Article pubs.acs.org/IC
Optical Spectra and Magnetic Behavior of a Wide Range of Europium(III) Oxo-Compounds: Analysis of the Ligand-Field Effects Anna Bronova, Nils Kannengießer, and Robert Glaum* Institute of Inorganic Chemistry, Rheinische Friedrich-Wilhelms-Universität, Gerhard-Domagk-Straße 1, 53121 Bonn, Germany S Supporting Information *
ABSTRACT: The europium−oxygen interaction in nine different europium(III) oxocompounds (including C-type Eu2O3) was investigated on the basis of powder reflectance spectra (near-IR/vis/UV) and temperature-dependent magnetic measurements. Computation of the transition energies and of the effective Bohr magneton numbers for Eu3+ in the different ligand fields were performed within the framework of the angular overlap model (AOM) using the computer program BonnMag. These calculations show that all electronic transition energies in the optical spectra, the magnetic susceptibilities as well as their temperature dependence, are very wellaccounted for by AOM. BonnMag provides a facile way to perform these calculations. Analysis of the obtained “best fit” AOM parameters eσ(EuIII−O) shows that these are significantly influenced by the further bonding partners of oxygen (“second-sphere ligand-field effect”). An increase of eσ, max(EuIII−O) from 404 cm−1 (EuPO4) to 687 cm−1 (EuSbO4), both normalized to d(EuIII−O) = 2.38 Å, is found. Correlation of this variation to oxide polarizability and optical basicity of the oxo-compounds is discussed.
■
INTRODUCTION The widespread technical application of compounds containing lanthanide ions (f n systems) has led to great interest in the chemical and physical properties of these ions.1−3 For deriving reliable structure−property relations a facile and comprehensive way to describe the various properties (ligand-field splitting, energies of the electronic states, magnetic susceptibilities, and, if necessary, g tensors) is needed. In recent papers we have shown the applicability of angular overlap modeling (AOM)4−7 (using computer program BonnMag8) for this purpose by analyzing the ligand-field in the chromophore [UIVO6Cl2] in UPO4Cl.9,10 Here we report on a detailed study of the ligand-field effects related to Eu3+ in oxo-compounds. The electronic structure of Eu3+ (f 6 system) is somewhat special in the series of RE3+ cations, since the ground-state 7F0 is “unmagnetic” (μeff/μB = 0) despite the six unpaired electrons. This behavior follows from strong spin−orbit coupling and is described by the Landé formula.11,12 Experimentally, μeff/μB = 3.4 to 3.6 is observed at room temperature with a strong decrease to μeff/μB ≈ 0.3 upon cooling to 2 K.13 In principle this temperature dependence can be understood by thermal population at T > 0 K of the first excited electronic state 7F1 with μeff/μB = 4.24. The energy difference between the “unmagnetic” ground state and the first excited state is reduced due to the splitting of the 7F1 state by the ligand field (Figure 1). The splitting between 7F0 and 7F1 in the free ion is Δν̃ ≈ 370 cm−1 and is a function of the spin−orbit-coupling constant. The 7F1 state is split by the ligand field into three states, with a separation of Δν̃ ≈ 80 cm−1 between the highest and the lowest sublevel. Thus, the magnetic moment of europium(III) compounds at room temperature can be qualitatively correlated © 2017 American Chemical Society
Figure 1. Splitting of the 7F state by spin−orbit coupling and ligand field. μcalc/μB(7F0) = 0, μcalc/μB(7F1) = 4.24. The higher states 7F2 to 7 F6 are omitted.
to the ligand-field splitting of the 7F1 state. Higher splitting leads to lowering of one or two 7F1 split states and therefore to a higher thermal population and a higher magnetic moment. For the free Eu3+ ion (no ligand field (LF)!) one would obtain at room temperature μcalc/μB = 3.20. The optical properties of Eu3+ ions in various types of compounds and host lattices have been reviewed only recently by Binnemans.13 We refer the reader to his paper for further details on the electronic structure of Eu3+ ions. A quantitative description of the temperature dependence of μcalc(Eu3+) requires obviously a thorough description of the ligand-field splitting of the 7F1 state. Only recently, with the introduction of BonnMag,8 a computer program that allows describing the ligand-field effects on all f n systems (1 ≤ n ≤ 13) within the AOM framework, has this become available. The lack of such a program prevented for a long time detailed Received: May 22, 2017 Published: July 21, 2017 9235
DOI: 10.1021/acs.inorgchem.7b01287 Inorg. Chem. 2017, 56, 9235−9246
Article
Inorganic Chemistry
depend on the metal−ligand distance (eq 1).36 Furthermore, throughout the paper isotropic π-interactions (eπ,x = eπ,y) with eπ,iso = 1/4 eσ were assumed.36 In effect, these assumptions leave just one parameter, for example, eσ,max(Eu3+−O2−) of the nearest ligand in each chromophore, for matching calculated against observed excited-state energies and magnetic susceptibilities.
ligand-field analysis for the more complicated f n electronic systems (like Sm3+ and Eu3+) despite a considerable amount of experimental data (see, e.g., references on EuXO4 (X: P, As, Sb, V),14 Eu2O3,15,16 EuOCl,17 EuP.18 In this paper we report temperature-dependent magnetic measurements for nine europium(III) oxo-compounds (see Table 1). In addition, we recorded powder reflectance UV/vis/
⎛ d(Eu − L)min ⎞7 em(d) = em(d(Eu − L)min ) ·⎜ ⎟ , ⎝ d(Eu − L) ⎠ (m: σ , π ; L: O, Cl)
Table 1. Europium(III) Oxo-Compounds Investigated in This Paper with the Chromophores Used for the AOM Calculations compound
space group
chromophorea
point group
ref
EuPO4 EuAsO4 EuVO4 EuOCl EuNbO4 EuTaO4 Eu2Ti2O7 EuSbO4 Eu2O3b
P21/n I41/amd I41/amd P4/nmm I2/a I2/a Fd3̅m P21/c Ia3̅
[EuO9] [EuO8] [EuO8] [EuO4Cl5] [EuO8] [EuO8] [EuO8] [EuO8] [Eu1O6] [Eu2O6]
C1 D2d D2d C4v C2 C2 D3d C1 S6 C2
19 20 21 22 23 23 24 25 26
(1)
