Li­thio­tantite, ideally LiTa₃O₈

Lithio­tantite (lithium tritantalum octa­oxide) and lithio­wodg­inite are natural dimorphs of LiTa₃O₈, corresponding to the laboratory-synthesized L-LiTa₃O₈ (low-temperature form) and M-LiTa₃O₈ (inter­mediate-temperature form) phases, respectively. Based on single-crystal X-ray diffraction data, this study presents the first structure determination of lithio­tantite from a new locality, the Murundu mine, Jenipapo District, Itinga, Minas Gerais, Brazil. Lithio­tantite is isotypic with LiNb₃O₈ and its structure is composed of a slightly distorted hexa­gonal close-packed array of O atoms stacked in the [-101] direction, with the metal atoms occupying half of the octa­hedral sites. There are four symmetrically non-equivalent cation sites, with three of them occupied mainly by (Ta⁵⁺ + Nb⁵⁺) and one by Li⁺. The four distinct octa­hedra share edges, forming two types of zigzag chains (A and B) extending along the b axis. The A chains are built exclusively of (Ta,Nb)O₆ octa­hedra (M1 and M2), whereas the B chains consist of alternating (Ta,Nb)O₆ and LiO₆ octa­hedra (M3 and M4, respectively). The average M1—O, M2—O, M3—O and M4—O bond lengths are 2.011, 2.004, 1.984, and 2.188 Å, respectively. Among the four octa­hedra, M3 is the least distorted and M4 the most. The refined Ta contents at the M1, M2 and M3 sites are 0.641 (2), 0.665 (2), and 0.874 (2), respectively, indicating a strong preference of Ta⁵⁺ for M3 in the B chain. The refined composition of the crystal investigated is Li_0.96Mn_0.03Na_0.01Nb_0.82Ta_2.18O₈.

Related literature

For lithio­tantite and isostructural materials, see: Voloshin et al. (1983); Lundberg (1971); Gatehouse & Leverett (1972). For lithio­wodginite and wodginite-type materials, see: Voloshin et al. (1990); Ferguson et al. (1976); Gatehouse et al. (1976); Santoro et al. (1977); Erict et al. (1992). For structural and phase-stability information on the LiTa₃O₈ system, see: Nord & Thomas (1978); Fallon et al. (1979); Hodeau et al. (1983, 1984); Allemann et al. (1996). For properties and applications of LiTa₃O₈ and LiNb₃O₈, see: Subasri & Sreedharan (1997); Akazawa & Shimada (2007); Zhang et al. (2008); Muller et al. (2011). For the definition of polyhedral distortion, see: Robinson et al. (1971).

Experimental

Crystal data

• Li_0.96Mn_0.03Na_0.01Nb_0.82Ta_2.18O₈

• M_r = 607.20

• Monoclinic, P 2₁ /c

• a = 7.4425 (4) Å

• b = 5.0493 (3) Å

• c = 15.2452 (7) Å

• β = 107.381 (3)°

• V = 546.75 (5) ų

• Z = 4

• Mo Kα radiation

• μ = 45.28 mm⁻¹

• T = 293 K

• 0.06 × 0.05 × 0.05 mm

Data collection

• Bruker APEXII CCD area-detector diffractometer

• Absorption correction: multi-scan (SADABS; Sheldrick, 2005) T_min = 0.172, T_max = 0.211

• 8136 measured reflections

• 1983 independent reflections

• 1683 reflections with I > 2σ(I)

• R_int = 0.027

Refinement

• R[F² > 2σ(F²)] = 0.028

• wR(F²) = 0.058

• S = 1.35

• 1983 reflections

• 71 parameters

• 1 restraint

• Δρ_max = 2.02 e Å⁻³

• Δρ_min = −1.92 e Å⁻³

Data collection: APEX2 (Bruker, 2004); cell refinement: SAINT (Bruker, 2004); data reduction: SAINT; program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008); molecular graphics: XtalDraw (Downs & Hall-Wallace, 2003); software used to prepare material for publication: publCIF (Westrip, 2010).

