research communications
0.64Fe2.15Ge0.21O4
of spinel-type LiaUniversity of Salzburg, Department Chemistry and Physics of Materials, Hellbrunnerstrasse 34, 5020 Salzburg, Austria
*Correspondence e-mail: guenther.redhammer@sbg.ac.at
Spinel-type Li0.64Fe2.15Ge0.21O4, lithium diiron(III) germanium tetraoxide, has been formed as a by-product during growth of an Li–Fe–Ge pyroxene-type material. In the title compound, lithium is ordered on the octahedral B sites, while Ge4+ orders onto the tetrahedral A sites, and iron distributes over both the octahedral and tetrahedral sites, and is in the trivalent state as determined from Mössbauer spectroscopy. The oxygen parameter u is 0.2543; thus, the spinel is close to having an ideal cubic closed packing of the O atoms. The title spinel is compared with other Li- and Ge-containing spinels.
Keywords: crystal structure; spinel; Mössbauer spectroscopy; geoscience.
CCDC reference: 1463892
1. Chemical context
The minerals of the spinel group are widely occurring compounds in the geosphere and are important not only in geoscience but also in many other disciplines. In recent years, in particular, Li-containing spinels like LiMn2O4 or Li0.5Fe2.5O4 have attracted much interest in battery technology as possible candidates for cathode materials in lithium ion secondary batteries (Liu et al., 2014; Patil et al., 2016; Thackeray et al., 1983). The ideal spinel structure consists of a closed packing of anions X, with one-eighth of the tetrahedral interstices and one-half of the octahedral interstices occupied by the cations. The vast majority of spinels crystallize in the Fdm. Here the cations in tetrahedral coordination occupy special position 8a (point symmetry 3m, at , , ), while the octahedrally coordinated cations reside on special position 16d (point symmetry m at , , ). The anions are at equipoint position 32e, which requires one positional parameter, often denoted as the u parameter. For u = 0.25, an ideal cubic closed packing of anions is realized and the octahedral bond length is 1.155 times larger than the tetrahedral one. Following Hill et al. (1979), variations in u reflect the adjustment of the structure to accommodate cations of different size in octahedral and tetrahedral positions. Increasing the value of u above 0.25 moves the anions away along [111] from the nearest tetrahedral cation, thereby increasing the size of the tetrahedron at the extent of the size of the octahedron. The majority of the spinels can be described with the general formula AB2O4, with the A and B cations having the formal charges A = 2 and B = 3 (2,3 spinels) or A = 4 and B = 2 (4,2 spinels). The perfect normal spinel is one in which the single A cation occupies the tetrahedral site and the two B cations reside at the two equivalent octahedral positions. Introducing parentheses, i.e. (…) and brackets, i.e. […], for tetrahedral and octahedral coordination, respectively, one may write the normal spinels in the form (A)[B2]O4. In contrast, the complete inverse spinel has a cationic distribution of (B)[AB]O4 (O'Neill & Navrotsky, 1983). More detailed reviews on the spinel structure and crystal chemistry can be found, for example, in Biagioni & Pasero (2014), Harrison & Putnis (1998), Hill et al. (1979) and O'Neill & Navrotsky (1983).
Germanium-containing spinels are considered to belong to the normal spinels, with a full ordering of Ge4+ onto the tetrahedral A site, while metal cations M order onto the octahedral B sites. This was demonstrated by, among others, Von Dreele et al. (1977) for GeMg2O4 and Welch et al. (2001) for the mineral brunogeierite (GeFe2O4). For LiMn2O4 and LiNi0.5Mn1.5O4, which represent excellent cathode materials, it was found that Li+ orders onto the tetrahedral site (Berg et al., 1998; Liu et al., 2014). Also for LiCrGeO4, Touboul & Bourée (1993) reported an almost exclusive ordering of Li+ for the tetrahedral site, while Cr3+ and Ge4+ occupy the octahedral sites. Different to this is the spinel Li0.5Fe2.5O4. This compound is an inverse spinel in which Fe3+ is ordered onto the tetrahedral site, while Li+ and the remaining Fe3+ are distributed over the octahedral site (Hankare et al., 2009; Patil et al., 2016; Tomas et al., 1983). This cationic distribution is thus similar to that in the inverse spinel magnetite, FeFe2O4 (Fleet, 1981).
