Perovskite Related Oxides: Development Of New Synthetic Methods, Materials And Properties
Abstract
Oxides of ABO3 composition (A = alkali, alkaline earth or rare earth metal in general, B = transition metal) constitute a large family of metal oxides of current interest to solid state and materials chemistry. Among the several structure types exhibited by ABO3 oxides (ilmenite, LiNbO3, perovskite, YAIO3/YMnO3, KSbO3, pyrochlore, among others), the perovskite structure is probably the most well known and widely investigated. The ideal perovskite structure consists of a three-dimensional (3D) framework of corner-sharing BO6 octahedra in which the A cation resides in the dodecahedral site surrounded by twelve oxide ions. The ideal cubic structure occurs when the Gold Schmidt’s tolerance factor, t = (rA + ro)/{V2 (rB + ro)}, adopts a value of unity and the A-O and B-O bond distances are perfectly matched. The flexibility of the perovskite structure towards a wide variety of substitutions at both A and B sites gives rise to a very large number (several hundreds) of perovskite derivatives with subtle variations in structure. The perovskite structure can also tolerate vacancies at both the A and O sites giving ordered superstructures. Members of y4BO3 oxides have numerous properties that find technological application, such as nonlinear optical response (LiNbO3), Ferro electricity (BaTiO3), piezoelectricity (PbZn_xTixO3), magneto ferroelectricity (YMnO3), superconductivity (Bai_xKxBi03)5 colossal magnetoresistance (La^xCaxMnO3) and ionic conductivity [(Lil_a)TiO3]
Ordering of cations at the A and B sites of the perovskite structure is an important phenomenon. Ordering of B site cations in double (/42BB'O6) and multiple (/43BB'2Og) perovskites gives rise to newer and interesting materials properties For example, 1*1 ordered Sr2FeMoO6 and Sr2FeReO6 are half-metallic ferrimagnets; Pb3MgNb2O9 is a relaxor ferroelectric; Ba3ZnTa2O9 is a low loss dielectric used in telecommunication and, last but not least, Ba3CoNb2O9 is a visible light driven photocatalyst. Realization of these properties in these materials depends crucially on the ordering/or otherwise of the B site cat ions in the perovskite structure. Furthermore, ordering of not only the metal atoms but also the oxygen/oxygen vacancies in the perovskite structure is equally important for the occurrence of superconductivity in the cuprate superconductor, YBa2Cu3O7.
The ideal perovskite structure gives way to hexagonal YMnO3/YAIO3 structure for smaller A cations (tolerance factor, t < 1). Oxides of this structure are attracting current attention for the realization of multiple magnetoferroic properties. On the other hand, for larger A cations (tolerance factor, t > 1), various perovskite polytypic structures are formed. For example, BaNiO3 forms a 2H polytypic structure, SrMnO3 and BaRuO3 adopts a 4H and 9R structures respectively, where the SO6 octahedra share faces or faces and corners.
Besides the foregoing 3D perovskites, a number of layered variants of the perovskite structure are also known. The most common layered perovskites are the Aurivillius phases, (Bi2O2)[A»-iBnO3n+iL the Ruddlesden-Popper phases, /4'2|7ln_iBnO3n+1], and the Dion-Jacobson phases, A[An^BnOzn+-\]' The two-dimensional (2D) perovskite unit, [^n-iBnOsn+i], which could be visualized as formed by slicing the 3D perovskite structure along <001>p is common for all the three layered perovskite series. The perovskite slabs are stacked alternately with various charge-balancing units, for example, with [Bi2O2]2+ in the Aurivillius phases and two alkali/alkaline earth cations (A+JA2+) in the Ruddlesden-Popper phases etc. Members of the layered perovskites are also important from the point of view of materials properties. For example, 2D magnetism (K2NiF4), superconductivity (La2-xSrxCuO4), ion exchange, Bronsted acidity, intercalation, exfoliation (K2La2Ti3Oio and CsCa2Nb3O10), photo catalysis (Rb2La2Ti30io) are some of the important materials properties found in layered perovskites. The high Tc-superconductors, Bi2Sr2CaCu2O8+XJ TI2Ba2Ca2Cu3Oi0, TIBa2Ca2Cu3O9 and HgBa2Ca2Cu3O8+x, also belong to the family of layered perovskites where the defective perovskite cuprate sheets are interleaved by other 2D entities like (Bi2O2), (TI2O2), (TIO) or (HgOx). In addition, Aurivillius phases, such as Bi2SrTa209 and Bi325Lao75Ti3Oi2, in thin film geometry are candidate materials for non-volatile ferroelectric memory devices.
