Raman Investigations of Strained Ferroelectrics

<p>This describes research done on a variety of ferroelectric systems over the course of three years during the Ph.D. programme at Victoria University of Wellington. The majority of the work involved using Raman spectroscopy to investigate the lattice dynamics of the ferroelectric materials studied, and by this method measuring the structural phase diagrams in the ferroelectrics. Three material systems were investigated. The first was bulk ceramics of the solid solution of Na₀.₅Bi₀.₅TiO₃ and BaTiO₃ (BNT-BT). The second material was PbTiO₃ (PT) nanowires prepared by annealing `PX' phase lead titanate in air. In these samples we found a tensile strain caused by nanoscopic voids in the wires. The third material studied was thin films of SrTiO₃ (STO) grown epitaxially on lattice mismatch substrates. This introduced strain into the system. In the BNT-BT system, temperature dependent Raman spectra were taken of the various samples. From the spectra, it was discovered that the structural phase transitions of the material did not perfectly correspond to the electrical phase transitions. For one significant phase boundary, no structural change occurs, only a loss of long-range order. The local sensitivity of the Raman spectroscopy technique allowed this to be found. As a consequence of this, no tricritical point is found in the phase diagram of BNT-BT. It was also found that poling the sample in electrical fields shifted the morphotropic phase boundary between 5% and 6% Ba-substitution about 1% towards the high-substitution side, but otherwise did not affect the phase diagram. In the PT nanowires system, temperature dependent Raman spectroscopy and scanning electron microscopy were performed to measure the spectra of single nanowires. A large enhancement of the ferroelectric phase transition temperature was discovered. The enhancement was found to be dependent on the nanowire diameter, with peak enhancements of over 100K measured in wires close to 125 nm. Wires both larger and smaller than this showed smaller degrees of enhancement. It is proposed that the enhancement is caused by tensile strain developed in the wires during their synthesis, where they were transformed from a low-density phase into a high-density phase. In the STO thin films system, temperature dependent ultraviolet Raman spectra and x-ray diffraction spectra were measured to establish a relationship between the biaxial strain developed in the films and their phase diagrams. The XRD found strain in the films of similar substrate which was inversely proportional to thickness up to a threshold point. Beyond that point, there is a discontinuity and additional thickness of film is grown without strain. The strain in the lower layer remains constant. The UV Raman spectroscopy method was able to enhance the signal such that thin films with weak signals could be measured. The spectra showed signs of a phase transition in all of the films. In one film enough of the spectral features were visible to characterise the low temperature phase as the orthorhombic ferroelectric phase of STO. The transition temperature varied from sample to sample, and a relationship between the biaxial strain and the transition temperature was seen.</p>

Raman Investigations of Strained Ferroelectrics

<p>This describes research done on a variety of ferroelectric systems over the course of three years during the Ph.D. programme at Victoria University of Wellington. The majority of the work involved using Raman spectroscopy to investigate the lattice dynamics of the ferroelectric materials studied, and by this method measuring the structural phase diagrams in the ferroelectrics. Three material systems were investigated. The first was bulk ceramics of the solid solution of Na₀.₅Bi₀.₅TiO₃ and BaTiO₃ (BNT-BT). The second material was PbTiO₃ (PT) nanowires prepared by annealing `PX' phase lead titanate in air. In these samples we found a tensile strain caused by nanoscopic voids in the wires. The third material studied was thin films of SrTiO₃ (STO) grown epitaxially on lattice mismatch substrates. This introduced strain into the system. In the BNT-BT system, temperature dependent Raman spectra were taken of the various samples. From the spectra, it was discovered that the structural phase transitions of the material did not perfectly correspond to the electrical phase transitions. For one significant phase boundary, no structural change occurs, only a loss of long-range order. The local sensitivity of the Raman spectroscopy technique allowed this to be found. As a consequence of this, no tricritical point is found in the phase diagram of BNT-BT. It was also found that poling the sample in electrical fields shifted the morphotropic phase boundary between 5% and 6% Ba-substitution about 1% towards the high-substitution side, but otherwise did not affect the phase diagram. In the PT nanowires system, temperature dependent Raman spectroscopy and scanning electron microscopy were performed to measure the spectra of single nanowires. A large enhancement of the ferroelectric phase transition temperature was discovered. The enhancement was found to be dependent on the nanowire diameter, with peak enhancements of over 100K measured in wires close to 125 nm. Wires both larger and smaller than this showed smaller degrees of enhancement. It is proposed that the enhancement is caused by tensile strain developed in the wires during their synthesis, where they were transformed from a low-density phase into a high-density phase. In the STO thin films system, temperature dependent ultraviolet Raman spectra and x-ray diffraction spectra were measured to establish a relationship between the biaxial strain developed in the films and their phase diagrams. The XRD found strain in the films of similar substrate which was inversely proportional to thickness up to a threshold point. Beyond that point, there is a discontinuity and additional thickness of film is grown without strain. The strain in the lower layer remains constant. The UV Raman spectroscopy method was able to enhance the signal such that thin films with weak signals could be measured. The spectra showed signs of a phase transition in all of the films. In one film enough of the spectral features were visible to characterise the low temperature phase as the orthorhombic ferroelectric phase of STO. The transition temperature varied from sample to sample, and a relationship between the biaxial strain and the transition temperature was seen.</p>