PSI - Issue 69

Carlo Alberto Biffi et al. / Procedia Structural Integrity 69 (2025) 121 – 126

123

Table 1: List of the principal process parameters investigated in this work Process parameters Values Power pumping [%] 32.5-35.0-37-5-40-42.5-45.0-47.5-50.0 Laser power [W] 35-45-55-65-75-85-95-105 Scanning speed [mm/s] 45-50-55

2.3 Functional characterization: The transformation temperatures of the NiTi wires were detected through DSC (mod. Q100 from TA Instruments); the DSC scans were carried out on the entire wire length with a heating/cooling rate of 10°C/min in the temperature interval from -150°C to 150°C. From the DSC scan, the characteristic temperatures Ms and Mf (start and end of martensite formation on cooling, respectively) and As and Af (start and end of austenite reformation on heating, respectively), were measured. Similarly, the heat exchanged were measured as the area under the transformation peak upon cooling and heating, respectively, and named as h cool and h heat . 3. Results The laser beam scan along the wire length, kept straight, was carried at varying the laser power and the scanning speed. The feasibility of the laser shape setting is depicted in Figure 2, where the three regions can be highlighted: (i) a first region below a threshold, in which the martensitic transformation (MT) cannot be promoted, due to insufficient energy irradiated to the NiTi wire (see the blue shaded area in Figure 2); (ii) an intermediate region in which the MT can be induced by the laser scanning, depending on the energy enough to change the initial microstructure; (see the green shaded area in Figure 2); and a last region in which the wires were melted by the laser beam, due to an excess of irradiated energy (see the pink shaded area in Figure 2). Figure 3 shows the evolution of the DSC scans performed at the lowest scan speed value investigated (v=45mm/s) for different values of laser power, able to promote the MT without the wire melting. This value of scanning speed was selected as representative to be deeply discussed in this work. Similar trends were observed at higher scanning speeds, though not shown here for brevity. With the slowest process speed selected and for the lowest power value (P=35%), the peaks that identify phase transformations in the material are of low intensity and barely evident. During the cooling phase, the first structure to appear is the rhombohedral one (R) which is very distant from the martensitic one (M) during the cooling phase. The notable width of the M peak indicates that the material is continuously transforming over a wide temperature range. The heating phase is instead characterized by two small, very close peaks, probably due to the coexistence of the austenitic (A) and R phases within the same temperature range.

Figure 2: Feasibility map for correlating the investigated process parameters and the features of the martensitic transformation