Supporting information

S1. Comment

The outstanding electro-optical properties of lithium tantalate, LiTa₃O₈, and lithium niobate, LiNb₃O₈, have made them leading functional materials for numerous applications, such as electro-optic modulators, surface acoustic wave devices, frequency-doubled lasers, second-harmonic generators, beam deflectors, waveguides, and holographic data processing devices (e.g., Subasri & Sreedharan, 1997; Akazawa & Shimada, 2007; Zhang et al., 2008; Muller et al., 2011). However, unlike LiNb₃O₈, which maintains the P2₁/c symmetry up to its incongruent melting point, LiTa₃O₈ is trimorphic, depending on its formation temperature (Allemann et al., 1996). Below 1063 K, LiTa₃O₈ crystallizes in the P2₁/c symmetry (designated as L-LiTa₃O₈). Above 1063 K, L-LiTa₃O₈ transforms irreversibly to the intermediate-temperature C2/c form (M-LiTa₃O₈), which further transforms irreversibly to the high-temperature Pmmn structure (H-LiTa₃O₈) above 1393 K.

The crystal structure of M-LiTa₃O₈ was first determined by Gatehouse et al. (1976), who showed it to be monoclinic with space group C2/c and unit-cell parameters a =9.413 (5), b = 11.522 (6), c = 5.050 (3) Å, β = 91.1 (1)°. This phase is isotypic with the mineral wodginite, MnSnTa₂O₈ (Ferguson et al., 1976; Erict et al., 1992). A further refinement of the M-LiTa₃O₈ structure by Santoro et al. (1977) using neutron powder-diffraction data located the Li atom, which was not found in the X-ray diffraction study by Gatehouse et al. (1976). In contrast, the structure of H-LiTa₃O₈ has been investigated quite intensively, confirming its real symmetry to be Pmmn with unit-cell parameters a =16.718 (2), b = 7.696 (1), c = 8.931 (1) Å (Hodeau et al., 1983, 1984), rather than Pmma with unit-cell parameters a =16.702 (8), b = 3.840 (4), c = 8.929 (5) Å (Nord & Thomas, 1978; Fallon et al., 1979). Yet, no structure study has been reported for L-LiTa₃O₈ thus far. From rotation and Weissenberg photographic measurements, Gatehouse & Leverett (1972) obtained unit-cell parameters a = 7.41 (5), b = 5.10 (6), c = 15.12 (10) Å, β = 107.2 (1)°, and space group P2₁/c for L-LiTa₃O₈, suggesting that this phase is the Ta-analogue of LiNb₃O₈ (Lundberg, 1971). Interestingly, Voloshin et al. (1983) described a new mineral, lithiotantite, with the empirical chemical formula Li_0.92(Ta_1.90Nb_1.10Sn_0.02)_Σ=3.02O₈ or ideally LiTa₃O₈, from granite pegmatites in Eastern Kazakhstan. The new mineral possesses space group P2₁/c and unit-cell parameters a = 7.444, b = 5.044, c = 15.255 Å, β = 107.18°. Although all reported crystallographic data suggest that lithiotantite is actually L-LiTa₃O₈, it is unclear how Nb in the natural sample is distributed among three cation sites found in the isostructural LiNb₃O₈ (Lundberg, 1971). On the basis of single-crystal X-ray diffraction data, this study reports the first structure refinement of lithiotantite found from a new locality, the Jenipapo District, Itinga, Minas Gerais, Brazil.

