Li_0.97Fe^II_0.79Fe^III_0.15(PO₄), a new oxidized triphylite-type phosphate

This paper reports a new partially oxidized triphylite-type phosphate (lithium iron phosphate), which has been synthesized hydro­thermally at 973 K and 0.1 GPa. The structure is similar to that of natural triphylite, LiFe(PO₄), and is characterized by two chains of edge-sharing octa­hedra parallel to the b axis. The weakly distorted M1 octa­hedra contain Li atoms, whereas the more strongly distorted M2 octa­hedra contain Fe^II and Fe^III atoms. Refined site occupancies and bond-valence analysis show the presence of Fe^III and vacancies on the M2 site, mainly explained by the substitution mechanism 3 Fe^II = 2 Fe^III + vacancies.

S1. Comment

Triphylite [Li(Fe,Mn)PO₄, a = 4.690, b = 10.286, c = 5.987 Å, space group Pbnm; Losey et al., 2004] is an accessory phosphate mineral occurring in granitic pegmatites, where it forms a solid solution with lithiophilite [Li(Mn,Fe)PO₄]. The crystal structure of these phosphates, which are isostructural with minerals of the olivine group, has been investigated from synthetic samples (Geller & Durand, 1960; Yakubovich et al., 1977) and natural minerals (Finger & Rapp, 1969; Losey et al., 2004; Fehr et al., 2007; Hatert et al., 2012). Since Padhi et al. (1997) reported the reversible electrochemical extraction of lithium from LiFePO₄, the olivine-type phosphates LiMPO₄ (M is Fe, Mn, Co or Ni) have received much attention as candidates for lithium batteries. The number of publications devoted to these compounds has increased linearly since 2001 and reached 300 publications per year in 2010. Recently, Kang & Ceder (2009) showed that LiFePO₄-based batteries can achieve ultrafast charging and discharging in 10–20 s, thus reaching the performance of supercapacitors. These features make LiFePO₄ the best candidate for producing batteries for many applications, such as electric bicycles, electric boats, electric cars or the storage of green energy.

This increasing interest in lithium iron phosphates prompted us to investigate the Li–Fe^II–Fe^III (+PO₄) system in detail using hydrothermal techniques, at temperatures between 400 and 973 K and under a pressure of 0.1 GPa (Dal Bo, 2011). In an experiment performed at 973 K (H.359), we observed black crystals which gave an X-ray powder diffraction pattern similar to that of triphylite. A single-crystal structure determination confirmed this hypothesis and indicated the presence of vacancies on both cationic sites of the structure, and of significant amounts of Fe^III on the M2 site. The present paper reports the crystal structure of this new partially oxidized triphylite-type phosphate, Li_0.97Fe^II_0.79Fe^III_0.15(PO₄).

The structure of Li_0.97Fe^II_0.79Fe^III_0.15(PO₄) was refined in space group Pnma (No. 62) and is similar to those of natural triphylite-type phosphates. It is characterized by two chains of edge-sharing octahedra parallel to the b axis (Fig. 1). The first chain contains weakly distorted M1 octahedra occupied by Li atoms [Li—O = 2.0900 (13)–2.1872 (13) Å], whereas the second chain contains more strongly distorted M2 octahedra occupied by Fe atoms [Fe—O = 2.0630 (13)–2.2456 (14) Å]. The chains are connected in the a direction by sharing edges of their octahedral sites, and the resulting planes are connected in the c direction by the PO₄ tetrahedra.

Refinement of the site-occupancy factors on the crystallographic sites of Li_0.97Fe^II_0.79Fe^III_0.15(PO₄) showed that vacancies occur on both M1 and M2 sites, which are occupied by 0.97 (3) Li and 0.935 (4) Fe, respectively. The charge deficit induced by these vacancies is compensated by a partial oxidation of Fe on the M2 site; charge transfers between Fe^II and Fe^III on that site explain the black colour of the phosphate. The substitution mechanism, responsible for 80% of the Fe^III insertion in the M2 site, corresponds to 3Fe^II = 2 Fe^III + vacancies and expresses an increasing high-temperature miscibility towards Li₃Fe^III₂(PO₄)₃ (Masquelier et al., 1998). A minor substitution mechanism, responsible for the insertion of 20% of Fe^III in the structure, is Li⁺ + Fe^II = vacancies + Fe^III; this mechanism has been widely described in natural and synthetic triphylite-type phosphates (Quensel, 1937; Mason, 1941; Padhi et al., 1997; Andersson et al., 2000).

The final site populations, calculated from the observed site occupancies and assuming charge balance, are 0.97 Li + 0.03 vacancies on M1, and 0.79 Fe^II + 0.15 Fe^III + 0.06 vacancies on M2. In order to confirm these site populations, bond-valence sums were calculated using the empirical parameters of Brown & Altermatt (1985). The P1 bond-valence sum is 4.91. The Li1 bond-valence sum is 0.95, in good agreement with the 0.97 Li atoms observed on the M1 site. The Fe2 bond-valence sum of 2.01 is very close to the ideal value of 2.03, calculated from the M2 site populations described above.

