PSI - Issue 77

Jiongyi Yan et al. / Procedia Structural Integrity 77 (2026) 135–142 J. Yan/ Structural Integrity Procedia 00 (2026) 000–000

137

3

2.2. Fibre orientation measurement The printed specimens were mounted in epoxy resin and cured in the air. The sectioning was prepared by an auto polisher. The samples were ground with <1200 grit sandpaper (300 rpm and 20 N) before being polished with 3 µm polycrystalline diamond suspension (MetaDi™Supreme) (150 rpm and 20 N) and 0.05 µm alumina suspension (BUEHLER MasterPrep™) (150 rpm and 15 N). For stereological measurement of elliptical fibre cross-sections, out of- plane Euler angle θ and in -plane deviation angle ф were measured. This allows the fibre orientation tensor to be identified in Eq. (1) (Randy S. et al., 1992), where 11 is the main-axis alignment along the printing direction; 22 is the lateral alignment transverse to the printing direction; 33 is the out-of-plane Z-direction alignment. The measurement averaged all fibres on a sectioned plane statistically. = � 11 12 13 21 22 23 31 32 33 � = � 2 2 2 ф ф ф 2 ф ф 2 2 ф ф ф ф 2 � (1) 2.3. Mechanical testing Tensile test was conducted on Instron ® 5944 system (Norwood, USA) for the printed samples, with load cell of 1 kN. The initial separation of two grips was 32 mm. In the static tensile test, corner specimens of three turn angles (30°, 90°, and 150°) with three replicas for each angle were tested until break. Considering the samples have different geometries at the gauge section (longer structural paths for the greater angles), the tensile strain rate was varied for each angle: 5, 5, and 40 mm.min -1 for 30°, 90°, and 150° respectively. For cyclic tensile test, the displacement-control mode was selected because it captures the features of deformation of corners where they could be straightened. In this mode, loading initiated from a baseline, increased to a set target, and then unloaded back to baseline for each cycle, which repeated for cyclic tensile. The 30° corner was chosen, with strain rate was set at 10 mm.min -1 to ensure effective and efficient deformation. The test had a boundary condition of a minimum displacement of 0 mm and a maximum displacement at 1.3 mm, as the effective path length when the 30° corner is nearly straightened. The specimen was tested with 50 cycles. 2.4. Fractography For the static tensile test, the fracture surfaces of representative specimens for three turn angles (30°, 90°, and 150°) were examined using scanning electron microscopy (SEM JEOL 7800). The fractured samples were coated by a sputter coater (Q150R S Quorum), with the non-oxidising metal coating Au:Pd = 80%:20%, under the protection of Ar. For SEM, all samples were tilted to 50 μm with accelerating voltage of 5 kV of the electron beam. The magnification varied from × 110 to × 1100, and the work distance varied from 9.5 to 11.6 mm, depending on the area of interest. 3. Results and Discussion 3.1. Fibre orientation of angular corners The fibre orientation was measured from micrographs of section specimens (Fig. 1). At corners, the 11 main axis alignment corresponds to the longitudinal direction before the path turns, and the 22 lateral alignment represents the transverse direction to the original longitudinal direction. They showed opposite trends with turning angles. The 11 was higher (~ 0.6) for flatter corners (30°, 150°) but lower by 32% for more orthogonal corners (60°, 90°, 120°), while the 22 was lower for flatter corners but low for more orthogonal corners, which increased by 143%. This is expected since the orthogonal corners hauled the fibres to the transverse direction away from the original direction, resulting in high transverse alignment. The strong nozzle shear has played a critical role in rapid response of fibres to be oriented in the new flow direction. It clearly shows the strong dependency of fibre orientation on flow directions. The 33 out-of-plane Z-alignment, though low for all angles (ranging from 0.16 to 0.23), continuously