首页   按字顺浏览 期刊浏览 卷期浏览 Crystal structure of Ba2Li2/3Ti16/3O13
Crystal structure of Ba2Li2/3Ti16/3O13

 

作者: Christian Dussarrat,  

 

期刊: Journal of Materials Chemistry  (RSC Available online 1997)
卷期: Volume 7, issue 10  

页码: 2103-2106

 

ISSN:0959-9428

 

年代: 1997

 

DOI:10.1039/a703252h

 

出版商: RSC

 

数据来源: RSC

 

摘要:

Crystal structure of Ba2Li2/3Ti16/3O13 Christian Dussarrat,a R. Alan Howie,b Glenn C. Mather,b Leticia M. Torres-Martinezc and Anthony R.Westb aInstitut de Chimie de laMatie`re Condense�e de Bordeaux, Cha�teau Brivazac, Avenue du Dr. Albert Schweitzer, 33608 Pessac CEDEX, France bDepartment of Chemistry, University of Aberdeen, Meston Walk, Old Aberdeen, Aberdeen, Scotland, UK AB24 3UE cUniversidad Autonoma de Nuevo L eon, Facultad de Quimica, San Nicolas de los Garza,Mexico Single crystals of Ba2Li2/3Ti16/3O13 have been isolated and its crystal structure solved from X-ray diVraction data.The structure is similar to that of Ba2Ti6O13 and Na2Ti6O13 [monoclinic, space group C2/m (no.12), a=15.171(13), b=3.8992(18), c=9.106(4) A ° , b=98.64(6)°, Z=2], but one of the three independent octahedral sites contains partial substitution of Ti by Li, whereas the others are uniquely occupied by Ti.This phase contains only Ti4+, in contrast to Ba2Ti6O13, which has a mixture of Ti3+ and Ti4+. The structure is described in terms of both an octahedral framework and a cubic close-packing model of [Ba2%O13] layers (%: vacancy). The BaO–Li2O–TiO2 system is currently of interest due given in Table 1.Atomic coordinates and bond distances are summarised in Tables 2 and 3, respectively.† to the discovery of two phases which show relatively high Li+-ion conductivity. One is a hollandite-like phase, Ba3xLi2x+4yTi8-2x-yO16;1 the other is of uncertain compo- Results and Discussion sition but shows good conduction properties.2 Recently, Torres- Martinez et al. have carried out a phase diagram study of the The crystal structure of Ba2Li2/3Ti16/3O13 is essentially ternary system BaTiO3–Li2TiO3–TiO2, in which a number of analogous to that of Na2Ti6O13 and Ba2Ti6O13.The Ti(1)O6, ternary phase regions were established.3 A significant region Ti/Li(2)O6 and Ti(3)O6 octahedra form edge-sharing trimers, of Ba4Ti13O30 ternary solid solution and a smaller area of as shown in Fig. 1. These trimers corner-share with other BaTi5O11 were found. Other single-phase areas included a trimers to form layers parallel to the ac plane. Between region centred on the Li+-ion conducting hollandite-like phase, successive layers, the trimers edge-share forming infinite riba phase centred on BaLi2Ti6O14, a phase based on a previously bons with a zigzag arrangement which run parallel to b.A reported solid solution Ba2Ti10-xLi4xO224 and a new phase projection of the structure along the b-axis is shown in Fig. 2. which formed over a range of compositions close to the The Ba atoms are found within the tunnels which are formed composition Ba3Li2Ti8O20. As a consequence of the crystal between these ribbons of edge-sharing octahedra.structure studies reported here, the actual composition of this The structure may also be described by a close-packing latter phase is found to be Ba2Li2/3Ti16/3O13. Its crystal struc- model in which [Ba2%O13] layers form a cubic close-packed ture is discussed with reference to a series of titanate phases array. The