PSI - Issue 25

S. Valvez et al. / Procedia Structural Integrity 25 (2020) 394–399

397

S.Valvez et al./ Structural Integrity Procedia 00 (2019) 000 – 000

4

rate of filament. The maximum carbon fibre content of 27% was obtained for a hatch spacing of 0.4 mm, and a significant decrease was observed for higher values. A very similar behaviour was observed for the feed rate of the filament, where the maximum carbon fibre content was obtained with the lowest value (60 mm/min). In terms of layer thickness, the carbon fibre content decreases with this parameter (from 15% to 12% between 0.3 and 0.8 mm), but it is not as expressive as that observed for hatch spacing (from 27% to 7% between 0.4 and 1.8 mm) and feed rate of the filament (from 16% to 7% between 80 and 160 mm/min). Tian et al. (2017) in another study developed an innovative process for recycling and remanufacturing 3D printed continuous carbon fibre reinforced PLA composites. The tensile strength was compared between neat PLA, originally printed and remanufactured composites specimens. While the PLA samples exhibited tensile strength and modulus of 62 MPa and 4.2 GPa, respectively, samples of originally printed composites present values around 4 and 5 times higher, respectively. However, when comparing samples of originally printed and remanufactured composites, very close values are observed due to unidirectional character of produced composites. In this case, the tensile load is mainly supported by the carbon fibres instead of the interfaces. Regarding the flexural properties, the lowest values were obtained again for the PLA samples, while some differences were observed for composite samples. Comparing the bending strength, remanufactured specimens present a 25 % improvement over the original composite samples. Since flexural strength is governed mainly by carbon fibre and interfaces, the remanufacturing process is responsible by extra melting impregnations and better interfaces are obtained (which explains the higher bending strength obtained). On the other hand, the modulus is slightly smaller (14.5 GPa for originally printed specimens and 13.3 GPa for remanufactured ones) due to degradation of the PLA matrix during the remanufacturing process. The Charpy impact strength was also compared, and values around 20 38.7 kJ/m2, 34.5 kJ/m2 and 38.7 kJ/m2 were obtained, respectively, for PLA samples, originally printed and remanufactured composites specimens. Finally, similar results were obtained for the interlaminar shear strength, which was around 5.7 MPa, 20.3 MPa and 22.7 MPa, respectively for neat PLA, originally printed and remanufactured composite specimens. Therefore, it is possible to conclude that, while flexural strength was significantly improved after the remanufacturing process, the tensile strength, impact strength and interlaminar shear strength more modestly improved. In addition to the benefits found on the mechanical properties, there were also high material recovery rates in terms of carbon fibre (100%) and PLA matrix (73%). According to open literature, the state of continuous fibre 3D printing is still very limited by the use of single nozzle printers. In this case, fibres and thermoplastic filaments pass through the same nozzle thus limiting control over fibre, poor overall surface finish and increased voids. Therefore, several researchers present different solutions, including nozzle modifications or the use of a dual extrusion printer, to increase process control. Omuro et al. (2017), for example, developed a new equipment in which a compaction roller was placed near the printer head to consolidate continuous fibre reinforced thermoplastics just after printing. In this study, a PLA filament and continuous carbon fibres were separately supplied to the 3D printer, and the fibres were impregnated with the filament in the heated nozzle. Therefore, the compaction roller reduces internal voids, promotes smooth surfaces, and prevents peeling of printed composite from the heated table. The samples were printed with the fibres aligned in the loading direction. Tensile and bending tests were performed to evaluate the mechanical properties of the 3D printed unidirectional composites, compacted and non – compacted, as well as neat PLA. The tensile strength and modulus were around 39.5 MPa and 4.7 GPa for neat PLA, while these values are, respectively, 9.9 and 10,4 times higher for non-compacted composites. However, when these properties are compared between non-compacted and compacted composites, the last ones reveal improvements of 36.4% and 30.1%, respectively. In terms of bending properties, the flexural strength and modulus were around 64.9 MPa and 2.5 GPa, respectively, for PLA samples. While flexural strength increased about 142% and 242% for non-compacted and compacted composites, respectively, the flexural modulus did not show this tendency. In this case, this property increased 10.1 times for non-compacted composites, while for compacted composites it was only 9.6 times higher than for PLA samples. This means that the flexural modulus decreased when composites were compacted. The failure modes observed for both composites showed that the tensile load was effectively transferred by all filaments, but significant delaminations were developed in bending mode for both composites, indicating low interfacial adhesion between filament/matrix. In order to improve the interfacial adhesion, compacted composites were, after printing, submitted to hot-pressing at 180 °C for 1 hour. Significant improvements were achieved with this postprocessing procedure, promoting flexural strength around 841 MPa and flexural modulus of 58,4 GPa against the 157 MPa and 25.1GPa obtained for non-compacted composites. This reveals that although a compaction load was applied, it was not enough to promote good fibre/matrix adhesion.