The new Ruddlesden-Popper-related phases An+1BnO3n+1 (n = 3) with the compositions Li2Eu2Nb3O10, Li2Eu1.5Ta3O10, Li2EuKNb3O10, and Li2EuKTa3O10 were synthesized by solid-state reactions from Li2[CO3] (+ K2[CO3]) and the corresponding refractory metals along with their oxides in a high-frequency furnace at temperatures above T = 1600°C. Their structures have been determined by single-crystal X-ray diffraction studies. Characteristic features are triple layers of corner-sharing [MO6]7− octahedra (M = Nb and Ta), which are connected via [LiO4]7− tetrahedra. The Eu2+ cations are cuboctahedrally surrounded by 12 oxygen atoms and according to the Eu–O distances of around 275 pm, they have the oxidation state +2, as confirmed by XPS measurements. In the potassium-containing samples they share their positions with K+ cations. The black compounds are stable in air at room temperature. Measurements of the magnetic susceptibilities in the range of T = 5–300 K revealed Li2Eu2Nb3O10, Li2Eu1.5Ta3O10 and Li2EuKTa3O10 to be paramagnetic without any ordering.
Perovskite-type oxides with the overall formula ABO3 have been studied for decades because of their interesting physical properties, and BaTiO3 with its very high dielectric constant is only one prominent example . For tuning the physical properties of such compounds numerous studies were carried out with magnetic cations on the B site. Investigations of perovskite-type compounds with highly paramagnetic Eu2+ cations on the A site have attracted particular attention. McCarthy and Greedan have proposed a few rules in 1974, which allow to predict the composition and properties of europium oxometalates (EuMO3) , and meanwhile a whole series of such oxides has been described, e.g. EuMO3 (M=Ti, Zr, Hf, Nb, etc.) , , , , . Phase-pure EuTiO3 was first synthesized in 1953 . It becomes antiferromagnetic below T=5.7 K. The respective phase transition is accompanied by a drop in the dielectric constant, owing to a strong spin-lattice coupling , and therefore a magnetostriction effect is observed at low temperatures . In addition, the development of 2D magneto-optical devices based on EuTiO3  appears possible. A common occupation of the A site by Sr2+ and Eu2+ leads to an interesting diluted magnetic system in the series EuxSr1−xTiO3 . More complex oxides like the double perovskites (A′AB2O6, e.g. Eu2ScTaO6 with elpasolite-type structure) , the layered perovskite analogues, such as Dion-Jacobson (RbLaNb2O7), Aurivillius ((Bi2O2)Bi2Ti3O10)  and Ruddlesden-Popper phases with the general formula An+1MnO3n+1, where n corresponds to the number of layers of vertex-connected [MO6]7− octahedra, have also been studied . With n=1, monolayers of corner-sharing [MO6]n− octahedra are formed, like in SrRuO4 . For n=2 double layers of fused [TiO6]8− octahedra occur, for example in Sr3Ti2O7 . For n=3 there are examples such as K2Nd2Ti3O10 and Ca4Ti3O10, which show triple blocks of layers of vertex-connected octahedra , . In the Ruddlesden-Popper phases for n>1 the An+ cations can have coordination numbers 9 or 12. The rare-earth elements prefer mostly the 12-fold coordination and alkali metals settle with the ninefold coordination, for example in K2SrTa2O7 .
Ruddlesden-Popper phases An+1BnO3n+1 with n=1, 2, and 3 are of particular relevance, because they often exhibit interesting properties, such as ionic conductivity , phase transitions , and magnetic ordering , and because they can be exfoliated . The substitution of an A2+ cation (A=alkaline-earth element) with two A′+ cations (A′=alkali element) in this structure type results in a series of compounds with the general formula A′An−1MnO3n+1 . Substitution with small Li+ cations leads to distorted [LiO4]7− tetrahedra connecting the layers of octahedral polyhedra, as e.g. in Li2SrTa2O7 , . Replacing Sr2+ with Eu2+ cations in this structure type results in the magnetically interesting system Li2Sr1−xEuxTa2O7 (0.8>x>1.0) . Besides other compositions derived from these series , Bhuvanesh et al. also obtained layered defect perovskites like Li4Sr3Nb6O20 . We have prepared a series of such phases with lithium in fourfold (tetrahedral) and europium (sometimes along with K+) in 12-fold (cuboctahedral) coordination, on which we report here.
2.1 Sample preparation
2.1.1 Synthesis of Li2Eu2Nb3O10 and Li2Eu1.5Ta3O10
The syntheses of Li2Eu2Nb3O10 and Li2Eu1.5Ta3O10 were performed by heating mixtures from stoichiometric quantities of Li2[CO3] (Acros Organics, Geel), Eu2O3 (ChemPur, 99.99%), Nb2O5 (Fluka, 99.9%) and Ta2O5 (ChemPur, 99.99%), respectively, and niobium (CIBA, reinst) and tantalum (Merck, 99.999%) metal powder, respectively, as described in the following equations:
These reactions were performed in cramped, but not soldered niobium or tantalum capsules (Plansee, Reutte), which were heated in a high-frequency furnace above T=1600°C for 10 min under an argon atmosphere. The melting cakes were cooled to room temperature within 10 min and black, crystalline products (Fig. 1) were obtained. The crystals remain unchanged in air for months.