During the synthesis of Li–Fe–Ge pyroxenes (Redhammer et al., 2009, 2010), black octahedral-shaped single crystals were frequently obtained, which turned out to be a spinel-type compound with significant Li+ and small Ge4+ concentrations. We present here the structure and 57Fe Mössbauer spectroscopic characterization of these crystals.
2. Structural commentary
The structure of the title compound is shown in Fig. 1. The site-occupation indicates that Li+ orders onto the octahedral B site, while Ge4+ is found on the tetrahedral A site, indicating a partial inverse spinel arrangement; iron is distributed over both sites. The derived crystal chemical formula of the title compound is thus (Fe3+0.79Ge4+0.21)[Li+0.64Fe3+1.36]O4, with the valence state of iron determined from 57Fe Mössbauer spectroscopy (see below). This formula is balanced in charge and agrees very well with the chemical composition determined from electron microprobe analysis. Generally, the title compound is similar to the Li0.5Fe2.5O4 spinel-type materials. The shift of Li+ to the octahedral site, for example, in comparison with LiCrGeO4 or LiMn2O4, can be explained by the strong preference of Fe3+ for the tetrahedral site. Based on the concept of crystal field stabilization energy, Miller (1959) theoretically calculated octahedral site preference energies which gave a stronger preference of Fe3+ for the tetrahedral site as compared, for example, to Li+ or Mn3+.
The lattice parameter of the title compound [8.2903 (3) Å] is smaller in comparison with, for example, magnetite Fe3O4 [a = 8.3941 (7) Å; Fleet, 1981], but larger than that observed in the Li spinels LiCrGeO4 [a = 8.1976 (1) Å; Touboul & Bourée, 1993] or LiMn2O4 and LiNi0.5Mn1.5O4 (a = 8.243 and 8.1685 Å, respectively; Liu et al., 2014). This is due mainly to the high amount of Fe3+ at the A sites, which has a larger ionic radius than Ge4+, Ni3+ or Mn3+/4+ (Shannon & Prewitt, 1969). The oxygen parameter u = 0.2543 is close to the ideal value for cubic closed packing, reflecting some distinct differences to the spinels which have the A site fully occupied by Li+. In the title compound, the bond length of the tetrahedrally coordinated site T is 1.857 (2) Å, which is distinctly smaller than in, for example, LiMn2O4, with the tetrahedral site being fully occupied by Li+. The T—O bond length is also smaller than in magnetite (Fleet, 1981) or Li0.5Fe2.5O5 (Tomas et al., 1983), with values of 1.8889 (9) and 1.880 (5) Å, respectively. In GeFe2O4, the T—O bond length is only 1.771 (2) Å and this smaller value of T—O compared to, for example, magnetite is due to the substitution of Ge4+ onto the A site and can be seen as additional proof for the correctness of the derived cationic distribution.
The bond length involving the octahedrally coordinated site M is 2.0373 (11) Å, which is 1.07 times larger than the bond length involving the tetrahedrally coordinated site. The M—O bond length is somewhat larger than 2.025 (3) Å in Li0.5Fe2.5O4 (Tomas et al., 1983). This agrees well with the observed higher Li content in the title compound, with the ionic radius for Li+ in an octahedral coordination (0.740 Å) being larger than that of Fe3+ (0.645 Å; Shannon & Prewitt, 1969), thus increasing the M—O distance. Magnetite has a mixed occupation of the octahedral sites, with both Fe2+ and Fe3+, thus having a larger M—O bond length of 2.0582 (9) Å, while in GeFe2O4, all the Fe atoms are in a divalent state and an M—O bond length of 2.132 (2) Å is observed.
In order to quantify the valence state of iron in the title compound, a 57Fe Mössbauer spectrum was recorded at 340 K. It shows a broad, slightly asymmetric, doublet, which can be evaluated with two Lorentzian-shaped doublets (Fig. 2). The first doublet shows an isomer shift (IS) of −0.053 (17) mm s−1 and a quadrupole splitting (QS) of 0.57 (3) mm s−1, and can be assigned to the ferric iron on the tetrahedral site. The second doublet has a larger IS of 0.115 (14) mm s−1 and an almost identical QS of 0.58 (2) mm s−1, and is assigned to ferric iron at the octahedral site. No indications for ferrous iron are present. The QS values suggest low polyhedral distortion, which is almost identical in both sites. The relative area ratio of tetrahedral to octahedral sites is 38.6 (8) to 61.4 (9)%. Assuming a total amount of 2.15 formula units Fe3+, the results of Mössbauer spectroscopy give a cation distribution of (Fe3+0.83)[Fe3+1.32], which is in good agreement with that obtained from the site-occupation of the X-ray data. At room temperature, the title compound is magnetically ordered, as revealed by its 57Fe Mössbauer spectrum.
