Dielectric Properties BaTiO3-δ

1 Dielectric Properties of BaTiO 3- A Thesis submitted to Indian Institute of Science Education and Research Kolkata in partial Fulfilment for the Requirements of the Degree of Master of Science In Physical Sciences by Monalisa Yadav 15MS168 Under the supervision of Dr.Barnali Ghosh(Saha) Co-supervisor: Prof Arup Kumar Raychaudhuri and Prof Bhavtosh Bansal Department of Physical Sciences Indian Institute of Science Education and Research Kolkata June, 2020 2 DECLARATION Date: 24 th June 2020 I, Ms. Monalisa Yadav Registration No. 15MS168 dated 24 th June 2020, a student of Department of Physical Sciences of the BS-MS Program of IISER Kolkata, hereby declare that this thesis is my own work and, to the best of my knowledge, it neither contains materials previously published or written by any other person, nor it has been submitted for any degree or any other academic award anywhere before. I also declare that all copyrighted material incorporated into this thesis is in compliance with the Indian Copyright (Amendment) Act, 2012 and that I have received written permission from the copyright owners for my use of their work Signature: Name: Monalisa Yadav Department of Physical Sciences Indian Institute of Science Education and Research Kolkata Mohanpur 741246, West Bengal, India 3 CERTIFICATE Date: 24 th June, 2020 This is to certify that the thesis entitled “Dielectric properties of BaTiO 3- ” submitted by Ms Monalisa Yadav Registration No. 15MS168 dated 24 th June 2020, a student of Department of Physical Sciences of the BS-MS Program of IISER Kolkata, is based upon her own research work under my supervision. This is also to certify that neither the thesis nor any part of it has been submitted for any degree or any other academic award anywhere before. In my opinion, the thesis fulfils the requirement for the award of the degree of Master of Science Signature: Name: Dr. Barnali Ghosh(Saha) Scientist-F Department of Condensed Matter Physics and Material Sciences S.N. Bose National Centre for Basic Sciences Kolkata 700106, West Bengal, India 4 ACKNOWLEDGEMENT I would like to take this opportunity to express my heartfelt gratitude to all those who helped me make my dissertation work a success. First and Foremost I express my utmost gratefulness to the almighty for the blessing showered on me throughout the study and providing me with the strength to do justice to my work and contribute my best to it, so that it turned out to be a successful venture. My deepest gratitude to my Co-Supervisor Prof Arup Kumar Raychaudhuri, Serb Distinguished Professor of Physics, CSIR-CGCRI, for his continuous encouragement, enthusiasm, ideas, valuable suggestions and comments, the efforts that he took to explain things in the simplest way possible and moral support in carrying out this dissertation work and guiding me through the project. I would also like to thank my supervisor Dr. Barnali Ghosh(Saha), Scientist-F, Department of Condensed Matter Physics and Material Sciences, S.N Bose National Centre for Basic Sciences, for her support, enthusiasm and guidance. I am grateful to Prof Bhavtosh Bansal, Department of Physical Sciences, IISER Kolkata for his valuable comments and generous help and all the respected teachers of Department of Physical Sciences, IISER Kolkata for their endless support and cooperation I am extremely thankful to Subhamita Sengupta, SRF Department of Condensed Matter Physics and Material Sciences, S.N. Bose National Centre for Basic Sciences for her help in doing my experiments, valuable suggestions and comments and her constant support throughout the project work. I am extremely thankful to my labmates Snehamoyee, Anirban, Vishal, Avisek, Chandan, Saikat, Paru, Sudipta, Dr. Ankita, Dr. Arun, Dr. Arnab, Soumya, Surojit, Soham, Ankita, Ravi and Shaili for their constant help and valuable suggestions. I am also grateful to XRD lab technician Dipayan, S.N Bose National Centre for Basic Sciences, Kolkata. I am extremely thankful to my parents, Masoom, Moonmoon, Ruchit and Aditya for their untiring support, love, care and attention. This journey would not have been easy without them. My special thanks to my friends Srishti, Sukanya and Kajal for their constant support and help. I would also like to express my thanks to all my classmates and seniors at IISER Kolkata. Monalisa Yadav 5 Abstract Dielectric Properties of BaTiO3- Barium titanate (BTO) has excellent electrical and optical properties. It is widely used as multilayer ceramic capacitors since it has a very high value of dielectric constant. BTO is a ferroelectric and undergoes many structural phase transitions. By changing the composition of the crystal structure or by doping or through introducing defects, it can be used either as insulators, semiconductors or even as metals. In spite of BTO being one of the most studied ceramics, many questions regarding the conduction mechanism in semiconducting BTO and experimental evidence of persistence of both metallicity and ferroelectricity in oxygen deficient BTO have remained unanswered. In this thesis, we have studied the properties of oxygen deficient Barium Titanate (BaTiO3-). We have studied the effect of oxygen vacancies and the effect of temperature variations on the electrical properties of BTO through Impedance Spectroscopy and have also tried to quantify the change in stoichiometry of pure BTO through Rietveld Refinement. 6 Contents Chapter 1 Introduction .......................................................................................................... 7 Chapter 2 Background and Overview .................................................................................... 8 2.1 Barium Titanate (BaTiO3) ............................................................................................ 8 2.1.1 Crystal Structure .................................................................................................... 8 2.1.2 Phase Transition of BaTiO3 ................................................................................. 10 2.1.3 Synthesis of BaTiO3 powder ................................................................................ 12 2.1.4 Synthesis of BaTiO3 pellet ................................................................................... 14 2.2 X-ray diffraction studies (XRD) and Rietveld Refinement of BaTiO3 ........................ 16 2.3 Dielectric Properties .................................................................................................. 21 2.4 Ferroelectrics ............................................................................................................. 23 2.5 Electrical properties of oxygen deficient BaTiO3 ....................................................... 24 Chapter 3 Objectives ........................................................................................................... 27 Chapter 4 Experimental Techniques .................................................................................... 28 4.1 Impedance Spectroscopy ............................................................................................ 28 4.2 Temperature dependent Impedance Spectroscopy ...................................................... 36 Chapter 5 Results ................................................................................................................ 39 5.1 Sample preparation of BaTiO3 pellet .......................................................................... 39 5.2 Sample preparation of oxygen deficient BaTiO3 powder and pellet ............................ 40 5.3 X-ray diffraction and Rietveld Refinement of pristine and oxygen deficient BaTiO3 .. 42 5.4 Temperature dependent Impedance Spectroscopy ...................................................... 