PSI - Issue 13

Jiaming Wang et al. / Procedia Structural Integrity 13 (2018) 560–565 J. Wang et al. / Structural Integrity Procedia 00 (2018) 000–000

564

5

Fig. 3. Model without (a) and with (b) ITZ under tension; Model without (c) and with (d) ITZ under compression.

Fig. 4. Mortar plasticity e ff ect in (a) Tension; (b) Compression.

While the stress-strain curves reveal moderate e ff ect of zCE on the composite behaviour, there is significant dif ference between crack patterns in models with and without cohesive ITZ. In models without, crack initiates and propagates near the loading surface (Figure 3 (a)). This does not match experimental observations of type 1 and 2 crack patterns, with correspondingly one or two dominant cracks. This suggests that omitting ITZs is not a viable strategy. In contrast, both type 1 and 2 crack patterns are observed in our models with cohesive ITZ. Figure 3 (b) shows fracture pattern of one dominant crack. In compression, experiments show cracks forming ’X’ or ’V’ shapes prior to failure. The model without ITZ yields a strange fracture pattern (Figure 3 (c)), while the crack pattern in Fig ure 3 (d) is very similar to the experimental ’V’ shape. Overall, the models with ITZ provide realistic crack evolution, with micro-cracks initiating at ITZ and macro-cracks forming after both ITZ and mortar experience damage. Further, plasticity is introduced to the mortar behaviour, with hardening commencing at 90% and 95% of peak strength. Figure 4 shows that this improves the agreement between simulated and measured responses in both tension and compression, when used in combination with zCE. It is noted, that the e ff ect of mortar plasticity is negligible in tension, as damage is initiated early along ITZ, but it should be considered in compression.