The europium(III) oxo-compounds under consideration in this paper crystallize in different structure types and show a wide variety of mostly low-symmetry chromophores [EuIIIOn] (6 ≤ n ≤ 9). Table 1 gives a summary of the chromophores and their point groups. Distances d(Eu−L) (L: O, Cl) as well as the orientation of the global (or “molecular”) coordinate systems are shown in Figure 2. In these compounds, the Eu3+ ions are coordinated only by oxygen or by oxygen and chlorine with a total coordination number of eight or nine except for Eu2O3. The C-type structure of Eu2O3 shows two different europium sites with coordination number c.n. (Eu3+) = 6.26 In AOM the orientation of the f-orbitals relative to the ligands is defined by the global coordinate system, which is shown for each chromophore in Figure 2. In principle, calculated energies of the f-electron states are independent of the orientation of the global coordinate system.35 In our calculations for convenience, the global z-axis has always been set along to the shortest europium−oxygen bond in a given chromophore. Matching Procedure. In the first step of ligand-field analysis for the europium(III) compounds in Table 1 validity of the aforementioned assumptions (β = 1.0, k = 1.0, eσ ≈ d(Eu− O)−7.0, eπ = 1/4eσ) was checked for EuPO4 and Eu2O3. The powder reflectance spectra for the two compounds are given in Figures 3 and 4, respectively. Introducing Slater−Condon− Shortley parameters F2, F4, and F6 and the spin−orbit coupling constant ζ for the free Eu3+ ion from ab initio calculations reported in the literature33 into BonnMag allows a generally good match of the calculated excited-state energies to the observed spectra for EuPO4 and Eu2O3. The match to those observed for the free Eu3+ ion33 and the very similar data compiled in Diekes diagram37 is equally good. Typically deviations of Δν̃ = ±300 cm−1 are found. The observed and calculated energies of the 7F0 → 5D2 transition of Eu3+ in EuPO4 and Eu2O3 were chosen to visualize the influence of β on the calculated energies (Figure 5). For EuPO4 β ≈ 1.002 and for Eu2O3 reduction of β ∼ 0.996 would allow a slightly better match to the experimental energies. These variations are, however, that small that based on currently available data we refrain from discussing them. The influence of the ligand-field on the spectra of EuPO4 and C-type Eu2O3 is shown by the splitting of the 5D2 state (Figures 3 and 4). For EuPO4 the splitting Δν̃(5D2)exp = 57 cm−1 is observed; for Eu2O3 we find Δν̃(5D2)exp = 157 cm−1. Variation of eσ,max(EuPO4), which is eσ(Eu−O) for dmin(Eu−O) = 2.380 Å in the chromophore [EuIIIO9], in the range from 1/3 to 3 times that of the “best fit” value leads to a significant variation of the calculated splitting of the 5D2 state (Figure 6). Obviously, Δν̃(5D2)calc increases with eσ,max(Eu−O). As expected, the calculated magnetic moments (Figure 7a−e) depend in the same way on eσ,max(Eu−O). The splitting of the 5D2 state as
a
All ligands closer to Eu3+ than the next cation were included. bCType.19
NIR (NIR = near-infrared) spectra for these compounds. To provide detailed experimental evidence for the prevailing ligand-field splittings, the electronic excitation 7F0 → 5D2 was measured with high resolution in the powder reflectance spectra. By matching the AOM calculations to the experimental data, we shall eventually obtain for the first time AOM parameters eσ(Eu3+−O2−) that allow rationalization of the ligand-field splitting in chromophores [EuIIIOn] within a chemically meaningful model.
■
ANGULAR OVERLAP MODELING WITH BONNMAG AOM Parameters. Since its first introduction in the early 1960s7,27,28 several accounts on the parametrization of ligandfield effects on dn (1 ≤ n ≤ 9) and f n (1 ≤ n ≤ 3 and 11 ≤ n ≤ 13) electronic systems by angular overlap modeling have been published.29,30 For details on AOM we refer to literature.4−7 Surprisingly, given the success of the AOM approach, only recently with BonnMag8 a computer program has become available that allows modeling of all f n systems (1 ≤ n ≤ 13). The energies of the free ion electronic states are introduced into the AOM calculations via the Slater−Condon−Shortley parameters F2, F4, F631 and the spin−orbit coupling constant ζ.32 Throughout this paper for the free Eu3+ ion F2 = 401 cm−1, F4 = 55.38 cm−1, F6 = 6.06 cm−1, and the spin−orbit coupling constant ζ = 1320 cm−1 were assumed according to literature.33 These numbers were not reduced (nephelauxetic ratio β = 1.0), and the Stevens orbital reduction factor34 k = 1.0 was maintained in all calculations. Nevertheless, for the sake of simplicity, comparability, and transferability of the parameters we refrained from introducing a larger number of independent parameters for the free ion as it is occasionally reported in literature (e.g., on Cs2NaLnCl647). In the AOM approach, which is based on molecular orbital theory,35 the global ligand-field is decomposed into individual contributions from the various metal−ligand interactions, which are described by the AOM parameters eσ and eπ. In accordance with literature, these parameters are assumed to 9236
DOI: 10.1021/acs.inorgchem.7b01287 Inorg. Chem. 2017, 56, 9235−9246
Article
Inorganic Chemistry
Figure 3. UV/vis/NIR spectra of EuPO4 (blue curves). Comparison of results from AOM (zero-phonon lines, black ticks at the bottom) and estimated intensities (Judd−Ofelt theory, height of black ticks) with experimental spectra. (inset) Transition 7F0 → 5D2 measured separately with higher resolution (0.05 nm steps, slit 0.015 mm).
Figure 4. UV/vis/NIR spectra of Eu2O3 (blue curves). Comparison of results from AOM (zero-phonon lines, black ticks at the bottom) and estimated intensities (Judd−Ofelt theory, height of black ticks) for the two crystallographically independent chromophores with experimental spectra. (inset) Transition 7F0 → 5D2 measured separately with higher resolution (0.05 nm steps, slit 0.015 mm).
= 404 cm−1 and the corresponding eπ,iso(Eu−O) = 101 cm−1. Variation of n in the relation eσ ≈ d(Eu−O)−n in the range of 3 ≤ n ≤ 9 (with best fit eσ) for EuPO4 shows no significant influence on the splitting of the 5D2 state (47 ≤ Δν̃(5D2)calc ≤ 50 cm−1 for 3 ≤ n ≤ 9). Neither are the temperature-dependent magnetic moments (Figures 7) significantly affected. Variation of the ratio eπ/eσ in the range from 0.1 to 0.5 (with best fit eσ) for EuPO4 shows also no influence (Figure 7f). The same variations of eσ,max and n (in the relation eσ ≈ d(Eu−O)−n) for Eu2O3 lead to same results as for EuPO4 (Figure 8). As expected from theory, the influence of the variation of β on the magnetic moment of Eu3+ and its temperature
Figure 2. ORTEP representation of the chromophores [EuIIIOn] and [EuIIIO4Cl5] used for the AOM calculations. The global orthogonal coordinate systems with the central atom as origin are indicated in red.
well as the observed temperature dependence are nicely accounted for by the “best fit” AOM parameters eσ,max(EuPO4) 9237
DOI: 10.1021/acs.inorgchem.7b01287 Inorg. Chem. 2017, 56, 9235−9246
Article
Inorganic Chemistry
Figure 5. Energies for the transition 7F0 → 5D2 at ν̃ ≈ 21 500 cm−1. Horizontal dashed lines indicate experimental energies. Energies calculated with variable β for EuPO4 (a) and Eu 2O3 with chromophores [Eu1O6] (b) and [Eu2O6] (c).
Figure 7. Variation of μ/μB vs T for EuPO4 with eσ,max and with the exponent n in the relation eσ ≈ d(Eu−O)−n (solid lines) in comparison to the observed temperature dependence. (a) 3eσ,max., (b) 2eσ,max., (c) best fit eσ,max, (d) 2/3eσ,max, (e) 1/3eσ,max, (f) 3 ≤ n ≤ 9.