2.1.2 Synthesis of Li2EuKNb3O10 and Li2EuKTa3O10
The syntheses of Li2EuKNb3O10 and Li2EuKTa3O10 were performed by heating mixtures from stoichiometric quantities of Li2[CO3], K2[CO3] (Merck, 99.0%), Eu2O3, Nb2O5 or Ta2O5 and niobium or tantalum metal powder (M=Nb or Ta), respectively, according to
These mixtures were heated for 10 min as described above and cooled to room temperature within 10 more min. Black compact products were obtained and removed mechanically from the metal containers (Fig. 1). Both compounds are stable in air for several months.
2.2 Structure determinations
For the single-crystal X-ray diffraction studies, well-faceted crystals were selected under a light microscope, packed in 0.1 mm diameter glass capillaries and sealed. The crystals were measured at room temperature on a κ-CCD diffractometer (Bruker-Nonius) with MoKα radiation (λ=71.07 pm). The crystallographic data for Li2EuKNb3O10, Li2Eu2Nb3O10, Li2Eu1.5Ta3O10, and Li2EuKTa3O10 and the results of the final refinements are listed in Table 1, the atomic coordinates appear in Table 2.
|Space group||I4/mmm (no. 139)|
|Formula units per unit cell||Z=2|
|Radiation; λ/pm||MoKα; 71.07|
|h=k, l||±5, ±34||±5, ±34||±5, ±34||±5, ±33|
|Rint; Rσ||0.063; 0.018||0.091; 0.030||0.057; 0.021||0.063; 0.017|
|R1; wR2||0.022; 0.054||0.026; 0.067||0.022; 0.044||0.020; 0.051|
|ρmax; min/e·Å−3||1.29; −0.90||1.47; −1.76||1.02; −0.71||2.34; −1.47|
|Atom||Wyckoff site||s.o.f.||x/a||y/b||z/c||Ueq /pm2||Atom||Wyckoff site||s.o.f.||x/a||y/b||z/c||Ueq /pm2|
Further details of the crystal structure investigations may be obtained from Fachinformationszentrum Karlsruhe, 76344 Eggenstein-Leopoldshafen, Germany (fax: (+49) 7247-808-666, http://www.fiz-karlsruhe.de/request_for_deposited _data.html, E-Mail: firstname.lastname@example.org), on quoting the depository numbers CSD-1949876 to CSD-1949879.
3.1 Crystal structures
Li2Eu2Nb3O10, Li2Eu1.5Ta3O10, Li2EuKNb3O10, and Li2EuKTa3O10 crystallize in the tetragonal space group I4/mmm typically found for Ruddlesden-Popper phases with lattice parameters ranging from a=393 to 396 pm and from c=2609 to 2688 pm (Fig. 2). The crystallographically independent Eu2+ cations are located in cuboctahedra formed by oxygen atoms. In all four compounds the Eu–O distances lie between 270 and 290 pm (Table 3), which is typical for europium atoms in the oxidation state +2 in agreement with the calculated bond-valence parameters (Table 4). These bond-valence parameters correspond well with the results from calculations of the Charge Distribution calculated with Chardi2015 and the Madelung Parts of the Lattice Energy (MAPLE) , , , , . It is remarkable that in Li2Eu2Nb3O10 the position of the europium atoms is fully occupied, in contrast to Li2Eu1.5Ta3O10, where only ¾ of this position is filled with europium (Table 2). Simple charge counting thus results in the oxidation state +5 for tantalum in Li2Eu1.5Ta3O10, whereas an average oxidation state of +4.667 is obtained for niobium in Li2Eu2Nb3O10. For Li2EuKNb3O10 and Li2EuKTa3O10 a full occupation of the corresponding crystallographic sites with europium and potassium is found (Table 2). This appears reasonable, because of the comparable ionic radii of Eu2+ (ri=148 pm) and K+ (ri=164 pm) for a 12-fold coordination , .
|Eu in 4e||1.85; 2.24||1.86; 2.25||1.77; 2.14||1.81; 2.19|
|M1 in 2a||5.21||5.44||5.10||5.25|
|M2 in 4e||4.98||5.08||4.90||5.11|
Taking the refined occupation factors of the crystallographic sites for europium and potassium in Li2EuKNb3O10 and Li2EuKTa3O10 into account, an oxidation of +5 results for both octahedrally coordinated atoms, niobium and tantalum, respectively. The polyhedra are vertex-connected forming layers in the ab plane. Along the c direction, these layers form triple blocks (Fig. 2, top), which are separated by layers (Fig. 2, bottom), where the Li+ cations are tetrahedrally coordinated (Fig. 3), but exhibit two extra oxygen contacts along .
