Effect of aggregate damage on self-healing characteristics of asphalt concrete: An image processing-based method


Atakan M., Yıldız K.

CONSTRUCTION AND BUILDING MATERIALS, vol.425, pp.1-14, 2024 (SCI-Expanded)

  • Publication Type: Article / Article
  • Volume: 425
  • Publication Date: 2024
  • Doi Number: 10.1016/j.conbuildmat.2024.135924
  • Journal Name: CONSTRUCTION AND BUILDING MATERIALS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Aerospace Database, CAB Abstracts, Communication Abstracts, Compendex, INSPEC, Metadex, Veterinary Science Database, Civil Engineering Abstracts
  • Page Numbers: pp.1-14
  • Gazi University Affiliated: Yes

Abstract

In the fracture-based healing test, asphalt specimens generally break into two halves and the aggregates in the crack path also break. In this study, the effect of aggregate damage on the self-healing performance of asphalt concrete was studied. To achieve this, semi-circle asphalt specimens were produced. Then, they were broken using the Semi-circle bending test (SCB) at different temperatures (i.e., − 20, − 10, 0, 10, and 20 ◦C) to create various aggregate damage situations. Next, the fracture surfaces of specimens were photographed and the percentages of damage types (i.e., adhesive, cohesive, and broken aggregates) were obtained by using image processing techniques. After the specimens were healed at a constant 55 ◦C, the SCB test was applied again to measure the strength of the specimens and to calculate healing performance. Finally, fracture surfaces were photographed again. According to findings, as breaking temperatures increase, broken aggregates and adhesive damage decrease. Cohesive damage, on the other hand, rose with the high breaking temperature. Healing performance was better at low breaking temperatures. A positive correlation was found between aggregate damage and healing levels, but it was explained as a spurious relationship. The real correlation was negative which was obtained from a second dataset.