■
RESULTS AND DISCUSSION Experimental Results. The optical transitions in the UV/ vis/NIR spectra of the europium(III) oxo-compounds reported here (Figures 3, 4, and 9) are in broad agreement with those summarized for Eu3+ ions (neglecting ligand-field splitting) in Diekes diagram.28 For EuPO4, EuAsO4, EuSbO4, EuVO4, Eu2O3, EuOCl electronic absorption spectra with f−f intrashell transitions have been reported before.14,39,40 Our spectra are in good agreement with these reports. The influence of the ligand field on the excited states of Eu3+ is shown by our high-resolution powder reflectance measurements of the transition 7F0 → 5D2 (Figures 3, 4, and 9), which have not been measured before. Quite variable splitting patterns are observed and will be discussed subsequently. The number of electronic transitions to the sublevels of 5D2 (Figure 4) in C-type Eu2O3 is in nice agreement with the presence of the two crystallographically independent chromophores [Eu1O6] and [Eu2O6]. We also want to point out the quite different intensities for these transitions, which are rather strong in the sesquioxide and below the detection limit in Eu2Ti2O7 (Figures 3, 4, and 9). Eventually, the spectra provide some hints on electronic excitation in addition to the f−f transitions centered on Eu3+. In case of EuVO4 two broad, prominent bands at ν̃ ≈ 25 000 cm−1 and ν̃ ≈ 32 500 cm−1 are observed, which are likely to originate from the transition from the Eu3+ ground-state 7F0 to the empty d-orbitals of V5+ in the tetrahedral vanadate(V) group. UV/vis spectra of Ca3(VO4)2 show two broad bands caused by ligand-
Figure 6. Variation of the calculated splittings Δν̃(5D2)calc with m·eσ,max for EuPO4 (●) and Eu2O3 (Eu1:■, Eu2:▲) in comparison to the observed splittings (dashed horizontal lines).
dependence is negliable. On the contrary, the influence of k on μ/μB versus T is high (see Figure S1 in Supporting Information). Significant reduction of k without reduction of β appears, however, unreasonable.35 For AOM of EuOCl the parameters eσ(Eu−Cl) with eπ,iso(Eu−Cl) = 1/4 eσ(Eu−Cl) were taken from Cs2NaEuCl6.38 The difference between the distances d(Eu−Cl) in EuOCl and Cs2NaEuCl6 were accounted for by relation 1. Matching the calculated optical transition energies and temperature-dependent magnetic moments of the remaining compounds (EuVO4, EuAsO4, EuOCl, Eu2Ti2O7, EuNbO4, EuTaO4, EuSbO4) to the experimental data was achieved just by variation of eσ, max(Eu−O) for each compound. “Best fit” parameters (Table 2 and S1) were derived by visual comparison of calculated and observed data. Table 2 provides a summary of the AOM parametrization used in these calculations. Table S1 provides a summary of all AOM parameters for the investigated europium compounds. In Figures 9 and 10 calculated and observed spectra and magnetic moments are compared. 9238
DOI: 10.1021/acs.inorgchem.7b01287 Inorg. Chem. 2017, 56, 9235−9246
Article
Inorganic Chemistry
is also caused by LMCT, in that case O2− → Sb5+. In the same way LMCT from O2− to empty d-orbitals of Ti4+, Nb5+, and Ta5+ may explain the rising “background” at higher wavenumbers in the corresponding compounds. The estimation of line strengths for the f−f transitions according to Judd−Ofelt theory42,43 as implemented in BonnMag is in rough agreement with the observations on the europium(III) compounds. The generally very strong absorption related to the transition 7F0 → 7F6 is always calculated to be the strongest. The spectra measured at ambient temperature show no hints on “hot bands” (excitation from the thermally populated 7F1 state). On closer inspection, the limits of Judd− Ofelt theory become apparent, as can be seen for the transitions 7 F0 → 5D1 and 7F0 → 5D2 at ν̃ ≈ 19 000 cm−1 and ν̃ ≈ 21 500 cm−1, respectively. As explained in literature13 the former are caused by a magnetic dipole mechanism, while the latter are described as hypersensitive with respect to the excitation mechanism. Both effects are not covered by the Judd−Ofelt theory. Magnetic Measurements. Rather surprisingly Bohr magneton numbers μexp/μB at 298 K of the oxo-compounds from EuOCl to EuSbO4 cover the unexpectedly wide range from 3.21 to 3.56 (see Table 3). On the basis of various considerations and observations (see Experimental Section) we estimate the accuracy of our μexp/μB to be better than ±0.03. This estimation is supported by comparison of our data for EuPO4, EuAsO4, EuSbO4, EuVO4, EuOCl, and Eu2Ti2O7 to magnetic measurements reported in literature.14,17,44,45 Several reports are published in literature on the magnetic behavior of C-type Eu2O3.15,16,46 These data are in good agreement with our measurements. For none of the oxo-compounds reported here hints on magnetic ordering were found. Ligand-Field Effects and Angular Overlap Modeling. The ticks representing the calculated excited-state energies for the various europium(III) oxo-compounds (“zero-phonon lines”, Figures 3, 4, and 9) match nicely the observed splitting of the 5D2 level of the Eu3+ ions by the various ligand fields. Absolute deviations (as a consequence of slightly deviating free ion parameters F2, F4, F6, and ζ) are generally smaller than 300 cm−1. The magnitude of the LF-splitting Δν̃(5D2)calc is generally reproduced by AOM with deviations less than 20 cm−1 (Table 3). Furthermore, the temperature dependence of the Bohr magneton numbers μexp/μB, which is related to the LF splitting of the 7F1 level are, too, very well-accounted for by AOM (Figures 7, 8, and 10). The deviations of μ/μB = ±0.12 are found. The match between observed and calculated data μ/ μB versus T is slightly less good for Eu2O3, EuSbO4, and Eu2Ti2O7, which show the highest Bohr magneton numbers at 298 K (the highest contribution of thermally populated 7F1 sublevels). Nevertheless, our AOM treatment provides the first ligand-field analysis of C-type Eu2O3 (Figures 4, 8, 10) allowing for the two quite different chromophores in contrast to work reported in literature.15,16,46 Overall, AOM describes the effects related to the splitting of the free ion levels under the ligand field with good accuracy. Obviously, the electronic states of Eu3+ in the europium(III) oxo-compounds considered here are well-described by using the Slater−Condon−Shortley parameters for the free Eu3+ ion from literature,33 the AOM parameters eσ and eπ reported in this paper (see Tables 2 and S1 in Supporting Information), and the angular dependent overlap integrals derived from the geometric structure of the chromophores. The radial part of the overlap integral is well-accounted for by the dependence of
Figure 8. Variation of μ/μB vs T for Eu2O3 (C-type) with eσ,max and with the exponent n in the relation eσ ≈ d(Eu−O)−n (solid lines) in comparison to the observed temperature dependence. Note that the experimental as well as the calculated values for μ/μB vs T are the weighted average for the two crystallographically independent chromophores [Eu1O6] and [Eu2O6]. (a) 3eσ,max, (b) 2eσ,max, (c) “best fit” eσ,max, (d) 2/3eσ,max, (e) 1/3eσ,max, (f) 3 ≤ n ≤ 9.