Related literature

For related literature, see: Andersson et al. (2000); Brown & Altermatt (1985); Dal (2011); Fehr et al. (2007); Finger & Rapp (1969); Fransolet (1975); Geller & Durand (1960); Hatert et al. (2012); Kang & Ceder (2009); Losey et al. (2004); Mason (1941); Masquelier et al. (1998); Padhi et al. (1997); Quensel (1937); Tuttle (1949); Yakubovich et al. (1977).

S2. Experimental

The title compound was synthesized under hydrothermal conditions. The starting material was prepared by mixing Li₃PO₄, FePO₄ and Fe in a 4:2:1 molar ratio, and then by homogenizing this mixture in an agate mortar. A small quantity of the mixture (about 25 mg) was sealed in a gold tube with an outer diameter of 2 mm and a length of 25 mm, containing distilled water (2 mg). The gold capsule was then inserted in a Tuttle-type pressure vessel (Tuttle, 1949) and maintained at a temperature of 973 K and a pressure of 0.1 GPa. After 7 d, the gold tube containing the sample was quenched in the autoclave to room temperature in a stream of cold air. The synthesized products were identified by X-ray powder diffraction (Panalytical PW-3710 goniometer using Fe Kα radiation, λ = 1.9373 Å); they consisted of black crystals of the title compound, associated with colourless Li₃PO₄ crystals.

A chemical analysis was performed using a CAMEBAX SX-100 electron microprobe (15 kV acceleration voltage, 5 nA beam current, analyst T. Theye, Stuttgart, Germany). The standard used to calibrate both Fe and P was graftonite from Kabira (sample KF16, Fransolet, 1975). The average of 11 point analyses gives P₂O₅ 46.30, Fe₂O₃* 13.24, FeO* 29.00 and Li₂O* 9.75, total 98.29 wt.% (* denotes values calculated to maintain charge balance, assuming one Li atom per formula unit). The chemical composition, calculated on the basis of one P atom per formula unit, corresponds to Li_1.00Fe^II_0.62Fe^III_0.25(PO₄), in fairly good agreement with the composition calculated from the structural data, Li_0.97Fe^II_0.79Fe^III_0.15(PO₄).

S3. Refinement

All atoms were refined anisotropically. The refined site-occupancy factors then indicated low electronic densities on the M1 and M2 sites, thus showing that these sites were not fully occupied by Li and Fe, respectively. Consequently, Li and vacancies were refined on the M1 site, whereas Fe and vacancies were refined on the M2 site.

Computing details

Data collection: CrysAlis PRO (Agilent, 2012); cell refinement: CrysAlis PRO (Agilent, 2012); data reduction: CrysAlis PRO (Agilent, 2012); program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008); molecular graphics: ATOMS (Dowty, 1993); software used to prepare material for publication: OLEX2 (Dolomanov et al., 2009).

Figures

Fig. 1. The crystal structure of Li0.97Fe^II0.79Fe^III0.15(PO4). Colour key: M1O6 octahedra are white, M2O6 octahedra are light grey and PO4 tetrahedra are dark grey.

Lithium iron phosphate

Crystal data

LiFeII0.619FeIII0.254(PO4)	Dx = 3.514 Mg m^−3
Mr = 615.69	Mo Kα radiation, λ = 0.7107 Å
Orthorhombic, Pnma	Cell parameters from 2820 reflections
a = 10.3060 (4) Å	θ = 3.4–32.3°
b = 6.0041 (2) Å	µ = 5.22 mm^−1
c = 4.69281 (15) Å	T = 293 K
V = 290.38 (2) Å^3	Isometric, black
Z = 1	0.15 × 0.10 × 0.08 mm
F(000) = 296

Data collection

Agilent Xcalibur Eos diffractometer	545 independent reflections
Radiation source: Enhance (Mo) X-ray Source	508 reflections with I > 2σ(I)
Graphite monochromator	Rint = 0.041
Detector resolution: 16.0087 pixels mm^-1	θmax = 32.4°, θmin = 4.0°
ω scans	h = −15→15
Absorption correction: analytical [CrysAlis PRO (Agilent, 2012), based on expressions derived by Clark & Reid (1995)]	k = −8→9
Tmin = 0.595, Tmax = 0.751	l = −7→6
5358 measured reflections

Refinement

Refinement on F^2	0 restraints
Least-squares matrix: full	Primary atom site location: structure-invariant direct methods
R[F^2 > 2σ(F^2)] = 0.028	Secondary atom site location: difference Fourier map
wR(F^2) = 0.072	w = 1/[σ^2(Fo^2) + (0.044P)^2 + 0.0599P] where P = (Fo^2 + 2Fc^2)/3
S = 1.20	(Δ/σ)max < 0.001
545 reflections	Δρmax = 0.56 e Å^−3
42 parameters	Δρmin = −1.26 e Å^−3

Special details

Experimental. Absorption correction: CrysAlis PRO, Agilent Technologies, Version 1.171.35.21 (release 20-01-2012 CrysAlis171 .NET) (compiled Jan 23 2012,18:06:46) Analytical numeric absorption correction using a multifaceted crystal model based on expressions derived by Clark & Reid [Clark, R. C. & Reid, J. S. (1995). Acta Cryst. A51, 887–897].

