Crystal Structure of A-site deficient La0.2Sr0.7-xCaxTiO3 Perovskite at ambient conditions and high temperatures: A neutron powder diffraction study

Please note that technical editing may introduce minor changes to the text and/or graphics, which may alter content. The journal’s standard Terms & Conditions and the Ethical guidelines still apply. In no event shall the Royal Society of Chemistry be held responsible for any errors or omissions in this Accepted Manuscript or any consequences arising from the use of any information it contains. Accepted Manuscript Dalton Transactions


Introduction
Solid oxide fuel cells are attracting much attention offering high conversion efficiency of chemical energy to electricity with an attractive possibility of using different types of fuel besides pure hydrogen.2][3][4] With the drawbacks of the widely used Ni-YSZ cermet anode in SOFCs, 2,5 alternative anode materials are actively being studied.These include perovskite-based materials which offer good stability during operation and tolerance to sulphur poisoning and carbon build up. 6revious studies on the perovskite system La 0.2 Sr 0.7−x -Ca x TiO 3 showed an increase in this compound's electrical conductivity with increased calcium substitution at anodic operational conditions; which was consistent with the decrease in the unit cell volume of this perovskite. 7However, this trend was reversed at much higher calcium content; i.e. x > 0.45; which required further structural investigations of this system.Since it is widely understood that electronic conduction involves electrons on the titanium oxygen sublattices, it is anticipated that local distortions of the TiO 6 octahedra 8-10 can be responsible for the drop in conductivity.Due to the low scattering power of oxygen in XRD, neutron diffraction can give a better picture due to the relatively large scattering length of oxygen ions. 11herefore, this paper outlines the results obtained using neutron powder diffraction to characterise the crystal structures of different compositions of the perovskite system La 0.2 Sr 0.7−x Ca x TiO 3 at ambient conditions, as well as, at high temperatures for the two samples discussed in previous work; 7 to investigate the origin of the slight drop in electrical conductivity.

Sample preparation
Samples of La 0.2 Sr 0.7−x Ca x TiO 3 with x = 0.1, 0.2, 0.3, 0.35, 0.4, 0.45, 0.5, 0.6 and 0.7 were synthesised using conventional solid state methods.Stoichiometric amount of high purity starting materials; i.e.La 2 O 3 (Sigma-Aldrich 99.99%), SrCO 3 (Alfa Aesar 99%), CaCO 3 (Alfa Aesar 99.5%) and TiO 2 (Alfa Aesar 99.5%) were dried prior to weight and mixed using a mortar and pestle.All mixtures were calcined in air at 1000 °C for a minimum of 15 hours.This was followed by ball milling the powders in a planetary ball mill in acetone for 4 hours to ensure a uniform particle size.These were then dried and pressed into ∼13 mm pellets using a uniaxial press.All pellets were sintered in air at 1500 °C for 15 hours.Pellets were crushed, ball milled for 4 hours and pressed, respectively, before sintering them for a second time at the same conditions.For La 0.2 Sr 0.25 Ca 0.45 TiO 3 which showed the highest electrical conductivity of this series, 7 a pellet was reduced in a a Mechanical and Materials Engineering Department, Institute Centre for Energy -iEnergy, Masdar Institute of Science and Technology, PO Box 54224, Abu Dhabi, United Arab Emirates tube furnace at 1050 °C for 72 hours under a constant flow rate of 5% H 2 -95% Ar gas mixture.All resulting pellets were crushed and ball milled, as above, into fine powders for neutron diffraction studies.

Neutron powder diffraction
Neutron powder diffraction patterns were collected using the D2B powder diffractometer at the Institute Laue-Langevin (ILL) 1.051( 5  facility in Grenoble, France.Data were collected from powder samples within vanadium containers over an angular range of 0 < 2θ < 160°, using a 0.05°step size, with a neutron radiation of λ ≈ 1.594 Å.Data collection; i.e. counting time; varied between 2 hours for normal resolution and 5 hours for high resolution runs.High temperature runs were collected at various temperatures from room temperature up to 900 °C.All patterns were analysed and refined using the Rietveld method using the software package Fullprof (version 2.05).Diffraction peaks were refined using a pseudo-Voigt function       and the background was refined with a 6-parameter polynomial.The initial unit cell parameters were set using the findings obtained from previous XRD studies and the atomic positions for the different sites were set according to the space group symmetry. 11B-site (Ti) atomic coordinates were kept fixed at the origin; while the other sites were allowed to vary.Occupancies were initially allowed to vary through the refinement process yielding insignificant variations from the initial stoichiometries, later these were kept fixed to nominal values to obtain more stable refinements.All models were refined to convergence with the best fits chosen by the agreement factors and stability of the refinement profiles.