Li and Ti cations lie in interstices between the of general formula AmM2nO4n+1 (A=alkali-metal, alkaline- close-packed layers.Fig. 3 shows the [Ba2%O13] layers which earth metal, M=octahedrally coordinated cation), and is also lie parallel to the (5 -1 -1) planes of the unit cell. Barium is compared with a number of similar structures in which the eleven-coordinate as a result of the oxygen vacancy in the octahedral cation composition is diVerent to that of the [Ba2%O13] layer which, if occupied, would confer a twelvetitle phase.coordinate cuboctahedral coordination on Ba, equivalent to the A-cation environment in the cubic perovskite structure. Fig. 4 shows the coordination environment of Ba; it is com- Experimental posed of an antiprism of oxygens together with five oxygens which lie in a [Ba2%O13] layer.Needle-shaped single crystals of the title phase were prepared by heating an intimate mixture of TiO2 and BaCO3 in 853 All three octahedra are distorted; the greatest distortion occurs for Ti(3)O6, in which the bond distances are spread molar ratio with an excess of Li2CO3 for one week at 1250 °C; the product was then slow-cooled to room temperature at a over the range 1.805 to 2.207 A ° .A large variation of TiMO distances is not uncommon in barium titanates, and is the rate of 5 °C h-1. A crystal suitable for single-crystal analysis was selected from the reaction mixture under a petrographic result of the surrounding O atoms experiencing diVering bond strengths.9 The mean TiMO distance, 1.982 A ° , is similar to microscope. Details of the data collection are summarised in Table 1.The atomic coordinates for Ba2Ti6O13 were adopted that found in most barium titanates. Similarly, the range of BaMO distances, 2.696–3.224 A ° , falls within the limits of as the starting point for the refinement,8 which was carried out by full-matrix least squares. On refinement, it became clear BaMO distances found in barium titanates. As mentioned previously, there are sizeable residual maxima that Ti(2) had a high (isotropic) thermal vibration parameter compared with Ti(1) and Ti(3) and it was concluded that the in the final diVerence map.One of these, 2.6 e A ° -3 in size, is situated at 0.84 A ° from Ba and is attributable to ripple. The Ti(2) site had undergone significant substitution of Li for Ti, compatible with the preparation conditions of the material.other, 3.7 e A ° -3 and 1.95 A ° from Ba, and almost directly above The refinement was then completed with variable occupancies of the Ti(2) site. All atoms were refined anisotropically, except † Atomic coordinates, thermal parameters, and bond lengths and for O(6) and O(7) which became non-positive definite when angles have been deposited at the Cambridge Crystallographic Data this was attempted.A further anomaly, discussed below, is the Centre (CCDC). See Information for Authors, J. Mater. Chem., 1997, presence in the final diVerence map of rather large residual Issue 1. Any request to the CCDC for this material should quote the full literature citation and the reference number 1145/46. maxima. Refinement conditions and final reliability factors are J.Mater. Chem., 1997, 7(10), 2103–2106 2103Table 1 Crystallographic data crystal data Dc=4.631 Mg m-3 Ba2Li2/3Ti16/3O13 Mo-Ka radiation Mr=742.76 l=0.710 73 A ° monoclinic cell parameters from 14 reflections space group