3. Synthesis and crystallization
The spinel formed as a by-product during the synthesis of pyroxene-type LiFeGe2O6 in flux-growth experiments (Redhammer et al., 2010). For the synthesis of the pyroxene, Li2CO3, Fe2O3 and GeO2 in the stoichiometry of the compound and Li2MoO4/LiVO3 as a (mass ratio sample to = 1:10) were mixed together, heated to 1473 K in a platinum crucible, covered with a lid, held at this temperature for 24 h and cooled afterwards at a rate of 1.5 K h−1 to 973 K. The experimental batch consisted of large pyroxene crystals and a distinct amount of black crystals with idiomorphic octahedral habit, up to 200 µm. Semi-quantitative EDX (energy-dispersive X-ray) analysis revealed iron and some germanium as the main elements; powder X-ray revealed the crystals as a spinel-type material. An electron microprobe analysis on polished/embedded crystals (three different grains with five measurement points each) yielded a chemical composition of 84.86 (30) wt% Fe2O3, 10.52 (25) wt% GeO2 and 4.62 wt% Li2O, with the latter calculated from the difference to 100 oxide%. There is no evidence for Mo or V from the nor for any other chemical elements. From the oxide percentage, a chemical formula of Li0.63 (2)Fe2.18 (1)Ge0.20 (2)O4 was calculated, which is in good agreement with that obtained from the structure Individual crystals are homogeneous in composition, with no significant systematic variation from rim-core; also, there is no systematic variation in composition from crystal to crystal.
4. Refinement
Crystal data, data collection and structure . In a first stage of only iron was considered on the A and B sites, thereby allowing unconstrained of the site-occupation factors. This gave a surplus of electron density (higher occupation than allowed by the multiplicity) at the tetrahedral site, while a lower occupation than possible was found for the octahedral site. From this it was concluded that Li enters the octahedral site and Ge enters the tetrahedral site. In the final refinements, it was assumed that both tetrahedral and octahedral sites are fully occupied, with Fe + Ge = 1 as a restraint for the tetrahedral site and Fe + Li = 1 for the octahedral site.
details are summarized in Table 1Supporting information
CCDC reference: 1463892
https://doi.org/10.1107/S205698901600414X/wm5279sup1.cif
contains datablocks global, I. DOI:Structure factors: contains datablock I. DOI: https://doi.org/10.1107/S205698901600414X/wm5279Isup2.hkl
Data collection: APEX2 (Bruker, 2012); cell
SAINT (Bruker, 2012); data reduction: SAINT (Bruker, 2012); program(s) used to solve structure: coordinates from an isotypic structure; program(s) used to refine structure: SHELXL2014 (Sheldrick, 2015); molecular graphics: DIAMOND (Brandenburg, 2006); software used to prepare material for publication: WinGX (Farrugia, 2012).Li0.64Fe2.15Ge0.21O4 | Dx = 4.744 Mg m−3 |
Mr = 203.5 | Mo Kα radiation, λ = 0.71073 Å |
