On the Factors Influencing the Stability of Phases in the Multiferroic System BiFeO3-PbTiO3
Rhombohedral perovskite BiFeO3 is a single phase multiferroic compound exhibiting both magnetic (Neel temperature ~370˚C) and ferroelectric (Curie point ~840˚C) ordering well above the room temperature. Ferroelectricity in BiFeO3 is due to stereochemically active 6slone pair in Biion which causes large relative displacements of Bi and O ions along the  direction. Long range spiral modulation of the canted antiferromagnetic spin arrangement in Feeffectively cancels the macroscopic magnetization due to Dzyaloshinskii–Moriya interaction and thereby prevents linear magneto-electric effect. Synthesizing dense pure BiFeO3 by conventional solid state method is difficult due to the formation of thermodynamically stable secondary phases such as Bi2Fe4O9, Bi25FeO39 and Bi46Fe2O72. To stabilize the perovskite phase and to suppress the cycloid several groups have adopted different strategies such as thin film growth, different synthesis methods and chemical substitution. Of the various substitutions reported in the literature, PbTiO3 substitution has shown very interesting features, such as (i) unusually large tetragonality (c/a~1.19), (ii) formation of morphotropic phase boundary (MPB) and (iii) high curie point Tc~650C. MPB ferroelectric systems such as lead zirconate titanate (PZT) are known to exhibit high piezoelectric response due to the coupling between strain and polarization. Hence the existence of magnetic ordering in BiFeO3-PbTiO3 offers an interesting scenario where polarization, strain and magnetization may couple together. The high Curie point also makes the system an interesting candidate for high temperature piezoelectric application. However its potential as a high temperature piezoelectric material has not been realized yet. A detailed review of literature suggests a lack of clear agreement with regards to the composition range of the reported MPB itself. Different research groups have reported different composition range of MPB for this system even for almost similar synthesis conditions. The present thesis deals with broadly two parts, firstly with the preparation of pure BiFeO3 by co-precipitation and hydrothermal methods and its thermal stability and secondly resolving the cause of discrepancy in range of MPB reported in BiFeO3-PbTiO3 solid solution. Detailed examination of this system (BiFeO3-PbTiO3) around the reported MPB composition by temperature dependent X-ray, electron and neutron diffraction techniques, in conjunction with a systematic correlation of sintering temperature and time with microstructural and phase formation behavior revealed the fact that the formation of MPB or the single ferroelectric phase is critically dependent on the grain size. This phenomenon is also intimately related to the abnormal grain growth in this system. Chapter 1 gives the brief overview of the literature on the topics relevant to the present study. The literature survey starts with a brief introduction about the perovskite oxides; their ferroelectric, magnetic and multiferroic properties were discussed in further sections. A brief outline on the grain growth mechanism is described. An overview of BiFeO3 and various synthesis methods, different chemical substitutions and their effect on properties are provided. A brief review of published literature on BiFeO3-PbTiO3 solid solution and its properties is also presented. Chapter 2 deals with the synthesis of pure BiFeO3, heat treatment and characterisation. BiFeO3 was synthesised by (a) co-precipitation and (b) hydrothermal methods. In co-precipitation method, calcination of precipitate at different temperature resulted in the formation of BiFeO3 along with secondary phases (Bi2Fe4O9 and Bi24FeO39). The optimum calcination temperature to prepare pure BiFeO3 was found to be 560C. The synthesized pure BiFeO3 exhibits weak ferromagnetic hysteresis at room temperature, the degree of which increases slightly at 10K (-263C). The hydrothermal treatment was carried out in (a) carbonate and (b) hydroxide precipitates with KOH as mineralizer. BiFeO3 prepared using hydroxide precipitate was stable till 800C whereas with carbonate precipitate it was stable only till 600C. Chapter 3 deals with the stability of phases in (1-x)BiFeO3 -(x)PbTiO3 solid solution. Samples prepared by conventional solid state route sometimes remain as dense pellet and on certain occasions it disintegrate completely into powder observed after sintering. Irrespective of the composition, sintering time and temperature, powder X-ray Diffraction (XRD) pattern of the survived pellet (crushed into powder) shows coexistence of rhombohedral (R3c) and tetragonal (P4mm) phases and the disintegrated powder (without crushing) show 100% tetragonal (P4mm) phase. Very high spontaneous tetragonal strain (c/a-1) ~0.19 at MPB is believed to be the origin for disintegration. But in all the survived pellets at least a minor fraction of rhombohedral phase (5-7%) is present. Systematic sintering studies with the time and temperature shows, decreasing the sintering temperature and time will increase the lifetime of the pellet and by increasing the sintering temperature and time the pellet will disintegrate. In this work we have conclusively proved that the wide composition range of MPB reported in the literature is due to kinetic arrest of the metastable rhombohedral phase and that if sufficient temperature and