PSI - Issue 28

Konstantinos Tserpes et al. / Procedia Structural Integrity 28 (2020) 1644–1649 Tserpes, Bazios, Pantelakis, Michailidis/ Structural Integrity Procedia 00 (2020) 000–000

1647

4

velocity of 16 nm/s. By computing the indentation depth and force at a specified reference point throughout the simulation, the load versus depth curve is derived. In the analysis, the base of the specimen was fixed to ensure accurate portrayal of the indenter penetration. Lateral displacements in the indenter were fixed to allow motion only

in the vertical direction. 4. Experimental results

The load ( P ) –depth ( h ) curves obtained for the two alloys are presented in Fig. 4. As can be seen, the cW-Cu alloy exhibits a higher penetration depth of about 400 nm at an indentation load of 15 mN compared to that of the W-Cu alloy which exhibits a penetration depth of about 250 nm at the applied penetration load of 15 mN. In Table 1, the Young’s modulus, and yield strength of the two alloys, derived using the models of Dao et al (2001) and Ogasawara et al (2006), are listed. The experimental results reveal a higher indentation resistance and higher mechanical properties for the W-Cu which is explained by the Hall-Petch effect. It is noted that the W-Cu shows a superior yield strength of the order of 3.0 GPa.

Fig. 4. A typical FE mesh of the solid specimen and the indenter.

Table 1. Mechanical properties of the cW-Cu and W-Cu alloys

Model of Dao et al

Model of Ogasawara et al

Alloy

E (GPa)

σ y (MPa)

E (GPa)

σ y (MPa)

cW-Cu W-Cu

200 240

1420 3150

195 225

1480 3420

5. Validation of the model In Fig. 5, compared are the numerical load-depth curves with the experimental ones for the two alloys. As in the literature, there was reported a significant effect of the indenter’s sharpness (Lichinchi, Lenardi, Haupt and Vitali, 1998), in the model, a sharp indenter’s tip (no radius) and a rounded tip (radius = 100 nm) were considered. The comparison shows a good correlation between the model and the tests. The small observed deviation may be attributed to several parameters related to the experimental process, to the accuracy of the input material properties in the model, to the accuracy of the contact algorithm, etc. As shown, the use of the rounded tip leads to a predicted load-depth curve which is closer to the tests; this is an indication of presence of roundness at the tip of Berkovich indenter.