Synthesis and Crystal Structure of The New Perovskite Cala Cati O (La) (Ca Ti) O)
Synthesis and Crystal Structure of The New Perovskite Cala Cati O (La) (Ca Ti) O)
www.elsevier.com/locate/ssscie
Abstract
The new CaLa2 CaTi2 O9 (= (Ca1/3 La2/3 )(Ca1/3 Ti2/3 )O3 ) perovskite was prepared by solid state reaction at 1300 ◦ C in air atmosphere.
The structure refinement by Rietveld analysis using X-ray powder diffraction (XRD) shows that the cationic distribution over the A and B site
of this perovskite can be illustrated by the crystallographic formula (Ca1/3 La2/3 )A (Ca1/3 Ti2/3 )B O3 . The diffraction pattern is interpreted to
exhibit splitting of the hkl and h0l lines indicative of a monoclinic (P 21 /n space group) derivative of the GdFeO3 -type structure with lattice
parameters: a = 5.5865(2) Å, b = 5.7077(2) Å, c = 7.9805(5) Å and β = 90.10(2)◦ . The monoclinic structural distortion involves partial
1:1 ordering between Ti4+ (in 2c site) and a random mixture (2/3 Ca2+ ; 1/3 Ti4+ ) (in 2d site) all at the B-perovskite sites.
2003 Éditions scientifiques et médicales Elsevier SAS. All rights reserved.
1. Introduction eral reports have shown that the ordering of the two B and
B cations is dependent on the size difference and charge
A majority of perovskite-like oxide have the general for- imbalance between them [6–11]. Note for example that the
mula ABO3 and have held the interest over many years. De- short-range ordering between Fe3+ and Nb5+ at the B-site
pending on the tolerance factor, t = (rA + rO )/21/2 (rB + rO ), observed in Ca(Fe1/2 Nb1/2)O3 [6] and the long-range order-
the structure is simple cubic (t ∼ 1) with P m3̄m space ing between Mg2+ and Ti4+ in Nd(Mg1/2Ti1/2 )O3 [7] and
group, or distorted (t > 1 or <1) with generally lower sym- between Ca2+ and U6+ in Ca(Ca1/2 U1/2)O3 [8] can be re-
metry. The ideal perovskite ABO3 structure consists of cor- lated to the charge difference between B and B cations,
ner linked BO6 octahedra with the large A cation in the cen- rather than difference in ionic radii [6–8]. The long-range
tre of the space between the octahedra. Many of the real cation ordering observed in Ba(Zn1/3Ta2/3 )O3 appears at
ABO3 compounds have a distorted perovskite-like structure. first sight to originate in the difference in the ionic size of
As a rule, the ionic radii do not match exactly (t = 1), which the two cations (Ta5+ (0.64 Å) and Zn2+ (0.74 Å)) [9,10]
is compensated for either by displacements of the cations, whereas in the case of Ca(Mn1/2W1/2 )O3 the driving force
within the BO6 octahedra, along certain directions and/or for Mn2+ (0.97 Å) and W6+ (0.74 Å) ordering is the size and
by tilting of the oxygen octahedra. Within the broad fam- charge difference between both kinds of cations. Gallasso
ily of complex metal oxides with perovskite structure there and Pyle studied the ordering of B cations in A(B1/3 B2/3 )O3
are compounds with several B-site cations as in the case compounds and found that B cations ordering is dependent
of A2 (B B )O6 and A3 (B B2 )O9 which can more clearly on the temperature of synthesis [12,13]. Thus, the structural
written as A(B1/2 B1/2 )O3 and A(B1/3 B2/3 )O3 (the primes analysis shows that Mg2+ and Ta5+ in Ba(Mg1/3Ta2/3 )O3
indicate the possibility of different ions or different oxida- calcined at 1600 ◦ C, are completely ordered (1:2 ordering,
tion state) [1–5]. Despite the large number of perovskite ox- trigonal P 3̄m1 space group) whereas they are in a dis-
ides prepared over the years, the problem of the ordering of ordered arrangement (cubic P m3m space group) for that
cations in the B-sublattice is only partly understood. Sev- prepared at 1300 ◦ C [15]. By varying the heat treatment,
four Ca(Ca1/3 Nb2/3)O3 polymorphs with perovskite-related
E-mail address: [email protected] (A. Aatiq). structure and different arrangements of B-site cations was
1293-2558/03/$ – see front matter 2003 Éditions scientifiques et médicales Elsevier SAS. All rights reserved.