46 5.4.1 Contact Optimization for BaTiO3 pellet ............................................................... 47 5.4.2 Dielectric Study of oxygen deficient BTO annealed at 900oC .............................. 48 Chapter 6 Conclusion .......................................................................................................... 55 References .......................................................................................................................... 56 7 Chapter 1 Introduction BaTiO3 (BTO) has been a widely studied ceramic due to its early discovery in the 1940’s and its simple crystal structure belonging to the perovskite family. BTO finds applications in numerous technological fields as a ferroelectric, dielectric, piezoelectric, pyroelectric and as a semiconductor. Besides its tremendous applications in industries, it shows interesting physics which includes the various structural phase transitions and the corresponding variations in electrical properties, the mechanism of conduction, the type of ferroelectricity etc. Many questions have been answered in the past, but still a few questions remain a topic of investigation. Till now it is not very clear, which mechanism is responsible for the conduction in oxygen deficient or electron doped BTO whether it is polaronic type or conduction band type? Recently, metal-insulator transition has been observed in oxygen deficient or electrostatic doped BTO and some papers have claimed the persistence of both ferroelectricity and metallicity in BTO. But the exact physics has still not been found that can support two contradictory mechanisms of localisation of polarised charges (ferroelectricity) and simultaneously the presence of itinerant electrons that are responsible for the metallic behaviour of Oxygen deficient BTO. In this thesis, we have studied the temperature dependent dielectric behaviour of both pristine and oxygen deficient BTO. We have tried to optimize the annealing conditions and also tried to quantify the deficiency of oxygen in BTO through Rietveld technique. This work is primarily experimental except the theoretical calculations in determining the quantity of oxygen deficiency through Rietveld Refinement. All theoretical results have been compared against experimental data whenever available. This brief introduction is followed by chapter 2 on literature review of BTO, the method of synthesis, X-ray diffraction, the Rietveld technique and the dielectric, ferroelectric, semiconducting and metallic properties. Chapter 3 discusses about the objectives of the thesis and the motivation behind the problem studied. Chapter 4 talks about all the experimental techniques used in this thesis. All results from sample preparation, to characterization, to temperature dependent Impedance Spectroscopy has been mentioned in chapter 5. Finally in chapter 6, conclusions have been summarized and future outlook has been mentioned. 8 Chapter 2 Background and Overview 2.1 Barium Titanate (BaTiO3) Barium Titanate (BaTiO3) is a ceramic belonging to the ABX3 (A and B are cations and X is an anion bonded to both the cations) perovskite family. It is the first ferroelectric ceramic discovered by Wainer and Salomon in 1942 [1].The material has interesting physics and undergoes numerous structural phase transitions. Above 120oC, BaTiO3 is stable in cubic structure. On lowering down the temperature below 120oC, it undergoes a transition from paraelectric cubic to ferroelectric tetragonal structure. On further lowering down the temperature, at 5oC, it transits to orthorhombic structure and below -90oC it is stable in the rhombohedral structure. Both the low temperature phases are ferroelectric in nature. BaTiO3 also has a high temperature transition to hexagonal structure at 1460oC [2]. BaTiO3 finds tremendous applications because it can be easily prepared, is chemically and mechanically very stable [3] and has excellent ferroelectric, dielectric, piezoelectric and pyroelectric properties. The method employed for preparation of the material has significant effect on the properties of the materials. The Curie temperature also fluctuates between 120oC to 130oC depending on the method used to make the material. The powder is mainly synthesized through traditional solid state method, various chemical methods and mechanochemical method discussed further in details in section 2.1.3. BaTiO3 (BTO) has high dielectric constant greater than 1000 and low dielectric loss [4]. Therefore, it is highly used as capacitors and in multilayer capacitors. Doped BaTiO3 is also used as semiconductors, positive thermal Coefficient (PTC) resistors [5]. BTO based ferroelectrics have high polarization, large permittivity and large strain which makes them excellent materials for use in piezoelectric transducers [6]. What makes BTO even more attractive for industrial applications is that, it is lead free and hence environment friendly [7]. . 2.1.1 Crystal Structure BTO has a perovskite structure with a general formula of ABX3, where A and B are metallic ions and X is a non-metallic ion. X can be F1-, Cl1-, O2but mostly it is the oxide anion. A is usually bigger than the B cation. A ions are monovalent, divalent or trivalent for example Na+, K+, Ca2+, Sr2+, Ba2+, Y3+ and B ions are pentavalent, tetravalent or trivalent respectively. The smaller cations are mostly Nb5+, Sn4+, Zr4+, Ti4+, Mn4+, In3+, and Ga3+. The undistorted, ideal perovskite structure is cubic in form with the B ion occupying the centre of the cube, the 9 A ions at the cube corners with the X anions at the centre of each face of the cube (figure 1a). The coordination number of A is 12 and that of B is 6. The structure can also be described as AX6 octahedron with B ion occupying the void of 8 such octahedra as shown in figure 1b. In case of BaTiO3, the Ba ions are large in size and the void formed by the BaO6 octahedra is too large as compared to the size of Ti ion. This causes the Ti ion to get off-centred towards the oxygen ions due to random thermal motion. When an electric field is applied, Ti gets aligned to the field direction from a random orientation and because valency of Ti ion is 4+, this causes large polarization and high dielectric constant. The interaction between the dipole moment leads to the formation of ferroelectric domains [8]. But the ideal structure is not very common and most of the perovskites found have a distorted structure with a reduced symmetry. It is this reduced symmetry which gives rise to many electric and magnetic properties that find great applications in industries. The distorted structure can be orthorhombic, rhombohedral, monoclinic, tetragonal or triclinic [9]. The distortion in the ideal structure is mainly caused by size effects, deviation from the ideal composition and the JahnTeller effect [10]. Many a times more than a single factor contributes to the distortion. In the cubic structure, the lattice parameter a can be written in the form of equation 1, 𝑎=(𝑟+𝑟 ). =2 (𝑟+𝑟) (1) The equation can be rearranged to be written in the form of equation 2, 𝑡= (𝑟+𝑟) (𝑟+𝑟). (2) where t is defined as the Goldschmidt’s tolerance factor. Deviation of t from ideal value of 1 gives rise to different crystal structures. But the size effect cannot be regarded as an exact description of the distortion since the perovskites are not purely ionic compounds and the t value depend on the value of ionic radius taken. Sometimes the B ions in ABX3 are Jahn-Teller active ions, for example in LaMnO3, the d orbitals of Mn ion undergo splitting into 3 tg and 1 eg levels which causes an elongation of the MnO6 octahedron. 