Lithiotantite is isotypic with LiNb₃O₈ (Lundberg, 1971; Gatehouse & Leverett, 1972). Its structure consists of a slightly distorted hexagonal close-packed array of oxygen atoms stacked in the [-1 0 1] direction, with the metal atoms occupying half of the octahedral sites. There are four symmetrically nonequivalent cation sites, M1, M2, M3, and M4 (Fig. 1), with the first three occupied mainly by (Ta + Nb) and the last one by Li. The four distinct octahedra share edges to form two types of zigzag chains (A- and B-chains) extending along the b axis (Fig. 2). The A-chains are built exclusively of (Ta,Nb)O₆ octahedra (M1 and M2), whereas the B-chains consist of alternating (Ta,Nb)O₆ and LiO₆ octahedra (M3 and M4, respectively). The refined Ta contents are 0.641 (2), 0.665 (2), and 0.874 (2) for M1, M2, and M3, respectively, indicating a relatively strong preference of Ta⁵⁺ on M3 over M1 or M2. The average M1—O, M2—O, M3—O, and M4—O bond distances are 2.011, 2.004, 1.984, and 2.188 Å, respectively. Among the four octahedra, M4 is the most distorted and M3 the least, as measured by the values of the octahedral angle variance (OAV) (Robinson et al., 1971), which are 88, 81, 35, and 98° for the M1, M2, M3, and M4 octahedra, respectively. This observation is the direct consequence of the octahedral linkage within the two different chains. In the B-chain, the M4 octahedron occupied primarily by Li⁺ is by far more weakly-bonded than M3 occupied by (Ta⁵⁺ + Nb⁵⁺). Hence, for the two octahedra to share edges to form the chains, the relatively large M4 octahedron has to make more adjustments to fit to the configuration of the M3 octahedron and to minimize the cation-cation repulsion across the shared edges, thus resulting in its greater distortion. For the M1 and M2 octahedra that are occupied by the similar ratios of Ta/Nb, the cation-cation repulsion between the two across the shared edges is markedly stronger than that between M3 and M4. Therefore, the M1 and M2 octahedra exhibit similar OAV values and are more distorted than M3.

Gatehouse et al. (1976) demonstrated that M-LiTa₃O₈ is isomorphous with the mineral wodginite and presented a comprehensive structural comparison between M-LiTa₃O₈ and LiNb₃O₈. The mineral lithiowodginite, ideally LiTa₃O₈, a member of the wodginite group, was later described by Voloshin et al. (1990). With the discovery of lithiotantite (Voloshin et al., 1983), isostructural with L-LiTa₃O₈ and LiNb₃O₈, one may postulate the possible existence of a natural LiTa₃O₈ phase with the H-LiTa₃O₈-type structure, as well as a natural Nb-analogue of lithiotantite.

Related literature

For lithiotantite and isostructural materials, see: Voloshin et al. (1983); Lundberg (1971); Gatehouse & Leverett (1972). For lithiowodginite and wodginite-type materials, see: Voloshin et al. (1990); Ferguson et al. (1976); Gatehouse et al. (1976); Santoro et al. (1977); Erict et al. (1992). For structural and phase-stability information on the LiTa₃O₈ system, see: Nord & Thomas (1978); Fallon et al. (1979); Hodeau et al. (1983, 1984); Allemann et al. (1996). For properties and applications of LiTa₃O₈ and LiNb₃O₈, see: Subasri & Sreedharan (1997); Akazawa & Shimada (2007); Zhang et al. (2008); Muller et al. (2011). For the definition of polyhedral distortion, see: Robinson et al. (1971).

S2. Experimental

The lithiotantite specimen used in this study is from the Murundu mine, the Jenipapo District, Itinga, Minas Gerais, Brazil and is in the collection of the RRUFF project (deposition No. R100165; http://rruff.info). Its chemical composition was analyzed with a Jeol JXA-8900 electron microprobe at the conditions of 20 kV and 25 nA. Counting times on peaks and backgrounds of the X-ray lines were 10 and 5 s, respectively. Raw data were corrected using the PRZ procedure, which gave (wt%) Ta₂O₅ = 78.5 (9), Nb₂O₅ = 17.1 (7), SnO₂ = 0.7 (3), MnO = 0.20 (10), FeO = 0.06 (3), Na₂O = 0.06 (3) (average of 9 analysis points). Li₂O was added to bring the total cation sums to 4.0 based on 8 O atoms, while maintaining the charge balance, yielding a chemical formula (Li_0.96Na_0.01Mn²⁺_0.02Fe²⁺_0.01)~Σ=1.00_(Ta2.18_Nb0.79_Sn0.03₎Σ=3.00_O8_.