doi:10.1016/S1293-2558(03)00076-1
746 A. Aatiq / Solid State Sciences 5 (2003) 745–749
reported [14,16]. The structure of Ca(Ca1/3 Nb2/3 )O3 poly- [14,16], Ba(La1/3 Ru2/3 )O3 [22] and La(Cu1/2Ti1/2 )O3 [17,
morphs is either disordered or exhibits chemical 1:1 order- 23,24]. In order to clarify the calcium–lanthanum distribu-
ing or 1:2 ordering of Ca and Nb cations [14,16]. Previously, tion over the available A and B perovskite sites, three crys-
the synthesis, structural and physical characterization of per- tallographic formulas were proposed. In the first formula
ovskite with general formula (A1/3 Ln2/3)A (B1/3 Ti2/3 )B O3 (Ca2/3 La1/3)A (La1/3Ti2/3 )B O3 all calcium atoms occupy the
(A = alkaline-earth, Ln = rare-earth, B = transition metal) A-site and lanthanum atoms were divided between A- and B-
has been realized [17]. Therefore, in a continuation of our site and in the second (Ca3/6 La3/6 )A (Ca1/6 La1/6Ti2/3 )B O3
search concerning perovskite-like structure and in order to both lanthanum and calcium were supposed to occupy the A
understand the effect of charge imbalance and size difference and B sites. In the third formula (Ca1/3 La2/3 )A (Ca1/3 Ti2/3)B -
of A and B cations on the structure and physical properties, O3 , lanthanum atoms were supposed to occupy only the
in the present paper the preparation and structural charac- A-site and calcium atoms were divided between A and
teristics of the new (Ca1/3 La2/3)A (Ca1/3Ti2/3 )B O3 phase by B-site. Rietveld refinement, in Pbnm space group, using
powder X-ray diffraction (XRD) study using the Rietveld (Ca2/3 La1/3)A (La1/3Ti2/3 )B O3 and (Ca3/6 La3/6 )A (Ca1/6 -
method are investigated. La1/6Ti2/3 )B O3 formulas with A-type cations in 4c site,
B-type cations in 4b one, O1 atom in 4c site and O2 one
in 8d leads to a high reliability factors (RWP = 21.8%,
2. Experimental RP = 14.9% for (Ca2/3 La1/3 )A (La1/3Ti2/3 )B O3 and RWP =
19.0%, RP = 13.1% for (Ca3/6 La3/6 )A (Ca1/6 La1/6Ti2/3 )B -
Polycrystalline CaLa2 CaTi2 O9 was prepared by classical O3 . The observed, calculated and difference patterns from
high temperature solid-state chemistry from powdered mix- Rietveld refinement in the two hypothesis of cationic dis-
tures of CaCO3 , TiO2 and La2 O3 in the adequate stoichio- tribution over the 2θ range from 20◦ to 50◦ are com-
metric ratio. The mixture was heated progressively with in- pared, in Fig. 1a and b, with those obtained in the case of
termittent grinding at 900 ◦ C (12 h), 1000 ◦ C (24 h), pressed (Ca1/3 La2/3)A (Ca1/3Ti2/3 )B O3 formula (Fig. 1c). As shown
into pellets and sintered at 1300 ◦ C for another 12 h in air by the fit between the experimental and calculated XRD pat-
atmosphere. The sample was cooled to room temperature terns (Fig. 1c), refinement using (Ca1/3 La2/3)A (Ca1/3 Ti2/3)B -
for re-grinding several times. The compound was character- O3 formula leads to a relatively good reliability factors
ized by X-ray diffraction (XRD) at room temperature with a (RWP = 15.2%, RP = 11.4%) in comparison to those ob-
Bruker D8 Advanced diffractometer using Cu Kα-radiation. tained for the two first hypothesis. Taking in account those
For the Rietveld refinement, the data were collected in the deduction, in the following only the (Ca1/3 La2/3 )A (Ca1/3 -
20–120◦ range (0.02 step/20 s counting time). The refine- Ti2/3 )B O3 formula was considered.
ment was carried out by the Rietveld method using the Full- Close examination of the diffraction profile shows that
prof program [18]. several diffraction peaks corresponding to the orthorhom-
bic (P bnm space group) type-perovskite structure became
broadening and the peak shape almost begins to split.