Table 2. Summary of AOM Parametrization for All Europium(III) Oxo-Compounds under Consideration Slater−Condon−Shortley parameters for Eu3+ (cm−1)33 F2 = 401.0, F4 = 55.38, F6 = 6.06 (β = F2/F2, f.i. = 1.0) spin−orbit coupling constant ζ = 1320.0 cm−133 Stevens orbital reduction factor k = 1.0 AOM parameters (all cm−1); eπ,iso(Eu−O) = 1/4 eσ(Eu−O) compound EuPO4 EuAsO4 EuVO4 EuOCl EuNbO4 EuTaO4 Eu2Ti2O7 EuSbO4 Eu2O3 (C-type)
dmin(Eu−O) 2.380 2.350 2.321 2.286 2.366 2.366 2.210 2.263 2.304 (Eu1) 2.287 (Eu2)
eσ,max(Eu−O) 404.0 441.5 481.6 535.7 589.5 589.5 882.4 977.5 862.0 (Eu1) 907.9 (Eu2)
eσ,norm(Eu− O)2.38 Ǻ 404 404 404 404 525 525 525 687 687
to-metal charge transfer (LMCT) O2− → V5+.41 The energy difference between these two bands for Ca3(VO4)2 is Δν̃ ≈ 7500 cm−1 as found here for EuVO4. We believe that the steep rise in the reflectance spectrum of EuSbO4 at ν̃ = 20 000 cm−1 9239
DOI: 10.1021/acs.inorgchem.7b01287 Inorg. Chem. 2017, 56, 9235−9246
Article
Inorganic Chemistry
Figure 9. UV/vis/NIR spectra (ambient temperature) of europium(III) oxo-compounds (blue curves). Comparison of results from AOM (zerophonon lines, black ticks at the bottom) and estimated intensities (Judd−Ofelt theory, height of black ticks). 9240
DOI: 10.1021/acs.inorgchem.7b01287 Inorg. Chem. 2017, 56, 9235−9246
Article
Inorganic Chemistry
Figure 10. Graphical representation of the experimental Bohr magneton numbers μ/μB vs T and comparison to the calculated numbers (solid lines). For Eu2O3 the weighted average for the two crystallographically independent chromophores [Eu1O6] and [Eu2O6] (b) as well as the individual contributions of Eu1 (a) Eu2 (c) are shown. 9241
DOI: 10.1021/acs.inorgchem.7b01287 Inorg. Chem. 2017, 56, 9235−9246
Article
Inorganic Chemistry
the values eσ(Ln−O) = 475 cm−1 (d(Nd−O) = 2.386 Å in NdAlO367), 450 cm−1 (d(Nd−O) = 2.4 Å in Nd2O367), 318 cm−1 (d(Er−O) = 2.402 Å in Er(OH)367), and 360 cm−1 (d(Yb−O) = 2.19 Å in CsYbO268). In case of EuOCl the variation of the eσ(Eu−Cl) obtained from the scaling of the value reported for Cs2NaEuCl6 shows only a rather small influence on the calculated transition energies and magnetic moments. The fitting procedure for the whole set of experimental data shows clearly that eσ,max(Eu−O) does not only depend on the distance d(Eu−O) but also on some additional effect. Thus, for matching the observed data for all oxo-compounds the AOM parameter eσ,max(Eu−O) (normalized to d(Eu−O) = 2.38 Å) had to be increased significantly in the series EuPO4 (EuVO4, EuAsO4, EuOCl) → Eu2Ti2O7 → EuNbO4 (EuTaO4) → Eu2O3 (EuSbO4) from 404 to 687 cm−1. Our data suggest an increase of eσ(Eu−O) not only with decreasing distance d(Eu− O) but also with the basicity of oxygen in the various compounds. Using the optical basicity Λ54,55 as a measure with Λ(EuPO4) = 0.62, Λ(EuNbO4) = 0.79, and Λ(Eu2O3) = 1.10, a roughly linear increase of eσ(Eu−O) on Λ is found. The dependence of eσ(Eu−O) on d(Eu−O) and Λ is visualized in Figure 11.
Table 3. Comparison of Observed and Calculated Splittings Δν̃(5D2)calc and of the Bohr Magneton Numbers μ/μB (at 298 K) to the Wybourne Ligand-Field Splitting Δ for the Europium(III) Oxo-Compounds under Consideration compound EuPO4 EuAsO4 EuVO4 EuOCl EuNbO4 EuTaO4 Eu2Ti2O7 EuSbO4 Eu2O3 (C-type)
Δν̃(5D2)calc (cm−1)
Δν̃(5D2)calc (cm−1)
μexp/μB (at 298 K)
μexp/μB (at 298 K)
Δc (cm−1)
57 (50)b (57)b 40 101 102 d 120 157a
48 85 (50) 84 (50) 49 96 96 146 139 174 (Eu1)
3.25 3.23 3.27 3.21 3.35 3.33 3.55 3.56 3.50
3.23 3.24 3.24 3.24 3.36 3.34 3.56 3.55 3.51e
606 663 656 550 1138 1102 1157 1440 2061
133 (Eu2)
1611
a
Energy difference between the highest and the lowest transition (see Figures 3, 4, and 9). bOne split term is not observed. cCalculated using definition by Wybourne. dNot observed. eWeighted average of μcalc/ μB(Eu1) = 3.73 and μcalc/μB(Eu2) = 3.43.
eσ(Eu3+−O2−) on d(Eu3+−O2−)−7 (eq 1). An even better fit of calculated to observed excited-state energies should be possible in allowing for a more elaborate model for the description of the free ions electronic states as it was described for Cs2NaEuCl6 containing the [EuIIICl6] chromophore.47 At the current state we refrain from this procedure for two reasons. In particular, for low-symmetry chromophores the fitting of a large parameter set (including Slater-Condon-Shortley, ς, Eave, α, β, γ, T i (i = 2,3,4,6,7,8), Mk (k = 0,2,4), and Pk (k = 2,4,6) is quite challenging. These parameters are defined according to the conventions used in refs 48−52. Second, it is reasonably safe to assume that the AOM (LF) parameters and the parameters for the free ion are independent of each other. Since our investigation focuses on the ligand-field effects and not on the perfect match of the free ion energies, only F2, F4, F6, and ζ are used for the parametrization of the free ion states to keep the model as simple as possible. Considering the higher-order parameters47 for the free ion would lead to a very difficult parametrization procedure. Furthermore, least-squares fitting of ligand-field and free-ion parameters would not be possible by any means. As already mentioned the eσ,norm(Eu−O) value derived here for d(Eu−O) = 2.38 Å was found by matching the calculated excited-state energies and (!) magnetic moments against the observed data for EuPO4. For a more precise estimate of the AOM parameter eσ the splitting of the transition 7F0 → 5D2 for EuPO4 at ν̃ ≈ 21 500 cm−1 was considered. This splitting Δν̃(5D2)calc = 57 cm−1 deviates by only 8 cm−1 from the calculated value (Table 3). Our eσ,norm(Eu−O) derived for EuPO4 agrees well with the AOM parameter eσ(Eu3+−O2−) = 549 cm−1, which has been reported for Eu3+ doped in YOCl53 for d(Eu−O) = 2.278 Å. This is the average of the distance values for EuOCl (d(Eu−O) = 2.286 Å) and YOCl (d(Y−O) = 2.27 Å). The eσ,norm(Eu−O) value for the doped compound converted to d(Eu−O) = 2.38 Å is 399 cm−1. The assumed d(Eu−O) for Eu3+ doped in YOCl appears to be reasonable. For Eu3+ doped in Y2O353 for d(Eu−O) = 2.250 Å and d(Eu− O) = 2.330 Å the values eσ(Eu3+−O2−) = 881 cm−1 and eσ(Eu3+−O2−) = 391 cm−1, respectively, have been reported in literature. The values for eσ(Eu−O) derived for EuPO4 range from 160 to 404 cm−1. For other lanthanide oxo-compounds
Figure 11. Dependence of eσ(Eu−O) on d(Eu−O) and optical basicity Λ.
