EDS and EBSD chemical, orientation, and phase mapping are techniques that are routinely applied on many materials. When a sample is capable of producing high EDS count rates or good EBSD patterns, this type of analysis is straightforward. But when the input data quality is not so good, due to a variety of reasons, the mapping quality will be poor and data collection may be time consuming. New averaging functions are proposed to improve map data quality without adding to the measurement time while minimising loss of detail in your map.
Currently in EDS spectral mapping, phase maps can be constructed in real time by comparing the spectra or ROI signal of selected elements for each point in the map and grouping pixels with equivalent chemical signature. Due to the low number of counts inside each spectrum It may take several passes before all points are assigned to a phase. The new method performs a continuous averaging of the spectra in each pixel with all pixels directly surrounding it in combination with a dynamic binning that automatically changes the effective map resolution during acquisition. This ensures that each data point quickly contains enough counts for successful element detection and phase assignment.
EBSD maps recorded on a hexagonal scan grid may be reprocessed off-line using recorded EBSD patterns. NPAR (Neighbour Pattern Averaging and Reindexing) takes the pattern at any point in the EBSD map point and then considers the 6 patterns in the hexagon around it. These original 7 patterns are put together and averaged onto the single pattern in the middle which greatly improves the signal to noise ratio and allows successful reindexing.
In this presentation the EDS and EBSD averaging techniques will be explained and the effects on the conservation of details and orientation precision explored.