmosphere and filled into crucibles of hexagonal boron nitride (Henze Boron Nitride Products GmbH, HeBoSint® P100, Kempten, Germany). These were compressed inside of a Walker-type multianvil assembly using a 1000 ton hydraulic press (Voggenreiter, Mainleus, Germany) and heated via resistive heating of a graphite tube surrounding the crucible. For detailed information of the assembly, the reader is referred to references [18–22]. For the synthesis of all three new compounds, the educts were compressed to 5.5 GPa in 140 min, and this pressure was maintained during the following heating period. The temperature was raised from room temperature to 1100 °C within 10 min and kept there for 20 min, before cooling down to 500 °C within 30/300/120 min for the La, Pr, and Nd phase, respectively. Afterwards, the samples were quenched to room temperature by switching off the heating. The decompression of the assemblies required about 7 h. The recovered octahedral pressure media were broken apart and the samples carefully separated from the surrounding boron nitride crucibles. The compounds RE2B8O15 (RE = La, Pr, Nd) were obtained as colorless (La2B8O15), green (Pr2B8O15), and alexandritic pink-gray (Nd2B8O15) crystalline products that are stable under ambient conditions.
Besides the boron oxide excess in the educt mixtures, the pressure applied during the syntheses played a crucial role for the formation of the desired products RE2B8O15 (RE = La, Pr, Nd). Lowering the pressure to 4 GPa, while keeping all other conditions of the syntheses identical, led to the formation of δ-RE(BO2)3 (RE = La, Nd) [4, 15] as the main product of the syntheses with RE = La, Nd, and to a high amount of λ-PrBO3  as side product of the syntheses of Pr2B8O15. Above 6 GPa, γ-RE(BO2)3 (RE = Pr, Nd)  were the main products and the attempt to synthesize La2B8O15 resulted in the formation of a yet unknown, very dark, and microcrystalline phase.
2.2 X-ray structure determinations
The products RE2B8O15 (RE = La, Pr, Nd) were identified by X-ray powder diffraction on flat samples in transmission geometry using a Stoe Stadi P powder diffractometer (MoKα1 radiation, Ge(111)-monochromatized, λ = 70.93 pm). All three new compounds RE2B8O15 (RE = La, Pr, Nd) could be identified by comparing the reflection patterns with that of the isotypic phase Ce2B8O15 . Apart from a slight shift caused by minor differences of the lattice parameters due to the lanthanide contraction, the reflections tallied perfectly. Figure 1 shows exemplarily the powder pattern of Pr2B8O15. As it was not possible to analyze the crystal structure of Pr2B8O15 by means of single-crystal X-ray structure determination, the experimental powder pattern of Pr2B8O15 (top) is compared with the theoretical powder pattern of Nd2B8O15 (bottom) based on single-crystal diffraction data. For best comparability, the theoretical pattern was calculated with Pr3+ as the rare earth cation and by applying the lattice parameters obtained from indexing of the experimental powder diffraction pattern of Pr2B8O15. By indexing the reflections of RE2B8O15 (RE = La, Pr, Nd), we obtained the parameters a = 919.0(2), b = 421.64(5), c = 1251.2(2) pm, β = 116.61(1)°, and a volume of 0.43349(7) nm3 for La2B8O15, a = 912.37(9), b = 420.19(4), c = 1246.3(2) pm, β = 116.82(1)°, and a volume of 0.42639(5) nm3 for Pr2B8O15, and a = 908.87(6), b = 419.32(3), c = 1244.5(2) pm, β = 116.84°, and a volume of 0.42319(4) nm3 for Nd2B8O15. The values for La2B8O15 and Nd2B8O15 are in good agreement with the lattice parameters received from the single-crystal X-ray diffraction data (Table 1) and the values obtained for Pr2B8O15 correspond well to the progression of the lattice parameters along the series of lanthanides (vide infra).