Room temperature studies
All patterns were successfully refined using the Rietveld method resulting in very good fits as shown in Fig. 1.A significant finding of this work by using neutron diffraction was the point where the transition from the tetragonal I4/ mcm to the orthorhombic Pbnm, see Table 1, which was previously reported to be at calcium content of x = 0.45. 7The new transition point now appears to take place at a calcium content of x = 0.4; i.e.La 0.2 Sr 0.3 Ca 0.4 TiO 3 .Hence, the updated phase map of this system is shown in Fig. 2.
With neutron diffraction, there was not a strong evidence of the existence of other intermediate phases like the ones reported for the system Ca 1−x Sr x TiO 3 ; e.g.Cmmm or Imma.Since the structure of the perovskite unit cell is highly affected by the A-site ionic radius mismatch, we believe that since our system incorporates a fixed stoichiometry of lanthanum; the variation of Sr/Ca ratio does not induce a severe disruption to the A-site lattice points.Thus, as it was evident from the absence of superlattice reflections in our neutron diffraction patterns that are indicative of the mentioned space groups, the system La 0.2 Sr 0.7−x Ca x TiO 3 undergoes a single first order phase transition with increasing calcium content; i.e. from I4/mcm to Pbnm symmetries.
Changes in symmetry in perovskites are manifested in changes to the tilt system of the BO 6 octahedra, which was apparent in this study.These tilt systems were analysed by having the O2 atomic coordinates as , hence, the anti-phase tilt angle along the [001] direction in the I4/ mcm phase is equal to φ = tan −1 4u.The out-of-phase tilt angles along the [100] and [010] directions in the Pbnm phase were calculated using the relation φ ¼ tan À1 4 ffiffiffi 2 p u. [11][12][13] These tilts have the Glazer notation a 0 a 0 c − and a − a − c + in the I4/mcm and Pbnm space groups, respectively. 14The calculated tilt angles from the refined atomic positions for the system studied here are plotted against composition in Fig. 3.This result shows that our system distortions are less severe than the parent perovskite CaTiO 3 ; 12 which can be attributed to a lower A-site ionic radius mismatch, due to the fixed stoichiometry of lanthanum throughout the range of the studied compositions.This is more apparent from the almost linear change of the anti-phase tilt angle with calcium content; indicating a more stable system and re-confirms that intermediate phases are non-existing in the compositions studied here.
Calcium introduction into SrCrO 3 resulted in a similar behaviour to that of the system La 0.2 Sr 0.7−x Ca x TiO 3 , in terms of structural transitions.With more calcium it was found that transitions occurred from Pm3 ˉm to I4/mcm to Pbnm. 15

High temperature studies
Two compositions; i.e.La 0.2 Sr 0.25 Ca 0.45 TiO 3 and La 0.2 Sr 0.2 Ca 0.5 -TiO 3 ; were studied at high temperatures with the results of Rietveld refinement listed in Tables 2 and 3, respectively.
Symmetry changes were evident from the diffraction patterns as shown in Fig. 4, where both compositions showed a similar behaviour as they were orthorhombic Pbnm at room temperature and evolved to the ideal cubic Pm3 ˉm symmetry around 900 °C through an intermediate tetragonal I4/mcm phase.
Fig. 5 and 6 show the behaviour of the lattice parameters at high temperatures of La 0.2 Sr 0.25 Ca 0.45 TiO 3 and La 0.2 Sr 0.2 Ca 0.5- TiO 3 , respectively.These lattice parameters can be seen to discontinuously change at the Pbnm-I4/mcm transition point, where at the I4/mcm-Pm3 ˉm transition point, the lattice parameters are changing in a continuous fashion as the samples were heated.This suggests that the first transition is a first order one while the latter can be a second order or a higher phase transition.Another aspect which was observed is that this system shows the high symmetry space group Pm3 ˉm at a much lower temperature compared to that of the parent compound CaTiO 3 , which according to Ali and Yashima 16 was found at temperatures over 1647 K.This shows that the degree of the distortion existing in La 0.2 Sr 0.7−x Ca x TiO 3 are much lower than those in CaTiO 3 ; which also explains the less severe distortions compared to Sr 1−x Ca x TiO 3 as was discussed in previous publication, 7 where the A-site ionic radius mismatch is greatly affected by increasing calcium content.
Comparing between the two compositions behaviour at high temperatures, not much differences can be seen structurally.As mentioned earlier, a slight drop in electrical conductivity was seen between the two compositions at 900 °C in reducing conditions.The results here have shown a significant difference in the isotropic atomic displacement parameters between the two samples as shown in Tables 2 and 3. La 0.2 Sr 0.2 Ca 0.5 TiO 3 showed slightly higher values at 900 °C compared to that of La 0.2 Sr 0.25 Ca 0.45 TiO 3 especially on the titanium site.This indicates that at this temperature, the titanium atoms are less stable; i.e. there is mort short range disorder; on their respective atomic site within the perovskite lattice.This must translates to the extent which the conducting d-orbitals overlap; which lowers the overall electrical conductivity of this oxide.Hence, La 0.2 Sr 0.25 Ca 0.45 TiO 3 was chosen as the optimal candidate of this series for a new type of perovskite  based anodes.As an anode backbone, this material is showing very encouraging performances upon further improvements. 22

