Material extrusion (MEX) additive manufacturing has successfully fabricated assembly-free structures composed of different materials processed in the same manufacturing cycle. Materials with different mechanical properties can be employed for the fabrication of bio-inspired structures (i.e., stiff materials connected to soft materials), which are appealing for many fields, such as bio-medical and soft robotics. In the present paper, process parameters and 3D printing strategies are presented to improve the interfacial adhesion between carbon fiber-reinforced nylon (CFPA) and thermoplastic polyurethane (TPU), which are extruded in the same manufacturing cycle using a multi-material MEX setup. To achieve our goal, a double cantilever beam (DCB) test was used to evaluate the mode I fracture toughness. The results show that the application of a heating gun (assembled near the nozzle) provides a statistically significant increase in mean fracture toughness energy from 12.3 kJ/m² to 33.4 kJ/m². The underlying mechanism driving this finding was further investigated by quantifying porosity at the multi-material interface using an X-ray computed tomography (CT) system, in addition to quantifying thermal history. The results show that using both bead ironing and the hot air gun during the printing process leads to a reduction of 24% in the average void volume fraction. The findings from the DCB test and X-ray CT analysis agree well with the polymer healing theory, in which an increased thermal history led to an increased fracture toughness at the multi-material interface. Moreover, this study considers the thermal history of each printed layer to correlate the measured debonding energy with results obtained using the reptation theory.

In this work, the effects of tool bias as a processing parameter on an additive friction stir deposition repair of AA7050-T7451 plates were investigated. In this study, the asymmetrical material flow effect was assessed by subjecting a machined groove plate to different tool biases during the AFSD process. The tool bias effectiveness on repairs was quantified through mechanical testing and destructive and non-destructive evaluation. Monotonic tensile tests conducted in parallel with a digital image correlation method captured the mechanical performance of the repairs and the local strain effects of the heat-affected zone. X-ray computed tomography images of the different tool bias repairs were analyzed for any voids or defects to measure the overall bonding effectiveness. The results suggest that the advancing side tool bias had adequate bonding with the substrate, which yielded improved mechanical performance compared to the centered and retreating side depositions.

This article presents a method of ultrasonic testing (UT) that detects and quantifies interlaminar delaminations in CFRP composites with high resolution in terms of both spatial resolution in the planar dimension and depth into the laminate. Unidirectional and woven CFRP laminates were fabricated for this study, with a PTFE film inserted at various depths throughout the laminate to act as intentional crack initiation sites. All samples were mechanically tested via a three-point, end-notched flexure (ENF) test, followed by a quantification of the extent of the induced interlaminar delaminations using UT and X-ray computed tomography (CT). UT analysis for unidirectional CFRP samples was able to show a clear contrast between the delaminated area and the non-delaminated area. UT analysis of the woven CFRP samples yielded comparable results but required finer tuning of analysis parameters due to the interlocking woven fabric. CT results revealed a significant contrast between the crack and composite; thus, fine geometrical features of the crack front could be observed. UT and CT measurements were then compared, revealing an average difference of 1.09% in the delamination area, with UT overestimating as compared to CT. A UT depth study was also performed to automatically locate the interlaminar delamination at different depths throughout the components, with the delamination being predicted within one lamina interface for all samples. These results demonstrate UT’s ability to accurately detect and quantify the extent and location of interlaminar delaminations due to bending.