|M, g mol−1||604.30||614.96|
|Cryst. size, mm3||0.022 × 0.037 × 0.041||0.031 × 0.045 × 0.050|
|Powder diffractometer||Stoe Stadi P|
|Radiation||MoKα1 (λ = 70.93 pm)|
|Single-crystal diffractometer||Bruker D8 Quest (Photon 100)|
|Radiation||MoKα (λ = 71.073 pm)|
|Dcalcd, g cm−3||4.62||4.81|
|hkl range||±15, ±7, ±21||±15, −7 ± 6, ±20|
|Refl. with I > 2 σ(I)||2131||1820|
|Absorption correction||multi-scan (Bruker Difabs 2014/5)|
|Absorption coefficient, mm−1||9.83||12.22|
|R1(F)/wR2(F2) (I > 2 σ(I))||0.0157/0.0371||0.0173/0.0377|
|R1(F)/wR2(F2) (all reflections)||0.0195/0.0385||0.0235/0.0396|
|Δρfin (max/min), e Å−3||1.01/−1.02||1.02/−1.40|
For La2B8O15 and Nd2B8O15, we were able to isolate small single crystals by mechanical fragmentation that could be analyzed via single-crystal X-ray diffraction. The intensity data were collected at room temperature with a Bruker D8 Quest diffractometer (Photon 100) equipped with an Incoatec Microfocus source generator (multi layered optics monochromatized MoKα radiation, λ = 71.073 pm). “Multi-scan” absorption corrections were applied with the program Bruker Difabs-2014/5. The positional parameters of Ce2B8O15  were used as starting values for the structure solution of La2B8O15 and Nd2B8O15. The parameter refinements (full-matrix least-squares against F2) were executed utilizing the Shelxs/l-97 software suite [23, 24]. For both compounds, all atoms could be refined with anisotropic atomic displacement parameters. Final difference Fourier syntheses did not reveal any significant residual peaks. All relevant details of the data collections and evaluations are given in Table 1. The atomic coordinates and isotropic equivalent displacement parameters, interatomic distances, and interatomic angles are listed in the Tables 2–4.
Ueq is defined as one third of the trace of the orthogonalized Uij tensor.
|Ø = 267.5||Ø = 264.3|
|Ø = 147.4||Ø = 147.3|
|Ø = 148.3||Ø = 147.9|
|Ø = 148.2||Ø = 148.0|
|Ø = 147.1||Ø = 146.8|
|Ø = 109.49||Ø = 109.49|
|Ø = 109.36||Ø = 109.38|
|Ø = 109.42||Ø = 109.41|
|Ø = 109.42||Ø = 109.43|
Further details of the crystal structure investigations may be obtained from Fachinformationszentrum Karlsruhe, 76344 Eggenstein-Leopoldshafen, Germany (fax: +49-7247-808-666; e-mail: firstname.lastname@example.org, https://icsd.fiz-karlsruhe.de/search/basic.xhtml) on quoting the deposition numbers CSD-430713 for La2B8O15 and CSD-430714 for Nd2B8O15.
2.3 Vibrational spectroscopy
The ATR-FT-IR spectra of RE2B8O15 (RE = La, Ce, Pr, Nd) were measured from small amounts of the grinded products utilizing a Bruker ALPHA-P “Platinum-ATR” FT-IR spectrometer with a spectral resolution of <2 cm−1. The IR radiation with a wavelength of 850 nm was produced by a HeNe-laser and the reflected beam was detected by a DTGS detector at room temperature. For the measurements, the samples were positioned between two small diamond crystals of the IR spectrometer. 60 scans of each sample and the background were acquired using the Opus 7.0  software.
3 Results and discussion
3.1 Crystal structures
As already mentioned above, all three new compounds RE2B8O15 (RE = La, Pr, Nd) crystallize isotypically to Ce2B8O15 . Figure 2 shows the crystal structures of RE2B8O15 (RE = La, Pr, Nd), which are built up solely from corner-sharing [BO4]5− tetrahedra that are strongly interconnected via threefold bridging oxygen atoms to form a complex three-dimensional framework that hosts the rare earth cations (Fig. 3). The main structural motif consists of four [BO4]5− tetrahedra, three of them building up a three-membered ring and the fourth tetrahedron being connected to this ring via the threefold bridging oxygen atom O3. The fundamental building block can therefore be described as 4□:[Φ]<3□>|□| (after Burns et al. ). The rare earth cations are coordinated by 11 oxygen atoms in the form of an irregularly shaped coordination sphere as depicted in Fig. 4. For a detailed description of the crystal structure, the reader is referred to the description of the isotypic compound Ce2B8O15 . Below, a comparison of the four isotypic compounds RE2B8O15 (RE = La, Ce, Pr, Nd) is given.