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)
Fe2	0.21795 (4)	0.7500	0.02445 (8)	0.00580 (15)	0.935 (4)
P1	0.40493 (6)	0.7500	−0.41852 (13)	0.0051 (2)
O1	0.04312 (17)	0.7500	−0.2049 (4)	0.0081 (4)
O2	0.40280 (17)	0.7500	0.2574 (4)	0.0087 (4)
O3	0.16569 (12)	0.4536 (2)	0.2149 (3)	0.0086 (3)
Li1	0.0000	0.5000	−0.5000	0.0161 (19)	0.97 (3)

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
Fe2	0.0056 (2)	0.0050 (2)	0.0068 (2)	0.000	0.00011 (12)	0.000
P1	0.0059 (3)	0.0055 (3)	0.0038 (3)	0.000	0.00006 (19)	0.000
O1	0.0058 (7)	0.0115 (9)	0.0069 (8)	0.000	−0.0003 (6)	0.000
O2	0.0110 (9)	0.0102 (8)	0.0048 (8)	0.000	−0.0005 (6)	0.000
O3	0.0111 (6)	0.0071 (6)	0.0076 (6)	−0.0021 (5)	−0.0009 (4)	−0.0007 (5)
Li1	0.021 (3)	0.014 (3)	0.014 (3)	0.004 (2)	0.0007 (17)	−0.0018 (19)

Geometric parameters (Å, º)

Fe2—O1	2.0988 (18)	P1—O3^i	1.5544 (13)
Fe2—O2	2.1963 (18)	P1—O3^iii	1.5543 (13)
Fe2—O3	2.0630 (13)	Li1—O1	2.0900 (13)
Fe2—O3^i	2.2456 (14)	Li1—O1^vi	2.0900 (13)
Fe2—O3^ii	2.0630 (13)	Li1—O2^iii	2.1714 (13)
Fe2—O3^iii	2.2456 (14)	Li1—O2^vii	2.1715 (13)
P1—O1^iv	1.5375 (18)	Li1—O3^viii	2.1872 (13)
P1—O2^v	1.5212 (19)	Li1—O3^v	2.1872 (13)
O1—Fe2—O2	178.99 (7)	O2^v—P1—O3^i	113.33 (6)
O1—Fe2—O3^iii	97.28 (5)	O2^v—P1—O3^iii	113.33 (6)
O1—Fe2—O3^i	97.28 (5)	O3^iii—P1—O3^i	103.71 (10)
O2—Fe2—O3^i	81.88 (5)	O1^vi—Li1—O1	180.0
O2—Fe2—O3^iii	81.88 (5)	O1—Li1—O2^iii	91.71 (5)
O3^ii—Fe2—O1	89.89 (4)	O1^vi—Li1—O2^iii	88.29 (5)
O3—Fe2—O1	89.89 (4)	O1—Li1—O2^vii	88.29 (5)
O3—Fe2—O2	90.62 (4)	O1^vi—Li1—O2^vii	91.71 (5)
O3^ii—Fe2—O2	90.62 (4)	O1—Li1—O3^v	109.32 (6)
O3^ii—Fe2—O3^iii	152.78 (6)	O1^vi—Li1—O3^viii	109.32 (6)
O3—Fe2—O3^iii	87.14 (3)	O1^vi—Li1—O3^v	70.68 (6)
O3—Fe2—O3^ii	119.23 (8)	O1—Li1—O3^viii	70.68 (6)
O3^iii—Fe2—O3^i	65.96 (7)	O2^iii—Li1—O2^vii	180.0
O3^ii—Fe2—O3^i	87.14 (3)	O2^vii—Li1—O3^v	96.20 (6)
O3—Fe2—O3^i	152.78 (6)	O2^iii—Li1—O3^v	83.80 (6)
O1^iv—P1—O3^i	106.38 (7)	O2^iii—Li1—O3^viii	96.20 (6)
O1^iv—P1—O3^iii	106.38 (7)	O2^vii—Li1—O3^viii	83.80 (6)
O2^v—P1—O1^iv	112.96 (10)	O3^v—Li1—O3^viii	180.0

Symmetry codes: (i) −x+1/2, y+1/2, z−1/2; (ii) x, −y+3/2, z; (iii) −x+1/2, −y+1, z−1/2; (iv) x+1/2, y, −z−1/2; (v) x, y, z−1; (vi) −x, −y+1, −z−1; (vii) x−1/2, y, −z−1/2; (viii) −x, −y+1, −z.