PSI - Issue 28

P. Santos et al. / Procedia Structural Integrity 28 (2020) 1816–1826 P. Santos/ Structural Integrity Procedia 00 (2019) 000–000

1823

8

phenomenon is based on physical phenomena and/or chemical phenomena (Sreekala et al. 2001; Varghese et al. 1994), where both expected in this study. However, when the fibres are incorporated into the matrix, they hinder the molecular flow in the matrix and, consequently, delay the relaxation process (Obaid et al. 2017). In this case, the interface properties play a relevant role, because relaxation occurs due to the breaking of bonds and their propagation. Another evidence from the open literature is that fibres are also subject to stress relaxation and this phenomenon is dependent on the fibre type (Oskouei and Taleie 2010; Reis et al. 2020). Carbon fibres exhibit a negligible viscoelastic behaviour, and when they are combined with polymeric matrices promote considerably small amount of creep or stress relaxation in the reinforcement direction (Miranda Guedes et al. 2000; Oskouei and Taleie 2010; Reis et al. 2020). On the other hand, and in same circumstances, glass and aramid composites show significant decreases when they are subjected to constant load (Pisani 1998; Oskouei and Taleie 2010). Regarding the aramid fibres, the relaxation is independent of the stress level applied (Gerritse 1993), and when they are subjected to axial compression or bending, aramid fibres may exhibit non-linear plastic deformation, as consequence of structural defects developed in the chain (Oskouei and Taleie 2010). Considering hybrid laminates, the stress relaxation is greater with the highest content of glass fibres. In relation to full carbon fibres, for example, the stress value after 180 min is 1.2% and 2.6% lower for 6C+2G and 4C+4G laminates, respectively, evidencing the most sensitive behaviour of the glass fibres to the stress relaxation.

1,00

0.969

8C

 /  0

0.968

0,96

0.967 0.944

0.943 0.941 4C+4G

0.957

0.956

0,92

6C+2G

0.955

0.879

0,88

0.873 0.871 8G

0,84

0

40 80 120 160 200

Time [min]

Fig. 6. Stress relaxation curves for all configurations studied.

In terms of creep behaviour, Fig. 7 shows typical curves obtained from the experimental tests, where the displacement is the value obtained at any instant of the test (D) divided by its initial value (D 0 ). For this study, a fixed bending stress was applied, with values corresponding to of 422 MPa, 327 MPa, 319 MPa and 317 MPa for 8C, 6C+2G, 4C+4G and 8G laminates, respectively. These values correspond to 50% of the maximum bending stress of each laminate. It is possible to observe that all curves present an instantaneous displacement, followed by the primary and secondary creep regimes that characterize the typical creep curves. In the present study, the third regime is expected to occur only for higher stress values or longer times. It is also noticed that, similar to the stress relaxation behaviour, full carbon composites are less sensitive to creep behaviour than full glass fibre composites. For example, the displacement after 180 min of the glass fibre laminates is about 3.8% higher than the value observed for full carbon laminates. This is explained by the creep phenomenon that occurs in polymers, even at room temperature and for stresses below their ultimate strength due to the molecular motion in backbone polymer arrangement (Park and Balatinecz 1998; Houshyar et al. 2005; Bouafif et al. 2013; Wang et al. 2015), and by the contribution of the fibres (Reis et al. 2020). In fact, both elastic deformation and viscous flow are retarded by the presence of fibres and, consequently, the creep process is delayed. However, as previously reported, carbon fibres exhibit negligible