Vol. 33, Issue 1, pp. 97-105
Keywords
sol-gel matrix, physical properties, repeatability of production, pattern recognition, dimensionality reduction, cluster analysis
Abstract
The methods based on the statistical pattern recognition are used for the evaluation of microscopic images of sol-gel matrices. The applied classification is a two steps scheme: dimensionality reduction of the data and cluster analysis. The obtained results have demonstrated that the molar ratio R (the number of solvent moles to the number of substrate moles) influences the stability of sol-gel materials, thus enabling us to reach the higher repeatability of the production process. It has been shown that the repeatability varies between 96.67% (for molar ratio R = 5) and 98.33% (for R = 15, 32, 50). This corresponds with the results of light scattering and transmission measurements.