C2/m h=7.7–14.5° a=15.171(13) A ° m=10.6 mm-1 b=3.8992(18) A ° T=298 K c=9.106(4) A ° platey needle b=98.64(6)° 0.30×0.12×0.04 mm V=532.5(6) A ° 3 colourless Z=2 data collection Nicolet P3 diVractometer h–2h scans absorption correction: y-scans5 Tmin=0.535, Tmax=0.648 h=-18�18 908 reflections with F>4s(F), n=4 k=0�5 Rint=0.027 l=0�12 hmax=30° 908 measured reflections 855 independent reflections frequency:c 1 in 50 refinement program used to refine structure: SHELX-76.7 refinement on F w=1/(s2F+0.0003F2) R=0.052 (D/s)max=0.001 wR=0.048 Drmax=3.74a, 2.62b GOF=2.092 Drmin=-4.64 776 reflections isotopic extinction coeYcient (SHELX76)7a 61 parameters scattering factors from ref. 7(b) data collection and cell refinement: Nicolet P3 software.6 a1.95 A ° from Ba, i.e. at x,D+y,z relative to Ba at x,y,z. b0.84 A ° from Ba. cStandard check frequency. Table 2 Atomic coordinates (×104) and thermal parameters (×103) for Ba2Li2/3Ti16/3O13 with e.s.d.s in parenthesesa atom x/a y/b z/c Ueq b occupancy Bac 514.8(6) 0 7676.7(8) 8.9(2) Ti(1) 3802(2) 0 9056(2) 7.9(6) Ti(2) 2583(2) 0 2284(3) 3.6(9) 0.640(13) Li(2) 2583(2) 0 2284(3) 3.6(9) 0.360(13) Ti(3) 3301(2) 0 5613(2) 4.9(6) O(1) 1306(8) 0 1091(10) 13(2) O(2) 2633(6) 0 7606(10) 6(2) O(3) 2010(6) 0 4313(9) 6(2) O(4) 3334(7) 0 845(10) 10(2) Fig. 1 Cation arrangement in trimer of edge-sharing octahedra O(5) 3724(8) 0 3863(11) 15(2) O(6) 4292(7) 0 7022(10) 10(2) sites are fully occupied with occupancies of 1/3 and 2/3, O(7) 5000 0 10000 16(3) respectively. The discrepancy is, however, only about 2×e.s.d. aAll sites are fully occupied unless otated. bUeq=1/3SiSj and may not be significant. It was not feasible to carry out UiUja*i a*j ai aj.cAll atoms are in position 4i, apart from O(7) which direct chemical analysis of the crystal, especially of its Li is in 2b of space group C2/m. content and so some slight ambiguity remains over the precise composition and structural model. Table 3 Bond distances (A ° ) for Ba2Li2/3Ti16/3O13 The stoichiometry of the title phase, Ba2Li2/3Ti16/3O13, diVers in its Li (and O) content from that proposed previously, Ti(1)MO(1) 1.9594(11) Li/Ti(2)MO(1) 2.074(12) Ba3Li2Ti8O20, but the Ba5Ti ratio is the same.This indicates Ti(1)MO(2) 2.046(9) Li/Ti(2)MO(2) 1.8920(18)×2 that loss of Li2O by volatilisation can be a significant problem Ti(1)MO(4) 1.873(9)×2 Li/Ti(2)MO(3) 2.157(9) during high temperature syntheses. This phase may exist over Ti(1)MO(6) 2.096(10) Li/Ti(2)MO(4) 1.861(10) Ti(1)MO(7) 1.890(2) Li/Ti(2)MO(5) 2.077(11) a limited stoichiometry range as indicated previously;3 one Ti(3)MO(2) 2.207(9) BaMO(1) 3.160(9) possibility is to have partial reduction of Ti4+ according to Ti(3)MO(3) 2.130(9) BaMO(1) 3.133(11) Li++2Ti4+u3Ti3+.This mechanism would retain full occu- Ti(3)MO(3) 2.009(3)×2 BaMO(2) 3.224(10) pancy of the octahedral sites, as is found in isostructural Ti(3)MO(5) 1.805(10) BaMO(4) 2.820(7)×2 Ba2Ti6O13.8 We think this mechanism to be unlikely in the Ti(3)MO(6) 1.824(10) BaMO(5) 2.755(8)×2 non-reducing conditions used here, however.BaMO(6) 2.696(7)×2 BaMO(7) 3.0632(6)×2 Two other means of creating non-stoichiometry can be imagined, assuming full occupancy of oxygen positions and the cation oxidation states to be +2(Ba), +1(Li), +4(Ti).