In literature occasionally an “effective ligand-field splitting” (better splitting of the 7F1 state by the ligand field) has been considered as a parameter in van Vlecks equation to describe empirically the contribution of the first excited state to the overall Bohr magneton number and thus the temperature dependence of μ/μB.56 Comparison of the various ligand-field splittings Δ (Table 3) as derived from the AOM parameters eσ and eπ and the splitting patterns calculated by AOM for the 7F1 states in the various chromophores (Figure 12) regarded here provides a much more detailed picture. By using the definition given by Wybourne57 (eq 2) the ligand-field strength Δ was calculated for all europium(III) compounds from the obtained “best-fit” AOM parameters eσ and eπ. These calculations were performed within BonnMag. Note: While Δ is directly related to the AOM parameters deduction of eσ and eπ from the Wybourne parameters is only possible by an indirect iteration procedure.57 9242
DOI: 10.1021/acs.inorgchem.7b01287 Inorg. Chem. 2017, 56, 9235−9246
Article
Inorganic Chemistry
of the 7F1 state leads to higher values μexp/μB at 298 K with μcalc/μB (298 K) = 3.21 for the free Eu3+ ion as lower limit. A detailed comparison (Table 3 and Figure 12) of μ/μB(298 K) with Δ shows, however, clear deviations from this simple relation. The calculated splitting of the 7F1 state for EuAsO4 and EuVO4, for example, is very small (ca. 1 cm−1). The calculated magnetic moment at 298 K and its temperature dependence (which matches the experimental data perfectly) are, however, almost the same as for EuPO4 with a calculated splitting of the 7F1 state of 77 cm−1. In contrast, the Wybourne ligand-field strengths Δ for EuAsO4 and EuVO4 (Table 3) are ∼10% higher than Δ(EuPO4). The calculated splitting of the 7F1 state for the chromophore [EuIIIO8] in Eu2Ti2O7 is huge (ca. 350 cm−1; Figure 12). Yet, the magnetic behavior of the titanate is very similar to that of EuSbO4 and Eu2O3 (C-type) for which much smaller splittings of the 7F1 state are calculated (Figures 10 and 12). In the context of this discussion it appears also quite interesting that the ligand-field splitting Δ as well as the calculated magnitude of splitting of the 7F1 state for presumably rather similar chromophores like [Eu1O6] and [Eu2O6] in Eu2O3 can be quite different. It appears therefore even more remarkable that AOM of the two chromophores leads to a good match between calculated and experimental μ/μB versus T. Inspection of the splittings calculated for the 7F1 states (Figure 12) in the various chromophores reveals further effects that will have some impact on μ/μB vs T. The three sublevels resulting from interaction of the 7F1 state with the ligand field follow roughly three different splitting patterns. These might be described as “2 above 1” (as for chromophore [Eu1O6] in Eu2O3), “three well-resolved sublevels” (as, e.g., EuPO4, EuTaO4, and EuSbO4), and “1 above 2” (as for chromophores [Eu2O6] in Eu2O3 and [EuO8] in Eu2Ti2O7). These splitting patterns are a direct consequence of the chromophore (ligandfields) symmetry and, not surprisingly, both symmetry and ligand-field strength Δ will affect the magnetic behavior. Figure 12 shows also nicely that configuration interaction between the 7F1 sublevels and the sublevels of appropriate symmetry of higher states can take on significant values. Without configuration interaction the center of gravity of the sublevels should be the same for the free Eu3+ ion, where the 7 F1 state is 373 cm−1 above the 7F0 ground state, and for all the chromophores. The fact that this is obviously not the case points to significant contributions of configuration interaction that will lower the 7F1 sublevels with respect to the ground state and will also lead to admixture of contributions from levels with even higher magnetic moment than 7F1. Our AOM points for the [EuO8] chromophore in Eu2Ti2O7 to a possible crossing in energy of the highest sublevel of 7F1 and the lowest of 7F2 on slight variation of eσ, max and of the ratio eπ/eσ. In particular, for the oxo-compounds with the highest values μexp/μB(298 K), namely, EuSbO4, Eu2O3, and Eu2Ti2O7, such effects might be responsible for the slightly less good match of calculated and observed data (Figure 10). To settle this question future measurement of emission spectra of the oxo-compounds considered here is planned. Via such measurements direct characterization of the splitting of 7F1 and 7F2 states would be possible.
Figure 12. Calculated (AOM, parameters in Table 2) splitting of the first magnetic state 7F1 of Eu3+ in the chromophores [EuIIIOn] and [EuIIIO4Cl5]. The energy (above the ground-state 7F0) assigned to the parental term 7F1 represents the center of gravity of the energies of the LF split terms. With the same SCS parameters (see Table 2) for the free Eu3+ ion E(7F1) = 373 cm−1 is obtained.
