PSI - Issue 13

Felipe C. da Silva et al. / Procedia Structural Integrity 13 (2018) 658–663 / Structural Integrity Procedia 00 (2018) 000 – 000

661

4

Table 2 shows the residual stresses determined in the constant N2 flow film deposited on Al, using the GIXRD experiments. The results suggest that compressive residual stresses are induced in the film, independent from the applied bias. It is apparent that the lowest bias condition (-40 V) resulted in the largest residual stresses, particularly in the (111) planes. Since the films are textured, with this plane parallel to the film surface, this contribution would correspond to the in plane residual stresses. This is inconsistent with the scratching test results, which shows better cohesive properties for the films obtained with higher bias. It is to be reminded that the scratching test results in the films deposited on Al are questionable, due to the higher compliance of the film/substrate system. Results are under way in the other conditions, and these may allow a more complete picture on the effect of bias over the production of residual stresses in these films Table 2. Results of the residual stresses (in [GPa]) determined in the GIXRD experiments for the constant N2 flow film deposited on Al. Condition Plane (111) Plane (200) Plane (220) Plane (311) Average Bias -40 V -6.0 -6.5 -4.3 -2.1 -4.7 +- 1.7 Bias -75 V -3.6 -7.6 -5.3 -3.6 -4.0 +- 1,7 Bias -100 V -4.6 -4.6 -4.8 -2.8 -4.2 +- 0.8 Table 3 shows the results of the scratching test for the films deposited on Brass and for the films deposited on Al. The results are considerably variable, but a general trend can be identified (particularly in the constant N2 flow films), in which an increased bias lead to more resistant films. In terms of the adhesive failure, the variability is so large that any trend is questionable. Apparently films deposited on Brass are more adherent on the substrate compared with the films deposited on Al. It must be reminded, however, that the Yield modulus results show that the films deposited on Al are more compliant than the films deposited on Brass, therefore the low load observed in the Al films may be simply a consequence of the larger deflection in the surface of the film, as the load is applied during the scratching tests. This would result in bending efforts applied to the film, resulting in a more complex stress state. So the results should be considered with care . Table 3. – Results of the scratching test in films deposited on both substrates. Condition Lc1 (cracks) [N] Lc2 (delamination) [N] (Brass) Constant N 2 flow/ -40 V bias 1.5 4.4

(Brass) Constant N 2 flow/ -75 V bias (Brass) Constant N 2 flow/ -100 V bias

5.5 4.0 3.9 2.4 4.0 1.7 2.2 2.1 1.4 2.0 2.0

8.1 1.5

(Brass) Graded/ -40 V bias (Brass) Graded/ -75 V bias (Brass) Graded/ -100 V bias

5.0 7.0 7.1 2.5 2.4 2.15

(Aluminum) Constant N 2 flow/ -40 V bias (Aluminum) Constant N 2 flow/ -75 V bias (Aluminum) Constant N 2 flow/ -100 V bias

2.7 2.4 2.1

(Aluminum) Graded/ -40 V bias (Aluminum) Graded/ -75 V bias (Aluminum) Graded/ -100 V bias

Figure 2 shows the final configuration of the films deposited on Brass (a-b) and Aluminum (c-d), after the tensile test. We may also assume also, as a first approximation, that the fracture strength of the film is independent of the deposition method. Within the class, however, the comparison is possible and as in the case of the films deposited under constant N2 flow, the application of a bias decreases the interface adhesion. The situation is different in the case of the films deposited on Al, as shown in Figure 1b. In particular in the case of the films deposited under constant N2 flow, it is evident that a higher bias leads to improved adhesion. In the cases of the graded films the situation is not so evident, but apparently the bias also has a positive effect over adhesion. Table 4 show the average distance between cracks, for all deposition conditions. The results conform the qualitative analysis of the images shown in Figure 1. Results comparing different substrates and different deposition conditions cannot be made, since the film thickness is not constant in these cases, qualitatively, however, it is possible to observe that increasing the bias has a positive effect over adhesion in the case of the films deposited on brass, while the contrary