S3. Refinement

For simplicity, during the structure refinement, the trace amount of Fe was treated as Mn, and Sn (0.03 apfu) as Nb, giving rise to a structural formula (Li_0.96Na_0.01Mn²⁺_0.03)_Σ=1.00(Ta_2.18Nb_0.83)_Σ=3.00O₈, which was used throughout the structure refinements. Because a preliminary refinement showed that anisotropic displacement parameters were non-positive defined for M4 (mainly Li) and two O atoms, due most likely to the obvious inhomogeneity of the studied samples (Fig. 3), M4 and all O atoms were refined with isotropic displacement parameters only. In Figure 3, the contrast in darkness reflects the relative distribution of Ta vs. Nb in the sample. The distributions of Ta and Nb among the three octahedral sites were refined with their total amounts constrained to the above simplified formula. The final refinement indicates relatively large GOF value. We attempted to refine the Li position with a split-site model (or disordered model). Although the final R factor was slightly reduced from 0.0273 to 0.0271, the GOF value is essentially unchanged (still above 1.3). We tried to omit six bad reflections with Fo²-Fc² > 7.0, but still failed to improve the GOF value. Even with the model of the split Li positions, the anisotropic displacement for Li is still non-positive definite. Thus, we believe that all of these may result from the obvious inhomogeneities of our natural sample. In the past, we have noticed that a structure refinement may give rise to a large GOF value when the sample is not homogenous (like our current case, or with fine exsolutions, or badly twinned). In addition, we also tried to allow all oxygen atoms to be refined with anisotropic displacements. Yet, that only reduced the R factor to 0.0268 and the Li atom is still non-positive definite. The highest residual peak in the difference Fourier maps was located at (0.4240, 0.2970, 0.2913), 0.78 Å from M3, and the deepest hole at (0.0221, 0.1663, 0.3115), 0.64 Å from M2.

Computing details

Data collection: APEX2 (Bruker, 2004); cell refinement: SAINT (Bruker, 2004); data reduction: SAINT (Bruker, 2004); program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008); molecular graphics: XtalDraw (Downs & Hall-Wallace, 2003); software used to prepare material for publication: publCIF (Westrip, 2010).

Figures

	Fig. 1. Crystal structure of lithiotantite viewed along [010]. Red spheres represent oxygen atoms. The purple, pale blue, yellow, and red octahedra represent M1, M2, M3, and M4 octahedra, respectively.
	Fig. 2. A slice of the lithiotantite structure showing the two types of zigzag edge-shared octahedral chains. All color coding and symbols are as in Figure 1.
	Fig. 3. A back scattering electron image of the lithiotantite sample, showing the obvious chemical inhomogeneity of the studied sample. The contrast in darkness reflects the relative distribution of Ta vs. Nb in the sample.

lithium tritantalum octaoxide

Crystal data

LiTa3O8	F(000) = 1042
Mr = 607.20	Dx = 7.377 Mg m^−3
Monoclinic, P21/c	Mo Kα radiation, λ = 0.71073 Å
Hall symbol: -P 2ybc	Cell parameters from 2589 reflections
a = 7.4425 (4) Å	θ = 2.8–32.6°
b = 5.0493 (3) Å	µ = 45.28 mm^−1
c = 15.2452 (7) Å	T = 293 K
β = 107.381 (3)°	Cuboid, red–brown
V = 546.75 (5) Å^3	0.06 × 0.05 × 0.05 mm
Z = 4

Data collection

Bruker APEXII CCD area-detector diffractometer	1983 independent reflections
Radiation source: fine-focus sealed tube	1683 reflections with I > 2σ(I)
Graphite monochromator	Rint = 0.027
ϕ and ω scan	θmax = 32.6°, θmin = 2.8°
Absorption correction: multi-scan (SADABS; Sheldrick, 2005)	h = −10→11
Tmin = 0.172, Tmax = 0.211	k = −7→7
8136 measured reflections	l = −22→23

Refinement

Refinement on F^2	Primary atom site location: structure-invariant direct methods
Least-squares matrix: full	Secondary atom site location: difference Fourier map
R[F^2 > 2σ(F^2)] = 0.028	w = 1/[σ^2(Fo^2) + (0.0077P)^2 + 6.6039P] where P = (Fo^2 + 2Fc^2)/3
wR(F^2) = 0.058	(Δ/σ)max = 0.002
S = 1.35	Δρmax = 2.02 e Å^−3
1983 reflections	Δρmin = −1.92 e Å^−3
71 parameters	Extinction correction: SHELXL97 (Sheldrick, 2008), Fc^*=kFc[1+0.001xFc^2λ^3/sin(2θ)]^-1/4
1 restraint	Extinction coefficient: 0.00065 (6)

Special details

Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes.