16 After making green compact, it is usually sintered to make pellets with a higher density and greater strength. Sintering is a process where the powder particles fuse together to produce a dense end product. It is usually done by heating the green compact at a suitable temperature called the sintering temperature. The free energy of the final dense sintered pellet is much lower than that of pellet before sintering because fusion of particles causes the surface area to decrease and hence lower the free surface energy. The green density of the pellet and also the variation in the green density greatly influence the sintering behaviour and the end microstructure. The bulk density of the pellet increases at a faster rate on increasing the green density. Besides densification, grain growth also occurs during sintering. Sintering the pellet also helps in removal of binder and other organic materials which are present in the pellet either in the form of lubricants used or in the form of impurities. If the binder remains after sintering it can adversely affect the microstructure and produce internal defects in the pellet thereby altering its properties. All these processes lead to the shrinkage of a green compact into final dense pellets. 2.2 X-ray diffraction studies (XRD) and Rietveld Refinement of BaTiO3 Since years people have studied X-ray diffraction of various materials in order to determine their crystal structure. All the optical, electronic, magnetic, dielectric, ferroelectric properties depend on the atomic arrangement of atoms within the unit cell; therefore the characterization of the material becomes a really important part of any study on materials. Each substance is known to produce a unique diffraction profile. The final XRD profile is a combination of all the substances in the material and each substance in the mixture produces a pattern independent of others. The diffraction profile helps us to get both qualitative and quantative information about the material. The constructive interference of the diffracted rays from the planes of the lattice gives rise to the Bragg peaks. The condition for constructive interference is given by Bragg’s law (equation 4) 2𝑑𝑠𝑖𝑛  =𝑛  (4) 17 Where d is the distance between the lattice planes,  is the angle between the X-ray source and the lattice plane, n is any positive integer and  is the wavelength of the monochromatic X-ray source. The calculated intensity is given by the formula in equation 5 𝑌() =𝑠𝐿|𝐹 |  (2  −2  )𝑃𝐴+𝑦() (5)  where s is the scale factor, Lk contains the Lorentz, polarization and multiplicity factors,  is the reflection profile function that models both instrumental and sample effects, Pk is the preferred orientation function, A is the absorption function and Fhkl is the structure factor given by the formula (equation 6) 𝐹 =∑𝑓   𝑒𝑥𝑝 2𝜋𝑖(ℎ𝑥+𝑘𝑦+𝑙𝑧) (6) Where fn is the atomic structure factor and is a measure of the scattering strength of the atoms, h, k, l are the miller indices of the scattering plane, x, y, z is the position coordinates of the atom n and N is the total number of atoms in the unit cell. From the above equations, we can see that calculated intensity is directly proportional to the square of the structure factor. Therefore the unit cell gives us various information of the crystal structure like the distance between the planes, lattice parameters, the crystallite size, various phases formed and the position of atoms in the unit cell. In order to understand the crystal structure, numerous methods have been used like the Le Bail method, the Rietveld method etc. Rietveld refinement technique is widely used and is considered to be more accurate. This technique matches the experimental intensity data with that of the structural model given by the user by refining various parameters till a close match is reached. According to the Rietveld method, the model is refined by the least-squares fitting process in equation 7 𝑆=∑𝑤𝑦()– 𝑦() (7)  Where s is the residue or the difference between the calculated intensity and the experimental intensity, the steps are denoted by i, yobs (i) is the observed or the experimental intensity at step i, ycalc(i) is the calculated intensity of the model at position i and the weights wi is given by the formula (equation 8) 𝑤= () ()  (8) 18 where 𝜎() is the standard deviation of the peak and 𝜎() is the standard deviation associated with background intensity. A number of quantative and qualitative information can be gathered from the XRD profile and Rietveld refinement. The Rietveld method removes the background noise and corrects for various instrumental errors like zero shift, displacement error and error in the wavelength of the incident wave. The lattice parameters are calculated from the d-spacing by the following formula (equation 9) 1 𝑑  = 1 𝑉 [(ℎ𝑏𝑐)sin  + (𝑘𝑐𝑎)sin  + (𝑙𝑎𝑏)sin  )] (9) Where d is the distance between the lattice planes a, b, c are lattice distances in the x, y, z direction respectively and , ,  are the angles between x and y, y and z and x and z axis respectively. h, k, l are the miller indices and V is the volume of the unit cell given by the following formula in equation 10 𝑉=𝑎𝑏𝑐[1+2𝑐𝑜𝑠  𝑐𝑜𝑠  𝑐𝑜𝑠  − (𝑐𝑜𝑠 2  + 𝑐𝑜𝑠 2  + 𝑐𝑜𝑠 2  )]. (10) The Rietveld method allows us to fit the peak shape from a number of options available from the software like the Gaussian, Lorentzian, Pseudo-Voigt, Pearson VII etc, The most widely used function is the Pseudo-Voigt method which uses a combination of both Gaussian and Lorentzian functions. The Pseudo-Voigt pV(x) is given by (equation 11) 𝑝𝑉(𝑥)=  𝐿(𝑥)+(1−  )𝐺(𝑥) ( 11) where L(x) is the Cauchy Lorentzian function and G(x) is the Gaussian function and  is the mixing parameter given by (equation 12).  =  +𝑋2  ( 12) where X is Lorentzian isotropic strain parameter and both X and o are refinable parameters. Full width at half maxima (FWHM) is given by the following formula in equation 13 𝐹𝑊𝐻𝑀 = (𝑈𝑡𝑎𝑛 2  + 𝑉𝑡𝑎𝑛 2  + 𝑊)+   (13) 19 where U, V, W are refinable half-width parameters and IG is the gaussian isotropic size parameter and can be refined. FWHM is helpful in calculating the crystallite size D given by the well-known Debye Scherrer’s formula in equation 14 𝐷= 0.94  𝐹𝑊𝐻𝑀 𝑐𝑜𝑠  (14) where  is the wavelength of the monochromatic X-ray source and  is peak position. Using the Rietveld method we can calculate the occupancy (Occu) of the atom in the unit cell given by the following formula in equation 15 𝑂𝑐𝑐𝑢= 𝑐∗𝑚 𝑀 (15) Where c is the chemical occupancy of the ion in the unit cell, m is the Wyckoff multiplicity of the site and M is the general multiplicity of the space group. Occupation number is not very stable during refinement and should only be refined after having a very stable model for our experimental data. Rietveld method allows us to correct for the change in intensity caused by atomic displacements of atoms from their ground state position. Thermal motion in a lattice gives rise to vibrations in atoms which can lead to the attenuation of the X-ray wave scattering from the ions. The Debye-Waller factor accounts for this change and can be refined in the form of overall B factor, isotropic and anisotropic B factor. Lower the B value implies more ordered is the atom in the crystal and higher the B value of the ions makes the structure more flexible. Like Occupation number, B value is also very sensitive to refinement and is not very stable, therefore should be refined only after getting a close accurate model for the experimental data. Non-random orientation of crystallites can arise during crystallization, growth, sample preparation which can cause the grains to prefer certain crystallographic direction. This can cause variation in intensity of the peaks of the spectrum as there are either more or less peaks with certain orientations. To correct for the preferred orientation, Rietveld technique provides us two refinable parameters G1 and G2. The asymmetry parameter act as a multiplier to the peak