Figure 5 shows a comparison of the lattice parameters of RE2B8O15 (RE = La, Ce, Pr, Nd) derived from indexing of the powder diffraction patterns. The decrease of the lattice parameters along the series La3+–Nd3+ corresponds to the decreasing ionic radii of the rare earth cations. The lanthanide contraction is also the cause for slightly decreasing interatomic distances RE–O from La2B8O15 to Nd2B8O15. The average RE–O distance equals 267.5 pm in La2B8O15, 266.3 pm in Ce2B8O15, and 264.3 pm in Nd2B8O15. The mean interatomic distances of the B–O bonds possess values of 147.7 pm in La2B8O15, 147.6 pm in Ce2B8O15, and 147.5 pm in Nd2B8O15. These values are equal within the range of the respective standard deviations. The mean distances B–O of each [BO4]5− group within the structures of RE2B8O15 (RE = La, Ce, Nd) are all in perfect agreement with the literature value of 147.6 pm [27, 28]. The results of n-MEFIR (Mean Fictive Ionic Radii) calculations with the MAPLE program  imply that La2B8O15 has the lowest atomic packing factor (64.2 %), followed by Ce2B8O15 (64.8 %) and Nd2B8O15 (64.8 %). This corresponds to the calculated densities of the compounds, namely 4.62 g cm−3 for La2B8O15, 4.68 g cm−3 for Ce2B8O15 , and 4.81 g cm−3 for Nd2B8O15. The interatomic angles O–B–O within the [BO4]5− tetrahedra show no significant variation within all three compounds RE2B8O15 (RE = La, Ce, Nd). With average values of 109.4 to 109.5° within each [BO4]5− group, they are all very close to the ideal tetrahedral angle of 109.47°.
The charge distribution of the atoms in the compounds RE2B8O15 (RE = La, Nd) was calculated according to the BLBS (bond length/bond strength, ΣV) [30–34] and the CHARDI (charge distribution in solids) concept (ΣQ) [32, 33, 35]. The results are listed in Table 5 and verify the formal valence states of the cations and anions.
Additionally, we calculated the MAPLE values (Madelung Part of Lattice Energy) according to Hoppe [29, 36, 37] of RE2B8O15 (RE = La, Nd), which were checked against the data of the binary compounds. We obtained a value of 102125 kJ mol−1 for La2B8O15 to be compared with 101986 kJ mol−1 (deviation: 0.13 %) based on the binary compounds [La2O3 (14234 kJ mol−1 ) + 4 × HP-B2O3 (21938 kJ mol−1 )] and a value of 102401 kJ mol−1 for Nd2B8O15 to be compared with 102316 kJ mol−1 (deviation: 0.08 %) based on the binary compounds [Nd2O3 (14564 kJ mol−1 ) + 4 × HP-B2O3 (21938 kJ mol−1 )]. The good accordance between the values of the products and the sum of the values of the binary compounds prove the plausibility of the crystal structure solutions.
3.2 FT-IR spectroscopy
Figure 6 shows the IR spectra of RE2B8O15 (RE = La, Ce, Pr, Nd) in the range of 400–1700 cm−1. The absorption bands correspond well for all four compounds with slight shifts in wavelengths and relative intensities. The spectra were recorded between 400 and 4000 cm−1 and showed no vibrational bands due to OH groups or water in the upper range (2000–4000 cm−1). As already reported for Ce2B8O15 , a distinct assignment of the detected absorption bands to certain vibrations is difficult due to the highly cross-linked framework of the crystal structure and the resulting complexity of absorption peaks in the spectra. However, a general assignment of various regions of the spectra can be done based on previous investigations and theoretical calculations. The strong bands observed in the spectral range between 800 and 1150 cm−1 are typical for stretching vibrations within [BO4]5− groups as in γ-RE(BO2)3 (RE = La–Nd), π-GdBO3, π-YBO3, or TaBO4 [16, 41–43]. Absorption peaks in the range of 1200–1500 cm−1 are usually assigned to stretching vibrations of trigonal [BO3]3− groups. As the crystal structures of RE2B8O15 (RE = La, Ce, Pr, Nd) do not contain any [BO3]3− groups, the absorptions in this spectral range have to be assigned to B–O–RE, O–B–O, B–O–B, and B–O stretching and bending vibrations. This was found by ab initio quantum chemical calculations on the oxoborates β-ZnB4O7 and β-CaB4O7 . Their crystal structure is similarly built up exclusively from [BO4]5− groups forming a complex anionic network structure and show IR absorption bands in the range between 1200 and 1500 cm−1 as well. Furthermore, the existence of strongly shortened B–O bond lengths caused by the tight interconnections of [BO4]5− groups in the crystal structures of RE2B8O15 (RE = La, Ce, Pr, Nd) can also be named as a reason for the finding of absorptions in the range of 1200–1500 cm−1. The interatomic distances B3–O7 of only 140.9(2) pm in La2B8O15 and Ce2B8O15 as well as of 141.1(2) pm in Nd2B8O15 are more in the range of classical [BO3]3− groups (137.0 pm ) rather than [BO4]5− groups (147.6 pm [27, 28]).