Conclusions
The perovskite La 0.2 Sr 0.7−x Ca x TiO 3 showed structural changes with composition and with temperature.Phase pure samples showed a drop in symmetry from the ideal cubic Pm3 ˉm of La 0.2 Sr 0.7 TiO 3 to the tetragonal I4/mcm phase for 0.1 ≤ x ≤ 0.35 and the orthorhombic Pbnm phase for samples with 0.4 ≤ x ≤ 0.7.Orthorhombic samples showed transition to higher symmetries with increasing temperature.These transitions were a discontinuous Orthorhombic Pbnm-Tetragonal I4/mcm transitions and a continuous Tetragonal I4/mcm-Cubic Pm3 ˉm phase transition.These studies have helped greatly in understanding some earlier findings regarding the performance of this material as an anode material for SOFCs.

Fig. 1
Fig. 1 Rietveld refinement patterns for neutron diffraction data of different compositions of La 0.2 Sr 0.7−x Ca x TiO 3 at ambient conditions fitted to the indicated symmetries.

Fig. 2
Fig. 2 Phase diagram of the ternary system CaTiO 3 -La 2/3 TiO 3 -SrTiO 3 .All numbers represent calcium content.Solid symbols represent the samples studied in this work and analysed using NPD data; where shaded ones represent compositions obtained from literature.17-21 ◆ refers to the tetragonal samples and • refers to the orthorhombic samples.
Fig. 2 Phase diagram of the ternary system CaTiO 3 -La 2/3 TiO 3 -SrTiO 3 .All numbers represent calcium content.Solid symbols represent the samples studied in this work and analysed using NPD data; where shaded ones represent compositions obtained from literature.17-21 ◆ refers to the tetragonal samples and • refers to the orthorhombic samples.
Fig. 2 Phase diagram of the ternary system CaTiO 3 -La 2/3 TiO 3 -SrTiO 3 .All numbers represent calcium content.Solid symbols represent the samples studied in this work and analysed using NPD data; where shaded ones represent compositions obtained from literature.17-21 ◆ refers to the tetragonal samples and • refers to the orthorhombic samples.

Fig. 3
Fig. 3 The out-of-phase tilt angle values for the different compositions of La 0.2 Sr 0.7−x Ca x TiO 3 at room temperature.

Table 3
Lattice parameters, atomic positions, thermal parameters and R-factors from Rietveld refinement results of NPD data for a reduced sample of La 0.2 Sr 0.2 Ca 0.5 TiO 3 at various temperatures

Fig. 4
Fig. 4 Rietveld refinement of NPD patterns of a reduced sample of La 0.2 Sr 0.25 Ca 0.45 TiO 3 at different temperatures showing the different symmetries.Note: pattern at 300 °C was collected with shorter counting time compared to the other two.

Fig. 5
Fig. 5 Reduced lattice parameters of a reduced sample of La 0.2 Sr 0.25- Ca 0.45 TiO 3 plotted against temperature.

Table 1
Lattice parameters, atomic positions, thermal parameters and R-factors from Rietveld refinement results of NPD data for the different samples of La 0.2 Sr 0.7−x Ca x TiO 3 at ambient conditions

Table 2
Lattice parameters, atomic positions, thermal parameters and R-factors from Rietveld refinement results of NPD data for a reduced sample of La 0.2 Sr 0.25 Ca 0.45 TiO 3 at various temperatures