S= ⎡ ⎢1 ⎢⎣ 3
∑ k
⎤1/2 1 k 2 k 2 k 2 ⎥ ((B0 ) + 2 ∑ ((ReBq ) + (ImBq ) )) ⎥⎦ 2k + 1 q>0 (2)
Δ ligand-field splitting Bk0 Wybourne parameters with q = 0 and k = 2, 4, 6 ReBkq real part of Wybourne parameters with q = 1−6 and k = 2, 4, 6 ImBkq imaginary part of Wybourne parameters with q = 1−6 and k = 2, 4, 6 Clearly, the Bohr magneton number is correlated to the splitting of the 7F1 state. In a first approximation larger splitting 9243
DOI: 10.1021/acs.inorgchem.7b01287 Inorg. Chem. 2017, 56, 9235−9246
Article
Inorganic Chemistry
■
prepared via solution combustion synthesis64 with urea as chelating agent and fuel. Therefore, at first Eu2O3 was solved in 2 M nitric acid. After that a stoichiometric amount of arsenic acid solution (c = 95.4 mmol L−1) and a threefold excess of urea compared to the amount of metal ion was added. The combustion took place in a 400 °C furnace. Further annealing at 1000 °C in air yielded a phase-pure product. All powders but EuSbO4 were single-phase according to their XRPD (IPGuinier technique). The EuSbO4 sample contains a trace of an unidentified crystalline impurity. XRPD (Guinier technique). Powder diffraction patterns for phase identification and purity control were collected at ambient temperature using an imaging plate Guinier camera (HUBER G670, Cu -Kα1 radiation, λ = 1.54059 Å). Powder patterns were recorded for 15 min in the range of 0 ≤ 2θ ≤ 100°. Powder Reflectance Spectra. The diffuse powder reflectance spectra (4000 ≤ ν̃ ≤ 35 000 cm−1) were recorded at ambient temperature using two modified CARY 14 and CARY 17 spectrophotometers (OLIS Inc., USA), which were equipped with integrating spheres. For the whole spectral range three different experimental setups (detector, slit, scan range, step width) were used. From 280 to 600 nm (UV range) a total of 640 data points were collected by a photomultiplier detector (PMT) system with a scan rate of 1.0 nm·sec−1, step width 0.5 nm, and the slit width of 0.1 mm. From 300 to 900 nm (Vis range) and from 600 to 2600 nm (NIR range) the total of collected data points was 600 and 500, respectively, with scan rates of 1.0 and 4.0 nm·sec−1, accordingly. For the Vis region a PMT detector (slit width of 0.06 mm) was used. In the NIR region the data were collected by a PbS detector at a variable slit width (1.4 to 2.2 mm). In every case the intensity of BaSO4 was measured as standard. The diffuse reflectance (K/S) was calculated using the Kubelka−Munk function, Equation 3.
CONCLUSIONS
(1) Our results show that AOM with the newly developed computer program BonnMag allows a very good match between calculated and observed optical transition energies and temperature-dependent magnetic moments for a series of chemically quite different europium(III) oxo-compounds. (2) For the first time detailed ligand-field analyses within the framework of molecular orbital (MO) theory with chemically meaningful parameters has been successfully performed for a series of europium(III) compounds. (3) Our analyses of the various (mostly low-symmetry) ligand-fields around the Eu3+ ions show that the angular part of the ligand field is nicely accounted for by the AOM approach. The radial part, as described by the eσ(Eu3+−O) depends to a good approximation on the distances d(Eu3+−O)−7.0. In addition, an unexpectedly strong influence of the second coordination sphere (next-nearest cations) on eσ(Eu3+−O) has been discovered. We attribute this effect to variation of the electronic polarizability of oxygen α(O2−). Qualitatively, this influence might be correlated to the optical basicity of the various europium(III) oxo-compounds. For a quantitative understanding of this effect data on α(O2−) from measurement of the dielectric constants would be most desirable. (4) The ligand-field analysis for C-type Eu2O3 not only accounts for the two quite different chromophores [Eu1O6] and [Eu2O6] but would be impossible without this differentiation. Such a simplification was suggested several times in literature.15,16 (5) With BonnMag ligand-field analysis for all “complicated” f n ions can now easily be accomplished, provided experimental data of sufficient quality for comparison are available. (6) Our results show the predictive power of AOM for describing ligand-field effects in f n systems. (7) Characterization of synthetically slightly more demanding compounds like the highly basic oxide BaEu2O458 and the rather acidic sulfate Eu2(SO4)359 would be very welcome to provide an extended experimental base for our suggestion of a close relation between eσ(Eu3+−O2−) and the optical basicity of a multinary oxide. Following this reasoning experimental data on the polarizability of the oxygen atoms in multinary europium(III) oxides (from measurements of dielectric constants) would also be of high interest to a more detailed understanding of chemical bonding between the environment and Eu3+, in particular, and all RE3+ in general.
■
K /S =
(1 − R diff )2 2R diff
(3)
where the diffuse reflectance Rdiff = (Isample/Istandard) Magnetic Susceptibility Measurements. All samples except EuSbO4 were single-phase according to their XRPD. The samples were packed in kapton foil and attached to the sample holder rod for measuring the magnetic properties in a VSM (Quantum Design Physical-Property-Measurement-System)65 in the temperature range of 2.0−300 K with magnetic flux densities up to 10 kOe. A diamagnetic correction was applied to the observed weight data according to the method of atom and group increments.66 With μexp/ μB = 2.8279·(T·χmol)1/2, the magnetic moment per Eu3+ ion was calculated. For EuPO4, Eu2O3, and EuSbO4 samples from various syntheses were measured including field-dependent measurements (100−50 000 Oe). Thus, obtained susceptibilities showed variations of less than ±0.01 μexp/μB.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.7b01287. AOM parameters eσ and eπ for different d(Eu−L). Variation of μ/μB versus T for EuPO4 with k in comparison to the observed temperature dependence (PDF)
EXPERIMENTAL SECTION
Synthesis. Powders of the oxo-compounds EuPO4,60 EuVO4,61 EuSbO4,25 Eu2Ti2O7,62 EuNbO4,23 and EuTaO423 were synthesized as reported in the respective literature. Single-phase C-type Eu2O3 was purchased from Strem Chemicals Inc., with a purity of 99.99% (with respect to other Ln3+) and controlled for phase purity by X-ray powder diffraction (XRPD). No other polymorph of Eu2O3 was found. Synthesis of EuOCl was realized based on the literature; however, the number of reaction steps that led to a phase-pure product could be reduced. Eu2O3 and NH4Cl with a mass ratio of 1:1.5 were dissolved in hydrochloric acid (35%). The water was removed by heating just under the boiling point of the solution. Further annealing of the dry residue at 400 °C led to a homogeneous product.63 EuAsO4 was
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Robert Glaum: 0000-0001-5805-1466 9244
DOI: 10.1021/acs.inorgchem.7b01287 Inorg. Chem. 2017, 56, 9235−9246
Article
Inorganic Chemistry
