FIB cross-sections revealed that th

FIB cross-sections revealed that the whole surface was covered with a thin—few 100 nm thick oxide layer, which was formed due to the interaction between oxygen in the air and molten copper. Beneath the oxide layers, there was a polycrystalline microstructure. It consisted of copper grains separated by high-angle grain boundaries. Twin boundaries were also observed within some grains. It was not possible to determine the thickness of the molten surface layer from the inspection of the microstructure. The waves could be observed on the copper plate, which was slightly exaggerated by the oxide layer. However, the height of waves was less than 1 μm, as estimated from the FIB cross-sections.
The copper plates were initially in the as-rolled condition. They contained elongated grains with a high dislocation density. In FIB-micrographs, this kind of microstructure possesses a mottled contrast. At the site of final waves, only a thin surface layer with a thickness of 2–3 μm possessed crystal grains with a sharp contrast. These grains can be either formed during solidification or by the recrystallization. The thickness of the grains with a sharp contrast was much greater closer to the crater; i.e. in the region with a higher fluence (Fig. 4). Since it is not possible to distinguish crystal grains formed by solidification or by recrystallization, the thickness of the molten area cannot be determined.
The periodicity was measured at several places. The periodicity at a distance of 50 μm from the crater centre was 2.9 ± 0.5 μm (Fig. 4b, d); and this value was typical for the most of the wavelike area. Only at the largest distances from the crater, the periodicity was larger 5.2 ± 1.9 μm (Fig. 4a, c).