Synthesis plays a key role in realizing new structures and materials properties for ABO3 oxides. The conventional synthetic methods (ceramic method) involve mixing and heating of solid reactants at elevated temperatures. Although this approach continues to be employed to synthesize new materials, it is often limited by the fact that it yields thermodynamically stable phases. Since many of the perovskite oxides showing useful materials properties are metastable in nature and are required in the form of fine particles (free-standing / monodisperse / submicron or nanometer dimensions) for application, the ceramic methods are of no avail for this purpose. Therefore, materials chemists constantly endeavor to develop alternate synthetic routes that enable them to synthesize novel oxides under mild conditions. Typical examples of metastable perovskites are: the super conducting cuprates (e.g. TlosPbosS^CaC^Og) and perovskite based lithium ion conductors (La2/3-xLi3XDi/3-2xTiO3). Also the control of oxidation states in double perovskites, such as Sr2FeMoO6 and Sr2FeRe06 and pyrochlores such as Pb2MnReC>6, cannot be achieved by conventional means. Therefore, the synthesis of such metastable phases requires special synthetic strategies that involve soft chemistry (chimie douce) methods where mild reactions/reaction conditions are employed to access metastable phases.
The present thesis is mainly devoted to an investigation of perovskite related oxides towards developing new synthetic strategies and materials as well as exploring hydrogen insertion - a novel materials property - in certain members of this family. Solid-state metathesis (SSM) reactions provide a convenient route for the synthesis of a wide variety of non-oxide ceramic materials such as, bondes, carbides, silicides, pnictides and chalcogenides. A typical metathesis reaction, for example,
M0CI5 + 5/2 Na2S -» MoS2 + 5 NaCI + 1/2 S (1) involves exchange of atoms/ions between the reactants and is accompanied by a large enthalpy change (AHm = - 890 kJ mol"1) and high adiabatic reaction temperature (Tm = 1413 °C). The reactions are often self-propagating and believed to be driven by the formation of stable salt byproducts such as alkali halides with high lattice energy. In our laboratory we have developed a different kind of metathesis reaction for the synthesis of perovskite related oxides, a typical example being,
K2La2Ti30io + 2 BiOCI -* [Bi2O2]La2Ti3O10 + 2 KCI.
A major difference between metathesis reactions (1) and (2) is that unlike (1), reaction (2) is not self-propagating, requiring longer duration. In this study, we have investigated metathesis reactions of the second kind at some length for the synthesis of perovskite related oxides. We found that rocksalt oxides such as UMO2 (M = Mn, Co) and Li2TiO3 constitute convenient precursors for the formation of v4BO3 perovskite oxides in metathesis reactions with appropriate reaction partners such as halides, oxyhalides or sulphates,
LiCoO2 + LaOCl -» LaCoO3 + LiCt (3)
LiMnO2 + LaOCl + x/2 O2 -> LaMnO3+x + LiCI (4)
Li2TiO3 + PbSO4 -» PbTiO3 + Li2SO4. (5)
We could synthesize not only well known ABO3 oxides but also functional perovskites such as PbZr0 4sTio 52O3 (PZT), La2/3Cai/3MnO3 as well as superconducting BaPbo75Bio2s03 by this method. We could also synthesize La2CuO4 and its superconducting analogues, La185^oi5Cu04 (A = Sr, Ba), by the same method using Li2CuO2 and LaOCl. For the synthesis of double perovskites A2BB%OQ by this method however, appropriate lithium containing rocksalt precursor oxides are not known in the literature. Therefore, we first synthesized rocksalt precursor oxides of the general formula Li4MWO6 (M = Mg, Mn, Fe, Ni) and established their identity. Using these precursor oxides, we could synthesize the double perovskite oxides Sr2MWO6 (M = Mg, Mn, Fe, Ni) in the metathesis reaction
Li4MWO6 + 2 SrCI2 -» Sr2MWO6 + 4 LiC
Significantly, the double perovskites are formed with an ordered structure at relatively low temperatures (750 - 800 °C) as compared to the high temperatures (up to 1400 °C) usually employed for the synthesis of these materials by conventional ceramic approach.