3.2 Physical properties
The individual magnetic susceptibilities of Li2Eu2Nb3O10, Li2Eu1.5Ta3O10, and Li2EuKNb3O10 were measured between T=5 and 300 K in a SQUID magnetometer MPMS3 from Quantum Design (San Diego) (Fig. 4). The susceptibilities follow a paramagnetic behavior in the measured temperature range (Table 5). Applying the Curie-Weiss law, the magnetic moments (μeff) were calculated to 6.31 μB (Li2Eu2Nb3O10), 7.89 μB (Li2Eu1.5Ta3O10), and 8.01 μB (Li2EuKTa3O10). The theoretical moment for Eu2+ cations is 7.94 μB, so for Li2Eu1.5Ta3O10 and Li2EuKTa3O10 it fits well, but for Li2Eu2Nb3O10 the moment is too small. As it is expected from the nearly full occupation of the europium site, niobium has to be partly reduced from +5 to +4. According to the findings from X-ray photo emission spectroscopy (XPS) studies, there are Nb4+ cations in the structure, which show van Vleck paramagnetism for a 4d1 configuration, leading to a calculated moment of 6.47 μB:
a1 Oe=7.96×103 A m−1.
The cation-defect variant Li2Eu1.86(2)□0.14(2)Nb3O10 has been characterized by a single-crystal structure determination based on X-ray diffraction (Section 3.1). Such a deficiency at the europium site must be accompanied by the presence of niobium atoms which have an average oxidation state lower than +5. For an additional analysis of this oxoniobate, XPS spectra have been measured on a commercial Kratos Axis Ultra DLD system (Shimadzu, Kyoto) with a monochromatized AlKα source (1486.6 eV) with a base pressure in the lower 10−10 mbar range. The binding energy (BE) was calibrated by setting the C 1s BE to 284.8 eV with respect to the Fermi level. High-resolution spectra were acquired with an analyzer pass energy of 20 eV. Analysis of the XPS data was performed with CasaXPS software. The measured polycrystalline samples were pressed into an indium foil. In their XPS spectra (Fig. 5) only europium atoms with the oxidation state +2 are observed and just minor impurities of Eu3+ can be detected, which probably stem from partial oxidation on the surface. However, as expected, besides Nb5+ a significant amount of Nb4+ is found, confirming the results from the structure refinement. The peak around 1200 eV in Fig. 5 (top) originates from small impurities of carbon arising from transportation processes of CO2 during the high-temperature syntheses.
The new multinary oxides Li2Eu2Nb3O10, Li2Eu1.5Ta3O10, Li2EuKNb3O10, and Li2EuKTa3O10 with triple layers of corner-sharing [MO6]7− octahedra (M=Nb and Ta) including layers of edge-sharing [LiO4]7− tetrahedra, have been synthesized for the first time using high-temperature solid-state reactions. They crystallize in the space group I4/mmm, typically found for Ruddlesden-Popper phases. A full occupation of the cation sites in Li2Eu2Nb3O10 leads to a mixed-valent oxoniobate(IV,V), whereas the structure of Li2Eu1.5Ta3O10 demonstrates the possibility of complex cationic non-stoichiometry that allows for the formation of an oxotantalate(V). Exfoliation of these compounds might lead to new low-dimensional magnetic materials with octahedral triple layers as found for the double-layer example Li2EuNb2O7.
Dedicated to: Professor Arndt Simon on the occasion of his 80th birthday.
The authors thank Kathrin Küster (Max-Planck-Institut für Festkörperforschung, Stuttgart) for the measurements of the XPS spectra, Dr. Björn Blaschkowski (Institut für Anorganische Chemie, Universität Stuttgart) for the magnetic measurements and Dr. Falk Lissner (Institut für Anorganische Chemie, Universität Stuttgart) for the collection of the single-crystal X-ray diffraction data.
 P. Reuvekamp, K. Caslin, Z. Guguchia, H. Keller, R. K. Kremer, A. Simon, J. Köhler, A. Bussmann-Holder, J. Phys.: Cond. Mat. 2015, 17, 262201.Search in Google Scholar
 Ch. Funk, E. Brücher, J. Köhler, Th. Schleid, Z. Kristallogr.2019, Suppl. 39, 98.Search in Google Scholar
 G. Amow, J. E. Greedan, Acta Crystallogr. 1998, C 54, 1053.Search in Google Scholar
 Ch. Funk, J. Köhler, Th. Schleid, Z. Kristallogr.2016, Suppl. 36, 82.Search in Google Scholar
 R. X. Fischer, E. Tillmanns, Acta Crystallogr.1988, C44, 775.Search in Google Scholar
 M. Allieta, M. Scavini, L. J. Spalek, V. Scagnoli, H. C. Walker, Ch. Panagopoulos, S. S. Saxena, T. Katsufuji, C. Mazzoli, Phys. Rev. B201285, 184107-1–8.Search in Google Scholar
©2020 Walter de Gruyter GmbH, Berlin/Boston