Quantifying Intra-tow Fiber Volume Fraction in GFRP: A comparison of 3D non-destructive X-ray computed tomography and destructive optical microscopy
Topic(s) : Special Sessions
Co-authors :
Jesper JESPER JOHN LISEGAARD (DENMARK), Debabrata DEBABRATA ADHIKARI (DENMARK), Jesper JESPER HENRI HATTEL , Sankhya SANKHYA MOHANTYAbstract :
The rapid expansion of wind energy demands innovative solutions to enhance the structural integrity and performance of wind turbine blades through accurate modelling of composite materials. The utilization of X-ray μ-Computed Tomography (XCT) has witnessed substantial growth in the characterization of composite materials. This non-destructive technique enables highly accurate three-dimensional inspections of fiber architectures, manufacturing defects, and other properties of interest. Still, the improvement of characterization techniques and post-processing methodologies are imperative for reliably quantifying the structural integrity of fiber-reinforced polymers as well as for implementing model-based approaches to optimize the corresponding manufacturing processes.
In this paper, a comparative analysis between the non-destructive XCT approach and the destructive optical microscopy approach is presented, focusing on the characterization of the microstructural attributes of a Non-Crimp Fabric (NCF) Glass Fiber Reinforced Polymer (GFRP) sample. The complex fiber architecture of the NCF fabric configuration poses challenges to efficient quantification of the inter- and intra-tow fiber volume fractions, which vary significantly from meso- to microscale. To address this, a statistical modelling method for the intra-tow fiber volume fraction using XCTs with varying resolutions is introduced, aligning with results obtained through destructive optical microscopy. First, the volumetric XCT data is acquired at varying resolutions, with each of the volumetric data sets being subsequently segmented to determine the intra-tow fiber volume fraction as a function of the resolution level, highlighting a converging trend for higher resolution levels. Subsequently, multiple cross-sections of the same sample are subjected to optical microscopy, with the corresponding results serving as a ground truth of the intra-tow fiber volume fractions, against which the proposed XCT model is validated.
The proposed model for the fiber volume fraction not only facilitates a more accurate intra-tow fiber volume fraction using low-resolution XCT, but also serves as a foundation for efficient as-manufactured model-based approaches towards optimizing manufacturing of large structural glass-fiber composites. Furthermore, while this work focuses on glass fibers (as the greater fiber diameter and the higher density allow using lower resolutions for XCT analysis compared to that required with carbon-fiber reinforced polymers (CFRP)), it also paves the way for similar methods for other fiber reinforced polymers.
In this paper, a comparative analysis between the non-destructive XCT approach and the destructive optical microscopy approach is presented, focusing on the characterization of the microstructural attributes of a Non-Crimp Fabric (NCF) Glass Fiber Reinforced Polymer (GFRP) sample. The complex fiber architecture of the NCF fabric configuration poses challenges to efficient quantification of the inter- and intra-tow fiber volume fractions, which vary significantly from meso- to microscale. To address this, a statistical modelling method for the intra-tow fiber volume fraction using XCTs with varying resolutions is introduced, aligning with results obtained through destructive optical microscopy. First, the volumetric XCT data is acquired at varying resolutions, with each of the volumetric data sets being subsequently segmented to determine the intra-tow fiber volume fraction as a function of the resolution level, highlighting a converging trend for higher resolution levels. Subsequently, multiple cross-sections of the same sample are subjected to optical microscopy, with the corresponding results serving as a ground truth of the intra-tow fiber volume fractions, against which the proposed XCT model is validated.
The proposed model for the fiber volume fraction not only facilitates a more accurate intra-tow fiber volume fraction using low-resolution XCT, but also serves as a foundation for efficient as-manufactured model-based approaches towards optimizing manufacturing of large structural glass-fiber composites. Furthermore, while this work focuses on glass fibers (as the greater fiber diameter and the higher density allow using lower resolutions for XCT analysis compared to that required with carbon-fiber reinforced polymers (CFRP)), it also paves the way for similar methods for other fiber reinforced polymers.