(1) A replacement mechanism 4Li+>Ti4+, with creation of Ba in the direction of b, is not so easy to explain. One plausible explanation is that it results as a consequence of stacking vacancies on the octahedral sites, consistent with the structure of isostructural Ba2Ti5.5O13.10 (2) A slight substoichiometry in faults in the creation of the layers perpendicular to b.The Li and Ti occupancy factors for the shared site [0.360(13) and barium sites (4Ba2+u4Li++Ti4+) constrained by the limit of full occupancy of the octahedral sites. Then, similarly to the 0.640(13), respectively] suggest that the composition of the material is slightly Li rich compared with the ideal where all hollandite-like phase, Ba3xLi2x+4yTi8-2x-2yO16,1 a general 2104 J.Mater. Chem., 1997, 7(10), 2103–2106Fig. 4 Coordination environment of Ba of Li for Ti apparently exclusively in the Ti(2) site. In this way, charge balance is attained without the need for any reduction of Ti4+ to Ti3+. A number of other structurally related phases show variations in the occupancies of the octahedral sites.In isostructural Ba2Ti5.5O13, for example, the departure from the ideal composition is accommodated by Fig. 2 Projection of the Ba2Li2/3Ti16/3O13 structure along b. Ba atoms the presence of vacancies in the central octahedron of the are represented as circles. trimeric unit.10 A small number of phases with more than one type of octahedrally coordinated cation have been reported.In the case of Ba2Fe2Ti4O13,11 Fe and Ti are distributed over all three octahedral sites, although Fe shows a strong preference for one of the end sites in the trimer of octahedra. The octahedral cation site distribution in Ba2Ti4Cr2O13 could not be determined unambiguously by XRD due to the similar scattering factors of Cr3+ and Ti4+.12 Lattice energy calculations, nevertheless, predict that Cr3+ would prefer an end site in order to minimize Coulombic cation–cation repulsion eVects.In Ba2Ti5ZnO13, there is also a preference for Zn to occupy an end site in the trimer.13 As far as we are aware, Ba2Li2/3Ti16/3O13 is the only analogue in which the nontitanium cation occupies a single crystallographic site; in the others, with the possible exception of the Cr analogue, the two octahedral cations are distributed, but non-statistically, over more than one site.The Li+ ions are, thus, isolated and in fully occupied sites. This would appear to explain why Ba2Li2/3Ti16/3O13 shows no significant levels of Li+-ion conduction. The crystal structure of Ba2Li2/3Ti16/3O13 belongs to a series of titanates characterised by the formula AmM2nO4n+1, where A is an alkali-metal or alkaline-earth metal cation which fills a large (ten- or eleven-coordinate) site andMis an octahedrally coordinated cation. These titanates may be thought of as a series of tunnel structures built up from a network of Ticontaining octahedra.In general, m is equal to n/2 or (n+1)/2 for n even or odd, respectively, in order to minimize cation– cation interactions in the tunnels.For all members of this series, octahedra share edges within a layer, creating a unit which is n octahedra wide. Each octahedral unit shares corners with two other units in the same layer and edges with units in the layers above and below, to form a network of infinite chains of edge-sharing octahedra; these chains have a zigzag arrangement and are parallel to the b direction of the unit cell.Fig. 3 A close-packing layer of composition [Ba2%O13] where % is The