Redetermination of the low-temperature polymorph of Li2MnSiO4 from single-crystal X-ray data

Crystals of dilithium manganese(II) silicate were grown under high-temperature hydrothermal conditions in the system LiOH—MnO2—SiO2. The title compound crystallizes in the βII-Li3PO4 structure type. The coordination polyhedra of all cations are slightly distorted tetrahedra (m symmetry for MnO4 and SiO4), which are linked by corner-sharing to each other. The vertices of the tetrahedra point to the same direction perpendicular to the distorted hexagonal close-packed (hcp) array of O atoms within which half of the tetrahedral voids are occupied by cations. In comparison with the previous refinement from powder X-ray data [Dominko et al. (2006 ▶). Electrochem. Commun. 8, 217–222], the present reinvestigation from single-crystal X-ray data allows a more precise determination of the distribution of the Li+ and Mn2+ cations, giving a perfectly site-ordered structure model for both Li+ and Mn2+.

Crystals of dilithium manganese(II) silicate were grown under high-temperature hydrothermal conditions in the system LiOH-MnO 2 -SiO 2 . The title compound crystallizes in the II -Li 3 PO 4 structure type. The coordination polyhedra of all cations are slightly distorted tetrahedra (m symmetry for MnO 4 and SiO 4 ), which are linked by corner-sharing to each other. The vertices of the tetrahedra point to the same direction perpendicular to the distorted hexagonal closepacked (hcp) array of O atoms within which half of the tetrahedral voids are occupied by cations. In comparison with the previous refinement from powder X-ray data [Dominko et al. (2006). Electrochem. Commun. 8,[217][218][219][220][221][222], the present reinvestigation from single-crystal X-ray data allows a more precise determination of the distribution of the Li + and Mn 2+ cations, giving a perfectly site-ordered structure model for both Li + and Mn 2+ .
Supplementary data and figures for this paper are available from the IUCr electronic archives (Reference: WM2658).
Li(Ni,Mn,Co)O 2 , LiMn 2 O 4 and LiMPO 4 (M = Fe, Mn), because of the natural abundance of silica, iron, and manganese, but also due to a possible high theoretical capacity through two electron delivery. In order to understand the intercalation mechanism of Li + ions in Li 2 MSiO 4 cathode materials, their crystal structures have been investigated mainly by means of powder methods using X-ray or synchrotron radiation. Summarized by structural studies up-to-date (Islam et al., 2011;Santamaría-Pérez et al., 2012), it could be concluded that the Li 2 MSiO 4 compounds (M = Fe, Mn, Co) belong to a large family of materials known as derivatives of Li 3 PO 4 , where oxygen atoms form arrays with a distorted hexagonal-closepacking (hcp) within which half of the tetrahedral voids are occupied by cations. Depending on which site (up or down) of the array the cations occupy, the material shows a rich polymorphism. Such compounds may be divided into two families, designated as β-and γ-forms after the notations used for Li 3 PO 4 polymorphs (West & Glasser, 1972). The γpolymorphs are built up of both corner-and edge-sharing tetrahedra with half of the tetrahedra pointing along one direction perpendicular to the hcp array and the other half pointing along the opposite direction, while the β-polymorphs are built up of only corner-sharing tetrahedra, with all the tetrahedra pointing to the same direction perpendicular to the hcp array. Detailed structural information, particularly for electrochemically active Li 2 MnSiO 4 , were available from the previous studies. Li 2 MnSiO 4 exhibit three polymorphs, namely a low-temperature form denoted as β II (Pmn2 1 ), an intermediate temperature form denoted as γ II (Pmnb), and a high-temperature form denoted as γ o (P2 1 /n) (Arroyo-de Dompablo et al., 2006Dompablo et al., , 2008Belharouak et al., 2009;Dominko et al., 2006;Kokalj et al., 2007;Politaev et al., 2007;Wu et al., 2009;Zhong et al., 2010). It should be noted that in almost all polymorphs a site disorder for cationic sites, particularly for Li + sites substituted by transition metal ions, was observed. Surprisingly, the structure models proposed for Li 2 MSiO 4 (M = Mn, Fe, Co) have all been determined and refined by powder diffraction methods except for that of Li 2 CoSiO 4 (Yamaguchi et al., 1979). In terms of the fact that lithium has quite low scattering factors for X-rays, this may be true even for neutron diffraction, the crystallographic information obtained for Li sites by powder diffraction should inevitably include ambiguity to some extent. Efforts to obtain single crystals for structure determination have not been rewarded for Li 2 MSiO 4 (M = Mn, Fe). Although Setoguchi (1988) succeeded to grow single crystals of Li 2 FeSiO 4 from a flux method using LiCl at elevated temperatures, he could not determine the structure because of suffering from twinned crystals. Here we describe the single-crystal growth of Li 2 MnSiO 4 by means of a high temperature hydrothermal method and its structure determination using single-crystal X-ray diffraction, confirming a perfectly site-ordered structure for Li 2 MnSiO 4 in its low-temperature β II (Pmn2 1 ) polymorph.
supplementary materials The first detailed report on the description of the structure model for Li 2 MnSiO 4 accompanied with numerical crystallographic data is probably that determined by Dominko et al. (2006) who performed Rietveld refinements in the space group Pmn2 1 with a = 6.3109 (9) Å, b = 5.3800 (9)  (pseudo) trigonal-bipyramidal MO 5 polyhedra, as shown in Fig. 1(b). The structure can also be described as a typical β II -Li 3 PO 4 structure if cation sites with low site-occupancies are removed, as shown in Fig. 1(c) and (d). This structure model, which has such excessive disorders, may have somewhat possible deficiencies. Furthermore, the environment around Li + ions is crucially important for understanding the lithium intercalation behavior during the charge/discharge process. Theroretical studies concerned with expectation of redox potentials and lithium migration paths for Li 2 MnSiO 4 cathodes have been accomplished by several groups (Kokalj et al. 2007;Kuganathan & Islam, 2009;Wu et al. 2009;Mali et al. 2010;Zhong et al. 2010;Duncan et al. 2011) based on the model by Dominko et al. (2006); most of these studies adopted the cation site disorder model or the idealized ordered one only with primary sites.
The structure refined in the present study is a perfectly site-ordered one for all cationic sites (Fig. 2) though the fundamental framework structure is the same as that previously reported. Notably, the displacement ellipsoids are relatively large not only for lithium atoms but also for manganese atoms (Fig. 3). This may reflect a high diffusibility both of Li + and Mn 2+ ions in this cathode material. The ordered structure model found in the present study is consistent with the results of an NMR study by Sirisopanaporn et al. (2011). Unexpectedly, information on atomic coordinates available for Li 2 MnSiO 4 is scarce in the literature, where the data were refined from powder diffraction analyses (Dominko et al., 2006) and obtained from an optimization by atomistic simulation (Arroyo-de Dompablo et al., 2008). Table 2 shows the results of the bond-valence sum (BVS) analysis (Brown & Altermatt, 1985) for cation tetrahedra estimated from refined atomic coordinates in Li 2 MnSiO 4 , together with those for the structure models proposed previously for comparison. The deviations from the formal valences of each ion are fairly large for the previous studies, in particularly for Si, while in the present study the values of the BVS calculation for all ions are in very good agreement with the theoretical ones. Moreover, it should be mentioned that the MO 4 tetrahedra in the present model have much more regular environments than the previous models.
No structural data based on single-crystal X-ray diffraction data have been reported for Li 2 MnSiO 4 , although recent works on positive electrode materials for rechargeable lithium batteries reported the electrochemical characterization of this cathode material. Our present structural study of Li 2 MnSiO 4 provides more accurate information of its crystal structure than has been available up to now.