3. Results and discussions Peak splitting observed in some XRD reflections of (Ca1/3 -
La2/3)A (Ca1/3 Ti2/3)B O3 is essentially similar to that ob-
3.1. Structural refinement served on the XRD patterns of many perovskites such
as Ca(Fe1/2Nb1/2 )O3 [21], Sm(Mn1/2Ti1/2 )O3 [25] and
XRD patterns of CaLa2 CaTi2 O9 , is able to be indexed RNiO3 (R = Ho, Y) [26]. Those considerations indicate
based on a similar orthorhombic GdFeO3 type-structure a lowering in the crystal symmetry with respect to the
(Pbnm space group) with a randomly cationic distribution orthorhombic Pbnm model. The Rietveld refinement was
within the A and B site of the ABO3 perovskite [19]. In therefore reconsidered in a monoclinic space group from
a first time of the Rietveld refinement, the structure refine- XRD data. P 21 /n is a subgroup of Pbnm, hence, the mono-
ment was performed in the space group Pbnm. The atomic clinic starting model was built using the refined orthorhom-
position of CaTiO3 were used as starting parameters for bic structural parameters. The monoclinic unit-cell derived
the Rietveld refinement of CaLa2 CaTi2 O9 [20,21]. It is well from the parental orthorhombic cell (P bnm or P nma) by
known that Ti4+ ion prefer to occupy the octahedral B-site 1:1 cation ordering at the B-site corresponds to the space
of ABO3 perovskite however for Ca2+ and La3+ cations, group P 21 /n (a ≈ b ≈ 21/2 ac , c ≈ 2ac , β ≈ 90 ◦ ). It should
where there is no significant difference between their ionic be noticed that the superstructure peaks h0l (h, l = odd)
3+ 2+
radii (rLa = 1.032 Å, rCa = 1.00 Å in octahedral site, i.e., are not observed on the XRD pattern, suggesting that there
3+ 2+
A-site) and (rLa = 1.36 Å, rCa = 1.34 Å in twelve coor- is no long-range ordering between Ca2+ and Ti4+ ions in
dination, i.e., B-site) [9], it is difficult to distribute these the B-site of ABO3 compounds. However, the hkl and h0l
atoms between A- and B-sites of the perovskite structure lines are broadened, reduced in intensity and flattened at
without ambiguity. Note that A and B sites of ABO3 per- the tip. This indicates further splitting of these lines into
ovskite can be occupied either by calcium and/or lanthanum hkl + hk − l and h0l + h0 − l components, and suggests
atoms as it was mentioned for example in Ca(Ca1/3 Nb2/3)O3 a monoclinic distortion of the crystal structure. The atomic
A. Aatiq / Solid State Sciences 5 (2003) 745–749 747
Table 1
Powder diffraction data of (Ca1/3 La2/3 )(Ca1/3 Ti2/3 )O3 (Cu Kα1 : 1.54060 Å)
h k l dobs. (Å) 100 × I /I0 h k l dobs. (Å) 100 × I /I0
1 1 0 3.992 15 1 3 1 1.756 4
0 0 2 3.989 8 3 1 −1 1.729 2
1 1 −1 3.572 7 3 1 1 1.728 2
1 1 1 3.568 8 1 3 −2 1.642 5
0 2 0 2.854 25 1 3 2 1.641 9
1 1 −2 2.824 86 0 2 4 1.635 14
1 1 2 2.820 100 2 0 −4 1.625 9
2 0 0 2.793 28 2 0 4 1.622 15
0 2 1 2.687 3 3 1 −2 1.619 21
2 1 0 2.509 1 3 1 2 1.617 21
1 2 −1 2.422 1 2 2 −3 1.597 2
1 2 1 2.421 1 2 2 3 1.595 2
0 1 3 2.411 1 1 3 −3 1.419 3
1 0 −3 2.403 2 1 3 3 1.421 3
2 1 −1 2.394 4 1 1 −5 1.483 1
2 1 1 2.392 4 0 4 0 1.427 2
0 2 2 2.321 9 2 2 −4 1.412 5
2 0 −2 2.290 9 2 2 4 1.410 8
2 0 2 2.286 10 0 4 1 1.405 3
1 1 −3 2.215 3 4 0 0 1.396 3
1 1 3 2.212 4 0 2 5 1.393 1
1 2 −2 2.143 1 3 1 4 1.323 1
2 2 0 1.996 36 3 3 −1 1.313 1
0 0 4 1.995 37 3 3 1 1.312 1
0 2 3 1.946 2 2 4 0 1.270 1
2 2 −1 1.937 5 3 3 −2 1.263 4
2 2 1 1.936 5 1 1 −6 1.262 8
2 2 −2 1.786 2 3 3 2 1.262 8
1 1 −4 1.785 3 1 1 6 1.261 8
2 2 2 1.784 3 0 4 3 1.257 4
1 1 4 1.784 3 2 4 −1 1.255 4
3 1 0 1.770 1 2 4 1 1.255 4
1 3 −1 1.757 4 4 2 0 1.254 4
Note: diffraction lines with Iobs. < 1 are omitted.