Next, we investigated ABO$ compositions corresponding to the formula for 6 = Cu and Ni, where we could obtain a YAIO3 superstructure consisting of triangular Cu clusters for 6 = Cu, whereas a perovskite phase for B = Ni. Moreover, the Cu-phase appears to be a unique line phase formed around LasCi^VOg composition, whereas a continuous series of GdFeO3-like perovskite oxides are formed for LaNii»xVxO3 (0 < x < 1/3)forS = Ni.
Considering the current interest in bringing different transition metal ions (d°/dn electronic configuration) in the same perovskite related structure towards developing multiferroic materials, we investigated the substitution of aliovalent cations in a typical Aurivillius phase, Bi2Sr2Nb2TiOi2. We have characterized new aliovalent cation substituted Aurivillius phases, Bi2SrNaNb2TaOi2, Bi2Sr2Nb2Zr012J Bi2Sr2Nb2 5Feo50i2 and Bi2Sr2Nb2 ezZno 33O12.
Lastly, we investigated the interaction of hydrogen with perovskite oxides, /\MnO3 (A = Ca, Sr, Ba) in an attempt to characterize possible existence of hydrogen-inserted oxide materials. An oxide-hydride of the formula LaSrCoO3H07 has recently been reported in the literature. Conventionally, the interaction of hydrogen with perovskite related oxides is known to result in either anion deficient phases (e.g. CaMnO3 -> Ca2Mn205), or hydrogen inserted materials, 'hydrogen bronzes', (e. g. HXWO3, HxBaRuO3), where hydrogen acts as an electron donor (H -^ H+ + e). We have characterized a new mode of hydrogen incorporation in Pt dispersed BaMnO3 and SrMnO3. Detailed investigation of the hydrogen sorption behaviour of 1 atom % Pt dispersed materials showed that about 1.25 mass % of hydrogen is inserted per mole of BaMnO3/Pt, corresponding to an insertion of - 3 hydrogen atoms giving 'BaMnOsHs'. While the exact nature of inserted hydrogen is yet to be established unambiguously, our results suggest that the inserted hydrogen is unlikely to be protonic (H+) in the hydrogen insertion product, BaMnO3H3.
The results of these investigations are presented in the thesis consisting of seven chapters. Chapter 1 gives an overview of perovskite related oxides - structure, properties and synthesis. Chapter 2 presents metathesis as a general route for the synthesis of ABO3 oxides and illustrates the method by transforming several rocksalt oxides such as LiCoO2, Li2Mn03 and Li2Ti03 to corresponding ABO3 oxides, LaCoO3, /\MnO3 and ATiO3 (A = Ca, Sr, Ba). Uniformly in all the cases, the perovskite oxides are obtained in the form of loosely connected submicron sized particles at considerably lower temperatures than those usually employed for their synthesis by ceramic methods. Thermodynamic calculations have also been carried out to probe into the driving force of metathesis reactions involved in the synthesis.
Chapter 3 describes an extension of the metathesis route for the synthesis of double perovskites, Sr2MWO6 (M = Mg, Mn, Fe, Ni). For this purpose, first we synthesized new rocksalt oxides of the general formula, Li4MWO6 (M = Mg, Mn, Fe, Ni). The oxides adopt rocksalt superstructures related to Li4MgReO6 (for M = Mg, Mn, Ni) and U4WO5 (for M = Fe). Metathesis reaction between Li4MWO6 and SrCi2 at 750 - 800 °C yields the corresponding double perovskites where the octahedral site M and W are ordered in the long range. Formation of ordered perovskite oxides at relatively low temperatures (750 - 800 °C) by the metathesis route is a significant result, considering that synthesis of these oxides by conventional ceramic method requires much higher temperatures (1300 - 1400 °C) and prolonged annealing.
Synthesis of La2CuO4, Nd2CuO4 and super conducting La-j 85>4oi5Cu04 (A = Sr, Ba) by the metathesis route is described in Chapter 4.