Refinement. Refinement of F² against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F², conventional R-factors R are based on F, with F set to zero for negative F². The threshold expression of F² > σ(F²) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F² are statistically about twice as large as those based on F, and R- factors based on ALL data will be even larger.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Ų)

	x	y	z	Uiso*/Ueq	Occ. (<1)
TAM1	0.74765 (4)	0.24400 (7)	0.07791 (2)	0.00508 (9)	0.641 (2)
NBM1	0.74765 (4)	0.24400 (7)	0.07791 (2)	0.00508 (9)	0.359 (2)
TAM2	0.98987 (4)	0.24084 (6)	0.33689 (2)	0.00466 (9)	0.665 (2)
NBM2	0.98987 (4)	0.24084 (6)	0.33689 (2)	0.00466 (9)	0.335 (2)
TAM3	0.50096 (4)	0.23890 (6)	0.333006 (18)	0.00528 (8)	0.874 (2)
NBM3	0.50096 (4)	0.23890 (6)	0.333006 (18)	0.00528 (8)	0.126 (2)
LIM4	0.2391 (10)	0.2561 (17)	0.0794 (5)	0.0030 (13)*	0.96
MNM4	0.2391 (10)	0.2561 (17)	0.0794 (5)	0.0030 (13)*	0.03
NAM4	0.2391 (10)	0.2561 (17)	0.0794 (5)	0.0030 (13)*	0.01
O1	0.0002 (7)	0.0608 (9)	0.0979 (3)	0.0064 (8)*
O2	0.4176 (7)	0.0670 (9)	0.2181 (3)	0.0059 (8)*
O3	0.7653 (7)	0.1008 (9)	0.3443 (3)	0.0073 (8)*
O4	0.5368 (7)	0.4210 (9)	0.1011 (3)	0.0064 (8)*
O5	0.9144 (7)	0.4112 (9)	0.2154 (3)	0.0054 (8)*
O6	0.6436 (7)	0.3931 (9)	0.4606 (3)	0.0078 (9)*
O7	0.8524 (7)	0.5637 (9)	0.0489 (3)	0.0073 (8)*
O8	0.2755 (7)	0.4123 (9)	0.3444 (3)	0.0063 (8)*

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
TAM1	0.00614 (14)	0.00390 (13)	0.00485 (14)	0.00065 (10)	0.00108 (10)	−0.00033 (10)
NBM1	0.00614 (14)	0.00390 (13)	0.00485 (14)	0.00065 (10)	0.00108 (10)	−0.00033 (10)
TAM2	0.00518 (14)	0.00380 (13)	0.00496 (14)	0.00056 (11)	0.00143 (10)	0.00038 (11)
NBM2	0.00518 (14)	0.00380 (13)	0.00496 (14)	0.00056 (11)	0.00143 (10)	0.00038 (11)
TAM3	0.00517 (12)	0.00520 (12)	0.00565 (13)	−0.00042 (10)	0.00190 (9)	−0.00014 (10)
NBM3	0.00517 (12)	0.00520 (12)	0.00565 (13)	−0.00042 (10)	0.00190 (9)	−0.00014 (10)

Geometric parameters (Å, º)

