Quantitative assessment on the orientation and distribution of carbon fibers in a conductive polymer composite using high-frequency ultrasound

Conductive polymer composites, typically fabricated from a mix of conductive fillers and a polymer substrate, are commonly applied as bipolar plates in a fuel cell stack. Electrical conductivity is a crucial property that greatly depends on the distribution and orientation of the fillers. In this study, a 50-MHz ultrasound imaging system and analysis techniques capable of nondestructively assessing the properties of carbon fibers (CFs) in conductive polymer composites were developed. Composite materials containing a mix of polycarbonate substrates and 0 to 0.3 wt% of CFs were prepared using an injection molding technique. Ultrasonic A-line signals and C-scan images were acquired from each composite sample in regions at a depth of 0.15 mm beneath the sample surface (region A) and those at a depth of 0.3 mm (region B). The integrated backscatter (IB) and the Nakagami statistical parameter were calculated to quantitatively assess the samples. The area ratio, defined as the percentage of areas composed of CF images normalized by that of the whole C-scan image, was applied to further quantify the orientation of CFs perpendicular to the sample surface. Corresponding to the increase in CF concentrations from 0.1 to 0.3 wt%, the average IB and Nakagami parameter (m) of the composite samples increased from -78.10 ± 2.20 (mean ± standard deviation) to -72.66 ± 1.40 dB and from 0.024 ± 0.012 to 0.048 ± 0.011, respectively. The corresponding area ratios were respectively estimated to be 0.78 ± 0.35%, 2.33 ± 0.66%, and 2.20 ± 0.60% in region A of the samples; those of CFs with a perpendicular orientation were 0.04 ± 0.03%, 0.08 ± 0.02%, and 0.12 ± 0.05%. The area ratios in region B of the samples were calculated to be 1.19 ± 0.54%, 2.81 ± 0.42%, and 2.64 ± 0.76%, and those of CFs with a perpendicular orientation were 0.07 ± 0.04%, 0.12 ± 0.04%, and 0.14 ± 0.03%. According to the results of the orientations and ultrasonic images, CFs tended to distribute more uniformly in the deeper regions of the samples. This study validates that the distribution and orientation of CFs in conductive polymer composites could be sensitively and quantitatively assessed by high-frequency ultrasound in conjunction with current analysis methods.