Chapter 5 deals with synthesis, structure and magnetic properties of mixed-metal oxides of ABO3 composition in the La-6-V-O (6 = Ni, Cu) systems. While the B = Ni oxides adopt GdFeO3-like perovskite structure containing disordered nickel and vanadium at the octahedral B site, La3Cu2VO9 crystallizes in a YAIO3-type structure. A detailed investigation of the superstructure of nominal La3Cu2VO9 by WDS analysis and
Rietveld refinement of powder XRD data reveals that the likely composition of the phase is Lai3Cu9V4O38 5, where the Cu and V atoms are ordered in a Vi3ah (ah = hexagonal a parameter of YAlCMike subcell) superstructure. Magnetic susceptibility data support the proposed superstructure consisting of triangular Cu3 clusters. The present work reveals the contrasting behaviour of La-Cu-V-O and La-Ni-V-0 systems, while a unique line-phase related to YAIO3 structure is formed around La3Cu2VO9 composition in the copper system, a continuous series of perovskite-GdFeO3 solid solutions, LaNi1.0CVxO3 for 0 < x < 1/3 seems to obtain in the nickel system.
The chapter also describes the formation of a new transparent Cu(l) oxide, Lai4V6CuO365, and its characterization. This oxide was obtained during attempts to grow single crystals of LasC^VOg. Single crystal structure determination of Lai4V6CuO36 5 showed that the structure contains isolated VO43" tetrahedra and [OCuO]3" sticks dispersed in a lanthanum oxide network. Films of Lai4V6CuO36 5 were grown on R-plane sapphire by using pulsed laser deposition. Rutherford backscattering spectroscopic and X-ray diffraction analyses of the films showed oriented growth of the title phase, with an optical band gap of -~ 5 eV and n-type conductivity
Chapter 6 presents the work on the flexibility of the Aurivillius structures for substitution of aliovalent/isovalent cations at both A and 6 sites of the perovskite slabs. For example, in a typical n = 3 member, Bi2Sr2Nb2TiOi2, substitution of both Sr and Na at the A site and Ta at the B site has enabled us to synthesize a new n = 3 member, Bi2SrNaNb2Ta0i2, where we see a preference of Nb for the terminal octahedral sheets. Similarly, aliovalent substitution only at the B site of the perovskite slabs of Bi2Sr2Nb2TiOi2 has yielded new members for specific compositions, Bi2Sr2Nb2ZrOi2, Bi2Sr2Nb2 5Feo50i2 and Bi2Sr2Nb2 67Zno33012 that tend to be oxygen-stoichiometric. The latter phases again show a preference of Nb for the terminal octahedral sites that are strongly distorted as compared to the middle octahedral site. This chapter also describes substitution of La3+ for Bi3+ in the perovskite slabs of Bi4Nb30i5 stabilizing a new series of n = 1/ n = 2 intergrowth Aurivillius phases of the formulas, Bi4LnNb3Oi5 (Ln = La, Pr, Nd) and Bi4LaTa30i5. The present work suggests that replacement of Bi3+: 6s2 lone pair ion by non-6s2 cations such as Sr2"* and La3+ in the perovskite slabs of Aurivillius phases tends to render the structure Centro symmetric and the materials lose NLOSHG response.
Chapter 7 describes our investigation of the interaction of hydrogen with alkaline earth manganites (IV) >AMnO3 (>A = Ca, Sr, Ba) dispersed with 1 atom % Pt. The result shows an unprecedented uptake of hydrogen by BaMnO3/Pt to the extent of - 1.25 mass % at moderate temperatures (190 - 260 °C) and ambient pressure. Gravimetric sorption isotherms and mass spectrometric analysis of the desorption products indicate that approximately three hydrogen atoms per mole of BaMnCVPt is inserted reversibly. The nature of hydrogen in the insertion product, BaMnO3H3, is discussed in the light of the structure of BaMnC>3.
The work presented in the thesis is carried out by the candidate as a part of the Ph. D. training programme and most of it has been published in the literature. He hopes that the studies reported here will constitute a worthwhile contribution to the materials chemistry of ABO3 oxides in general.