The structure of lithium barium silicate, Li2BaSiO4, has been determined from synchrotron radiation powder data. The title compound was synthesized by high-temperature solid-state reaction and crystallizes in the hexagonal space group P63cm. It contains two Li atoms, one Ba atom (both site symmetry ..m on special position 6c), two Si atoms [on special positions 4b (site symmetry 3..) and 2a (site symmetry 3.m)] and four O atoms (one on general position 12d, and three on special positions 6c, 4b and 2a). The basic units of the structure are (Li6SiO13)5- units, each comprising seven tetrahedra sharing edges and vertices. These basic units are connected by sharing corners parallel to [001] and through sharing (SiO4)4- tetrahedra in (001). The relationship between the structures and luminescence properties of Li2SrSiO4, Li2CaSiO4 and the title compound is discussed.
Supporting information
LBSO was synthesized by a high-temperature solid-state reaction using
stoichiometric quantities of high purity oxides, namely lithium carbonate
(Li2CO3, Sigma Aldrich, 99.99%), barium carbonate (BaCO3, Kojundo
Chemicals Ltd, 99.9%) and silicon oxide (SiO2, Kojundo Chemical Ltd, 99.9%).
A stoichiometric mixture was weighed and mixed thoroughly in an agate mortar
with the addition of acetone, which was then triturated several times for
homogenization. The preparation was then placed in alumina crucibles and
preheated in air at 573 K for 3 h, then ground again and heated at 1123 K
under a reducing atmosphere of a 25%H2–75%N2 stream in a horizontal tube
furnace for 16 h, so that the desired oxidation state of the activator ion
could be attained. After firing, the samples were cooled to room temperature
in the furnace and were ground again with an agate mortar for further use. The
powder pattern for the obtained powder was collected using synchrotron
radiation at the 8 C2 high-resolution powder diffraction beamline of the
Pohang Light Source. Small amounts of Ba2SiO4, Li2SiO3 and Li4SiO4
were detected as impurities.
The diffraction pattern also includes peaks originated from Ba2SiO4
(Pmcn, a = 5.8115 Å, b = 10.2135 Å and c
=7.5035 Å), Li2SiO3 (Cmc21, a = 9.4008 Å, b = 5.4073 Å and c = 4.6529 Å) and Li4SiO4 (P21/m,
a = 5.1469 Å, b = 6.1051 Å, c = 5.2960 Å and β =
90.3546°), but the peaks of Li2BaSiO4 and those of the impurities are
well separated from one another in the high-resolution synchrotron radiation
powder pattern, at least in the low and medium scattering angle region (ICSD,
2008). We could easily exclude the peaks of impurities from the peaks
used for
unit-cell determination. The Li2BaSiO4 powder diffraction pattern was
indexed in the hexagonal system using TREOR (Werner et al.,
1985) with merit M30 = 271.0 (F30 = 448.0). The 2θ
difference
between the positions of observed and calculated peaks was less than 0.002°.
The space group P63cm (No. 185) was chosen from the systematic
absences and confirmed by the subsequent structure refinement. The positions
of Ba and Si atoms were determined by direct methods using the integrated
intensities in the FULLPROF suite (Rodriguez-Carvajal, 1990).
The
positions of the O atoms were determined by the simulated annealing method in
FULLPROF. The positions of the Li atoms were determined by the maximum
entropy method using the ENIGMA package (Sakata & Sato, 1990).
Fig. 4
shows
the positions of Li atoms on the plane with z = 0.06. The Rietveld
refinement was initiated by using the atomic positions obtained from the
combination of the direct method, the simulated annealing method and the
maximum entropy method.
Data collection: PLS HRPD Beamline Software (reference?); cell refinement: FULLPROF (Rodriguez-Carvajal, 1990); data reduction: FULLPROF (Rodriguez-Carvajal, 1990); program(s) used to solve structure: FULLPROF(Rodriguez-Carvajal, 1990) and ENIGMA (Sakata & Sato, 1990); program(s) used to refine structure: FULLPROF (Rodriguez-Carvajal, 1990); molecular graphics: Balls&Sticks (Ozawa & Kang, 2004); software used to prepare material for publication: FULLPROF (Rodriguez-Carvajal, 1990).