time is given, the metastable phase disappears. The suppression/formation of minor rhombohedral phase is expected due to the play of local kinetic factors during the transformation process. This makes the system behave in an unpredictable way with regard to the fraction of rhombohedral phase that is observed at room temperature. A systematic X-ray and neutron powder diffraction study of the giant tetragonality multiferroic (1-x)BiFeO3 -(x)PbTiO3 have shown that the compositions close to the morphotropic phase boundary of this system present two different structural phase transition scenarios on cooling from the cubic phase: (i) Pm3m P4mm(T2)+P4mm(T1) P4mm (T1) and (ii) Pm3m P4mm(T2) + P4mm(T1) + R3c P4mm (T1) + R3c. The comparatively larger tetragonality of the T1 phase as compared to the coexisting isostructural T2 phase is shown to be a result of significantly greater degree of overlap of the Pb/Bi-6s and Ti/Fe-3d with the O-2p orbitals as compared to that in the T2 phase. High temperature electron diffraction studies show that the metastable rhombohedral phase is present in the cubic matrix well above the Curie point as nuclei. Life time of the metastable R3c nuclei is very sensitive to composition and temperature, and nearly diverges at x → 0.27. MPB like state appears only if the system is cooled before the metastable R3c nuclei could vanish. Issue of the metastable rhombohedral state is developed further in Chapter 4. A one-to-one correlation was found between the grain size and phase formation behavior. Fine grained (~1µm) microstructure (usually pellets) shows phase coexistence (R3c+P4mm) and the disintegrated coarse grains (~10µm) show tetragonal (P4mm) phase. Microstructural analysis revealed the disintegration was caused by abnormal grain growth along with the disappearance of metastable rhombohedral phase. Abnormal grain growth starts at the periphery/crack i.e., at the free surface and move towards the canter of the pellet. Size reduction of disintegrated coarse grains (~10µm) to fine grains (~1µm) by crushing the sample showed that the system switching form pure tetragonal (P4mm) state to the MPB state comprising of tetragonal and rhombohedral phases (R3c+P4mm). In another approach the smaller sized particles of x=0.20 were synthesized by sol gel method. It was reported that in conventional solid state route x=0.20 exhibits pure rhombohedral phase. The sol-gel sample calcined at 500C (particle size ~15nm) stabilizes tetragonal metastable phase along with the stable rhombohedral phase, the morphotropic phase boundary state. Samples calcined at higher temperature, 800C (particle size ~50nm) also showed stable rhombohedral phase. Ferromagnetic behavior was observed in the sample having phase coexistence and the sample with pure rhombohedral phase showed antiferromagnetic behavior. Hence this material is a promising candidate which can be tuned to exhibit different behavior just by adopting different grain size. Chapter 5 deals with the magnetic structure of (1-x)BiFeO3 -xPbTiO3 solid solution with change in composition and temperature. Magnetic structure was studied using powder neutron diffraction in the composition range x=0.05 -0.35. Rietveld analysis was carried out for the nuclear and magnetic phases, by considering R3c phase for the nuclear structure. To account for the magnetic Bragg peak at d=4.59Å, three antiferromagnetic models were considered for the magnetic structure: (i) helical spin arrangement as in BiFeO3, (ii) commensurate G-type antiferromagnetic ordering with moments in the a-b plane (of the hexagonal cell), and (iii) commensurate G-type ordering with moments parallel to the c-axis (of the hexagonal cell). The third model was found to be suitable to explain the magnetic peak accurately and the better fitting of magnetic peak was observed in this model compared to others. At room temperature the MPB compositions have rhombohedral and tetragonal nuclear phases along with the rhombohedral magnetic phase. Addition of PbTiO3 in BiFeO3 not only changes the magnetic structure but also reduces the magnetic moment due to the substitution of Ti in Fesite. High temperature neutron diffraction studies reveal the magnetic transition at ~300C for x=0.20, ~95C for x=0.27 and ~150C for x=0.35. The Neel temperature observed in neutron diffraction studies were also confirmed by DSC and by temperature dependent dielectric studies. For x=0.20, anomalous variation in the lattice parameters and the octahedral tilt angle was observed across the magnetic transition temperature. In the magnetic phase, the c-parameter was contracted and the octahedral tilt angle slightly increased. This result suggests a coupling between spin, lattice and structural degrees of freedom around the transition temperature. Temperature dependent powder neutron diffraction study at low temperature from 300K (27C) to 4K (-269C) in x=0.35 shows the evolution of tetragonal magnetic phase at 200K (-73C) whose intensity is increasing with decrease in temperature. Below 200K, x=0.35 has rhombohedral and tetragonal magnetic and nuclear phases. While in x=0.27 at low temperature, rhombohedral magnetic and nuclear phases are present along with the tetragonal nuclear phase alone (the tetragonal magnetic phase is absent). We propose this discrepancy in the Neel temperature and the magnetic phase formation can be due to the probabilistic nature of the existence of metastable rhombohedral phase which was discussed earlier.