The primary objective of this research is to investigate the effects of process variables on microstructure in several fluoride and oxide thin films prepared by vapor deposition, in order to predict the properties and behaviors of nanocrystalline thin film materials.

There are three distinct stages of this research. The first stage focuses on measuring of the porosity in polycrystalline thin films of a variety of fluorides as a function of the substrate temperature during deposition, and discussing the mechanism by which the porosity varies as a function of the process variables. We have measured the porosity in thin films of lithium fluoride (LiF), magnesium fluoride (MgF₂), barium fluoride (BaF ₂) and calcium fluoride (CaF₂) using an atomic force microscope (AFM) and a quartz crystal thickness monitor. The porosity is very sensitive to the substrate temperature and decreases as the substrate temperature increases. Consistent behavior is observed among all of the materials in this study.

The second stage is to understand the film microstructure including grain growth and texture development, because these factors are known to influence the behavior and stability of polycrystalline thin films. This study focuses on grain growth and texture development in polycrystalline lithium fluoride thin films using dark field (DF) transmission electron microscopy (TEM). It is demonstrated that we can isolate the size distribution of <111> surface normal grains from the overall size distribution, based on simple and plausible assumptions about the texture. The {111} texture formation and surface morphology were also observed by x-ray diffraction (XRD) and AFM, respectively. The grain size distributions become clearly bimodal as the annealing time increases, and we deduce that the short-time size distributions are also a sum of two overlapping peaks. The smaller grain-size peak in the distribution corresponds to the {111}-oriented grains which do not grow significantly, while all other grains increase in size with annealing time. A novel feature of the LiF films is that the {111} texture component strengthens with annealing, despite the absence of growth for these grains, through the continued nucleation of new grains.

The third stage focuses on the evaluation of triple junction energy in nanocrystalline ZrO₂ thin films. Grain boundaries and triple junctions are important aspects of the microstructure of most crystalline materials, and it is necessary to understand them to be able to predict the behavior of bulk polycrystals and polycrystalline thin films. Triple junctions, where three grains or grain boundaries meet, become increasingly important in nanocrystalline materials where they occupy an increasing fraction of the total volume of the material. It would therefore be of great significance to know whether, and if so how triple junction energy varies. In this study we evaluate triple junction energies in nanocrystalline ZrO₂ thin films using thickness mapping images produced by energy filtered transmission electron microscopy (EFTEM), which enable us to measure the surface topography associated with grain boundaries and triple junctions. In our films, the triple junction energy is deduced to be either zero (within the accuracy of the measurement) for most, but significantly positive for a few of the junctions.