Significance of paperRecent efforts in the composites industry towards less-expensive means to produce high-performance parts have often involved optimisation of liquid composite moulding processes such as resin transfer moulding (RTM). The most significant gap in part quality between RTM manufactured parts and traditional autoclave processes is the usually higher void content in the former, arising from the entrapment of bubbles during infusion, and the lower consolidation pressures used during such processes. Many laboratories around the world are working on understanding bubble entrapment and subsequent bubble mobility, so as to optimise RTM processes to reduce the concentration and size of voids. This paper contributes to that understanding with threedimensional examination using CT imaging of the morphology, size, clustering and location of the individual voids in composite parts made with RTM. Most important / novel contributionsThis paper presents visualisation of the voids in a composite part with significantly more statistical information regarding size distribution, orientation and location than has been previously shown in the literature. It also addresses a reinforcement architecture more typical to RTM, an un-balanced weave, than the unidirectional fabrics usually studied in previous work on void formation. Analysing a complex reinforcement along with the enhanced visualisation abilities of the CT-imaging technique allowed novel observations and conclusions regarding voids: 1) As the orientation angle between a reinforcement layer and the resin flow direction increases from parallel to perpendicular, larger voids and a greater number of voids were observed in that layer. This was linked to the resulting greater propensity for fatigue crack propagation between voids in layers transverse to the loading direction. A simple optimisation strategy is thus to infuse in a direction transverse to the expected primary load direction, thus creating the fewest voids in the transverse-to-load direction. 2) Voids accumulate around both the layer interfaces and yarns and are nearly completely absent from the layer thickness between those interfaces and away from yarns. 3) Void distribution follows the orientation of fibres in adjacent layers, suggesting that out-of-plane flow is a significant mechanism in void formation and mobility. 4) Observations 1-3 above all imply that current void formation and mobility models must be expanded from a microscale approach to the laminate scale, in order to focus on out-of-plane bubble movement and yarn bubble entrapment. M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT AbstractTomographic imaging using both microfocus radiation and synchrotron radiation was performed to assess the void defects in resin transfer moulded woven carbon fibre composites. The focus of this study is on characterising the void homology (e.g. local void size and spatial distribution) in relation to weave orientation, infusion direction and potential effects on damage formation in tensile loadin...

The focus of this study is the influence of voids on the damage behaviour in quasi-static loading of resin-infused carbon fibre-reinforced polymers. Experimental results are presented for quasi-static loading in combination with high-resolution tomographic imaging and statistical analysis (homology of pores or voids and induced cracks). Three distinct mechanisms were observed to control delamination growth in the presence of sharp and blunt voids. Delamination cracks interact with the supporting yarns, especially in combination with air pockets trapped in the resin in the form of long, sharp voids. This resulted in crack growth that coalesces with delamination cracks from neighbouring yarn-voids during increased out-of-plane load–displacement, with almost no presence of intralaminar transverse cracks. This highlights the benefits and drawbacks of the supporting yarn during out-of-plane loading.

Microstructure design is a vital enabling technology for modern industry that promises to short circuit the traditional empirical-based methods for design of new materials. However, inverse design methods that underlie microstructure design are generally computationally intensive, thus limiting their widespread adoption. This paper presents an inverse design approach that is based upon spectral methods in order to achieve a highly effi cient framework for rapid design of materials. Some initial applications of the approach are briefl y outlined.

No abstract

Liquid composite molding (LCM) is growing in importance as an alternative to traditional prepreg-autoclave methods for manufacture high-performance composites. The most significant roadblock to industry’s implementation of LCM is the usually higher void content compared with prepreg processing. One tool for reducing void levels in LCM involves optimization of flow velocity, which requires models to be developed to describe void formation at a given velocity. To help solve this problem, the following research illustrates the first known method for optical void measurement in situ during infusion in a carbon fiber reinforcement. Similar to previous studies on glass fiber, this work utilizes fluorescent dye and a digital camera to produce sufficient contrast and resolution for image analysis. Visible bubbles are photographed against the opaque carbon fiber background. An automated method of image analysis is outlined, which was used to analyze 230 images for three different flow orientations of a single fabric, producing the highest amount of experimental data seen so far on in situ void measurement. The resulting data identifies a minimum velocity threshold for minimal macro-void formation. The resultant void characterization framework will better enable optimization of LCM processing for high-performance composites based on carbon reinforcements.