shape. The asymmetry can be corrected with the help of four refinable parameters P1, P2, P3 and P4 which are all independent from each other. Their starting values should be very 20 less around 0.01 and should not be kept fixed during refinement as it can lead to bad refinement or divergence[25]. Although the Rietveld method allows for the correction for a number of parameters, one has to be careful while refining the parameters. Refining a large numbers of parameters at the same time can cause a singular matrix. A prior judgement should be made to check which parameters should be kept fixed and which should be allowed to vary. It is better to start the refinement with the parameters that are stable and independent like the scale factor, zero shift, instrumental displacement or incident wave wavelength error, lattice parameters, background points followed by peak shape and preferred orientation parameters. The thermal displacement parameters and occupancy parameters should only be refined after getting a very accurate, reliable and stable model for our data otherwise; refining these parameters can give highly unphysical values. Also the parameters which are correlated should be refined separately like the thermal parameters and occupancy parameters. The overall scale parameter and the occupation numbers are also highly correlated. During the refining process instead of being in the global minima, the model might get trapped in the local minima of the refinable parameter. This can lead to unphysical values of the parameters and wrong matching of the model with the experimental data. In order to check for the accuracy of our model or how close our model describes the actual experimental data, Rietveld technique provides a number of agreement factors [25] like the profile factor (Rp), weighted profile factor (Rwp), expected weighted profile factor (Rexp), Bragg factor (RB), Goodness of fit indicator (S), reduced chi-square (2) and crystallographic RF-Factor (RF) given by the following formulae (equations 16-21). 𝑅𝑝=100∑|𝑦−𝑦|   ∑𝑦   (16) Where yi is the observed intensity at point i and yci is the calculated intensity at point i and n is the total number of points 𝑅 =100󰇩∑𝑤|𝑦−𝑦|   ∑𝑤𝑦   󰇪  (17) 𝑅 =100󰇩 𝑛−𝑝 ∑𝑤𝑦  󰇪  (18) 21 Where p is the total number of refined parameters 𝑆= 𝑅 𝑅 (19) =𝑆 (20) 𝑅=100∑|𝐼 −𝐼| ∑|𝐼| (21) Where Iobs and Icalc are the integrated observed and integrated calculated intensities respectively. Rexp gives the value of best fitting of the experimental profile that can be possible. A lower Rexp is an indicator of a high quality data. Rp and Rwp describe how close the theoretical data is to the actual experimental data [26]. According to equation 20 and 21, S and  should ideally reach a value 1, in case of the best match obtained. Often this is not the case, but during refining we try to get as close to 1 as possible. RB is useful because its value depends only on the structural parameters and not on the profile parameters. Although XRD is extremely helpful in understanding the crystal structure of the material, one of its main disadvantages is that it cannot detect smaller atoms with atomic number less than 13 accurately. As X-ray rays interact with the electron cloud of the ions, these smaller ions with less number of electrons almost remain invisible to the rays. 2.3 Dielectric Properties Dielectrics are those materials that do not conduct electricity on the application of an Electric Field. The field causes shift in the centre of the positive and negative charges in the material, such that the material acquires an electric dipole moment. Dipole moment per unit volume is called the polarization. The polarization is always proportional to the applied electric field given by the following formula in equation 22. 𝑷=    𝑬 (22) Where P is the polarization,   is the dielectric permittivity (degree of polarization experienced by the dielectric under the influence of Electric field) in vacuum E is the applied electric field and   is the dielectric susceptibility (measure of the dielectric’s ability to form dipoles in response to the electric field) given by the following equation 23 22  =  − 1 (23) where   is the relative permittivity (ratio of dielectric permittivity in the medium and dielectric permittivity in vacuum. Value of dielectric constant in vacuum is 8.85 x 10-14 Farad/cm). However the molecule in the interior of the dielectric experiences an electric field different from that applied; therefore, the final polarization of the material depends on the local electric field and the polarizability of the molecule. At the microscopic level, polarization can be of many types given below-  Electronic polarization: This kind of polarization arises in the dielectric medium due to the displacement of the centre of positive charges from negative charges  Ionic Polarization: Materials that have ionic bonds in them, on application of the electric field, the distance between the positive and the negative ion further increases giving rise to a dipole moment.  Orientational Polarization: Some materials have permanent dipole moments but due to thermal motion they remain randomly oriented. On application of an external electric field, they orient in the direction of the field, given rise to a net dipole moment.  Spontaneous Polarization: In ferroelectric materials, dipole moment arises due to shifting of ion from the centre of the unit cell.  Space Charge or Interfacial polarization: This kind of polarization occurs due to accumulation of charges at the interface of two materials or between two regions within the same material or between the electrode and the dielectric medium on the application of an external electric field. The degree of charge storage in a dielectric medium is measured by its dielectric constant. When an electric field is applied between two metallic plates that have a dielectric medium between them, one of the plates become positively charged and the other becomes negatively charged. Thus the electric field causes polarization in the dielectric medium. Depending on the kind of polarization in the dielectric medium, different materials show variations in the dependence of dielectric constant on temperature. Electronic polarization is almost insensitive to temperature. In ionic polarization, with increase in temperature, number of charge carriers and mobility increases leading to an increase in the polarization value. In orientational polarization since the permanent dipoles are randomly oriented due to thermal motion; they show a greater polarization at low temperatures. Dielectric Constant also 23 depends on the frequency of the applied electric field. The polarization requires time to respond to the ac field with continuously changing direction. The frequency after which the dipoles cannot respond to the fast changing field is called the relaxation frequency. Electronic polarization can respond to high frequency fields while the rest of the polarization only respond at lower frequencies since molecules do not have enough time to reorient and align parallel to the fast changing fields. Materials that have low dielectric constant are used as electric insulators and those with high dielectric constant are used as capacitors [28]. Dielectrics do have a limit to which electric field can be applied. High electric field causes the material to break down and conduct electricity which destroys the purpose of the usage of the dielectric material. The maximum field that the material can sustain is called dielectric strength and varies with different materials. Old capacitors, impurities in the dielectric or very strong field can generate electrons in the conduction band which on applying high electric field accelerate very strongly, they can further transfer their high energies to valence electrons, thereby creating large number of conduction electrons which in the long run can cause the dielectric to burn or melt. A group of dielectrics can be classified as piezoelectric, materials that develop polarization on the application of stress. A sub group of piezoelectrics are pyroelectric materials that have polarization in one crystallographic direction. On the application of heat, the material expands leading to a mechanical deformation which increases the pre-existing polarization. A sub group of pyroelectrics are ferroelectrics which have been discussed in detail in the next section. 