(18) Jones, E. D. Temperature Dependence of the Eu3+ Spin ⟨S(T) in EuP. J. Appl. Phys. 1968, 39, 1090−1091. (19) Mullica, D. F.; Grossie, D. A.; Boatner, L. A. Coordination Geometry and Structural Determinations of SmPO4, EuPO4, and GdPO4. Inorg. Chim. Acta 1985, 109, 105−110. (20) Golbs, S.; Cardoso-Gil, R.; Schmidt, M. Crystal structure of europium arsenate, EuAsO4. Z. Kristallogr. - New Cryst. Struct. 2009, 224, 169−170. (21) Mullica, D. F.; Sappenfield, E. L.; Abraham, M. M.; Chakoumakos, B. C.; Boatner, L. A. Structural investigations of several LnVO4 compounds. Inorg. Chim. Acta 1996, 248, 85−88. (22) Bärnighausen, H.; Brauer, G.; Schultz, N. Darstellung und Kristallstruktur der Samarium-, Europium- und Ytterbium-oxidbromide LnOBr und Ln3 O4Br. Z. Anorg. Allg. Chem. 1965, 338, 250−265. (23) Keller, C. Ü ber ternäre Oxide des Niobs und Tantals vom Typ ABO4. Z. Anorg. Allg. Chem. 1962, 318, 89−106. (24) Chtoun, E.; Hanebali, L.; Garnier, P.; Kiat, J. M. X-rays and neutrons rietveld analysis of the solid solutions (1-x)(A2Ti2O7) x(MgTiO3)(A = Y or Eu). Eur. J. Solid State Chem. 1997, 34, 553−561. (25) Gerlach, S.; Cardoso-Gil, R.; Milke, E.; Schmidt, M. Zum chemischen Transport und zur Kristallstruktur von Seltenerdmetallantimonaten(V) - RESbO4. Z. Anorg. Allg. Chem. 2007, 633, 83−92. (26) Heiba, Z. K.; Akin, Y.; Sigmund, W.; Hascicek, Y. S. X-ray structure and microstructure determination of the mixed sesquioxides (Eu1‑xYbx)2O3 prepared by a sol-gel process. J. Appl. Crystallogr. 2003, 36, 1411−1416. (27) Urland, W. The application of the angular overlap model in the calculation of paramagnetic principal susceptibilities for fn-electron systems. Chem. Phys. Lett. 1977, 46, 457−460. (28) Urland, W. The interpretation of the crystal field parameters for fn-electron systems by the angular overlap model. The rare-earth hydroxides. Chem. Phys. Lett. 1977, 50, 445−450. (29) Glaum, R.; Gruehn, R.; Moeller, M. Beiträge zum thermischen Verhalten wasserfreier Phosphate. I. - Darstellung und Struktur von αCrPO4. Z. Anorg. Allg. Chem. 1986, 543, 111−116. (30) Schöneborn, M.; Glaum, R. Chemical vapour transport, single crystal structure refinement and electronic absorption spectra of MTi2O2(PO4)2 (MII = Fe, Co, Ni). Z. Anorg. Allg. Chem. 2007, 633, 2568−2578. (31) Condon, E. U.; Shortley, G. H. The Theory of Atomic Spectra; Cambridge Univ. Press, 1967. (32) Lueken, H. Magnetochemie; Teubner: Stuttgart, Germany, 1999. (33) Ofelt, G. S. Structure of the f6 Configuration with Application to Rare-Earth Ions. J. Chem. Phys. 1963, 38, 2171−2180. (34) Stevens, K. W. H. On the Magnetic Properties of Covalent XY{6} Complexes. Proc. R. Soc. London, Ser. A 1953, 219, 542−555. (35) Figgis, B. N.; Hitchman, M. A. Ligand Field Theory and Its Applications; Wiley-VCH: New York, 2000. (36) Urland, W. The assessment of the crystal field parameters for f n electron systems by the angular overlap model. Rare-earth ions in CaWO4. Chem. Phys. 1979, 38, 407−412. (37) Dieke, G. H. Spectra and energy levels of rare earth ions in crystals; Interscience Publishers: New York, 1968. (38) Urland, W. The assessment of the crystal field parameters for fnelectron systems by the angular overlap model. rare-earth ions M3+ in Cs2NaMCl6. Chem. Phys. Lett. 1981, 83, 116−119. (39) Bazzi, R.; Flores, M. A.; Louis, C.; Lebbou, K.; Zhang, W.; Dujardin, C.; Roux, S.; Mercier, B.; Ledoux, G.; Bernstein, E.; Perriat, P.; Tillement, O. Synthesis and properties of europium-based phosphors on the nanometer scale: Eu2O3, Gd2O3:Eu, and Y2O3:Eu. J. Colloid Interface Sci. 2004, 273, 191−197. (40) Stump, N. A.; Dai, S.; Del Cul, G. D.; Peterson, J. R. Absorption Spectrum from Europium Ions in Europium Oxychloride, EuOCl. Spectrosc. Lett. 1992, 25, 1003−1010. (41) Qiu, K.; Li, J.; Li, J.; Lu, X.; Gong, Y.; Li, J. Luminescence property of Ca3(VO4)2:Eu3+ dependence on molar ratio Ca/V and solution combustion synthesis temperature. J. Mater. Sci. 2010, 45, 5456−5462.
Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
■
REFERENCES
We thank N. Wagner and V. Dittrich (both at Univ. of Bonn) for the magnetic measurements and recording of the UV/vis/ NIR spectra. We also thank Dr. M. Schmidt (MPI, Dresden) for magnetic measurements and single crystals of EuSbO4. Many enlightening discussions with W. Urland (Muralto, Switzerland) on AOM and chemical bonding of f n systems are particularly gratefully acknowledged.
(1) Ramanantoanina, H.; Kuri, G.; Daul, C.; Bertsch, J. Core electron excitations in U4+: modelling of the nd105f2 → nd9 5f3 transitions with n = 3, 4 and 5 by ligand field tools and density functional theory. Phys. Chem. Chem. Phys. 2016, 18, 19020−19031. (2) Gatteschi, D.; Sessoli, R.; Cornia, A. Single-molecule magnets based on iron(III)oxo clusters. Chem. Commun. 2000, 725−732. (3) Ishikawa, N.; Sugita, M.; Ishikawa, T.; Koshihara, S.; Kaizu, Y. Lanthanide double-decker complexes functioning as magnets at the single-molecular level. J. Am. Chem. Soc. 2003, 125, 8694−8695. (4) Jørgensen, C. K.; Pappalardo, R.; Schmidtke, H.-H. Do the Ligand Field Parameters in Lanthanides Represent Weak Covalent Bonding? J. Chem. Phys. 1963, 39, 1422−1430. (5) Richardson, D. E. The angular overlap model as a unified bonding model for main group and transition metal compounds: A version suitable for undergraduate inorganic students. J. Chem. Educ. 1993, 70, 372−380. (6) Larsen, E.; LaMar, G. N. The angular overlap model. How to use it and why. J. Chem. Educ. 1974, 51, 633−640. (7) Urland, W. On the ligand-field potential for f electrons in the angular overlap model. Chem. Phys. 1976, 14, 393−401. (8) Bronova, A. planned Ph.D. thesis, University of Bonn, 2017. (9) Bronova, A.; Droß, T.; Glaum, R.; Lueken, H.; Speldrich, M.; Urland, W. Comprehensive Characterization of the Electronic Structure of U(4+) in Uranium(IV) Phosphate Chloride. Inorg. Chem. 2016, 55, 6848−6852. (10) Bronova, A.; Bredow, T.; Glaum, R.; Urland, W. The Electronic States of U4+ in U(PO4)Cl: An Example for Angular Overlap Modeling of 5fn Systems. Inorg. Chem. 2016, 55, 6853−6860. (11) Hund, F. Zur Deutung der Molekelspektren. Eur. Phys. J. A 1927, 40, 742−764. (12) Landé, A. Termstruktur und Zeemaneffekt der Multiplets. Eur. Phys. J. A 1923, 15, 189−205. (13) Binnemans, K. Interpretation of europium(III) spectra. Coord. Chem. Rev. 2015, 295, 1−45. (14) Golbs, S.; Schappacher, F. M.; Pöttgen, R.; Cardoso-Gil, R.; Ormeci, A.; Schwarz, U.; Schnelle, W.; Grin, Y.; Schmidt, M. Europium phosphate, europium arsenate and europium antimonate Correlation of Crystal structure and Physical Properties. Z. Anorg. Allg. Chem. 2013, 639, 2139−2148. (15) Yang, Y.; Liu, N. L. H. Orbitally Anisotropic Exchange and Crystal Field Effects on the Magnetic Susceptibility of (EuxY1‑x)2O3. AIP Conf. Proc. 1971, 5, 801−805. (16) Huang, N. L.; van Vleck, J. H. Effect of the Anisotropic Exchange and the Crystalline Field on the Magnetic Susceptibility of Eu2O3. J. Appl. Phys. 1969, 40, 1144−1146. (17) Aitasalo, T.; Hölsä, J.; Lastusaari, M.; Legendziewicz, J.; Lehto, L.; Lindén, J.; Marysko, M. Structural, magnetic and spectroscopic investigations of europium oxychloride, EuOCl. J. Alloys Compd. 2004, 380, 296−302. 9245
DOI: 10.1021/acs.inorgchem.7b01287 Inorg. Chem. 2017, 56, 9235−9246
Article
Inorganic Chemistry
(66) Weiss, A.; Witte, H. Magnetochemie; Wiley-VCH: Weinheim, Germany, 1973. (67) Goerller-Walrand, C.; Binnemans, K. In Handbook on the Physics and Chemistry of Rare Earths; Eyring, L., Gschneidner, K. A., Eds.; North-Holland: Amsterdam, 1995; Vol. 21. (68) Urland, W. Magnetische Eigenschaften der Normaltemperaturform von CsYbO2. Z. Naturforsch., A: Phys. Sci. 1980, 35, 247−251.