lithium barium silicate
top
Crystal data top
Li2BaSiO4 | F(000) = 648 |
Mr = 243.3 | Dx = 4.023 (1) Mg m−3 |
Hexagonal, P63CM | Synchrotron radiation, λ = 1.54960 Å |
Hall symbol: P 6c -2 | T = 298 K |
a = 8.10040 (1) Å | white |
c = 10.60052 (1) Å | flat sheet, 20 × 0.5 mm |
V = 602.38 (1) Å3 | Specimen preparation: Prepared at 1123 K |
Z = 6 | |
Data collection top
Pohang Light Source 8C2 HRPD Beamline diffractometer | Data collection mode: reflection |
Radiation source: synchrotron, synchrotron | Scan method: step |
Si 111 monochromator | 2θmin = 10°, 2θmax = 131°, 2θstep = 0.005° |
Specimen mounting: packed powder pellet | |
Refinement top
Refinement on Inet | 24201 data points |
Least-squares matrix: full with fixed elements per cycle | Profile function: pseudo-Voigt |
Rp = 0.090 | 50 parameters |
Rwp = 0.121 | Weighting scheme based on measured s.u.'s |
Rexp = 0.072 | (Δ/σ)max < 0.001 |
RBragg = 0.037 | Background function: spline |
χ2 = 2.890 | Preferred orientation correction: none |
Crystal data top
Li2BaSiO4 | V = 602.38 (1) Å3 |
Mr = 243.3 | Z = 6 |
Hexagonal, P63CM | Synchrotron radiation, λ = 1.54960 Å |
a = 8.10040 (1) Å | T = 298 K |
c = 10.60052 (1) Å | flat sheet, 20 × 0.5 mm |
Data collection top
Pohang Light Source 8C2 HRPD Beamline diffractometer | Scan method: step |
Specimen mounting: packed powder pellet | 2θmin = 10°, 2θmax = 131°, 2θstep = 0.005° |
Data collection mode: reflection | |
Refinement top
Rp = 0.090 | χ2 = 2.890 |
Rwp = 0.121 | 24201 data points |
Rexp = 0.072 | 50 parameters |
RBragg = 0.037 | |
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top | x | y | z | Uiso*/Ueq | |
Ba | 0.57143 (6) | 0.00000 | 0.2283 (9) | 1.104 (8)* | |
Si1 | 1.00000 | 0.00000 | 0.3117 (9) | 0.67 (9)* | |
Si2 | 0.66666 | 0.33333 | 0.4788 (11) | 0.60 (4)* | |
O1 | 0.7774 (5) | 0.2260 (5) | 0.0332 (9) | 1.20 (9)* | |
O2 | 1.00000 | 0.1934 (5) | 0.2648 (10) | 0.86 (12)* | |
O3 | 0.66666 | 0.33333 | 0.3282 (11) | 1.74 (15)* | |
O4 | 0.00000 | 0.00000 | 0.475 (2) | 0.66 (13)* | |
Li1 | 1.00000 | 0.19438 | 0.07271 | 1.00000* | |
Li2 | 0.00000 | 0.75289 | 0.94752 | 1.00000* | |
Geometric parameters (Å, º) top
Ba—O1 | 2.713 (10) | Li1—O1 | 1.989 (5) |
Ba—O3 | 2.631 (6) | Li1—O2 | 2.036 (11) |
Si1—O2 | 1.644 (6) | Li1—O4iii | 1.885 (12) |
Si1—O4i | 1.73 (2) | Li2—O1iv | 1.963 (5) |
Si2—O1ii | 1.635 (6) | Li2—O2v | 1.985 (10) |
Si2—O3 | 1.596 (16) | Li2—O4vi | 2.023 (3) |
| | | |
O1vii—Ba—O3 | 74.54 (10) | | |
Symmetry codes: (i) x+1, y, z; (ii) x, x−y, z+1/2; (iii) x−y+1, x, z−1/2; (iv) −y, x−y, z+1; (v) −x+1, −y+1, z+1/2; (vi) x−y, x+1, z+1/2; (vii) −x+y, −x+1, z. |
Experimental details
Crystal data |
Chemical formula | Li2BaSiO4 |
Mr | 243.3 |
Crystal system, space group | Hexagonal, P63CM |
Temperature (K) | 298 |
a, c (Å) | 8.10040 (1), 10.60052 (1) |
V (Å3) | 602.38 (1) |
Z | 6 |
Radiation type | Synchrotron, λ = 1.54960 Å |
Specimen shape, size (mm) | Flat sheet, 20 × 0.5 |
|
Data collection |
Diffractometer | Pohang Light Source 8C2 HRPD Beamline diffractometer |
Specimen mounting | Packed powder pellet |
Data collection mode | Reflection |
Scan method | Step |
2θ values (°) | 2θmin = 10 2θmax = 131 2θstep = 0.005 |
|
Refinement |
R factors and goodness of fit | Rp = 0.090, Rwp = 0.121, Rexp = 0.072, RBragg = 0.037, χ2 = 2.890 |
No. of data points | 24201 |
No. of parameters | 50 |
No. of restraints | ? |
Selected bond lengths (Å) topBa—O1 | 2.713 (10) | Li1—O1 | 1.989 (5) |
Ba—O3 | 2.631 (6) | Li1—O2 | 2.036 (11) |
Si1—O2 | 1.644 (6) | Li1—O4iii | 1.885 (12) |
Si1—O4i | 1.73 (2) | Li2—O1iv | 1.963 (5) |
Si2—O1ii | 1.635 (6) | Li2—O2v | 1.985 (10) |
Si2—O3 | 1.596 (16) | Li2—O4vi | 2.023 (3) |
Symmetry codes: (i) x+1, y, z; (ii) x, x−y, z+1/2; (iii) x−y+1, x, z−1/2; (iv) −y, x−y, z+1; (v) −x+1, −y+1, z+1/2; (vi) x−y, x+1, z+1/2. |