2.4 Ferroelectrics Some materials have spontaneous polarization without an application of an Electric field. The temperature at which these materials lose their spontaneous polarization is called the Curie temperature. Spontaneous polarization can be switched in the opposite direction on the application of a strong electric field. We know that in BTO, the shifting of the Ti ion from the centre of the TiO6 octahedra causes symmetry breaking and give rise to the dipole moment in the unit cell. Usually the thermal interaction dominates over the exchange interaction between the dipole moments and so we get randomly oriented domains with different polarization directions. But all the dipole moments point in the same direction within each domain. Figure 4 describes the Polarization versus electric field (P-E loop). On increasing the 24 electric field, the polarization increases as the domains start aligning parallel to the electric field till it reaches a saturation value (P s ) and the polarization does not change further even on increasing the electric field (path O-C). Once we start decreasing the electric field, the polarization starts decreasing as due to thermal motion few domains are able to orient along the electric field direction (path C-D). However even when the electric field is 0, the polarization is not zero as few domains remain oriented in the direction of applied field. This is called remnant polarization (P r ). The value of the field required for the polarization to become 0 is called the coercive field (E c ). On applying the field in the opposite direction causes the domains to switch direction leading to polarization in the opposite direction (path E-F). As we further increase the electric field the polarization also increases till it reaches a saturation value. Thus a hysteresis (dependence of the state of the system on its history) loop can be observed. Therefore at a particular value of the electric field, more than 1 possible value of polarization is possible depending on how the field changed in the past. Above the Curie temperature for tetragonalcubic phase transition the hysteresis loop disappears. Figure 4: Polarization versus Electric Field (P-E) loop showing the Saturation polarization (Ps), Remnant polarization (Pr), Coercive field (Ec) [28]. 2.5 Electrical properties of oxygen deficient BaTiO 3 Electrical properties of undoped BaTiO 3 (BTO) have been a topic of investigation since years. It is known that stoichiometric BTO loses some amount of oxygen on annealing at high temperature. The amount of oxygen loss depends on a number of factors like annealing 25 temperature, atmosphere, heating and cooling rate, microstructure and the pellet density [29]. Depending on the amount of oxygen loss, the ceramic can be insulating, semiconducting, metallic or can be electrically heterogeneous, with different regions corresponding to different conductivities. Partial or complete reoxidation can occur on cooling the sample from its annealing temperature which can lead to gradients in oxygen content and hence Ti3+ ions [30]. Losing a small amount of oxygen makes it conducting, but the exact mechanism of electronic conduction whether it is polaronic type [31] or conduction band mechanism type [32] has still been a question of debate. On annealing at a high temperature, oxygen ions diffuse out of the unit cells, leaving behind oxygen vacancies and electrons which can be expressed as 𝑂 ⟶𝑉 .. + 0.5 𝑂+2𝑒 𝐵𝑎𝑇𝑖𝑂 −𝑥𝑂=𝐵𝑎𝑇𝑖  𝑇𝑖 𝑂  where VO.. is the oxygen vacancy created. The electron is taken by the Ti4+ ion and small amount of it changes to Ti3+. The transfer of electrons from Ti4+ to Ti3+ brings about conduction in oxygen deficient BTO. On the other hand undoped BTO is an insulator with a band gap of around 3.0 eV. Few papers have reported the n-type of conductive behaviour in polycrystalline BaTiO3 reduced in vacuum [33] and single crystal BaTiO3 reduced in hydrogen atmosphere [34] through thermoelectric measurements. Eror and Smyth [35] reported p-type conductivity on annealing BTO single crystals between 800oC an 1200oC at high partial pressure of oxygen and n-type conductivity on annealing at low partial pressure of oxygen. Singly ionised Ba vacancies give rise to p-type conductivity and doubly ionised O vacancies give rise to n-type conductivity. Also, n-type conductivity shows stronger temperature dependence than p-type conductivity. BTO shows positive temperature coefficient of resistivity (PTCR) effect above ferroelectric phase transition and the resistivity increases to several orders of magnitude [36]. PTCR effect mostly arises due to potential barrier formation at the grain boundaries arising from oxygen absorption or cation vacancies like Ba [37, 38]. Weakly reduced BTO shows an anomalous PTCR effect but if the reduction is very strong, the PTCR effect becomes very poor or in some cases may not be observed at all [39]. 32 Figure 8: Complex part of impedance (Z” or X) has been plotted versus frequency. The circles represent the observed data points and fit represents the fitted curve of the experimental data. The peak gives the value of relaxation frequency beyond which the bound charges are no longer able to orient themeselves with the fast changing applied ac electric field. Graph of Z” versus Z’ is called nyquisit plot (shown in figure 9). Nyquisit plot gives us a lot of information about the properties of the material. It is very sensitive to changes and it is quite easy to understand as most of the parameters can be directly read from the curve. A semicircle is a representaive of a single time constant. But in almost all materials, mostly more than one semicircles are found corresponding to different relaxation mechanisms. The radius of the semicircle is equal to half the value of parallel resistance. The intercept of the semicircle towards higher frequency gives us the value of series resistance and the the intercept towards the lower frequency side gives us the value of sum of both series and parallel resistance. Each point in the Nyquisit plot also known as the Cole-Cole plot is a representative of a single frequency. Most of the times we do not get complete semicircles, instead only a portion of the semicircle is obtained. The semicircle equation can be derived in the following way 𝑍󰆒=𝑅𝑠 + 𝑅𝑝 1+ 𝜔𝑅𝑝𝐶𝑝 𝑍󰆒󰆒 = − 𝑅𝑝𝜔𝐶𝑝 1 + 𝜔𝑅𝑝𝐶𝑝 𝑍󰆒−𝑅𝑠+𝑅𝑝 2= 𝑅𝑝 1+𝜔𝑅𝑝𝐶𝑝− 𝑅𝑝 2 𝑍󰆒−𝑅𝑠+𝑅𝑝 2 = 𝑅𝑝 (1 + 𝜔𝑅𝑝𝐶𝑝)+ 𝑅𝑝 4−𝑅𝑝 (1 + 𝜔𝑅𝑝𝐶𝑝) 100 1000 10000 100000 1000000 0 200000 400000 600000 800000 Zimag fit X (  ) frequency (Hz) Equ y = 󶂢Adj 0.99995 Value Sta E R 1.66006E 506.15696 E C 1.12342E 5.92577E-1 33 𝑍󰆒󰆒  = 𝑅𝑝𝜔𝐶𝑝 (1 + 𝜔𝑅𝑝𝐶𝑝) 𝑍󰆒−𝑅𝑠+𝑅𝑝 2+(𝑍󰆒󰆒) = 𝑅𝑝 (1 + 𝜔𝑅𝑝𝐶𝑝)+ 𝑅𝑝 (1 + 𝜔𝑅𝑝𝐶𝑝)+ 𝑅𝑝 4− 𝑅𝑝 (1 + 𝜔𝑅𝑝𝐶𝑝) =𝑅𝑝 2 The above equation represents the equation of a semicircle with centre at 󰇡Rs+ 󰇢 and radius of   Figure 9: Z” (X) versus Z’ (Rs) also known as the Cole-Cole plot has been shown. The curve is a semicircle with each point describing a data at a particular frequency. The arrow shows the direction of increasing frequency. The circles are measured data and fit the curve fit of the experimental value. We studied some parallel RC circuits with different time constants mentioned in table 