Table 2
Structural parameters for (Ca1/3 La2/3 )(Ca1/3 Ti2/3 )O3
Atom Site x y z B (Å2 ) Occupancy
Ca(1)/La 4e 0.5090(4) 0.5417(2) 0.2509(11) 0.55(2) 0.33/0.67
Ti(1) 2c 0 0.5 0 0.25(2) 1
Ti(2)/Ca(2) 2d 0.5 0 0 0.25(2) 0.33/0.67
O(1) 4e 0.2059(9) 0.2292(11) −0.0778(8) 0.35(3) 1
O(2) 4e 0.2702(9) 0.6945(7) −0.0458(6) 0.35(3) 1
O(3) 4e 0.4276(4) 0.9778(3) 0.2602(7) 0.35(3) 1
(Ca1/3 La2/3 )A (Ca1/3 Ti2/3 )B O3 space group (P 21 /n).
Constraints: for all B-sites B(Ca) = B(Ti), B(O(1)) = B(O(2)) = B(O(3)).
a = 5.5865(2) Å, b = 5.7077(2) Å, c = 7.9805(5) Å, β = 90.10(2) ◦ .
RWP = 12.2%, RP = 8.7%, RI = 5.6%.
position of Nd(Mg1/2Ti1/2)O3 (P 21 /n space group) were Fig. 2 and the results of refinement, the atomic positions and
used as starting parameters for the Rietveld refinement [7]. selected bonds distances are summarized in Tables 2 and 3.
The refinement using the space group P 21 /n achieves bet-
ter fit between the experimental and calculated XRD pat- 3.2. Structural description
terns than the refinement based on the orthorhombic Pbnm
model. A good resolution between the component peaks are The (Ca, La) cations are statistically distributed on A-
therefore obtained (Table 1). Intensities given in Table 1 are site in (Ca1/3La2/3 )A (Ca1/3 Ti2/3 )B O3 compounds. The type
obtained from the “observed intensities” of the Rietveld re- of B-cations ordering results in a partially ordered struc-
finement. The experimental and fitted profiles are shown in ture with alternating {111}c plane occupied exclusively by
748 A. Aatiq / Solid State Sciences 5 (2003) 745–749
Table 3
Selected interatomic distances (Å) for (Ca1/3 La2/3 )A (Ca1/3 Ti2/3 )B O3
B-sites
Ti(1)–O(1) 2.025(3) × 2 (Ca(2)/Ti(2))–O(1) 2.189(3) × 2
Ti(1)–O(2) 1.909(2) × 2 (Ca(2)/Ti(2))–O(2) 2.196(2) × 2
Ti(1)–O(3) 1.959(2) × 2 (Ca(2)/Ti(2))–O(3) 2.120(2) × 2
Average 1.964 2.168
A-sites
(Ca(1)/La)–O(1) 2.337(4) (Ca(1)/La)–O(2) 2.649(3)
(Ca(1)/La)–O(1) 2.482(3) (Ca(1)/La)–O(2) 2.853(3)
(Ca(1)/La)–O(1) 3.066(3) (Ca(1)/La)–O(3) 2.468(3)
(Ca(1)/La)–O(2) 2.455(4) (Ca(1)/La)–O(3) 2.532(3)
Average 2.605
Fig. 2. Observed (• • •), calculated (—) and difference profiles of the XRD
of (Ca1/3 La2/3 )(Ca1/3 Ti2/3 )O3 .
