Case Studies in Construction Materials (Dec 2024)
Experimental and numerical study on the effect of friction between crack faces on fracture parameters of hot mix asphalt concrete
Abstract
This article investigates the effect of friction between crack faces in ASCB (asymmetric semi-circular bend) specimen on fracture parameters of hot mix asphalt concrete. To achieve this, fracture tests were conducted on ASCB specimens containing cracks with different angles (0, 10, 20, 30, and 40°). In addition to the fracture tests, friction tests were also performed by designing a simple fixture. Subsequently, finite element analyses were carried out, and friction coefficients between crack faces were calculated based on the results of fracture and friction tests. Furthermore, two fracture criteria, ASED (average minimum strain energy density) and MTS (maximum tangential stress), were examined under two different scenarios: assuming zero-friction and non-zero friction between crack faces to predict fracture load and fracture strength. The fracture tests showed that with an increase in the crack angle, the fracture load initially decreased up to the crack angle of 20°, and then increased. Furthermore, the friction tests revealed that an increase in the normal load led to an increase in the friction coefficient. As per finite element results, the crack face contact occurred at a crack angle of 13.1°. Hence, crack faces in the ASCB specimens with the crack angles of 20, 30, and 40° had friction, which significantly affected the fracture parameters. The results also indicated that the average error values for the MTS and ASED criteria compared to the experimental results were 22.8 % and 18.8 %, respectively, when the friction` between crack faces is ignored, and 12.4 % and 4.0 % for the scenario that the friction effect is taken into account. The ASED criterion demonstrated higher accuracy than MTS criterion. Additionally, the average error value of fracture strength predicted by the ASED criterion with assuming non-zero friction was 4 %, indicating good predictive accuracy for fracture strength.