(42) Judd, B. R. Optical Absorption Intensities of Rare-Earth Ions. Phys. Rev. 1962, 127, 750−761. (43) Ofelt, G. S. Intensities of Crystal Spectra of Rare-Earth Ions. J. Chem. Phys. 1962, 37, 511−520. (44) Cashion, J. D.; Cooke, A. H.; Leask, M. J. M.; Thorp, T. L.; Wells, M. R. Crystal Growth and Magnetic Susceptibility of Some Rare-Earth Compounds Part II. J. Mater. Sci. 1968, 3, 402−407. (45) Dasgupta, P.; Jana, Y. M.; Nag Chattopadhyay, A.; Higashinaka, R.; Maeno, Y.; Ghosh, D. Low- temperature measuremens of magnetic susceptibility and specific heat of Eu2Ti2O7 - An XY pyrochlore. J. Phys. Chem. Solids 2007, 68, 347−354. (46) Caro, P.; Porcher, P. The paramagnetic suceptibility of C-Type europium sesquioxide. J. Magn. Magn. Mater. 1986, 58, 61−66. (47) Richardson, F. S.; Reid, M. F.; Dallara, J. J.; Smith, R. D. Energy levels of lanthanide ions in the cubic Cs 2 NaLnCl 6 and Cs2NaLnCl6:Ln3+ (doped) systems. J. Chem. Phys. 1985, 83, 3813− 3830. (48) Crosswhite, H. M.; Crosswhite, H.; Kaseta, F. W.; Sarup, R. The spectrum of Nd3+:LaCl3. J. Chem. Phys. 1976, 64, 1981−1985. (49) Carnall, W. T.; Crosswhite, H.; Crosswhite, H. M.; Conway, J. G. The spectrum of Nd3+:LaCl3. J. Chem. Phys. 1976, 64, 3582−3586. (50) Crosswhite, H. M.; Crosswhite, H.; Edelstein, N.; Rajnak, K. Parametric energy level analysis of Ho3+: LaCl3. J. Chem. Phys. 1977, 67, 3002−3010. (51) Carnall, W. T.; Beitz, J. V.; Crosswhite, H.; Rajnak, K.; Mann, J. B. In Systematics and the Properties of Lanthanides; Sinha, S. P., Ed.; Reidel: Boston, MA, 1983. (52) Carnall, W. T.; Crosswhite, H. M.; Crosswhite, H. Argonne National Laboratory Report; Argonne National Laboratory, 1977. (53) Faucher, M.; Garcia, D.; Jørgenson, C. K. Crystal field strength: Angular overlap analysis of ligand field effects on some rare-earth compounds. Chem. Phys. Lett. 1986, 129, 387−391. (54) (a) Duffy, J. A. Bonding. Energy Levels & Bands in Inorganic Solids; Longman Scientific & Technical, 1990. (b) Duffy, J. A.; Ingram, M. D. Establishment of an Optical Scale for Lewis Basicity in Inorganic Oxyacids, Molten Salts, and Glasses. J. Am. Chem. Soc. 1971, 93, 6448−6454. (55) (a) Lebouteiller, A.; Courtine, P. Improvement of a Bulk Optical Basicity Table for Oxidic Systems. J. Solid State Chem. 1998, 137, 94− 103. (b) Zhao, X.; Wang, X.; Lin, H.; Wang, Z. Electronic polarizability and optical basicity of lanthanide oxides. Phys. B 2007, 392, 132−136. (56) Hulliger, F. Magnetic properties of rare-earth pcnivtides. J. Magn. Magn. Mater. 1978, 8, 183−205. (57) Goerller-Walrand, C.; Binnemans, K. In Handbook on the Physics and Chemistry of Rare Earths; Eyring, L., Gschneidner, K. A., Eds.; North-Holland: Amsterdam, 1996; Vol. 23. (58) Besara, T.; Lundberg, M. S.; Sun, J.; Ramirez, D.; Dong, L.; Whalen, J. B.; Vasquez, R.; Herrera, F.; Allen, J. R.; Davidson, M. W.; Siegrist, T. Single crystal synthesis and magnetism of the BaLn2O4 family. Prog. Solid State Chem. 2014, 42, 23−36. (59) Wickleder, M. Wasserfreie Sulfate der Selten-Erd-Elemente: Synthese und Kristallstruktur von Y2(SO4)3 und Sc2(SO4)3. Z. Anorg. Allg. Chem. 2000, 626, 1468−1472. (60) Schwarz, H. Ü ber Verbindungen des Typs MeII0.5Me II0.5 XVO4 mit Huttonit-oder Zirkonstruktur. I. Phosphate und Arsenate. (MeII=Cd, Ca, Sr, Pb, Ba; MeIV=Th; Xv = P, As). Z. Anorg. Allg. Chem. 1963, 323, 44−54. (61) Milligan, W. O.; Watt, L. M.; Rachford, H. H., Jr. X-Ray Diffraction Studies on Heavy Metal Orthovanadates. J. Phys. Colloid Chem. 1949, 53, 227−234. (62) Brixner, L. H. Preparation and Properties of the Ln2Ti2O7-Type Rare Earth Titanate. Inorg. Chem. 1964, 3, 1065−1067. (63) Taylor, M. D.; Carter, C. P. Preparation of anhydrous lanthanide halides, especially iodides. J. Inorg. Nucl. Chem. 1962, 24, 387−391. (64) Patil, K. C.; Aruna, S. T.; Mimani, T. Combustion synthesis: in Update. Curr. Opin. Solid State Mater. Sci. 2002, 6, 507−512. (65) Buckel, W. Supraleitung, 5th ed.; VCH Verlagsgesellschaft: Weinheim, Germany, 1963. 9246
DOI: 10.1021/acs.inorgchem.7b01287 Inorg. Chem. 2017, 56, 9235−9246