This distortion results from short-range ordering between [11] A.K. Azad, S.A. Ivanov, S.-G. Eriksson, J. Eriksen, H. Rundlöf,
Ti(1) in 2c (0 12 0) sites and Ti(2)/Ca(2) in 2d ( 12 0 0) sites R. Mathieu, P. Svedlindh, Mater. Res. Bull. 36 (2001) 2485.
all in the B-sites. This partial ordering at the B-site may be [12] F. Galaasso, J. Pyle, Inorg. Chem. 2 (1963) 482.
[13] F. Galaasso, J. Pyle, J. Phys. Chem. 67 (1963) 1561.
related to the difference in ionic radii rather than the charge [14] I. Levin, L.A. Bendersky, J.P. Cline, R.S. Roth, T.A. Vanderah, J. Solid
difference between Ca2+ and Ti4+ cations. State Chem. 150 (2000) 43.
[15] S. Janaswamy, G.S. Murthy, E.D. Dias, V.R.K. Murthy, Mater. Lett.
(2002), in press.
[16] I. Levin, J.Y. Chan, R.G. Geyer, J.E. Maslar, T.A. Vanderah, J. Solid
References State Chem. 156 (2001) 122.
[17] A. Aatiq, Etude cristallochimique, diélectriques et magnétiques de
[1] L.H. Brixner, J. Phys. Chem. 64 (1960) 165. quelques perovskites de type (A, Ln)(B, Ti)O3 ; A = Ca, Sr, Ba; Ln =
[2] F.S. Galasso, W. Darby, J. Phys. Chem. 66 (1962) 131. La, Pr, Nd, Sm, Eu; B = Co, Ni, Cu, Zn; DES de 3ème Cycle de
[3] F.S. Galasso, in: R. Smoluchouski, N. Kurti (Eds.), International Series spécialité Faculty of Sciences; Rabat, Morocco, 1987.
of Monographs in Solid State Physics, Vol. 5, Pergamon, London, [18] J. Rodriguez-Carvajal, Fullprof, Program for Rietveld Refinement,
1969. Laboratoire Léon Brillouin (CEA-CNRS), Saclay, France, 1997.
[4] J.B. Goodenough, J.M. Longo, Crystallographic and Magnetic Prop- [19] M. Marezio, J.P. Remeika, P.D. Dernier, Acta Crystallogr., Sect. B 26
erties of Perovskite and Perovskite Related Compounds, New Se- (1970) 2008.
ries, Group III, Vol. 4a, Springer-Verlag, New York, 1970, (Landolt- [20] S. Sasaki, C.T. Prewitt, J.D. Bass, W.A. Schulze, Acta Crystallogr.,
Börnstein Springer link). Sect. C 43 (1987) 1668.
[5] A.J. Jacobson, B.M. Collins, B.E.F. Fender, Acta Crystallogr., [21] A.R. Chakmouradian, R.H. Mitchell, J. Solid State Chem. 138 (1998)
Sect. B 32 (1976) 1083. 272.
[6] A.R. Chakmouradian, R.H. Mitchell, J. Solid State Chem. 138 (1998) [22] Y. Doi, K. Matsuhira, Y. Hinatsu, J. Solid State Chem. 165 (2002) 317.
272. [23] N. Ramadass, J. Gopalakrishnan, M.V.C. Sastri, J. Inorg. Nucl.
[7] W.A. Groen, F.P.F. Van Berkel, D.J.W. Ijdo, Acta Crystallogr., Chem. 40 (1980) 261.
Sect. C 42 (1986) 1472. [24] S. Sefian-Lamarti, F. Weill, B. Darriet, P. Gravereau, G. LeFlem,
[8] H.C. Van Duivenboden, D.J.W. Idjo, Acta Crystallogr., Sect. C 42 J. Solid State Chem. 113 (1994) 398.
(1986) 523. [25] N. Kamegashira, N. Nakajima, K. Watanabe, M. Kobayashi, J. Alloys
[9] R.D. Shannon, Acta Crystallogr., Sect. A 32 (1976) 751. Compd. 311 (2000) 74.
[10] A.J. Jacobson, B.M. Collins, B.E.F. Fender, Acta Crystallogr., [26] J.A. Alonso, M.J. Martinez-Lope, M.T. Casais, M.A.G. Aranda,
Sect. B 32 (1976) 1083. M.T. Fernandez-Diaz, J. Am. Chem. Soc. 121 (1999) 4754.