2.2 | Impact testing
Impact tests were performed on samples A-C using an electro-pneumatic projectile launcher referred to as the Single Point Impact Fatigue Tester (SPIFT) [25]. The SPIFT was developed to repeatedly impact a target sample in a single point to be able to initiate and progress damage at high impact speeds up to 170 m/s. The purpose of the device is to mimic the impact speeds of rain droplets on the leading edge of wind turbine blades to understand the damage mechanisms of the coating materials at the relevant strain rates. Spherical nitrile rubber projectiles with a diameter of 6 mm and mass of approximately 0.1 g was used. The impacted surface was monitored continuously with a camera to observe the damage. In the current work, samples A and B were tested in the SPIFT until failure was visible on the surface, as seen in Figure 1(a,b). For sample C, the test was stopped before coating failure was visible from the surface, so only a slight difference in the shading of the coating is visible at the point of impact.
2.3 | X-ray CT reference scanning
XCT scanning was performed of the sample GZ3-01 for validation of the OCT images. The scan was performed on a Zeiss Xradia 520 Versa with a polychromatic X-ray beam with energies up to 60keV, generated by a tungsten target. A total of 4501 projections with an exposure time of 10 s were acquired on a 2k × 2k CCD detector during a full 360° rotation of the sample. Density maps were reconstructed by a Feldkamp algorithm for cone beam reconstruction [26] to 3D pixel volumes with a pixel size of 9.0 μm. The standard Zeiss Xradia procedure was followed for selecting the appropriate energy and source filter to tune the X-ray energy spectrum for the sample. The reconstruction of X-ray CT data provides 3D density maps that allow detection of small density differences in the bulk material, such as the presence of crack networks or air bubbles. Figure 2(b) shows slices from different directions through the reconstructed volume and a segmentation where air bubbles are colored blue and the cracks are red. The segmentation was made using a global intensity threshold value and visualized using Avizo 3D software for materials science.