Analysis of pore structure of activated carbon fibers using high resolution transmission electron microscopy and image processing

Activated carbon fibers (ACF's), already used widely as absorbent materials, are now expected to be useful as new electrical and electronic materials, for their very large specific surface areas (SSA). Chemical adsorption as well as x-ray diffraction have been mainly used for characterizing the ACF structure. While TEM observations reveal the texture of ACF's, such observations have not yet yielded quantitative information about the microstructure. To promote the quantitative interpretation of the TEM images, computer image analysis is used in this work to clarify the pore structure of ACF's. The microstructures of three samples, which are all isotropic pitch-based ACF's but with different SSA values, have been investigated. Operations such as noise reduction, low frequency cut-off filtering, and binary image formation are used to clarify the pore images of the ACF's. The distribution of the ACF porosity size is clearly shown by a frequency analysis of the two-dimensional fast Fourier transform (FFT). The results suggest that TEM images include contributions from many different pore sizes. Pores in different size ranges are extracted by the inverse FFT (IFFT) operation by selecting the specific frequency range, and by-this analysis the pore structure is shown to have fractal characteristics.