1. We calculated the theoretical time constant and the measured time constant by fitting the curves for the cole-cole plot. We can clearly see that as the frequency increases the time difference between theoretical and measured time constant decreases. A graph of time constant versus frequency has been shown in figure 10. The squares represent the theoretical time constant as 34 calculated from the value of resistors and capacitors taken and the Circles represent the measured time constants as calculated from the value of resistance and capacitance given by the LCR meter. Frequency (Hz) Theoretical 𝜏 (s) Measured 𝜏 (s) Error in 𝜏 (s) 70 Hz 2.2E-3 2.20E-3 0.07E-3 91 Hz 1.75E-3 1.86E-3 -0.11E-3 508 Hz 3.13E-4 3.19E-4 -0.06E-4 1000 Hz 1.59E-4 1.61E-4 -0.02E-4 96574 Hz 1.65E-6 1.42E-6 0.23E-6 682483 Hz 2.33E-7 2.30E-7 0.33E-7 984005 Hz 1.62E-7 1.62E-7 0.00 Table 1: Various Combinations of Resistors and capacitors have been taken with different time constants or corresponding frequency shown in first coloumn. The theoretical time constant (𝜏) has been calculated and shown in second coloumn. The experiemental 𝜏 has been shown in the third coloumn and difference in theoretical and experimental time constant has been shown in the last coloumn. 35 Figure 10: Plot of logarithm of time constant (𝜏) verus log of frequency (Hz) has been shown. The black squares show the theoretical calculated value of time constant and the red circles show the experimental time constant. The error bars shows the difference between the calculated and measured value. Impedance Spectroscopy can be used to study the dielectric behaviour of a material. The complex dielectric constant ( * ) is defined as below  ∗ =  󰆒+ 𝑗  󰆒󰆒 where ’ is the real part of the permittivity and ” is the imaginary part of the dielectric constant also known as the dielectric loss. We calculated the dielectric constant from the complex impedance (Z * ) using the following formula  ∗=1 𝑗𝜔𝐶𝑍∗ where C O is the capacitance without the sample. Hence the dielectric permittivity can be written as  󰆒= 𝑍󰆒󰆒 𝜔𝐶𝑍󰆒+𝑍󰆒󰆒 An the dielectric loss (Z”) can be written as 36  󰆒󰆒 = 𝑍󰆒 𝜔𝐶𝑍󰆒+𝑍󰆒󰆒 We can calculate the dielectric loss tangent (tan ) defined as tan  =  󰆒󰆒  󰆒 Using the dielectric parameters values, we can calculate the ac conductivity (σac ) defined as 𝜎 =  󰆒󰆒∗𝜔∗   where o is the dielectric permittivity of vacuum. Its value is 8.85 x 10-12 F/m We have defined the impedance and dielectric relations for a simple RC circuit. But usually most materials are heterogeneous and have more than one relaxation mechanism and have various contributions to dielectric data coming from different sources like the grain boundaries and grains. In that scenario we see two arcs in the complex impedance rather than one. The higher frequency arc represents the grain or the bulk effect and the lower frequency arc represent the grain boundary effect. In many cases, only a single arc will be observed if the resistance of the grain boundary dominates the grain resistance i.e. Rgb >> Rg . In the Z” versus frequency plot, we get two peaks corresponding to two relaxations of the grain boundaries and the grains. The higher frequency peak describes the relaxation of the bulk and the lower frequency peak describes the behaviour of the grain boundary. The capacitance of the grain boundary is also much higher than the capacitance of the grain since the thickness of the grain boundary is much smaller than the size of the grain. 4.2 Temperature dependent Impedance Spectroscopy To study the temperature dependent impedance behaviour of BTO, we have used Closed Cycle Refrigerator (CCR) that can cool samples down to 10K. The whole system consist of Lakeshore temperature controller, Rotary pump, Vacuum jacket, Radiation shield, Compressor and cryogen safe pipelines (shown in figure 11). A variety of electrical experiments can be done with temperature ranging from 10K to 415K. CCR uses Helium gas which is compressed cooled and then expanded. On expanding it takes away the heat from the sample mounting stage and returns back to the compressor where it is again compressed and cooled. CCR is safer and very easy to handle since it does not use liquid cryogens like Nitrogen or Helium. The only disadvantage is that it consumes a lot of power. 37 Figure 11: Temperature dependent Impedance spectroscopy set-up. 10K CCR, Compressor, gas lines and rotary pump have been shown in the figure A schematic of the CCR has been shown in figure 12 explaining the working of each part of the CCR. Helium gas goes from the compressor to the expander through one of the gas lines and comes back from the other. He-3 is used to cool down the temperature to 10K as it can remain in liquefied state almost till absolute 0 due to weak Vander Waal’s forces. This CCR does not require filling Helium again. Oil based Vacuum pump is used to remove unwanted gases and particles. It reduces pressure inside the apparatus which causes the gases to vaporize. Vacuum helps in creating a pressure gradient so that flow of cryogen can be controlled. Usually 10E-5 mbar is enough to start the compressor. The sample mounting stage is covered by a radiation shield which protects it from the thermal radiation given by the Vacuum Shroud. The radiation shield is covered by a Vacuum jacket which reduces the heat load on the expander caused by the conduction and convection from the instrumentation skirt. 38 Figure 12: Schematic of the inside of a Closed Cycle refrigerator (CCR) 39 Chapter 5 Results 5.1 Sample preparation of BaTiO3 pellet We have taken 0.2g of BaTiO3 powder to make our pellets. Density of BaTiO3 is 6.02 g/ cm3 and the radius of the pellets were 1cm in diameter which implies that the thickness of the pellet was small approximately around 0.3 mm. Such small thickness not only helps in forming homogenous pellet in the die with lesser density variation but also helps in reducing the coercive strength of the material. The smaller thickness also helps in diffusing out oxygen from the bulk and the surface which can decrease the resistance of the material. We have used polyvinyl alcohol (PVA) as binder. PVA is water soluble, has great binding strength, undergoes a glass transition around 85oC and boils off around 200oC to 300oC. We used 1% PVA dissolved in water because higher the water content the glass transition temperature of PVA decreases and helps in binding the particles in the compact strongly. Binder was mixed with BaTiO3 powder and crushed and grounded in a mortar and pestle to form homogenous agglomerates. We used hydraulic press to provide the pressure and a uniaxial die to form the pellet. We applied a pressure of 60 kN/mm2, the effective pressure on the powder is less because of the friction present between the walls. We used paraffin oil as a lubricant to reduce the friction and to form homogenous pellet with lesser density variation. The boiling point of the paraffin oil is around 300oC, which is lower than the temperature at which the pellet is sintered. After forming the green compact, we again crushed the pellet and ground the powder. The step was repeated so as to decrease the density variation and increase the homogeneity of the material. Finally, we sintered the pellet in the furnace in air at a temperature of 600oC and maintained the temperature for 2 hours. At such high temperature, the binder, lubricants, organic and volatile impurities boil off leading to the formation of a dense pellet with greater strength. Sintering helps in fusion and compaction of the particles by reducing the surface area and hence the surface energy. The melting point of BaTiO3 is around 1600oC so sintering should be done at a lower temperature because too high temperature and longer sintering time can lead to undesirable grain growth. 