Cubic, Fd3m | Cell parameters from 3046 reflections |
Hall symbol: -F 4vw 2vw 3 | θ = 7.0–41.9° |
a = 8.2903 (3) Å | µ = 12.85 mm−1 |
V = 569.78 (6) Å3 | T = 298 K |
Z = 8 | Octahedron, black |
F(000) = 771 | 0.13 × 0.12 × 0.12 mm |
Bruker SMART APEX CCD diffractometer | 118 independent reflections |
Radiation source: 3-circle diffractometer | 114 reflections with I > 2σ(I) |
Graphite monochromator | Rint = 0.021 |
ω–scan at 4 different φ positions | θmax = 41.9°, θmin = 7.0° |
Absorption correction: multi-scan (SADABS; Bruker, 2012) | h = −15→14 |
Tmin = 0.83, Tmax = 0.94 | k = −14→10 |
3046 measured reflections | l = −15→13 |
Refinement on F2 | 1 restraint |
Least-squares matrix: full | w = 1/[σ2(Fo2) + (0.0139P)2 + 2.542P] where P = (Fo2 + 2Fc2)/3 |
R[F2 > 2σ(F2)] = 0.018 | (Δ/σ)max < 0.001 |
wR(F2) = 0.042 | Δρmax = 0.36 e Å−3 |
S = 1.37 | Δρmin = −0.67 e Å−3 |
118 reflections | Extinction correction: SHELXL2014 (Sheldrick, 2015) |
10 parameters | Extinction coefficient: 0.0051 (6) |
Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes. |
x | y | z | Uiso*/Ueq | Occ. (<1) | |
Fe1 | 0.5 | 0.5 | 0.5 | 0.00795 (17) | 0.678 (4) |
Li1 | 0.5 | 0.5 | 0.5 | 0.00795 (17) | 0.322 (4) |
Fe2 | 0.125 | 0.125 | 0.125 | 0.00573 (17) | 0.795 (3) |
Ge2 | 0.125 | 0.125 | 0.125 | 0.00573 (17) | 0.205 (3) |
O2 | 0.25434 (14) | 0.25434 (14) | 0.25434 (14) | 0.0095 (3) |
U11 | U22 | U33 | U12 | U13 | U23 | |
Fe1 | 0.00795 (17) | 0.00795 (17) | 0.00795 (17) | −0.00100 (11) | −0.00100 (11) | −0.00100 (11) |
Li1 | 0.00795 (17) | 0.00795 (17) | 0.00795 (17) | −0.00100 (11) | −0.00100 (11) | −0.00100 (11) |
Fe2 | 0.00573 (17) | 0.00573 (17) | 0.00573 (17) | 0 | 0 | 0 |
Ge2 | 0.00573 (17) | 0.00573 (17) | 0.00573 (17) | 0 | 0 | 0 |
O2 | 0.0095 (3) | 0.0095 (3) | 0.0095 (3) | 0.0010 (3) | 0.0010 (3) | 0.0010 (3) |
Fe1—O2i | 2.0373 (11) | Fe1—Fe1ii | 2.9311 (1) |
Fe1—O2ii | 2.0373 (11) | Fe2—O2vii | 1.857 (2) |
Fe1—O2iii | 2.0373 (11) | Fe2—O2viii | 1.857 (2) |
Fe1—O2iv | 2.0373 (11) | Fe2—O2ix | 1.857 (2) |
Fe1—O2v | 2.0373 (11) | Fe2—O2 | 1.857 (2) |
Fe1—O2vi | 2.0373 (11) | ||
O2i—Fe1—O2ii | 180 | O2ii—Fe1—O2vi | 87.96 (7) |
O2i—Fe1—O2iii | 87.96 (7) | O2iii—Fe1—O2vi | 92.04 (7) |
O2ii—Fe1—O2iii | 92.04 (7) | O2iv—Fe1—O2vi | 87.96 (7) |
O2i—Fe1—O2iv | 92.04 (7) | O2v—Fe1—O2vi | 180.00 (7) |
O2ii—Fe1—O2iv | 87.96 (7) | O2vii—Fe2—O2viii | 109.5 |
O2iii—Fe1—O2iv | 180 | O2vii—Fe2—O2ix | 109.5 |
O2i—Fe1—O2v | 87.96 (7) | O2viii—Fe2—O2ix | 109.5 |
O2ii—Fe1—O2v | 92.04 (7) | O2vii—Fe2—O2 | 109.4710 (10) |
O2iii—Fe1—O2v | 87.96 (7) | O2viii—Fe2—O2 | 109.5 |
O2iv—Fe1—O2v | 92.04 (7) | O2ix—Fe2—O2 | 109.4710 (10) |
O2i—Fe1—O2vi | 92.04 (7) |
Symmetry codes: (i) x+1/4, y+1/4, −z+1; (ii) −x+3/4, −y+3/4, z; (iii) x+1/4, −y+1, z+1/4; (iv) −x+3/4, y, −z+3/4; (v) −x+1, y+1/4, z+1/4; (vi) x, −y+3/4, −z+3/4; (vii) −x+1/4, y, −z+1/4; (viii) x, −y+1/4, −z+1/4; (ix) −x+1/4, −y+1/4, z. |
