Novel method for quantitative starch penetration analysis through iodine staining and image analysis of cross - sections of uncoated fine paper

This paper presents a novel method for determining starch penetration through iodine staining and microscopic imaging of surface-sized fine paper cross-sections and image analysis of the resulting grayscale images. The more specific aim was to obtain quantitative starch penetration information in the z-direction of the sheet. The penetration information was determined through an image analysis procedure as a z-directional distribution of pixels classified to contain starch based on their grayscale values. This analysis was conducted on a number of cross-sectional images to improve the accuracy of the analysis since the variation of starch penetration within the scope of a single cross-sectional image varies across a large window. The numerical penetration information obtained was used in further processing, such as determining a dimensionless penetration number Q. The penetration number can be quantitatively used in comparing the penetration of starch in different paper samples produced under different process conditions, for example, that affect surface size starch penetration. Therefore, the penetration number Q can be used in characterizing penetration-related paper properties – such as the internal strength of the sheet – to help in the optimization of starch penetrationrelated papermaking process parameters. The penetration information was also processed to represent a starch amount distribution in the z-direction, defined here as a Simulated Starch Content (SSC) distribution. This SSC distribution gives a good estimate of the starch distribution across the z-direction. The method developed in this work is fast and it will give quantitative and comparable information on z-directional starch penetration. Since the complete procedure was defined in detail from the handling and iodine staining of the specimens to the microscopic photographing of the cross-sections and computational image analysis of the cross-sectional images, the results are not dependent on individual personal factors of the personnel conducting the analysis. The starch penetration curves were also compared against Li–distributions obtained through ToF-SIMS, that represented the distribution of the starch solution in the zdirection,using LiCl2 as a marker for starch.