Quantitative Morphology of Aluminum Silicate Nanoaggregates
- PDF / 175,998 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 54 Downloads / 195 Views
G15.3.1
Quantitative Morphology of Aluminum Silicate Nanoaggregates GIOVANNI F CROSTA¹,², CHANGMO SUNG², BONGWOO KANG², CAROLINA OSPINA², PETER J STENHOUSE³ ¹ ² ³
Dept. of Environmental Sciences, Università degli Studi Milano - Bicocca, 1, Piazza della Scienza, I 20126 MILAN, Italy; contact: [email protected] Center for Advanced Materials, University of Massachusetts - Lowell, LOWELL, MA US Army Natick Soldier Center, NATICK, MA.
ABSTRACT Aluminum silicate nanoaggregates grown on organic multilayer templates were imaged by a transmission electron microscope. Images were processed by fractal and FOURIER analysis. The estimated mass fractal dimension suggested that aggregate formation was diffusion limited. Nonlinear filtering of FOURIER spectra, which included comparison with a model spectral density, yielded «enhanced power spectra». Some morphological descriptors were extracted from the latter. The main result, materials classification, was attained by a two-scale procedure. Some descriptors were related to the material properties such as nanoparticle size distribution and sharpness of aggregate boundaries. INTRODUCTION The goal of quantitative morphology is to perform image analysis and classification eventually aimed at predicting to which class a given image belongs. Image analysis relies on methods which in turn can be classified by their dependence on spatial scale. On one end there are methods based on an absolute scale, such as the FOURIER transform. On the other end there is fractal analysis, which is scale invariant by definition. Morphological (i.e., FOURIER, wavelet and/or fractal) analysis lends itself to complex, automated procedures which start with the image under test and return some numerical descriptors suitable for subsequent statistical interpretation. As far as image classification is concerned, morphological analysis followed by the statistical one can be limited to the descriptive level i.e., duplicating the results obtained by an experienced human observer. A more ambitious goal is to have the classifier assign a new image to one of the previously identified sets of images. MATERIALS, SPECIMEN PREPARATION A solution of aluminum isopropoxide (AIPO) and tetraethylorthosilicate (TEOS) in tbutanol was the source of the reacting species. The molar ratios of AIPO to TEOS was 1:3. Solute concentration was 50 mM. The substrates onto which aluminum silicate was to form were carbon coated 200 mesh Formvar - Cu grids used in TEM. One set of substrates, labelled U, did not undergo any further treatment and was used in the control experiment. The other set of substrates, labelled C, was coated by an organic multilayer template based on d-PDAC (poly diallyl dimethyl ammonium chloride) and a polyphenol red - polyaniline complex. These grids were finally coated by a polyamidoamine dendrimer terminated by sulfonic groups. The expected role of the organic template was to promote the formation of aluminum silicate nanoparticles. Mineralization i.e., reaction and precipitation of the solutes onto t
Data Loading...