References
Berg, H., Thomas, J. O., Liu, W. & Farrington, G. C. (1998). Solid State Ionics, 112, 165–168. CAS Google Scholar
Biagioni, C. & Pasero, M. (2014). Am. Mineral. 99, 1254–1264. Web of Science CrossRef Google Scholar
Brandenburg, K. (2006). DIAMOND. Crystal Impact GbR, Bonn, Germany. Google Scholar
Bruker (2012). APEX2, SAINT and SADABS. Bruker AXS Inc., Madison, Wisconsin, USA. Google Scholar
Farrugia, L. J. (2012). J. Appl. Cryst. 45, 849–854. Web of Science CrossRef CAS IUCr Journals Google Scholar
Fleet, M. E. (1981). Acta Cryst. B37, 917–920. CSD CrossRef CAS Web of Science IUCr Journals Google Scholar
Hankare, P. P., Patil, R. P., Sankpal, U. B., Jadhav, S. D., Lokhande, P. D., Jadhav, K. M. & Sasikala, R. (2009). J. Solid State Chem. 182, 3217–3221. Web of Science CrossRef CAS Google Scholar
Harrison, R. J. & Putnis, A. (1998). Surv. Geophys. 19, 461–520. Web of Science CrossRef Google Scholar
Hill, R. J., Craig, J. R. & Gibbs, G. V. (1979). Phys. Chem. Miner. 4, 317–339. CrossRef CAS Web of Science Google Scholar
Liu, D., Zhu, W., Trottier, J., Gagnon, C., Barray, F., Guerfi, A., Mauger, A., Groult, H., Julien, C. M., Goodenough, J. B. & Zaghib, K. (2014). RSC Adv. 4, 154–167. Web of Science CrossRef CAS Google Scholar
Miller, A. (1959). J. Appl. Phys. 30, S24–S25. CrossRef Google Scholar
O'Neill, H. S. C. & Navrotsky, A. (1983). Am. Mineral. 68, 181–194. CAS Google Scholar
Patil, R. P., Patil, S. B., Jadhav, B. V., Delekar, S. D. & Hankare, P. P. (2016). J. Magn. Magn. Mater. 401, 870–874. Web of Science CrossRef CAS Google Scholar
Redhammer, G. J., Cámara, F., Alvaro, M., Nestola, F., Tippelt, G., Prinz, S., Simons, J., Roth, G. & Amthauer, G. (2010). Phys. Chem. Miner. 37, 685–704. Web of Science CrossRef CAS Google Scholar
Redhammer, G. J., Roth, G., Treutmann, W., Hoelzel, M., Paulus, W., André, G., Pietzonka, C. & Amthauer, G. (2009). J. Solid State Chem. 182, 2374–2384. Web of Science CrossRef CAS Google Scholar
Shannon, R. D. & Prewitt, C. T. (1969). Acta Cryst. B25, 925–946. CrossRef CAS IUCr Journals Web of Science Google Scholar
Sheldrick, G. M. (2015). Acta Cryst. C71, 3–8. Web of Science CrossRef IUCr Journals Google Scholar
Thackeray, M. M., David, W. I. F., Bruce, P. G. & Goodenough, J. B. (1983). Mater. Res. Bull. 18, 461–472. CrossRef CAS Web of Science Google Scholar
Tomas, A., Laruelle, P., Dormann, J. L. & Nogues, M. (1983). Acta Cryst. C39, 1615–1617. CrossRef CAS Web of Science IUCr Journals Google Scholar
Touboul, M. & Bourée, F. (1993). J. Mater. Chem. 3, 623–626. CrossRef CAS Web of Science Google Scholar
Von Dreele, R. B., Navrotsky, A. & Bowman, A. L. (1977). Acta Cryst. B33, 2287–2288. CrossRef CAS IUCr Journals Web of Science Google Scholar
Welch, M. D., Cooper, M. A. & Hawthorne, F. C. (2001). Mineral. Mag. 65, 441–444. Web of Science CrossRef CAS Google Scholar
This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.