Results and discussion
The impact of the detector dead-time on the quantitative analysis of nominally pure boron (B), Si, and for B-implanted Si (NIST-SRM2137) with a retained 10B dose of about 1×1015atomscm−2 was investigated. Ion correlation analyses were used to graphically demonstrate the impact of detector dead-time on multi-hit detection events. Given the important role of detector dead-time as a signal loss mechanism, three different methods for estimating the detector dead-time were presented. The following findings resulted from this research:
Polycrystalline films are of great importance to many technological applications, for example as protective coatings , device interconnects , sensors , surface acoustic wave devices  or transparent electrodes . The polycrystalline nature of the films has a strong effect on film properties, which may therefore differ significantly from those of their single crystal counterparts. It is thus important to establish growth models that explain how processing parameters influence grain size, shape and orientation in these films.
In many cases polycrystalline film growth starts from non-epitaxial nucleation of closely spaced nano-sized grains forming a layer of nuclei at the substrate. If, during film thickening, the atomic mobilities are low enough for the grain boundaries in the bulk to remain immobile, growth Go 6983 between neighboring grains occurs at the free surface. Grains with their fastest growth axis close to the film growth direction overgrow otherwise oriented grains and so coarsen as the film thickens. This leads to a film with a preferential orientation composed of dominant, columnar or V-shaped grains . The understanding of this growth competition and the resulting microstructure evolution during film thickening has been an important topic for polycrystalline film growth and several models and simulations have been proposed [7–10]. To validate such proposed models and simulations for a given film it is necessary to have quantitative data on how the film’s microstructure and preferential orientation develops throughout its thickness.
Scanning electron microscopy (SEM) and atomic force microscopy (AFM) imaging of the surface of a film series of varying thicknesses are common methods for the characterization of the microstructural evolution. However the precise and reliable quantification of grain size from images of the free surface is often not simple and sometimes not even possible, and additionally no direct information on preferential orientation is obtained. X-ray diffraction (XRD) is often used to evaluate preferential orientation, but with the diffracting volume usually extending through the entire film thickness, it is not straightforward to correlate the measurements to the microstructure. Alternatively, conventional transmission electron microscopy (TEM) provides coupled information on preferential orientation and microstructure even at the nanoscale, as required for ultrafine-grained materials. However quantifiable observations, for instance on grain size, are often only possible with tedious and time-consuming manual outlining, due to the presence of complicated image contrast. A possible way to solve this problem is by using automated crystal orientation mapping (ACOM). In recent years there has been a surge of orientation mapping techniques working at a nanoscale resolution [11–13]. One very successful method is based on scanning nano-beam diffraction followed by matching of recorded diffraction patterns to pre-calculated templates [13–16]. While powerful, TEM-based ACOM requires there to be a minimal amount of grain overlap in the specimen, a condition that is stringent on sample preparation, especially when studying nano-sized grains.
Here we use TEM-based ACOM to quantify the microstructure and preferential orientation evolution of polycrystalline low pressure metal–organic chemical vapor deposition (LP-MOCVD) grown ZnO thin films, currently used as transparent electrodes in photovoltaic applications . First, in order to illustrate how the amount of grain overlap is affected by the chosen specimen geometry, we compare ACOM on standard cross-section and plan-view sample geometries and show the advantages of plan-view sections for quantification of a film’s microstructure.