TAM1—O6^i	1.857 (5)	TAM3—O2	1.886 (5)
TAM1—O7	1.901 (5)	TAM3—O8	1.946 (5)
TAM1—O4	1.929 (5)	TAM3—O4^iii	1.957 (5)
TAM1—O1^ii	2.034 (5)	TAM3—O2^iv	2.000 (5)
TAM1—O8^iii	2.088 (5)	TAM3—O3	2.045 (5)
TAM1—O5	2.257 (5)	TAM3—O6	2.070 (5)
TAM2—O3	1.849 (5)	LIM4—O7^vi	2.079 (9)
TAM2—O1^iv	1.887 (5)	LIM4—O3^iv	2.099 (10)
TAM2—O5	1.965 (5)	LIM4—O1	2.124 (9)
TAM2—O7^v	1.997 (5)	LIM4—O6^iii	2.194 (9)
TAM2—O5^v	2.063 (5)	LIM4—O4	2.296 (9)
TAM2—O8^ii	2.266 (5)	LIM4—O2	2.338 (9)
O6^i—TAM1—O7	100.1 (2)	O2—TAM3—O8	103.7 (2)
O6^i—TAM1—O4	102.8 (2)	O2—TAM3—O4^iii	92.4 (2)
O7—TAM1—O4	93.5 (2)	O8—TAM3—O4^iii	93.6 (2)
O6^i—TAM1—O1^ii	94.3 (2)	O2—TAM3—O2^iv	94.20 (11)
O7—TAM1—O1^ii	89.79 (19)	O8—TAM3—O2^iv	91.82 (19)
O4—TAM1—O1^ii	161.7 (2)	O4^iii—TAM3—O2^iv	170.2 (2)
O6^i—TAM1—O8^iii	99.5 (2)	O2—TAM3—O3	87.9 (2)
O7—TAM1—O8^iii	157.2 (2)	O8—TAM3—O3	168.3 (2)
O4—TAM1—O8^iii	93.33 (19)	O4^iii—TAM3—O3	87.65 (19)
O1^ii—TAM1—O8^iii	77.25 (18)	O2^iv—TAM3—O3	85.38 (19)
O6^i—TAM1—O5	171.61 (19)	O2—TAM3—O6	168.6 (2)
O7—TAM1—O5	75.44 (18)	O8—TAM3—O6	87.8 (2)
O4—TAM1—O5	84.71 (19)	O4^iii—TAM3—O6	86.45 (19)
O1^ii—TAM1—O5	78.70 (18)	O2^iv—TAM3—O6	85.68 (19)
O8^iii—TAM1—O5	83.60 (17)	O3—TAM3—O6	80.72 (19)
O3—TAM2—O1^iv	100.9 (2)	O7^vi—LIM4—O3^iv	95.9 (4)
O3—TAM2—O5	102.5 (2)	O7^vi—LIM4—O1	106.0 (4)
O1^iv—TAM2—O5	94.4 (2)	O3^iv—LIM4—O1	99.2 (4)
O3—TAM2—O7^v	94.7 (2)	O7^vi—LIM4—O6^iii	98.6 (4)
O1^iv—TAM2—O7^v	90.2 (2)	O3^iv—LIM4—O6^iii	157.0 (4)
O5—TAM2—O7^v	161.0 (2)	O1—LIM4—O6^iii	93.9 (4)
O3—TAM2—O5^v	97.94 (19)	O7^vi—LIM4—O4	90.4 (3)
O1^iv—TAM2—O5^v	158.6 (2)	O3^iv—LIM4—O4	78.1 (3)
O5—TAM2—O5^v	91.34 (11)	O1—LIM4—O4	163.5 (4)
O7^v—TAM2—O5^v	78.15 (19)	O6^iii—LIM4—O4	84.0 (3)
O3—TAM2—O8^ii	173.9 (2)	O7^vi—LIM4—O2	165.3 (4)
O1^iv—TAM2—O8^ii	75.98 (18)	O3^iv—LIM4—O2	86.4 (3)
O5—TAM2—O8^ii	83.12 (18)	O1—LIM4—O2	87.9 (3)
O7^v—TAM2—O8^ii	80.15 (19)	O6^iii—LIM4—O2	75.2 (3)
O5^v—TAM2—O8^ii	84.29 (18)	O4—LIM4—O2	75.8 (3)

Symmetry codes: (i) x, −y+1/2, z−1/2; (ii) x+1, y, z; (iii) −x+1, y−1/2, −z+1/2; (iv) −x+1, y+1/2, −z+1/2; (v) −x+2, y−1/2, −z+1/2; (vi) −x+1, −y+1, −z.