REPOSITORY > RESULTS

Abstract

(1-x)Pb(Mg1/3Nb2/3)O3–xPbTiO3 (PMN-PT) thin films with three compositions (x = 0.25, x = 0.33 and x = 0.40) close to the morphotropic phase boundary (MPB) were grown on SrTiO3 (STO) substrates by means of pulsed-laser deposition. Motivated by the higher potential power output of the d33-mode energy–harvesters (EHs), the first part of the thesis is focused on the growth of PMN-PT (x = 0.33) thin films directly on STO substrates. Due to Pb-loss, pyrochlore-free growth could not be realized using a stoichiometric target. Therefore, Pb-rich targets were prepared in order to find the exact amount of Pb excess required for the growth of pure-perovskite films. I found that the addition of 20 mol. % was sufficient to ensure single-phase growth; however, the process window was extremely narrow, particularly regarding the substrate temperature and ambient pressure. X-ray diffraction measurements revealed a splitting of the PMN-PT peaks in the film prepared at 0.13 mbar. I found that the splitting was related to a high density of structural defects, such as out-of-phase boundaries (OPB), which was determined by atomic-resolution scanning-transmission electron microscopy. I have managed to significantly reduce the defect density by raising the process pressure to 0.27 mbar. By conducting an experiment with an Ar/O2 mixture I showed that the total ambient pressure, rather than the partial pressure of oxygen, was the determining factor for the structural quality of the layers. I argued that for the given system a transition in the plasma plume expansion regime occurs between 0.13 mbar and 0.27 mbar. I achieved a twofold increase of the relative dielectric permittivity and the longitudinal piezoelectric response upon raising the process pressure to 0.27 mbar. Nevertheless, both films were found to grow in a combined 2D and 3D manner, forming Ruddlesden-Popper-type (RP-type) defects between the as-grown islands, to compensate for the small surplus of Pb. A slight Mg-deficit was also observed. I achieved a vast expansion of the process window by growing the films on LaNiO3 (LNO) bottom electrodes, which is the focus of the second part of the thesis. Based on the results obtained by using other types of intermediate layers, I attributed the improved stability of the perovskite phase to the higher surface roughness of the LNO/STO system as compared to the bare STO substrates. The increased roughness improves the sticking probability for PbO, thereby suppressing the formation of the pyrochlore phase. Therefore, I selected the substrates with LNO electrodes for further investigations regarding the influence of the target composition on the structure and properties of the PMN-PT films. Due to the Mg deficit observed in films grown from targets with only PbO excess, I also used targets with both PbO and MgO excess. The film prepared from the target with 20 mol. % PbO and 10 mol. % MgO was very close to stoichiometric, its formula being Pb1.04 ± 0.01Mg0.203± 0.004Nb0.44 ± 0.01Ti0.33 ± 0.01O3. As the perovskite ABB’B’’O3 structure is unable to accommodate the surplus of Pb on the A site, additional PbO layers are grown, forming RP-type defects. The improved overall stoichiometry of the film, as well as the slightly larger crystallite size are believed to be responsible for the superior piezoelectric properties of this film, with a |𝑑33,𝑒𝑓𝑓𝑚𝑎𝑥 | of 135 pm V-1. Comparison of the films with a different PT content showed that the piezoelectric response in films with x = 0.33 and x = 0.40 was similar, implying a broader region of enhanced response as compared to bulk PMN-PT.