40 5.2 Sample preparation of oxygen deficient BaTiO 3 powder and pellet The major motivation of our thesis is to understand the characteristics, dielectric and transport properties of oxygen deficient BaTiO 3 . In many papers it has been reported that heating the ceramic oxide in air or vacuum causes the material to become oxygen deficient [42-43]. Vacuum annealing creates oxygen vacancy and helps in removal of chemisorbed oxygen from the surface and grain boundary [43]. Annealing at a higher temperature activates diffusion process of oxygen and desorption of oxygen from the grains and grain boundaries respectively. We annealed the pristine BaTiO 3 powder and the pristine BaTiO 3 pellet in vacuum at temperature of 800 o C and 900 o C for 3 hours. But the annealed pellet/powder did not show a change in colour or appeared darker. When compared to the X-ray diffraction (XRD) data of the pristine BaTiO 3 sample, the oxygen deficient sample showed no change in peaks or peak positions or the occupancy of the ions in the unit cell calculated through Reitveld Refinement (Figure 13). Figure 13: XRD data of pristine BTO and vacuum annealed BTO at 800oC and 900oC. The inset shows the peak (111). We clearly see that there is no peak shift between pristine and oxygen deficient BTO data. 41 Therefore, we next tried annealing the pristine sample in reducing atmosphere of 95% Ar and 5% H 2 in a tube at temperature of 900 o C and 1000 o C for 3 hours. Since H 2 is a combustible gas, Ar is added for stability. We further studied the XRD of the oxygen deficient sample and compared it with BaTiO 3 pristine pellet (figure 14). We find that there is a peak shift towards higher angle in oxygen deficient sample. According to Bragg’s law (equation 4), increase in angular peak position implies a decrease in the interplanar distance d, therefore we can say that there are structural changes in the lattice and the lattice constant decreases. The decrease in lattice constants can be due to oxygen vacancies created [44] or it can be due to barium vacancy or A-site defect in the lattice [45].The details of XRD study along with Reitveld refinement have been mentioned in section Figure 14: XRD data for pristine BTO and oxygen deficient BTO annealed at temperatures of 900oC and 1000oC in Ar/H2 atmosphere. The inset shows the (111) peak. There is clearly a marked shift in the peak position of pristine and oxygen deficient sample. 48 metallic contacts. But the Z’’ or X (imaginary part of complex impedance) peak for pristine BTO was shifted towards the lower frequency side and therefore not visible. Since we wanted to study the relaxation mechanism and find the relaxation frequency, we chose Cr/Au as electrical contacts for our pellets. Pristine BTO with Cr/Au contact has both lower resistance and visible relaxation peak. Figure 18: Complex Impedance study of pristine BTO with different metallic Contacts at room temperature. We took Ag paste, Cr/Au, Cu/Au and Ag/Al as metallic contacts. 5.4.2 Dielectric Study of oxygen deficient BTO annealed at 900 o C We studied the temperature dependent dielectric behaviour of oxygen deficient BTO annealed at 900 o C using the four probe technique with Cr/Au as metallic contacts (shown in figure 17) from 10K to 415K. The real part of complex impedance versus frequency has been drawn in figure 19. The lower frequency region (|) represents the contribution of the grain boundaries. The middle region shown by || gives the value of bulk resistance. We can clearly see that as the temperature increases the bulk resistance decreases from 10K to 50K to 100K finally becoming vanishingly small as the temperature is increased further. This is because as the temperature increases the dipoles in the bulk become more thermally active, thus lowering the value of the resistance. At very high frequencies (|||), the Z’ falls sharply because at such high frequencies, 49 all polarization, mobility mechanisms fail to follow the rapidly changing direction of the applied ac electric field. Figure 19: The real part of complex impedance (Z’) versus frequency at temperature of 10K, 50K, 100K and 300K. Three different regions of the curve have been marked in the graph shown by (|) in the lower frequency region, (||) in the middle region and (|||) at higher frequency region. 50 Figure 20: Nyquist plot for BTO sample drawn for temperatures 10K, 150K, 200K. The arrow marks the direction of increasing frequency. Z’’ is the imaginary part of complex impedance and Z’ is the real part From figure 20 we can clearly see that there are two relaxation processes one related to the relaxation of grain boundaries and other to the relaxations of grains. The semicircular arc at high frequency end is a clear indication of the presence of bound charges like the dipoles in the grains. The semicircle at the higher frequency end is mainly attributed to the grains and the end towards lower frequency mainly attributed to the grain boundaries. This is because at lower frequencies, the mobile charges accumulate at the grain boundaries and stay there for a long period of time before reversing their direction to the opposite side on changing the direction of ac electric field. At higher frequencies, the mobile charges do not get enough time to reach the grain boundary and hence at higher frequencies only bound charges or dipoles of the grain boundaries and the grains contribute [51]. Also, the lower end shows a very high value of resistance and capacitance. This is because when the frequency of the applied electric field is low, the free charges accumulate at the graingrain boundary surface and since the thickness of the grain boundary is very small, it leads to overall high capacitance. The grains are smaller in size than the grain boundaries; therefore they show low 51 capacitive behaviour. With temperature increase we can see that the resistance and capacitance of the grain boundary increases as more charges accumulate at the grain boundary region due to thermal motion. The resistance of the grains decrease with increasing temperature because the dipoles get more thermally active and is able to respond well to the changing electric field. The free charge carriers, defects, disorders, electrons and ions mainly contribute to the resistance and capacitance of grain boundary region. As our samples were oxygen deficient, oxygen vacancies could have been created at the grain boundaries leading to an increase in the resistivity of the grain boundaries. Figure 21: Dielectric loss tangent versus frequency shown at different temperatures given in Kelvin . Dielectric loss tangent (ratio of imaginary part of dielectric permittivity and real part of dielectric permittivity) versus frequency has been plotted in figure 21. Dielectric constant is the loss of energy in the material and occurs because the polarization lags behind the applied electric field mainly caused by the grain boundaries [52]. We can clearly see that as the frequency increases the dielectric loss decreases at all temperatures. This is because at lower frequencies more types of polarization occur like ionic, orientational, electronic, dipolar, interfacial but as the frequency increases, the molecules are not able to reorient quickly to the fast changing field and hence they relaxes leading to a decrease in the dielectric loss value. This kind of behaviour is very typical of every dielectric material. Higher value at low 52 frequencies can be due to increased resistivity of the grain boundaries as compared to the grains [52] Dielectric loss mainly arises due to impurities, imperfections and defects in the crystal structure. Materials that contain a large number of pores have low dielectric constant and high dielectric loss. Conductivity in a material arises due to electric conduction of mobile charges and weakly bound charges on the application of an external electric field and is a characteristic of dielectric loss in the medium. It mostly depends on the size of the grains and their distributions, impurities present and its structural symmetry. Ac conductivity dependence on frequency follows a simple power law behaviour given by A.K Jonscher [53] shown in equation 24 𝜎 = 𝜎+𝐴𝜔 (24) Where σac is the ac conductivity , σo is the frequency independent quantity and is usually related to the dc conductivity of the sample, s is an exponent that takes the value from 0 to 1 and A is a temperature dependent quantity. Value of s less than 1 implies hopping mechanism of charge carriers with translational motion. Ac conductivity dependence on frequency has been shown in figure 22. We mentioned earlier that at low frequencies since there are large number of carriers present, the overall conductivity of the sample decreases, since these carriers act as barriers to the passing current. But as the frequency keeps on increasing the conductivity also increases because of the relaxation of mobile charge carriers and release of space charges. 