Experimental
In order to synthesize Li 2 MnSiO 4 single crystals, a high-temperature, high-pressure hydrothermal synthetic method was performed in a silver ampoule contained in a home-made autoclave made of stainless steel (SUS304) with 6 cm in outer diameter, 0.8 cm in inner diameter, and 1.8 cm 3 in volume. The pressure was provided by water. diagram of pure water. The autoclave was then cooled to 323 K at 5 K/h and quenched to room temperature by removing the autoclave from the furnace. The product was filtered off, washed with water, rinsed with ethanol, and dried at ambient temperature. The reaction produced light-green rod-shaped crystals of Li 2 MnSiO 4 that were obtained as a major product along with some quartz crystals.
The surface of the single crystals was observed by using optical (Olympus BX-60) and scanning electron microscopy (SEM, Jeol JSM-5310LVB). The elemental composition of the crystals was characterized by energy dispersive X-ray spectroscopy (EDS) attached to SEM (SEM/EDS, Nippon Denshi JED-2140).

Refinement
The structure was solved by direct methods and refined by subsequent Fourier syntheses, leading to wR2 = 4.17% in the early stages of refinement. The relatively high value of the Flack parameter, x = 0.167, pointed to a possible twinned crystal. The examination using ROTAX (Cooper et al., 2002) indicated two possible rotation twin axes about [010] and [001]. In addition to these rotation twin formations, a racemic twin (inversion twin) formation can be also possible for the non-centrosymmetric Pmn2 1 space group. Subsequent refinements using the twin laws for the three cases yielded a satisfactory solution with wR2 = 3.67% for all cases and a Flack parameter x = 0.00 (2) for the rotation twin cases. The twin fraction ratio is 82.9: 17.1. In non-centrosymmetric space groups where mirror planes and/or glide planes exist, an inversion twin is equivalent to the rotation twin through the 2-fold rotation axis perpendicular to the mirror and/or the glide planes. This is true for the present case.

Computing details
Data collection: CrystalClear (Rigaku, 2010); cell refinement: CrystalClear (Rigaku, 2010); data reduction: CrystalClear (Rigaku, 2010); program(s) used to solve structure: SIR97 (Altomare et al., 1999); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008); molecular graphics: VESTA (Momma & Izumi, 2011); software used to prepare material for publication: WinGX (Farrugia, 1999  Structure model of Li 2 MnSiO 4 determined by Dominko et al. (2006). The original structure model is projected (a) along    where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max < 0.001 Δρ max = 0.25 e Å −3 Δρ min = −0.62 e Å −3 Extinction correction: SHELXL97 (Sheldrick, 2008), Fc * =kFc[1+0.001xFc 2 λ 3 /sin(2θ)] -1/4 Extinction coefficient: 0.392 (13) Absolute structure: Flack (1983), 189 Friedel pairs Flack parameter: 0.171 (15) Special details Geometry. All s.u.'s (except the s.u. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell s.u.'s are taken into account individually in the estimation of s.u.'s in distances, angles and torsion angles; correlations between s.u.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell s.u.'s is used for estimating s.u.'s involving l.s. planes. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > 2σ(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.