The effects of curing temperature on fracture energy and cohesive parameters for the adhesive Araldite 2015

ÖZER H., Erbayrak E.

JOURNAL OF ADHESION SCIENCE AND TECHNOLOGY, vol.32, no.12, pp.1287-1312, 2018 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 32 Issue: 12
  • Publication Date: 2018
  • Doi Number: 10.1080/01694243.2017.1410080
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.1287-1312
  • Keywords: Cohesive zone modeling, curing temperature, glass transition temperature, fracture properties, Araldite 2015, GLASS-TRANSITION TEMPERATURE, BONDED JOINTS, MODEL, BEHAVIOR, SIMULATION, TOUGHNESS, CONCRETE, REPAIRS, CURVE, LAW
  • Yıldız Technical University Affiliated: Yes


This paper attempts to investigate the effects of curing temperature on the fracture energy, the glass transition temperature (T-g) and cohesive parameters for the adhesive Araldite 2015. Relationship between curing temperature and the glass transition temperature was taken into account. Tensile tests were performed on the dogbone-shaped bulk specimens to evaluate the effect of curing temperature on the mechanical properties of the adhesive. DCB test results were used to obtain the cohesive laws of the adhesive Araldite 2015. The exponential and PPR cohesive zone models were used to obtain some of the fracture properties. Inverse analyses were also performed, if the experimental softening curves are incompatible with the numerical ones. It was seen that softening behavior of the adhesive can be easily controlled by the shape parameters available in the PPR cohesive zone model. It is seen from the DCB test results that curing the adhesive about the temperature at which the T-g is obtained caused the adhesive to have more ductility, higher load-carrying capacity and higher fracture energy than curing it below or above the temperature at which T-g is attained. Here, the T-g is the T-g of the fully cured network. Experimental and numerical R curves were obtained to account for deviations between experiments and simulations. A good agreement between the numerical and experimental load-displacement curves was achieved showing the adequacy of the cohesive model used.