53 Figure 22: Ac conductance versus frequency shown at different temperatures Next we studied the Capacitance (Cp) versus frequency curve of BTO which gave us very different values as measured from the 10K CCR apparatus and another high temperature setup (shown in figure 23). The capacitance value of 10K CCR was of the order ranging from 10 -8 to 10 -10 F. But the capacitance value from the high temperature set-up was of the order of 10 -11 F. This made us conclude that besides the capacitance of the material, we have some additional capacitance coming from the 10K CCR temperature dependent IS set-up. 54 Figure 23: Capacitance (Cp) versus frequency of oxygen deficient BTO has been shown. a,b) The measurement was taken from the 10K CCR ranging from 10K to 415K. c) The measurement was taken from the high temperature set-up ranging from 70oC to 150oC a b c 55 Chapter 6 Conclusion In summary, we studied the Pristine BaTiO3 and oxygen deficient BaTiO3. We tried to make oxygen deficient BTO by annealing in Vacuum at 800oC and 900oC and by annealing in reducing atmosphere of Ar/H2 at 900oC and 1000oC. We indeed proved that the vacuum treated pellet was not oxygen deficient. The Ar/H2 treated pellet was oxygen deficient which was proved through Rietveld Technique and XRD analysis. Also, the Ar/H2 treated pellet appeared darker. We did the characterization of our sample through X-ray diffraction and Rietveld Refinement and calculated various results such as the crystallite size, lattice parameters and occupancy. We studied Impedance Spectroscopy (IS) and modelled the dielectric behaviour of a parallel RC circuit. We studied the working of a 10K CCR temperature set-up. We optimized the metallic contact for pristine BTO pellet and found out that the most efficient contact would be Cr/Au out of Cu/Au, Ag paste and Ag/Al. We also studied the dielectric behaviour of oxygen deficient BTO annealed at 900oC and studied its Complex impedance, resistance, capacitance, dielectric loss and ac conductivity. In future, we intend to take the dielectric measurements through lock-in amplifier which would nullify the extra contribution to capacitance coming from the experimental set-up. We would take EDAX measurement to determine the elemental composition. We would also like to take the SEM of our data to validate our assertion of formation of grain boundaries that gives a major contribution to the Impedance data. At the end we would like to study the Metal-Insulator transition in oxygen deficient BTO. Since years numerous studies conducted on both doped and undoped BTO as films, single crystals, polycrystals have shown that it has numerous industrial applications owing to its interesting physics. Our work on pristine and oxygen deficient BTO will help us to understand more about the temperature (both low and high) and doping effect on the properties of BTO. 56 References [1] Von Hippel, A. R. Dielectrics and Waves; John Wiley & Sons: New York, 1954; 202 213. [2] W. S. Cho,Structural evolution and characterization of BaTiO3 nanoparticles synthesized from polymeric precursor. J. Phys. Chem. Solids, 1998, 59, 659. [3] Jona, F; Shiran, G; Ferroelectric crystals, Dover Publications, INC., New York, 1993 chapter V11 [4] Duran, P; Gutierrez, D; Tartaj, J; Moure, C. Densification behaviour, microstructure development an dielectric properties of pure BaTiO3 prepared by thermal decomposition of (Ba, Ti)-citrate polyester resins. Ceram. Int. 2002, 28, 283-292 [5] Vijatovic, M.M; Bobic, J.D; Stojanovic, B.D. History and Challenges of Barium Titanate: Part I. Science of Sintering, 2008, 40, 155-165: [6] Patel, R.K; Kumar, P; Prakash, C; Agrawal, D.K. Low temperature synthesis and dielectric, ferroelectric and piezoelectric study of microwave sintered BaTiO3 ceramics, Ceramics International, 2012, 38, 1585–1589. [7] Simon-Seveyrat, L; Hajjaji, A; Emziane, Y; Guiffard, B; Guyomar, D; Re-investigation of synthesis of BaTiO3 by conventional solid-state reaction and oxalate coprecipitation route for piezoelectric applications, Ceramics International, 2007, 33, 35–40. [8] Woodward, P.M; Octahedral tilting in perovskites. I. Geometrical considerations, Acta Crystallographica Section B: Structural Science,1997, 53, 32-43. [9] Johnsson, M.; Lemmens, P. Crystallography and Chemistry of Perovskites. In Handbook of Magnetism and Advanced Magnetic Materials; John Wiley & Sons, Ltd: Hoboken, NJ, 2007. [10] www.3dchem.com/inorganicmolecule.asp?id=1618 9. [11] Zulueta, Y; Lim, T; Dawson, J. Defect clustering in Rare-Earth-Doped BaTiO3 and SrTiO3 and the influence on dopant incorporation. The Journal of Physical Chemistry C, 2017, 121. 57 [12] Boulos, M; Guillement-Fritsch, S; Mathieu, F; Durand, B; Lebey, T; V. Bley, V. Hydrothermal synthesis of nanosized BaTiO3 powders and dielectric properties of corresponding ceramics. Solid State Ionics, 2005, 176, 1301. [13] Potdar, H.S; Deshpande, S.B; Date, S.K. Chemical coprecipitation of mixed (Ba + Ti) oxalates precursor leading to BaTiO3 powders. Mater. Chem. Phys.1999, 58, 121. [14] Sasirekha, N; Rajesh, B. Hydrothermal Synthesis of Barium Titanate: Effect of Titania Precursor and Calcination Temperature on Phase Transition. Industrial & Engineering Chemistry Research 2008 47 (6), 1868-1875 [15] Xu, H; Gao, L. Tetragonal nanocrystalline barium titanate powder: preparation, characterization, and dielectric properties J. Am. Ceram. Soc.,2003, 86, 203. [16] Wang, L; Liu, L; D. Xue, D; Kang, H; Liu,C. Wet routes of high purity BaTiO3 nanopowders. J. Alloys Comp. 2007, 440, 78. [17] Prasadarao, A.V; Suresh, M; Komarneni, S. pH dependent coprecipitated oxalate precursorsthermal study of barium titanat. Materials Letters 1999, 39, 359. [18] Cho, W.S; Hamada,E; Synthesis of ultrafine BaTiO3 particles from polymeric precursor: their structure and surface property. J. Alloys Comp. 1998, 266, 118 [19] Lazarevic, Z; Stojanovic, B.D; Varela, J.A; An approach to analysing synthesis, structure and properties of bismuth Titanate ceramics. Sci. Sintering, 2005, 37, 199. [20] Stojanovic, B.D; Simoes, A.Z; Paiva-Santos, C.O; Jovalekic, C; Mitic, V.V; Varela, J.A; Mechanochemical synthesis of barium Titanate. J. Eur. Ceram. Soc. 2005, 25, 1985. [21] Ristic,M.M; Milosevic, S.DJ. Mechanical Activation of Inorganic Materials, Serbian Academy of Sciences and Arts, Belgrade, 1998, 5. [22] Stojanovic, B.D; Jovalekic, C; Vukotic, V; Simoes, A.Z; Varela, J.A. Ferroelctric properties of mechanically sunthesized nanosized barium Titanate. Ferroelectrics, 2005, 319, 65. [23] N Ozkan. Compaction and sintering of ceramic powders.(Imperial College of Science, Technology and Medicine, London, 1994)