Deconvolution of scanning transmission electron microscopy images of ionomers

Previously, we studied a variety of ionomer morphologies with scanning transmission electron microscopy (STEM). Other groups have found that deconvoluting STEM images dramatically improve the overall image quality and the detection of sub‐nanometer‐scale features. In this study, STEM images of nanometer‐scale ion‐rich aggregates were deconvolved via the Pixon method with a simulated electron probe. The image models are considerably sharper with significantly decreased noise levels, thus making the size and shape of the ionic aggregates easier to distinguish relative to those in the raw STEM images. Raw and deconvoluted images of Zn‐neutralized poly(styrene‐ran‐methacrylic acid) ionomers containing spherical ionic aggregates indicate that the electron density varies smoothly from the edge to the center of the aggregates. Deconvolution also clarifies the issue of aggregate overlap in the STEM images. Furthermore, line scans across deconvoluted STEM images suggest that the three‐dimensional density distribution of these nanoaggregates compares favorably with a radially symmetric Gaussian distribution as opposed to a uniformly dense sphere. The overall result of this work is that deconvolution of STEM images provide ways in which to better investigate the morphologies of ionomers. © 2003 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 41: 319–326, 2003