large A cations are found in the tunnels which are formed a vacancy. Ba atoms are shown as shaded circles. by the network of infinite chains. The b parameter corresponds approximately to the diagonal of an octahedron (i.e. the width of a layer) and is about 3.7–3.8 A° in every structure of the formula covering the solid solution area can be written Ba2-xLi2/3+x-4yTi16/3+x/4+yO13. series (Table 4).The size of the a parameter is approximately 15 A ° in every structure, with the exception of Cs0.61Ti1.844O4. The results of this structure determination point to a comparatively simple explanation of the ease of preparation of this In this instance, the large size of caesium results in an a parameter that is greater than expected.Only the c parameter variant of the Ba2Ti6O13 structure type, namely the substitution J. Mater. Chem., 1997, 7(10), 2103–2106 2105Table 4 Unit-cell parameters and space group for members of the series AmM2nO4n+1 n phase space group a/A ° b/A ° c/A ° b/° ref. 2a BaTi4O9 C2/m 14.77 3.79 6.29 100.3 9 2+3b K2SrTi10O22 C2/m 15.317 3.787 15.439 102.68 15 3 Ba2Li2/3Ti16/3O13 C2/m 15.17 3.90 9.11 98.64 this work 3+4c Na2Ti7O15 C2/m 14.90 3.74 20.9 96.5 16 4 K2Ti8O17 C2/m 15.678 3.775 11.991 95.67 14 2 Cs0.61Ti1.844O4 Immm 17.012 3.829 2.962 90 17 aBaTi4O9 is a hypothetical structure.bIntergrowth between the members n=2 and n=3. cIntergrowth between the members n=3 and n=4. 8 W. H. Baur, E. Tillmanns and W. Hofmeister, Crystal Struct. depends on the number of octahedra in the units and corre- Commun., 1982, 11, 2021. sponds to the distance between opposite corners in a unit. 9 E. Tillmanns, W. Hofmeister and W. H. Baur, J. Solid State Chem., Assuming the octahedra are regular, the length of c can be 1985, 58, 14. estimated from the relation: c=Ó2(n2+1)dMMO. 10 V. W. Hofmeister and E. Tillmanns, Acta Crystallogr., Sect. B, 1979, 35, 1590. 11 T. A. Vanderah, Q. Huang, W. Wong-Ng, B. C. Chakoumakos, References R. B. Goldfarb, R. G. Geyer, J. Baker-Jarvis, R. S. Roth and A. Santoro, J. Solid State Chem., 1995, 120, 121. 1 C. Suckut, R. A. Howie, A. R. West and L. M. Torres-Martinez, 12 S. Mo�hr and Hk. Mu�ller-Buschbaum, Z.Naturforsch. T eil B, 1994, J.Mater. Chem., 1992, 2, 993. 49, 911. 2 W. J Zheng, R. Okuyuma, T. Esaka and H. Iwahara, Solid State 13 R. S. Roth, C. J. Rawn, C. J. Lindsay and W. Wong-Ng, J. Solid Ionics, 1989, 35, 235. State Chem., 1993, 104, 99. 3 L. M. Torres-Martinez, C. Suckut, R. Jimenez and A. R. West, 14 M. Le Granvalet and L. Brohan, J. Solid State Chem., 1993, 107, J.Mater. Chem., 1994, 4, 5. 127. 4 E. Tillmanns and I.Wendt, Z. Kristallogr., 1976, 144, 16. 15 A. D.Wadsley and W. G. Mumme, Acta Crystallogr., Sect. B, 1968, 5 A. C. T. North, D. C. Phillips & F. S. Mathews, Acta Crystallogr., 24, 392. Sect. A, 1968, 24, 251. 16 T. Sasaki, M. Watanabe, Y. Fujiki, Y. Kitami and M. Yokoyama, 6 Nicolet P3/R3 Data Collection Operator’s Manual, Net XRD J. Solid State Chem., 1991, 92, 537. Corporation, 1980. 17 I. E. Grey, C. Li, I. C. Madsen and J. A.Watts, J. Solid State Chem., 7 (a) G. M. Sheldrick, SHELX-76 Program for Crystal Structure 1987, 66, 7. Determination, University of Cambridge, 1976; (b) International T ables for X-Ray Crystallography, Kynock Press, Birmingham, 1974, vol. 4. Paper 7/03252H; Received 12th May, 1997 2106 J. Mater. Chem., 1997, 7(10), 2103–2106

 



返 回