An interference caused by small amounts of amorphous side products that could not be detected by means of powder diffractometry cannot be excluded. However, it is worth mentioning that the IR spectra measured on the bulk sample of Ce2B8O15 correspond well to those of the single-crystal measurement of Ce2B8O15 as reported in Ref. .
We have synthesized three new boron-rich rare earth polyborates RE2B8O15 (RE = La, Pr, Nd) crystallizing isotypically to Ce2B8O15. The crystal structures of La2B8O15 and Nd2B8O15 were determined by means of single-crystal X-ray diffraction, and Pr2B8O15 was characterized via powder X-ray diffractometry. The structures are exclusively built up from corner-sharing [BO4]5− tetrahedra that form a complex three-dimensional anionic framework containing the rare earth cations in channels along the b axis. A comparison of the lattice parameters of all four isotypic compounds RE2B8O15 (RE = La, Ce, Pr, Nd) shows the expected decrease due to the lanthanide contraction.
Dedicated to: Professor Wolfgang Jeitschko on the occasion of his 80th birthday.
We would like to thank Dr. G. Heymann for collecting the single-crystal data and Dr. K. Wurst for his support with the crystal structure refinement. The research was funded by the Austrian Science Fund (FWF): P 23212-N19.
 H. Emme, H. Huppertz. Z. Anorg. Allg. Chem.2002, 628, 2165.10.1002/1521-3749(200209)628:9/10<2165::AID-ZAAC11112165>3.0.CO;2-LSearch in Google Scholar
 H. Emme, H. Huppertz, Acta Crystallogr.2005, C61, i29.Search in Google Scholar
 M. Glätzle, G. Heymann, H. Huppertz, Z. Kristallogr.2013, 228, 449.Search in Google Scholar
 D. Walker, M. A. Carpenter, C. M. Hitch, Am. Mineral.1990, 75, 1020.Search in Google Scholar
 D. Walker, Am. Mineral.1991, 76, 1092.Search in Google Scholar
 G. M. Sheldrick, Shelxs/l-2014, Program Suite for the Solution and Refinement of Crystal Structures, University of Göttingen, Göttingen (Germany) 2014.Search in Google Scholar
 G. M. Sheldrick, Acta Crystallogr.2015, C71, 3.Search in Google Scholar
 Opus (version 7.0), Bruker Optik GmbH, Ettlingen (Deutschland) 2011.Search in Google Scholar
 P. C. Burns, J. D. Grice, F. C. Hawthorne, Can. Mineral.1995, 33, 1131.Search in Google Scholar
 F. C. Hawthorne, P. C. Burns, J. D. Grice in Boron: Mineralogy, Petrology, and Geochemistry, Vol. 33, 2nd ed., (Eds.: E. S. Grew, L. M. Anovitz), Mineralogical Society of America, Washington, 1996, p. 41.10.1515/9781501509223-004Search in Google Scholar
 R. Hübenthal, M. Serafin, R. Hoppe, Maple (version 4.0), Program for the Calculation of Distances, Angles, Effective Coordination Numbers, Coordination Spheres, and Lattice Energies, University of Gießen, Gießen (Germany) 1993.Search in Google Scholar
 N. E. Brese, M. O’Keeffe, Structure and Bonding, Springer-Verlag, Berlin, 1989.Search in Google Scholar
 R. Hoppe, Angew. Chem., Int. Ed. Engl.1966, 5, 96.Search in Google Scholar
©2016 by De Gruyter