There are several ways of constituting white light based on the LED technique, each of which involves various phosphors, which emit visible light by absorbing soft UV or blue light produced by LEDs. The representative oxide phosphors are Y3Al5O12:Ce3+ (YAG) and its variants, and divalent europium-doped orthosilicates. Based on their potential for use as white LEDs, Li2SrSiO4 (LSSO), Li2CaSiO4 (LCSO) and Li2BaSiO4 (LBSO) have recently been reported (Toda et al. 2006). In the present study, the structure of LBSO was determined and refined from synchrotron powder diffraction data (Fig. 1). The structure consists of two types of tetrahedra, (SiO4)4- and (LiO4)7-, with Ba2+ atoms located at the centers of open spaces formed by the tetrahedra (Fig. 2). These tetrahedra share edges to form (Li6SiO13)5- units that are connected by sharing corners (O atoms) parallel to [001] and through sharing (SiO4)4- tetrahedra in (001) (Fig. 3). The bond-valence sum (Brese & O'Keeffe, 1991) of the Ba2+ ion is close to the ideal value [S = 1.95 valence units (v.u.)]. The Si24+, Li11+ and Li21+ ions also reveal ideal bond-valence sums [S = 3.99, 1.02 and 0.99 v.u.], respectively, while the bond-valence sum of the Si14+ ion (S = 3.62 v.u.) deviates from the ideal value. This deviation is due to the longer bond length between atoms Si1 and O4 compared with those with the three O2 atoms (Table 1). Atom O4 is common to the Li1, Li2 and Si1 tetrahedra. Because of the Li—O interactions, the SiO4 tetrahedron is distorted and the Si14+ ion is underbonded.
The structures of LSSO (P3121; isostructural with Li2EuSiO4; Haferkorn & Meyer, 1998) and LCSO (I42m; Gard & West, 1973) also consist of (SiO4)4- and (LiO4)7- tetrahedra sharing edges; however, these tetrahedra do not form a compact (Li6SiO13)5- unit like that in the title compound.
When LSSO, LCSO and LBSO serve as phosphors, activator ions, such as Eu2+ in the present case, have to occupy the alkaline earth sites. In this regard, the local structure around the alkaline earth site has a great influence on the luminescent behavior. Dorenbos (2000a, 2000b, 2003) summarized the crystal field splitting in various Ce3+-doped phosphors and suggested a systematic model that describes the crystal field splitting in terms of polyhedron shape and ligand distance and also found that the experimentally measured centroid shift, crystal field splitting, red shift and Stokes shift values for Ce3+ ions are in a linear relationship with those of Eu2+ ions (Dorenbos, 2003). The crystal field splitting is proportional to 1/Reff2, with a constant designating the polyhedral shape, wherein Reff is the effective average ligand distance. The Sr, Ca and Ba polyhedra in LSSO, LCSO and LBSO have site symmetries 2, 42m and m, respectively, and different coordination numbers. Measured crystal field splittings of LSSO, LCSO and LBSO phosphors (Kulshreshtha, Sharma & Sohn, 2009) show that higher-symmetry, smaller coordination number and larger polyhedron size (ligand distance) lead to greater crystal field splitting.
On the other hand, the centroid shift is proportional to Nαsp/Reff6 according to the ligand polarization model, where N is the coordination number and αsp is the anion polarizability of the nearest anion neighbors (Dorenbos, 2000a, 2000b). Using experimentally measured centroid shift and ligand distance data for LSSO, LCSO and LBSO (Kulshreshtha, Sharma & Sohn, 2009), we confirmed that the greater the cation size, the higher the polarizability.
The energy transfer among the Eu2+ ions located at the alkaline earth site is also affected significantly by the local structure around the site. The angular class model proposed by Vasquez (1996) was found to be in a good agreement with the energy transfer rate obtained from the decay measurement for LSSO, LCSO and LBSO phosphors (Kulshreshtha, Shin & Sohn, 2009). The site symmetry information was incorporated into the model via several parameters, such as position generating vector of an angular class, maximum number of activators in an angular class and Riemann's zeta function of dipole–dipole interaction. It was found that the higher the site symmetry of the activator sites, the higher the energy transfer rate.