PSI - Issue 41

Kushal Mishra et al. / Procedia Structural Integrity 41 (2022) 248–253

249

2

Mishra & Singh/ Structural Integrity Procedia 00 (2022) 000 – 000

case of uniaxial compression tests allows it to be used to understand the constitutive behavior of epoxy and examine the effect of loading rate and temperature (Arruda et a l., 1995; Jordan et al., 2012; More lle et al., 2017; Mu lliken and Boyce, 2006; Naik et a l., 2011; Nguyen et al., 2016; Srivastava, 2010). With an increase in the compression loading rate, the yield stress and the flow stress shift to higher values both under quasi-static and dynamic loading (Chen et al., 2002; Goglio et al., 2008; Jordan et al., 2008). However, a standard tensile test on epoxy does not typically show a four-stage stress-strain response. Here the sample fa ils at relatively lower strains with the stress -strain curve showing only two distinct regions; an initia l linear stress -strain region followed by a non-linear region. Also, the stress at fracture can vary drastically depending on the specimen thickness. There is a decrease in the t ri-axial state of stress for a thin specimen and a s maller probability of containing a ‘would be critica l’ defect (Fied ler et al., 2001). When the tensile strain rate was varied, Chen et al (Chen et al., 2002) did not observe a clear dependence of failure stress on strain rate, but a significant increase in the flow stress and failure stress have been reported in other studies (Goglio et a l., 2008; Litte ll et al., 2008). However, in a ll the cases, the specimens fractured at lower strains as the strain rate was increased. The tensile behavior of amorphous epoxy is very sensitive to the free volume distribution, pore distribution and local heterogeneity which depend on the specimen preparation condition and aging history (Fiedler et al., 2001; Goglio et al., 2008; Van Breemen et al., 2011). Deformat ion in highly cross -linked epoxy can happen either by a shear yield phenomenon where the entire materia l partic ipates in the deformation process or by tensile stress -dominated crazing phenomenon where the deformation is loca lized to sharp notches or weak regions of the matrix (Chevalier et al., 2019; Littell et a l., 2008; Morelle , 2015). The former is ma inly observed when the epoxy is tested under comp ression, whereas crazes form under tension stresses. In the present work, we examine the effect of strain rate on the uniaxia l loading response of epoxy under compression versus tension. 2. Materials and methods

2.1. Specimen preparation

100 parts by weight of low viscosity DGEBA (Bisphenol A diglycidyl ether) epoxy was mixed wit h 38 parts by weight of a polyamine hardener. The mixtu re was well stirred, followed by vacuum degassing for 30 minutes to remove any trapped air. The mixture was then poured into molds made of silicone rubber and cured in two steps; first for 24 hours at room temperature, followed by post-curing at 100°C for 150 minutes.

2.2. Tensile and compression test

The specimens used for the tensile test had a gauge cross -section of 6 mm × 2 mm and a gauge length of 25 mm. The compression test specimens had a diameter and length of 7 mm each. The tensile tests were carried out using a 5 kN capacity pneumatica lly actuated Instron UTM (Universal Testing Machine). The compression tests were carried out in a 250 kN servo-hydraulic UTM. The mach ine was tuned at low loads for achiev ing high accuracy. The compression plates of the UTM were properly lubricated to minimize friction. Both the tests were carried out at 4 different strain rates (0.0003 s -1 , 0.005 s -1 , 0.02 s -1 , 0.2 s -1 ) spanning four order of magnitudes (Table 1) The fracture surface analysis of the tensile tested specimen was done using a tungsten filament Hitachi S3400N scanning electron microscope (SEM). The fracture surface was coated with carbon to make it conductive. 3. Results and Discussion Fig. 1(a) shows the uniaxial tensile stress -strain curve for specimens tested at strain rates as listed in Tab le 1. It is observed that for strain rates up to 0.005 s -1 , the tensile ductility decreases, whereas the ultimate tensile strength (UTS) rema ins very similar. For a slightly higher strain rate of 0.02 s -1 , both the UTS and the tensile ductility 2.3. Fracture surface analysis