Procuring the fiber size distribution of nanofiber filter media

According to the United States Environmental Protection Agency, the size of particles is directly associated with their ability to damage the human health. Hence, human exposure to nanoparticles is becoming increasingly a concern. Air filtration can be an effective way to reduce markedly the amount of nanoparticles inhaled in several instances. Nanofiber media is a promising media which can provide a greater filtration efficiency and higher performance than conventional fibrous filter media. A fiber is considered a nanofiber generally when its diameter is smaller than 500 nm. Fibrous media made with cellulose can be very effective in removing nanoparticles when coupled with a nanofiber layer on its top or at the bottom of the cellulose layer. The purpose of this study is to obtain a series of fiber diameters to be fitted in a lognormal distribution. Such distribution would be able to describe the size distribution of the fibers making up the nanofiber layer. The purpose of the whole process is to build a numerical model of the nanofiber filter media using another specific software package. The images of the nanofiber filter media were obtained with a Scanning Electron Microscope (SEM). Here we evaluate two different methods for obtaining the series of diameters. In both cases the image was divided in twelve equal parts to obtain a good number of intersection between the scribed lines and the fibers of the media. The entire image was analyzed starting from the left vertical line to the right vertical line. In the first method, we used the Image Pro Plus software package. We drew perpendicular lines to fiber axis and automatically measured the fiber diameter. In the second method, the fibers are directly matched to the widths of the rectangles on the scale bar. The scale can be printed in various sizes. The widths of the rectangles labeled 1 through 15 are proportional to a series of appropriate numbers. We discuss the results obtained with the two different approaches and how they can be used to generate the pursued statistical distributions.